Cells of the 16-cell mouse embryo endocytose horseradish peroxidase (HRP) which becomes localized in most cases to a juxtanuclear position. Cells that have ingested HRP in intact embryos, and cells dissociated from embryos prior to culture in HRP, showed similar patterns of cytoplasmic distribution of the ingested enzyme. Cells in the embryo in situ were incubated in HRP, and then labelled with fluorescent antibody either before (to label the outside surface of the embryo) or after (to reveal populations of outer polar and inner apolar cells) their disaggregation into single cells. The population of polar outside cells from the morula includes more cells with a highly restricted localization of HRP-containing vesicles than does the population of inside cells, and this restricted localization underlies the exposed surface or pole of the cell. A 2/16 couplet formed by division in vitro of a 1/8 cell is comparable to the pairs of cells dissociated from 16-cell embryos; most couplets from either source consisted of a larger cell that showed polarized surface binding of fluorescent ligand (fluorescent pole) and a smaller cell with a uniform distribution of bound ligand. The incidence of restricted patterns of HRP staining was highest among populations of both larger and polar cells. When 1 /8 cells labelled with HRP are observed during division to 2/16, the previously clustered vesicles of ingested HRP become more dispersed throughout the cytoplasm and, although the two cells of some couplets can stain differently very soon after their formation, the patterns of distribution of HRP take about 1 h after division to stabilize. These observations are consistent with cells of the 16-cell embryo inheriting different features of cytoplasmic organization.

Cellular position within the compacted mouse morula has long been thought to be critical for the differentiation of the inner cell mass (ICM) and trophecto-dermal lineages of the blastocyst (Tarkowski & Wroblewska, 1967; Herbert & Graham, 1974; Ducibella & Anderson, 1975; Kelly, Mulnard & Graham, 1978). The mechanism by which position is translated to fate has been unclear. Recently, it has been proposed that the fate of cells within the 16-cell morula does not depend solely on their position within the embryo (Johnson, Pratt & Handyside, 1981). Each cell of the 8-cell embryo becomes polarized along the radial axis of the embryo, and a differential inheritance at division to the 16-cell stage generates two cell types (Johnson & Ziomek, 1981): one present on the inside, the other on the outside of the morula (Handyside, 1981; Reeve & Ziomek, 1981). The polarity of individual cells of the 8-cell embryo has been shown by asymmetric distributions of surface-bound fluorescent ligand (Handyside, 1980; Ziomek & Johnson, 1980) and microvilli (Reeve & Ziomek, 1981), the high incidence of mitochondria and microtubules orientated parallel to the cell surface near areas of cell apposition (Ducibella, Ukena, Karnovsky & Anderson, 1977), alkaline phosphatase activity confined to apposed cell surfaces (Mulnard & Huygens, 1978), the basal location of the nucleus of each cell (Reeve, unpublished) and the restricted cytoplasmic localization of endocytosed vesicles containing horseradish peroxidase (HRP) (Reeve, 1981).

Sixteen-cell embryos can be labelled with a variety of fluorescent ligands either before or after their disaggregation into single cells. Immunofluorescent labelling of embryos before disaggregation marks those cell surfaces which are part of the external surface of the embryo, whereas, after disaggregation, some isolated cells show a polarity of ligand binding associated with a heterogeneity of microvillous distribution. Thus, in the 16-cell embryo, the inside cells have an even distribution of sparse microvilli and bind fluorescent ligand uniformly, while each outside cell has a region of dense microvilli (microvillous pole) facing outwards and associated with a high level of ligand-binding sites (Handyside, 1981 ; Reeve & Ziomek, 1981 ; Johnson & Ziomek, 1981). Most polarized cells of the 8-cell embryo divide, both in the embryo in situ (Handyside, 1981; Johnson & Ziomek, 1981) and in isolation (Johnson & Ziomek, 1981), to give one polar cell and one apolar cell.

This paper extends the description of the patterns of cytoplasmic localization of endocytosed HRP in the 8-cell embryo (Reeve, 1981) to the 16-cell stage at which it is demonstrated that inside and outside cell populations show different patterns of distribution of the ingested HRP. The tightly localized pattern of HRP distribution is more frequent in polar outside cells than in apolar inside cells, and occurs on the same axis as the surface polarity, a correlation described previously for the 8-cell embryo (Reeve, 1981). The patterns of HRP-vesicle localization are affected by both cell size and the presence of a microvillous pole, and the difference in staining patterns between polar and apolar cells is shown by the cells in the 16-cell embryo in situ, and by 2/16 couplets generated by division in vitro of 1/8 cells. Thus, cells from the 16-cell stage appear intrinsically different from the time of their formation at division, and these differences can occur in the absence of the inside and outside environmental cues previously considered critical to differentiation in the preimplantation mouse embryo (Ducibella & Anderson, 1975).

