Trophectoderm growth and bilateral symmetry of the blastocyst in the mouse

R.L. Gardner,1 and T.J. Davies

Mammalian Development Laboratory, University of Oxford, Department of Zoology, South Parks Road, Oxford OX1 3PS, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The present study was undertaken to ascertain whether the polarized flow of cells from polar to mural trophectoderm is related to the axis of bilateral symmetry of the blastocyst in the mouse, and whether trophectoderm cells can initiate new cycles once they have left the polar region. METHODS AND RESULTS: Two different approaches were used to investigate the relationship of polar to mural flow of trophectoderm cells to the bilateral axis. One was to mark peripheral polar trophectoderm cells at one or both ends of the bilateral axis in early blastocysts and examine the distribution of their clonal descendants after further growth in culture. The other was to mark the two ends of the bilateral axis with small oil drops in the zona pellucida in blastocysts whose polar trophectoderm was then labelled globally with fluorescent latex microspheres before culture. In both cases, marking of additional blastocysts orthogonal to the bilateral axis was also done. The results show that the direction of polar to mural flow of cells is not random, and that the most distal mural trophectoderm cell could yield up to eight descendants during 45 h of culture. CONCLUSION: The findings are consistent with the polar to mural flow of trophectoderm cells being aligned with the bilateral axis. Moreover, trophectoderm cells can embark on new cycles even when remote from the inner cell mass.

Key words: bilateral symmetry/mouse blastocyst/mural trophectoderm/polarized growth/polar trophectoderm


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The mouse conceptus is already bilaterally symmetrical about its embryonic–abembryonic (Em–Ab) axis by the early blastocyst stage and, according to the location of the surviving 2nd polar body, its bilateral axis normally coincides with the animal–vegetal (AV) axis of the zygote (Gardner, 1997Go). However, while the bilateral axis is clearly polarized in late blastocysts (Smith, 1980Go; Gardner, 1990Go), in early blastocysts it is not. Furthermore, because blastocysts temporarily lose their bilateral form during expansion, it is uncertain whether the orientation of the bilateral axis is conserved throughout their growth.

Among the questions raised by the early acquisition of bilateral symmetry is its significance for subsequent morphogenesis of the conceptus and, consequently, of the fetus itself (Gardner et al., 1992Go). Of interest in this context are findings from an earlier clonal analysis of growth of the trophectoderm which suggested that the flow of surplus polar cells into the mural region (Copp, 1978aGo; Cruz and Pedersen, 1985Go) was polarized rather than radially symmetrical (Gardner, 1996Go). This was confirmed by more recent experiments in which the entire polar trophectoderm was labelled selectively through endocytosis of fluorescent microspheres. Polar to mural movement of cells was found thus to be restricted circumferentially during the subsequent growth of blastocysts, typically yielding a single coherent patch of label that extended well into the mural region (Gardner, 2000Go). Moreover, in the few cases where there were two patches rather than one, these were diametrically opposite each other. This suggested that the direction of flow of polar trophectoderm cells might be related to the blastocyst's axis of bilateral symmetry. If the outflow of polar cells were aligned with the bilateral axis, it would argue that this axis was conserved during blastocyst expansion. It might also explain how the bilateral axis acquires polarity. To account for the latter it has been proposed that trophectoderm cells become less deformable once they have resided in the mural region for some time, possibly through deposition of extracellular matrix components on their blastocoelic surface (Biggers et al., 2000Go; Summers et al., 2000Go). Hence, the hydrostatic pressure of the fluid in the blastocoele (Watson, 1992Go) should cause relatively greater stretching and attenuation of cells that have entered the mural trophectoderm most recently. The consequent deformation of a coherent patch of cells in the proximal part of the mural region could account for the observed tilting of one side of the inner cell mass (ICM)/polar trophectoderm complex with respect to the blastocyst's Em–Ab axis (Gardner, 1998Go). Tilting of this complex is the most obvious manifestation of polarity of the bilateral axis, and Smith has defined the point where it is furthest from the abembryonic pole as the anterior end of the bilateral axis, and the diametrically opposite point as its posterior end (Smith, 1980Go).

A further question is the extent to which growth of the mural trophectoderm depends on proliferation of cells residing within it as opposed to recruitment from the mitotically more active polar region. This is of interest in view of evidence that fibroblast growth factor (FGF) signalling is required to sustain the cycling of all cells beyond 5th cleavage in the mouse (Chai et al., 1998Go), and that mural trophectoderm cells become postmitotic by the late blastocyst stage (Gardner and Davies, 1993Go). Although mitoses occur within the mural trophectoderm during blastocyst growth (Copp, 1978bGo), these might represent the completion of cycles to which cells were committed whilst they were still in the polar region.

The direction of movement of polar trophectoderm cells with respect to the bilateral axis of the early blastocyst was examined both clonally and by labelling the tissue globally. It was found by global labelling to be non-random and, within the limits of the resolution that was attainable, closer to parallel than orthogonal to the bilateral axis. The results of clonal analysis were consistent with this conclusion, and also revealed that new cycles could be initiated in mural cells that were remote from the polar region.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Mice and blastocysts
Mice of the closed-bred PO strain were kept under standard or altered lighting regimes in which the dark period was 1900–0500 and 1300–2300 h respectively. Females were selected for estrus by external inspection (Champlin et al., 1973Go), paired with males overnight, and checked for vaginal plugs the following morning. Early and expanding blastocysts were flushed from excised uterine horns with mouse tubal fluid (MTF)–HEPES medium (Gardner and Davies, 2000Go), either in the morning (altered lighting) or early afternoon (standard lighting) of the 4th day post coitum. Blastocysts were stored at room temperature and manipulated in MTF–HEPES. Nascent and very early blastocysts were cultured to the appropriate stage in the {alpha}-modification of Dulbecco's minimum essential medium (DMEM) supplemented with 10% (v/v) fetal calf serum, and all blastocysts were cultured in this medium following manipulation. Where culture was prolonged, the dishes were sometimes coated with 2% agar (Difco) in Dulbecco A phosphate-buffered saline (PBS; Oxoid, Basingstoke, UK) to prevent the blastocysts attaching and outgrowing.

Marking the bilateral axis
Early blastocysts were immobilized on a standard holding pipette (Gardner and Davies, 2000Go) with their Em–Ab axis vertical and embryonic pole uppermost. Small drops of soya oil were injected into the substance of the zona pellucida (ZP) via a very fine-tipped pipette pulled from a microelectrode capillary (GC100F-15; Harvard, Edenbridge, UK) and connected to a microinjector system (IM-6 Narishige, Japan) filled with heavy paraffin oil (Gardner, 2001Go). The drops were sited either at both ends of the bilateral axis, which was taken as their greater diameter (GD) or, as a control for the resolution of marking, at the both ends of the lesser diameter (LD).

Polar trophectoderm labelling
The entire polar trophectoderm was labelled selectively by injecting ~1 µm fluorescent latex microspheres (Fluoresbrite; YG Carboxylate, Polysciences, Warrington, PA, USA) that are readily endocytosed by trophectoderm cells locally under the ZP, as described elsewhere (Gardner, 2000Go). Individual trophectoderm cells were labelled with 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI; Molecular Probes Inc., Eugene, OR, USA), horseradish peroxidase (HRP; Boehringer-Ingelheim UJK, Bracknell, UK), 40 kDa tetramethylrhodamine dextran lysine (TMRDL) or fluorescein dextran lysine (FDL; Molecular Probes, Inc.) or, in some cases, a mixture of both HRP and TMRDL or FDL. DiI was prepared as a saturated stock in ethanol, with undissolved material being sedimented by centrifugation, before aliquots were diluted 1:5 with soya oil. Drops of the DiI in oil were injected against or into the cell to be labelled using the type of pipette and injector system described earlier for marking the ZP. Labelling of cells with HRP, either alone or together with TRMDL or FDL, was done ionophoretically (Gardner, 1996Go, 1997Go). All three labels were prepared at a final concentration of 6–8% (w/v) in 0.1 mol/l KCl. Staining for HRP activity was carried out essentially as described elsewhere (Beddington and Lawson, 1990Go). Where a fluorescent label was injected alone or together with HRP, the targeted cell was inspected briefly by fluorescence microscopy shortly thereafter to confirm labelling and to determine whether the label had spread to another cell. Such spreading is not uncommon in early mouse development because intercellular bridges between sister cells often remain patent for many hours following cytokinesis (Goodall and Johnson, 1984Go).