Embryo collection

Female HC-CFLP mice (4–5 weeks; Hacking & Churchill) were superovulated with intraperitoneal injections of 5 i.u. pregnant mare’s serum (PMS: Folligon, Intervet) followed after 44–48 h by 5 i.u. of human chorionic gonadotrophin (hCG: Chorulon, Intervet). Females were paired with HC–CFLP males, and the presence of vaginal plugs taken as an indication of mating. Embryos were flushed from the oviducts at 66–70 h post-hCG with phosphate-buffered medium 1 supplemented with 0·4% (w/v) bovine serum albumin (PB1 + BSA) (Whittingham & Wales, 1969), and were cultured at 37 °C in medium 16 with 0-4% (w/v) BSA (M16 + BSA) (Whittingham, 1971) under paraffin oil in 5% CO2 in air.

Zonae pellucidae were removed by a 15–30 s incubation in prewarmed (37 °C) acid Tyrode’s solution (pH 2·5)+ 0·4% (w/v) polyvinylpyrrolidone (Nicolson, Yanagimachi & Yanagimachi, 1975).

Terminology

Throughout this paper, the individual cells of 8-cell embryos are called 1/8 cells, and those of 16-cell embryos, 1/16 cells. Thus, an isolated 1/8 cell divides in vitro to form a 2/16 couplet.

Disaggregation and decompaction

Disaggregation into single cells was accomplished by pipetting embryos with a flame-polished micropipette after incubation for 10–30 min in calcium-free medium 16+0·6% (w/v) BSA, pre-equilibrated for at least 30 min at 37 °C in 5% CO2 in air. After disaggregation, cells were restored immediately to the culture medium.

2/16 couplets that had compacted were decompacted by incubation for 5 min in calcium-free medium 16 + 0·6% (w/v) BSA.

Horseradish peroxidase (HRP)

Embryos and isolated cells were incubated in 2 mg/ml HRP (Sigma Type II) in M16 + BSA for 3–10 h in 5 % CO2 in air. Embryos were then disaggregated for examination of the patterns of HRP staining of cells cultured in HRP in the embryo in situ. Intact embryos and isolated cells were rinsed in PB1 + BSA, and fixed in 4% (w/v) paraformaldehyde (Anderson & Co. Ltd) in phosphate-buffered saline (PBS) at 4 °C for 1 h, before further washing and storage in PB1 +BSA at 4 °C. Cells were stained histochemically for HRP by the aminoethylcarbazole (AEC; Sigma) method (Pearse, 1968; Reeve, 1981), and then mounted in wells of a tissue-typing slide (Baird & Tatlock) in drops of PB1 + BSA under oil.

Couplets

A population of dissociated 1 /8 blastomeres was cultured for at least 3 h in 2 mg/ml HRP in M16 + BSA, and 2/16 couplets were then harvested at 20 min or 1 h intervals. The couplets remained in the HRP-containing medium for various defined times until later immunofluorescent labelling and fixation. The division in vitro of a 1/8 cell to a 2/16 couplet has been shown to be comparable to the equivalent division in the intact embryo, as assessed by comparison of surface features (Johnson & Ziomek, 1981).

Indirect immunofluorescence

HRP-treated embryos and cells were incubated for 5 min in 25 μl drops of rabbit antiserum (RAMS) to mouse species antigens (Gardner & Johnson, 1975) diluted 1 in 15 in PB1 + BSA + 0· 02% (w/v) sodium azide, followed by thorough washing in PB1 + BSA 4-azide, and a similar incubation in fluorescein-conjugated goat anti-rabbit IgG (FITC-GAR IgG; Miles Labs.) diluted 1 in 15 in PB1 + BSA + azide. Embryos and cells were washed again, and embryos disaggregated into single cells. The couplets and single cells were fixed in 4 % (w/v) paraformaldehyde in PBS at 4 °C for 1 h, before being stained for HRP activity.