Preparation of labelled blastocysts for scoring following culture
Blastocysts whose polar trophectoderm had been labelled globally with fluorescent microspheres were fixed before being examined following culture (Gardner, 2000Go), as were those in which HRP was used as a cell label (Gardner, 1996Go). For examination, all blastocysts were placed in hanging drops in Puliv chambers (Gardner and Davies, 2000Go) so that they could be oriented optimally for immobilization on a holding pipette during photography. For assessing the relationship of polar trophectoderm cell movement to the early bilateral axis, blastocysts were immobilized with their embryonic pole uppermost and, depending on which had been marked, either their GD or LD parallel to the long axis of the holding pipette. Brightfield and epifluorescence images were then taken of each specimen, and the resulting prints from both the GD and LD series were coded by one author for scoring by the other.

Scoring
Analysis of clones was based on the expectation that cells lying at the site of outflow from the polar trophectoderm should be the first, and those diametrically opposite this site among the last, to leave this region (Gardner, 1996Go). The distribution of clones should therefore differ in two related respects, depending on whether or not their progenitors were aligned with the direction of outflow. Aligned progenitors should yield a higher proportion of clones that are retained within the polar trophectoderm (Figure 1Go). Moreover, clones from opposite pairs of aligned progenitors should more often lie at different levels along the Em–Ab axis than those from non-aligned progenitors (Figure 1Go). Clones were recorded as polar if none of their cells lay entirely within the mural region. Pairs of clones were recorded as being disparate in level along the Em–Ab axis if they differed by >=2 cell diameters in the position of their proximal or distal boundary.



View larger version (29K):
[in this window]
[in a new window]
 
Figure 1. Diagram of blastocyst in embryonic polar view to show minimum relative distances that cells at the periphery of the polar trophectoderm (grey) must travel before entering the mural trophectoderm (white) if polarized outflow corresponds with the greater diameter (GD). The distances differ markedly for cells located at opposite ends of the GD (1, 3) whilst being similar for those at opposite ends of the lesser diameter (LD) (2, 2). Hence, an obvious disparity in position along the Em–Ab axis of the blastocyst should be more common for pairs of clones resulting from labelling of cells aligned with greater diameter (GD) than with the LD. Also, individual clones should more often be confined to the polar region after GD than LD cell labelling.

 
The following method was used to determine the direction of the polar to mural flow of cells relative to the GD or LD. Tracing paper was first placed over the print of the brightfield image and the outline of the blastocyst plus the line of its GD or LD drawn on it. The trace was then aligned over the corresponding fluorescence print and the centre of the region of outflow of polar cells marked at the periphery. A straight line was drawn from this mark to the centre of the blastocyst and the angle between this line and the GD or LD measured with a protractor. Outflow was recorded as `on axis' if the angle was <45° and `off axis' if it was >45° (Figure 2Go). Finally, the pairs of prints were decoded so that the number of cases of `on' versus `off' axis flow could be established for both the series where the GD and the series where the LD were marked.



View larger version (36K):
[in this window]
[in a new window]
 
Figure 2. Diagrams illustrating oil-in-ZP marking of greater diameter (GD) versus lesser diameter (LD) (A, B), measurement of angle of departure of centre of polar to mural cell flow from GD and LD after global microsphere-labelling of the polar trophectoderm (C, D), and number of cases of `on axis' versus `off axis' flow for GD and LD series (E, F).

 

    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Validation of marking GD versus LD with oil drops in ZP
Single small drops of soya oil were injected into the core of the ZP of early expanding blastocysts, either equatorially or more proximally, and the immediately underlying trophectoderm cell was lineage-labelled thereafter. Blastocysts were then cultured for varying periods before the spatial relationship between the oil and labelled cell(s) was re-examined. The results, which are presented in Table IGo, show that the oil drop in the zona remained over the label in the trophectoderm in most blastocysts, regardless of the duration of culture. Therefore, oil drops in the zona were used to mark the GD or LD in the following two series of experiments in which the direction of polarized flow of cells from polar to mural trophectoderm was investigated.


View this table:
[in this window]
[in a new window]
 
Table I. Relationship post culture between oil drop in zona pellucida (ZP) and immediately subjacent trophectoderm cell
 
Labelling of peripheral polar trophectoderm cells aligned with the GD versus LD
Both ends of the GD or LD in obviously oval early blastocysts were marked by injecting small drops of soya oil into the ZP over the periphery of the polar trophectoderm. Following culture for not more than 4.5 h, the blastocysts had individual peripheral polar trophectoderm cells located at one or both ends of the GD or LD injected with lineage label. The blastocysts were then returned to culture for a further 16–18 h.

Five out of a total of 97 clones were discounted in the GD series because they included binucleate cells or debris. A single scorable clone was present in 16 blastocysts and pairs of clones in a further 38. Three clones were discounted in the LD series in which 13 blastocysts with single and 38 blastocysts with pairs of clones were scored. The results of analysis of the distribution of all the scorable clones is summarized in Table IIGo, from which it is evident that both polar retention and axial disparity were significantly more common among GD than LD clones. The mean size of clones was similar for the two series (see footnote to Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Distribution of clones derived from peripheral polar cell labelling in alignment with greater diameter (GD) versus lesser diameter (LD)a
 
Resolution afforded by marking single trophectoderm cells
This was assessed using 13 living blastocysts that were judged according to their degree of expansion (axial ratio of blastocoele to ICM of ~4/1) to be at the stage when clones were analysed in the foregoing experiments. The circumference of such blastocysts at the level of the polar–mural junction was found to be ~80% of that at the equator. Counts of the number of cells comprising the circumference of the blastocysts at the equator were made using differential interference contrast (DIC) microscopy to achieve the necessary optical sectioning. The mean number was 8.7 (± 0.21) and the range 8–10.

Global labelling of polar trophectoderm in blastocysts with the GD or LD marked with oil drops in ZP
Early oval blastocysts had a small oil drop injected into the ZP at each end of the GD or LD before their entire polar trophectoderm was labelled with fluorescent microspheres. The blastocysts were then cultured overnight, and recovered before they hatched so that the direction of polar to mural flow relative to the axes defined by opposing oil drops in the ZP could be determined (Figures 2 and 3GoGo). Flow was scored as `on axis' if its centre was within 45° of the axis defined by the oil drops, and as `off axis' if it was outside this angle. From the results summarized in Figure 2Go [on axis/off axis: GD 29/17 (63%), LD 12/33 (27%)] it is clear that the GD and LD series differed significantly in the frequency with which the flow was `on axis' versus `off axis' ({chi}2 = 12.15 for 1 d.f.; P < 0.001). However, while there were significantly more `off' than `on axis' cases in the LD series, in the GD series the `on axis' did not outnumber the `off axis' to the same extent (Figure 2Go). Nonetheless, for the combined data there was a significant bias in the direction of flow towards the GD rather than the LD ({chi}2 = 6.18 for 1 d.f ; P < 0. 02). Among 14 blastocysts in the GD series retaining an intact 2nd polar body at analysis, this persisting marker of the animal pole of the zygote (Gardner, 1997Go) lay towards the side of the polar to mural cell flow in eight cases and away from it in the remaining six.



View larger version (78K):
[in this window]
[in a new window]
 
Figure 3. Corresponding brightfield (A, C, E, G) and epifluorescence (B, D, F, H) images of blastocysts from the greater diameter (AD) and lesser diameter (EH) series showing `on axis' (A, B and E, F) versus `off axis' (C, D and G, H) polar to mural flow.