Fluorescence microscopy

A Zeiss Universal microscope, fitted with incident source HBO 50 and Zeiss filter set 487709, plus additional excitation filter LP 425, was used to examine cells for patterns of HRP staining and F1TC labelling. Bright-field and fluorescent micrographs were taken with Kodak Tri-X 35 mm film.

(1) The distribution of ingested HRP in cells of intact embryos

Intact 16-cell embryos were cultured in HRP, and then labelled by indirect immunofluorescence either before or after disaggregation to single cells or couplets of cells. Fluorescent labelling before disaggregation permitted comparison of the HRP staining patterns of inside (unlabelled by fluorescence) and outside (labelled by fluorescence) cells, whilst labelling of dissociated cells allowed examination of the HRP localization patterns of apolar (inside) and polar (outside) cells. A mean of 6·0 cells/embryo occupied an inside position as revealed by lack of fluorescence on cells from embryos labelled with fluorescent ligand before their disaggregation to single cells; whereas, for those embryos disaggregated completely before fluorescent labelling, 6·7 cells/embryo occupied an inside position as shown by a non-polarized binding of ligand (Table 1 ; lines 1 and 2). These figures are similar to those reported previously (Johnson & Ziomek, 1981; Handyside, 1981).

Table 1.

Cytoplasmic staining patterns in cells from intact 16-cell embryo (70–74 h post-hCG) and 2/16 couplets stained with HRP

Cytoplasmic staining patterns in cells from intact 16-cell embryo (70–74 h post-hCG) and 2/16 couplets stained with HRP
Cytoplasmic staining patterns in cells from intact 16-cell embryo (70–74 h post-hCG) and 2/16 couplets stained with HRP

Several patterns of HRP distribution in individual cells were identified (Fig. 1). The patterns included (Fig. 1a) a ‘tight localization’ with HRP-vesicles concentrated in one large well-defined mass; (Fig. 1b) ‘loose localization’ in which diffuse aggregates of staining occupied a small cytoplasmic region; (Fig. 1 c) ‘horseshoe’ in which approximately half of the nucleus was surrounded by staining; (Fig. 1d) ‘2-poles’ shown by two well-defined small masses of HRP on opposite sides of the nucleus; (Fig. 1e) ‘nuclear cap’ in which a small area of intense staining was juxtaposed to the nucleus, and (Fig. lf)a ‘nuclear ring’ revealed by staining concentrated around the nucleus. ‘Aggregates’ or uniform patterns (Fig. 1g) appeared as diffused distributions of vesicles throughout the cytoplasm. The tight and loose localization patterns represent large restricted areas of staining, and were more common in outside polar cells than in inside apolar cells (Table 1; lines 1 and 2). Cells showing these two patterns are hereafter defined as showing ‘restricted’ localization. The other patterns of stain distribution were grouped as ‘non-restricted’. In outside cells, the restricted cytoplasmic localization of HRP occurred on the same arxis as, and underlay, the fluorescent surface labelling on the outward-facing surface of the cell (Fig. 2).

Fig. 1.

The scale bar = 30 μm. The cells of the 16-cell embryo showed several staining patterns, (a) Tight localization. (b) Loose localization, (c) Horseshoe. (d)2-poles. Two aggregates of HRP occur on opposite sides of the nucleus, (e) Nuclear cap. (f) Nuclear ring, (g) Aggregates or uniform.

Fig. 1.

The scale bar = 30 μm. The cells of the 16-cell embryo showed several staining patterns, (a) Tight localization. (b) Loose localization, (c) Horseshoe. (d)2-poles. Two aggregates of HRP occur on opposite sides of the nucleus, (e) Nuclear cap. (f) Nuclear ring, (g) Aggregates or uniform.

Fig. 2.

The scale bar = 30 μm. All cells of a 16-cell embryo incubated in (a) HRP and (b) labelled with fluorescent ligand before embryo disaggregation. The isolated cells were then fixed and stained histochemically. The 11 outside cells tend to show greater localization of HRP than the 5 inside cells, and this localization underlies the pole of surface fluorescent labelling.

Fig. 2.

The scale bar = 30 μm. All cells of a 16-cell embryo incubated in (a) HRP and (b) labelled with fluorescent ligand before embryo disaggregation. The isolated cells were then fixed and stained histochemically. The 11 outside cells tend to show greater localization of HRP than the 5 inside cells, and this localization underlies the pole of surface fluorescent labelling.