 
Clonal analysis of proliferation of abembryonic polar mural trophectoderm cells
A potential problem is overestimating clone size because of transfer of the lineage label to the pre-existing sister of an injected cell (Goodall and Johnson, 1984Go). The likelihood of such transfer was assessed by checking for spread of injected fluorochrome, both at the time of labelling and again after 25–165 min of culture. This was done both in trial labellings and in some of the definitive experiments. The relevant findings are presented in Table IIIGo. Spread of the label was scored as complete when the indirectly labelled cell was judged to be as brightly fluorescent as the directly labelled one, and partial where it was clearly less so. Partial spread was usually readily discernible at final scoring with some positive cells being obviously more weakly stained than others. Provided that indirectly labelled clones were stained sufficiently to enable their cell number to be counted precisely, they were included in the final analysis since they served as a useful check on whether microinjection affected cell proliferation adversely. The critical data in Table IIIGo concern the majority of labelling experiments where, despite the lack of any spread of label at the time of injection, 15 cases of seemingly complete spread were recorded following a limited period of culture. In five out of 15, it was clear from the shape of the labelled cells that they were products of a recent mitosis. This is the most likely explanation for the remaining 10, since, even where partial spread was evident at the time of labelling, it failed to go to completion in the majority of cases (Table IIIGo). Even if it is assumed that the label spread horizontally rather than vertically in these ten cases, the proportion of clones whose size would be overestimated thereby is only 7%.


View this table:
[in this window]
[in a new window]
 
Table III. Spread of rhodamine–dextran lysine or DiI between clonally related trophectoderm cells before versus after brief culture
 
Hence, cases where spread was complete immediately after labelling were excluded from the analysis: those showing partial spread were included only if positive cells could be assigned unambiguously to one or other of a sister pair of clones on the basis of differential staining intensity. Finally, where no spread of label was discernible at the time of labelling, all positive cells were scored as belonging to a single directly labelled clone if they were stained similarly and as belonging to a sister pair of clones if they showed two distinct categories of staining intensity.

The results of experiments in which blastocysts were cultured for 1 or 2 days following lineage-labelling of single abembryonic mural versus polar trophectoderm cells are summarized in Table IVGo. Mural cells yielded clones composed of up to five cells after ~1 day in culture, and up to eight cells after 2 days (Figure 4Go). Both the maximum and mean sizes of polar clones were greater than mural at both intervals. Because single cell clones occurred only after direct labelling (Table IVGo), this may have occasionally resulted in damage that was sufficient to prevent or delay further proliferation (e.g. Cruz and Pedersen, 1985Go). However, the great majority of directly labelled cells could not have been affected adversely since their mean clone size was not consistently smaller than that of indirectly labelled cells.


View this table:
[in this window]
[in a new window]
 
Table IV. Size of mural versus polar trophectoderm clones in blastocysts cultured for 1 versus 2 days after lineage-labelling
 


View larger version (38K):
[in this window]
[in a new window]
 
Figure 4. Camera lucida drawing of a clone of eight cells formed after culture for 45 h of an early blastocyst whose abembryonic polar trophectoderm cell had been labelled with horse-radish peroxidase (HRP). That this clone was initiated after, rather than before, labelling was evident from the lack of spread of co-injected rhodamine dextran lysine before, or after brief, culture, and the similar intensity of staining of all eight cells for HRP activity.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
How the direction of flow of cells from polar to mural trophectoderm relates to the axis of bilateral symmetry during blastocyst growth was investigated in two different ways, both of which depended on marking the GD or LD of blastocysts with small oil drops injected into the ZP. The validity of this method of marking was assessed by lineage-labelling the trophectoderm cell immediately underlying an oil drop in the zona in a series of early blastocysts. Even after overnight culture the oil drop remained over the labelled trophectoderm in most cases, arguing, in conformity with earlier findings, that blastocysts seldom rotate extensively in the zona (Gardner, 2001Go).

The relationship of polar to mural flow to the bilateral axis was first investigated by examining the distribution of clones resulting from labelling of peripheral polar cells that were either aligned with, or orthogonal to, this axis. This approach was based on the expectation that clonal descendants of a peripheral polar cell lying at the site of egress should be the first, and those of a cell lying diametrically opposite this site the last, to enter the mural region. Consequently, polar retention should be more frequent among aligned than non-aligned clones, as also should a disparity between members of pairs of clones in their location along the Em–Ab axis (Figure 1Go). Both expectations were fulfilled and position-dependent differences in growth could be discounted as a confounding factor. The extent to which `on' and `off' axis clones differ in location should first increase and then decrease according to the amount of polar growth that intervenes between labelling and scoring. There is no reason to suppose the difference should be greatest after the 16–18 h of culture employed in the present study. However, in view of the considerable asynchrony in cell cycles both within and between conceptuses (Barlow et al., 1972Go; Chisholm et al., 1985Go), a more optimal interval might prove hard to attain. The resolution achievable by lineage-labelling single cells obviously depends on the proportion of the blastocyst's circumference orthogonal to the Em–Ab axis that is occupied by the resulting clone. This is clearly limited by the proportion of the circumference of the blastocyst orthogonal to its Em–Ab axis occupied by clonal descendants of the labelled cell. According to the cell counts on living blastocysts optically sectioned by DIC, this would not be <10% even where clones were oriented strictly proximo-distally, and possibly up to 25% where they were not.