(2) The restricted localization of HRP does not depend on the restricted exposure of the cell surface

The incidences of the different patterns of staining did not differ significantly whether cells were incubated in HRP in embryos in situ before disaggregation of the embryos to single cells, or dissociated from embryos before their culture in HRP (Table 2). Thus, the difference in HRP localization patterns is not obviously related to differential accessibility of the cell surfaces to HRP, and this conclusion is supported by observations on 2/16 cells generated by division of isolated 1 /8 cells in vitro.

Table 2.

Incidence of staining patterns in cells from 16-cell mouse embryos (70–74 h post-hCG) incubated in HRP before or after disaggregation to single cells

Incidence of staining patterns in cells from 16-cell mouse embryos (70–74 h post-hCG) incubated in HRP before or after disaggregation to single cells
Incidence of staining patterns in cells from 16-cell mouse embryos (70–74 h post-hCG) incubated in HRP before or after disaggregation to single cells

(3) The cytoplasmic staining patterns of 1/16 cells from intact embryos cultured in HRP are similar to those of 1/16 cells generated by division of isolated 1/8 cells in vitro

The 1/16 cells generated by division in vitro of isolated polarized 1/8 cells and the 1/16 cells recovered after disaggregation of 16-cell embryos display similar features of surface organization (Ziomek & Johnson, 1981). The present study confirms this observation, the incidence of cells with ligand-binding poles being 58 % after disaggregation from 16-cell embryos (Table 1 ; line 2), and 59 % after generation by division in vitro of isolated 1/8 cells (Table 1 ; line 3). Likewise, the patterns of cytoplasmic staining were in general similar for a population of 1/16 cells labelled with HRP before disaggregation of whole embryos and for a population of 1/16 cells formed from the division in vitro of isolated 1/8 cells (Table 1; compare lines 1 and 2 with line 3). The only significant difference was that the area of HRP localization appeared slightly more restricted in the cells of in vitro generated couplets, in which the incidence of the tight localization pattern was greater than in cells from intact morulae (Table 1 ; Fig. 1). The difference between the staining patterns shown by the two cells of a 2/16 couplet may arise artificially during scoring since the HRP localization patterns are directly compared in the cells of a couplet, whereas cells obtained from embryos pulsed with HRP prior to disaggregation are scored individually, and direct comparisons will tend to exaggerate small differences.

In all couplets, the enzyme in polar cells was localized in the cytoplasm underlying the pole, but in apolar cells the enzyme localization could not be related consistently to the point of contact with, or the site of HRP in, the polar cell (Fig. 3).

Fig. 3.

The scale bar = 30 μm. Polarized 1/8 cells were cultured, and allowed to divide, in HRP. In established 2/16 couplets (> 1 h after division), five different combinations of relative size and fluorescent pole presence among the two cells were identified, (a) Smaller apolar cell (nuclear cap) and larger polar cell (tight localization), (b) Smaller polar cell (tight localization) and larger apolar cell (nuclear cap), (c) The cells are of different sizes; both have poles and show loose localization of HRP. (d) The cells are of similar sizes, with the right possessing a fluorescent pole. Both show tight localization of HRP. (e) The cells are of similar sizes, and both have fluorescent poles. Both show loose localization of HRP. (See Johnson & Ziomek, 1981).

Fig. 3.

The scale bar = 30 μm. Polarized 1/8 cells were cultured, and allowed to divide, in HRP. In established 2/16 couplets (> 1 h after division), five different combinations of relative size and fluorescent pole presence among the two cells were identified, (a) Smaller apolar cell (nuclear cap) and larger polar cell (tight localization), (b) Smaller polar cell (tight localization) and larger apolar cell (nuclear cap), (c) The cells are of different sizes; both have poles and show loose localization of HRP. (d) The cells are of similar sizes, with the right possessing a fluorescent pole. Both show tight localization of HRP. (e) The cells are of similar sizes, and both have fluorescent poles. Both show loose localization of HRP. (See Johnson & Ziomek, 1981).

(4) The two cells of a couplet stain differently

After division in vitro of a polarized 1/8 cell, the two cells of a 2/16 couplet can usually be distinguished by the criteria of size, pattern of fluorescent ligand binding and microvillous distribution (Johnson & Ziomek, 1981). Commonly, the larger cell has a restricted area, or pole, of intense fluorescent ligand binding that coincides with a defined area of dense microvilli. The smaller cell binds ligand uniformly over its surface of sparse microvilli. A difference between the cells of a 2/16 in vitro generated couplet can also be demonstrated by staining for the distribution of previously ingested vesicles of HRP, and can be observed whether division has occurred in the presence of HRP, or after removal from HRP and restoration to control medium.

In this study, the in vitro generated 2/16 couplets were classified according to whether they contained one or two cells with a pole of ligand binding, and whether or not the cells of a couplet showed an obvious size difference. A size difference between the cells was detectable in 304 (84 %) out of 361 couplets. The relative cell sizes within a couplet affect the patterns of cytoplasmic staining; the larger cells show a significantly greater incidence of the restricted patterns of staining than do the smaller cells.

In couplets containing different sized cells, over 60% of the larger cells showed restricted localization of HRP, whereas a similar pattern was shown usually by less than 30% of the smaller cells (Figs. 3ac; Tables 3 and 4). In most couplets, only the larger cell was polar (Fig. 3a) and in those cases 78 % of polar larger cells, compared with only 33 % of smaller apolar cells, had a restricted localization of HRP staining (Table 3). Out of 146 antibody-labelled couplets, only four couplets examined between 1 and 2 h after division contained a smaller polar cell and a larger apolar cell (Fig. 3b). The smaller and larger cell types in these couplets stained similarly, each group containing 3 restricted and 1 non-restricted localization patterns. Seventeen couplets that were examined between 1 and 3 h after their formation, and in which the cells were of different sizes, consisted of two polar cells (Fig. 3c). Among this group, 14 of the larger polar cells had restricted and 3 had non-restricted localizations, compared with the incidences among the 17 smaller polar cells of 12 restricted and 5 non-restricted localizations.

Table 3.

Cytoplasmic staining patterns in 1/16 cells when the two cells of a couplet derived from in vitro division of a 1/8 cell could be distinguished by size, with only the larger cell showing polarized surface binding of fluorescent ligand

Cytoplasmic staining patterns in 1/16 cells when the two cells of a couplet derived from in vitro division of a 1/8 cell could be distinguished by size, with only the larger cell showing polarized surface binding of fluorescent ligand
Cytoplasmic staining patterns in 1/16 cells when the two cells of a couplet derived from in vitro division of a 1/8 cell could be distinguished by size, with only the larger cell showing polarized surface binding of fluorescent ligand
Table 4.

Cytoplasmic staining patterns in cells when couplets were not labelled with fluorescent ligand, but when the two cells of a couplet could be distinguished by size*

Cytoplasmic staining patterns in cells when couplets were not labelled with fluorescent ligand, but when the two cells of a couplet could be distinguished by size*
Cytoplasmic staining patterns in cells when couplets were not labelled with fluorescent ligand, but when the two cells of a couplet could be distinguished by size*

When the cells of a couplet were of similar sizes, two classes of couplet were identified, according to whether one (Fig. 3d) or both (Fig. 3e) cells showed polarized ligand binding. For 16 couplets in which the two cells were of similar sizes and only one showed polar binding, the polar population contained 13 restricted and 3 non-restricted localizations; the 16 apolar cells included 10 restricted and 6 non-restricted localizations. Nine couplets contained cells of similar sizes, both of which showed polar ligand binding. Among the 18 cells, there were 15 restricted and 3 non-restricted localizations.

The results suggest that the pattern of cytoplasmic localization of ingested HRP is affected by cell size and the presence of a microvillous pole, but the difference in staining between the two cells of a couplet does not depend on a difference in their endocytotic activities. The patterns of staining were examined after polarized 1 /8 cells had been cultured in HRP for at least 3 h, and restored to control medium just before cell division. Of the newly formed 2/16 couplets that were examined, 30 were considered to contain both a polar and an apolar cell, using criteria of differences in cell size and fluorescent intensity of bound ligand. Both cells of 2 couplets showed restricted staining, and both those of 6 couplets had non-restricted patterns of staining. However, in most cases, in 22 couplets, the polar cell showed restricted staining, and the apolar cell, non-restricted staining.

The microvillous pole of a polarized cell is re-established during the first hour after division, when fluorescent ligand is bound uniformly (Johnson & Ziomek, 1981). Throughout this period, the orientation of the restricted localization of HRP cannot be related to the axis of surface polarity, and therefore 2/16 couplets examined within this time of their formation were not labelled with fluorescent ligand. Examination of couplets at later times showed that the patterns of HRP staining did not differ whether (Table 3) or not (Table 4) cells had been labelled by indirect immunofluorescence. The staining patterns of HRP appear to have stabilized within about 1 h of division (Tables 3, 4). Prior to 1 h, the distribution of HRP is more dispersed; the larger cells of couplets have a lower incidence of the tight localization pattern, and the smaller cells show a high incidence of the aggregate staining distribution (Table 4). Cells undergoing division showed a low level of dispersed HRP (Fig. 4).