The second approach was to determine the direction of bulk movement into the mural region of surplus polar cells relative to pairs of oil drops in the ZP that marked either the GD (bilateral axis) or the LD at the early blastocyst stage. Within the limits of resolution noted above, the centre of the outflow of surplus polar cells was found to be `off axis' significantly more often for the LD than for the GD. However, because blastocysts had to be scored whilst they still retained the ZP, the bilateral axis was seldom discernibly polarized. This meant that it was possible to record only the size of the angles by which the centre of flow departed from the bilateral axis and not to which side of this axis they lay. Hence, the possibility that the mean direction of flow is at some angle of <45° to the bilateral axis rather than coincident with it cannot be discounted. Nonetheless, when considered in conjunction with those from the clonal analysis, these findings are consistent with the notion that the polar to mural flow of cells in the trophectoderm corresponds with the axis of bilateral symmetry of the blastocyst and hence with the AV axis of the zygote. However, because the flow could be either towards or away from a persisting 2nd polar body, it clearly cannot depend on information in the zygote for its polarity as opposed to orientation.

Finally, the mural trophectoderm cell furthest from the ICM was lineage-labelled in early blastocysts to determine how many times it could divide during their subsequent growth in vitro. Since macromolecular labels can pass from an injected cell to its sister, not all labelled cells are necessarily descendants of the injected one. Sufficient transfer of label for an uninjected cell to stain as strongly as its injected sister can only occur if there is extensive cytoplasmic continuity between them as, for example, before the intercellular bridge formed during cytokinesis becomes attenuated. In such circumstances, full spread of a co-injected fluorochrome should be discernible almost immediately after labelling. While partial spread of label may go undetected at the time of labelling, it can be identified at scoring as differential staining among positive cells. Thus, in one blastocyst, four cells were uniformly well stained and a further four uniformly less well stained for HRP activity. These were therefore classified as a pair of sister clones. In all other cases where more than two labelled cells were present they were similarly stained, and there was no reason to doubt that they were the clonal descendants of a single labelled cell that had divided after the start of culture. Hence, notwithstanding evidence that FGF signalling is needed for all cells of the mouse conceptus to complete 5th cleavage (Chai et al., 1998Go), trophectoderm cells can initiate at least two new cycles after they have become remote from the ICM. This might be because such obligatory signalling can license more than one cell cycle or that it initially occurs throughout the blastocyst, and only later becomes confined to the ICM.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We thank Andy Forkner and Inger Chilton for technical assistance, Ann Yates for help in preparing the manuscript, and the Royal Society and Wellcome Trust for support.


    Notes
 
1 To whom correspondence should be addressed. E-mail: richard.gardner{at}zoology.ox.ac.uk Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Barlow, P., Owen, D.A. and Graham, C. (1972) DNA synthesis in the preimplantation mouse embryo. J. Embryol. Exp. Morph., 27, 431–445.[ISI][Medline]

Beddington, R.S.P. and Lawson, K.A. (1990) Clonal analysis of cell lineages. In Copp, A.J. and Cockroft, D.L. (eds), Postimplantation Mammalian Embryos: A Practical Approach. Oxford University Press, Oxford, pp. 267–292.

Biggers, J.D., McGinnis, L.K. and Raffin, M. (2000) Amino acids and preimplantation development of the mouse in protein-free potassium simplex optimized medium. Biol. Reprod., 63, 281–293.[Abstract/Free Full Text]

Chai, N., Patel, Y., Jacobson, K., McMahon, J., McMahon, A. and Rappolee, D.A. (1998) FGF is an essential regulator of the fifth cell division in preimplantation mouse embryos. Dev. Biol., 198, 105–115.[ISI][Medline]

Champlin, A.K., Dorr, D.L. and Gates, A.H. (1973) Determining the stages of the estrous cycle in the mouse by the appearance of the vagina. Biol. Reprod., 8, 491–494.[ISI][Medline]

Chisholm, J.C., Johnson, M.H., Warren, P.D., Fleming, T.P. and Pickering, S.J. (1985) Developmental variability within and between mouse expanding blastocysts and their ICMs. J. Embryol. Exp. Morphol., 86, 311–336.[ISI][Medline]

Copp, A.J. (1978a) Interaction between inner cell mass and trophectoderm of the mouse blastocyst. II. The fate of the polar trophectoderm. J. Embryol. Exp. Morph., 51, 109–120.[ISI]

Copp, A.J. (1978b) Interaction between inner cell mass and trophectoderm of the mouse blastocyst. I. A study of cellular proliferation. J. Embryol. Exp. Morph., 48, 109–125.[ISI][Medline]

Cruz, Y.P. and Pedersen, R.A. (1985) Cell fate in the polar trophectoderm of mouse blastocysts as studied by the microinjection of cell lineage tracers. Dev. Biol., 112, 73–83.[ISI][Medline]

Gardner, R.L. (1990) Location and orientation of implantation. In Edwards, R.G. (ed.), Establishing a Successful Human Pregnancy: Serono Symposia Publications, Vol. 66. Raven Press, New York, pp. 225–238.

Gardner, R L. (1996) Clonal analysis of growth of the polar trophectoderm in the mouse. Hum. Reprod., 11, 1979–1984.[Abstract]

Gardner, R.L. (1997) The early blastocyst is bilaterally symmetrical and its axis of symmetry is aligned with the animal–vegetal axis of the zygote in the mouse. Development, 124, 289–301.[Abstract/Free Full Text]

Gardner, R L. (1998) Axial relationships between egg and embryo in the mouse. Curr. Top. Dev. Biol., 39, 35–71.[ISI][Medline]

Gardner, R.L. (2000) Flow of cells from polar to mural trophectoderm is polarized in the mouse blastocyst. Hum. Reprod., 15, 694–701.[Abstract/Free Full Text]

Gardner, R.L. (2001) Specification of embryonic axes begins before cleavage in normal mouse development. Development, 128, 839–847.[Abstract/Free Full Text]

Gardner, R.L. and Davies, T.J. (1993) Lack of coupling between onset of giant transformation and genome endoreduplication in the mural trophectoderm of the mouse blastocyst. J. Exp. Zool., 265, 54–60.[ISI][Medline]

Gardner, R.L. and Davies, T.J. (2000) Mouse chimeras and the analysis of development. In Tuan, R. and Lo, C. (eds), Developmental Biology Protocols, Vol. 11, Methods in Molecular Biology. Humana Press, Totowa, New Jersey, pp. 397–424.

Gardner, R.L., Meredith, M.M. and Altman, D.G. (1992) Is the anteriorposterior axis of the fetus specified before implantation in the mouse? J. Exp. Zool., 264, 437–443.[ISI][Medline]

Goodall, H. and Johnson, M.H. (1984) The nature of intercellular coupling within the preimplantation mouse embryo. J. Embryol. Exp. Morph., 79, 53–76.[ISI][Medline]

Smith, L.J. (1980) Embryonic axis orientation in the mouse and its correlation with blastocysts' relationship to the uterus. Part 1. Relationship between 82 h and 41/2 days. J. Embryol. Exp. Morph., 55, 257–277.[ISI][Medline]

Summers, M.C., McGinnis, L.K., Lawitts, J.A., Raffin, M. and Biggers, J.D. (2000) IVF of mouse ova in a simplex optimized medium supplemented with amino acids. Hum. Reprod., 15, 1791–1801.[Abstract/Free Full Text]

Watson, A.J. (1992) The cell biology of blastocyst development. Mol. Reprod. Dev., 33, 492–504.[ISI][Medline]

Submitted on January 15, 2002; accepted on March 6, 2002.