Fig. 4.

A 1/8 cell in division has a dispersed distribution of HRP.

Fig. 4.

A 1/8 cell in division has a dispersed distribution of HRP.

There has been little examination of cytoplasmic reorganization during preimplantation development of the mouse embryo. Although the rat embryo has been examined extensively by transmission electron microscopy (Schlafke & Enders, 1967; Dvorak, 1978), studies on the mouse embryo have either been very preliminary (Calarco & Brown, 1969; Van Blerkom & Motta, 1979) or have tended to focus on areas of cell contact (Ducibella et al. 1977). The immunofluorescent labelling of a range of early preimplantation stages has indicated distinct cytoplasmic localizations for several filamentous systems, but their organization did not appear to vary with the different developmental stages up to the morula (Lehtonen & Badley, 1980). Some information on cellular organization has arisen from cell lineage investigations involving the injection of marker oil drops into the cytoplasm of cleavage-stage embryos. The eventual location of the drop in the blastocyst could often be predicted from the site of intracellular injection (Wilson, Bolton & Cuttler, 1972; Graham & Deussen, 1978), and any droplets that were formed by fragmentation of the injected drop remained clustered (Wilson et al. 1972). Although the oil had been introduced passively into the cell, and may possibly have acted as an inert substance, these observations nonetheless suggest that some areas at least of cytoplasm are relatively stable. In contrast, Graham & Deussen (1978) interpreted movement of the oil droplet with respect to the cleavage furrow as suggesting some redistribution of the cytoplasm.

The results reported here using HRP as a cytoplasmic marker do not offer support for a rigid cytoplasmic organization. Thus, although cells of 8-cell embryos endocytose HRP, which becomes markedly restricted to a juxta-nuclear position underlying the external surface of the cell (Reeve, 1981), the HRP-vesicle distribution becomes diffuse at division to 2/16 cells and then becomes progressively more organized. Stable patterns of HRP distribution are evident within one hour of division (Tables 3, 4), but are more variable than those described previously for the 8-cell embryo (Reeve, 1981). In the 16-cell embryo the population of outside cells shows a greater incidence of restricted staining patterns of HRP than does the population of inside cells, and this restricted localization also underlies the fluorescent pole. Populations of inside cells show a higher incidence of a diffuse distribution of HRP and any localizations that are observed are not systematically related to the point of contact with an outer cell. What might the basis for these differences be?

Firstly, although couplets degrade HRP very rapidly, and show negligible staining within 30 min of their restoration to control medium, there is no evidence to relate the variation in patterns of staining to cellular differences in degradative rates. Secondly, the differences in patterns of staining of the two populations might be due to variations in cellular apposition or access to HRP (Hastings & Enders, 1974). This possibility can be excluded since the distinct patterns are also present both in populations of 1/16 cells isolated from intact embryos prior to culture in HRP (Table 2), and in cells of 2/16 couplets generated by the division in vitro of polarized 1/8 cells (Tables 3, 4). Thirdly, differences in patterns of staining between polar and apolar cells could reflect differences in the endocytotic activities of the two cell types. For example, it is possible that a polar cell, which has a heterogeneous surface containing a dense microvillous region and an area of few microvilli, does not ingest HRP uniformly over its entire surface. Thus, preferential uptake might occur at the region of the microvillous pole, and this could cause an aggregation of HRP-containing vesicles in the underlying cytoplasm. In contrast, apolar cells have a uniform distribution of sparse microvilli, and endocytosis of HRP might occur uniformly over the surface, resulting in a low level of ingested HRP over a widespread cytoplasmic distribution. Thus, variation in cell surface features could cause differences that are observed in the cytoplasmic organization of HRP-treated cells. However, it seems unlikely that the difference in staining patterns between polar and apolar cells depends on differences in endocytotic activities of the two cell types, since of 31 polarized 1/8 cells cultured in HRP for several hours, and restored to control medium just before cell division, 22 divided to give a population of couplets in which the polar cells showed a greater incidence of restricted staining than did the population of apolar cells. This result makes it seem likely therefore that, at division of 1 /8 cells, the larger polar cells generated must inherit either more endocytosed vesicles or some features responsible for the tighter aggregation of vesicles under the pole in an area of the cytoplasm known in other cell types to be occupied by the Golgi complex (Steinman, Silver, & Cohn, 1974; Piasek & Thyberg, 1979).

In conclusion, the evidence from this study, taken with that from other studies referred to above, suggests that although there may be considerable cytoplasmic disturbance during cleavage divisions, some elements of stability within the cytoplasm may occur. The data are not therefore inconsistent with a model for blastocyst development that ascribes an important role for differential inheritance of various cellular features at the 8- to 16-cell division (Johnson et al. 1981), although they do not in themselves constitute adequate proof of such a model.

Finally, although the population of polar outer cells shows a greater incidence of restricted localization of HRP than does the population of inner apolar cells, the differences between the two populations are not absolute and thus, unlike the differences in surface phenotype (Handyside, 1981; Johnson & Ziomek, 1981), cannot be diagnostic for position. It is possible, but improbable, that two discrete populations exist in situ, but that the sharp differences in HRP localization become blurred during cell analysis as a result of misscoring or technical manipulations. For example, some inaccuracies will arise from misassignment of cells to one of the seven categories of staining owing to problems of interpreting 3-dimensional distributions in a 2-dimensional analysis. It is unlikely that the patterns represent an artifact induced during cell preparation since observations on intact stained embryos, although not easy to make, indicate that the range of patterns among inside and outside cells in situ is similar to that observed in isolated cells. Moreover, among isolated cells the range of patterns does not vary with the method of cell isolation and the time of labelling. There is heterogeneity within each population, and it seems probable that there is a genuine overlap of HRP localization patterns between each population. It is not clear whether this heterogeneity is of developmental significance.

I wish to thank Drs C. A. Ziomek, H. P. M. Pratt and M. H. Johnson for constructive criticism, and Gin Flach, Mike Parr and John Bashford for technical help. The work was supported by grants from the Ford Foundation, the Cancer Research Campaign and the Medical Research Council to Dr. M. H. Johnson, and from the Medical Research Council and the Cambridge Philosophical Society to the author.

Calarco
,
P. G.
&
Brown
,
E. H.
(
1969
).
An ultrastructural and cytological study of preimplantation development of the mouse
.
J. exp. Zool
.
171
,
253
284
.
Ducibella
,
T.
&
Anderson
,
E.
(
1975
).
Cell shape and membrane changes in the eight-cell mouse embryo: prerequisites for morphogenesis of the blastocyst
.
Devi Biol
.
47
,
45
58
.
Ducibella
,
T.
,
Ukena
,
T.
,
Karnovsky
,
M.
&
Anderson
,
E.
(
1977
).
Changes in cell surface and cortical cytoplasmic organization during early embryogenesis in the preimplantation mouse embryo
.
J. Cell Biol
.
74
,
153
167
.
Dvorak
,
M.
(
1978
).
The differentiation of rat ova during cleavage
.
Advances in Anatomy, Embryology and Cell Biology
55
, Fase. 2.
Springer-Verlag
.
Gardner
,
R. L.
&
Johnson
,
M. H.
(
1975
).
Investigation of cellular interaction and deployment in the early mammalian embryo using interspecific chimaeras between the rat and mouse
.
In Cell Patterning, C1BA Fdn. Symp
.
29
,
183
200
.
Amsterdam
:
ASP
.
Graham
,
C. F.
&
Deussen
,
Z. A.
(
1978
).
Features of cell lineages in preimplantation mouse development
.
J. Embryol. exp. Morph
.
48
,
53
72
.
Handyside
,
A. H.
(
1980
).
Distribution of antibody- and lectin-binding sites on dissociated blastomeres from mouse morulae: evidence for polarization at compaction
.
J. Embryol. exp. Morph
.
60
,
99
116
.
Handyside
,
A. H.
(
1981
).
Immunofluorescence techniques for determining the numbers of inner and outer blastomeres in mouse morulae
.
J. Reprod. Immunol
.
2
,
339
350
.
Hastings
,
R. A.
&
Enders
,
A. C.
(
1974
).
Uptake of exogenous protein by the preimplantation rabbit
.
Anat. Rec
.
179
,
311
330
.
Herbert
,
M. C.
&
Graham
,
C. F.
(
1974
).
Cell determination and biochemical differentiation of the early mammalian embryo
.
Current Topics in Devi. Biol
.
8
,
151
178
.
Johnson
,
M. H.
,
Pratt
,
H. P. M.
&
Handyside
,
A. H.
(
1981
).
The generation and recognition of positional information in the preimplantation mouse embryo
.
In Cellular and Molecular Aspects of Implantation
(ed.
S. R.
Glasser
&
D. W.
Bullock
), pp.
55
74
.
Plenum Press
.
Johnson
,
M. H.
&
Ziomek
,
C. A.
(
1981
).
The foundation of two distinct cell lineages within the mouse morula
.
Cell
24
,
71
80
.
Kelly
,
S. J.
,
Mulnard
,
J. G.
&
Graham
,
C. F.
(
1978
).
Cell division and cell allocation in early mouse development
.
J. Embryol. exp. Morph
.
48
,
37
51
.
Lehtonen
,
E.
&
Badley
,
R. A.
(
1980
).
Localization of cytoskeletal proteins in preimplantation mouse embryos
.
J. Embryol. exp. Morph
.
55
,
211
225
.
Mulnard
,
J.
&
Huygens
,
R.
(
1978
).
Ultrastructural localization of non-specific alkaline phosphatase during cleavage and blastocyst formation in the mouse
.
J. Embryol. exp. Morph
.
44
,
121
131
.
Nicolson
,
G. L.
,
Yanagimachi
,
R.
&
Yanagimachi
,
H.
(
1975
).
Ultrastructural localization of lectin-binding sites on the zonae pellucidae and plasma membranes of mammalian eggs
.
J. Cell Biol
.
66
,
263
274
.
Pearse
,
A. G.
(
1968
).
Histochemistry, Theoretical and Applied
, 3rd ed.
EdinburghandLondon
:
Churchill Livingstone
.
Piasek
,
A.
&
Thyberg
,
J.
(
1979
).
Effects of colchicine on endocytosis and cellular inactivation of horseradish peroxidase in cultured chondrocytes
.
J. Cell Biol
.
81
,
426
437
.
Reeve
,
W. J. D.
(
1981
).
Cytoplasmic polarity develops at compaction in rat and mouse embryos
.
J. Embryol. exp. Morph
.
62
,
351
367
.
Reeve
,
W. J. D.
&
Ziomek
,
C. A.
(
1981
).
Distribution of microvilli on dissociated blastomeres from mouse embryos: evidence for surface polarization at compaction
.
J. Embryol. exp. Morph
.
62
,
339
350
.
Schlafke
,
S.
&
Enders
,
A. C.
(
1967
).
Cytological changes during cleavage and blastocyst formation in the rat
.
J. Anat
.
102
,
13
32
.
Steinman
,
R. M.
,
Silver
,
J. M.
&
Cohn
,
Z. A.
(
1974
).
Pinocytosis in fibroblasts. Quantitative studies in vitro
.
J. Cell Biol
.
63
,
949
969
.
Tarkowski
,
A. K.
&
Wróblewska
,
J.
(
1967
).
Development of blastomeres of mouse eggs isolated at the 4- and 8-cell stage
.
J. Embryol. exp. Morph
.
18
,
155
180
.
Van Blerkom
,
J.
&
Motta
,
P.
(
1979
).
The Cellular Basis of Mammalian Reproduction
.
Munich and Baltimore
:
Urban & Schwarzenberg
.
Whittingham
,
D. G.
(
1971
).
Culture of mouse ova
.
J. Reprod. Fert. (Suppl
.)
14
,
7
21
.
Whittingham
,
D. G.
&
Wales
,
R. G.
(
1969
).
Storage of two-cell mouse embryos, in vitro
.
Austr. J. biol. Sci
.
22
,
1065
1068
.
Wilson
,
I. B.
,
Bolton
,
E.
&
Cuttler
,
R. H.
(
1972
).
Preimplantation differentiation in the mouse egg as revealed by microinjection of vital markers
.
J. Embryol. exp. Morph
.
27
,
467
479
.
Ziomek
,
C. A.
&
Johnson
,
M. H.
(
1980
).
Cell surface interaction induces polarization of mouse 8-cell blastomeres at compaction
.
Cell
21
,
935
942
.
Ziomek
,
C. A.
&
Johnson
,
M. H.
(
1981
).
Properties of polar and apolar cells from the 16-cell mouse morula
.
Wilhelm Roux Arch devl Biol, (in press)
.