|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
First published online 30 January 2008
doi: 10.1242/dev.014316
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 MRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 0QH, UK.
2 Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2
3EJ, UK.
3 The Gurdon Institute, University of Cambridge, Tennis Court Road, Cambridge
CB2 1QN, UK.
Author for correspondence (e-mail:
mzg{at}mole.bio.cam.ac.uk)
Accepted 12 December 2007
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key words: Blastocyst, Mouse embryos, Pluripotency
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
).
Despite its importance, it is still unclear whether there is any spatial or
temporal pattern to the distribution of symmetric and asymmetric cell
divisions. If there is, does such pattern relate to particular lineages of
early blastomeres or is it independent of these? It also remains unclear
whether differential positioning of cells, inside versus outside, is an
essential prerequisite for any first differences to appear between mouse
embryo cells. Might some early pattern, meaning a propensity for blastomeres
to divide with specific orientations and/or order, exist prior to setting up
the inside and outside cell populations? If so, how might this early pattern
relate to the series of symmetric and asymmetric cleavage divisions that
position cells?
Two distinct models have been put forward to account for early mouse
development. One stresses that the mouse embryo is entirely symmetric, does
not have an animal-vegetal (AV) axis or show any other pre-patterning and
consequently develops as a ball of identical cells dividing with random
orientations (Alarcon and Marikawa,
2003; Hiiragi and Solter,
2004; Motosugi et al.,
2005). According to this view, the first differences between cells
can appear only when inside and outside cell populations are established after
the fourth cleavage divisions. This model also concludes that the blastocyst
cavity forms at a random site and so the orientation of the
embryonic-abembryonic axis does not relate to any earlier developmental event
(Motosugi et al., 2005). This
view is based on some lineage tracings of 2-cell blastomeres indicating that
their allocation to embryonic or abembryonic parts of the blastocyst is often
unpredictable, and on an idea that the regulative development of embryos
argues against any form of pattern (Alarcon
and Marikawa, 2003; Motosugi
et al., 2005; Chroscicka et
al., 2004). A second model proposes that some differences between
cells can be detected before cells adopt differential, inside or outside,
positions and whether these differences appear early depends on the
orientation of cell divisions along the AV axis
(Gardner, 1997;
Gardner, 2001;
Gardner, 2002;
Piotrowska et al., 2001;
Piotrowska and Zernicka-Goetz,
2001; Piotrowska-Nitsche et
al., 2005). The first evidence leading to this view was the
finding that the orientation of the first cleavage division along the AV axis
tends to be perpendicular to the embryonic-abembryonic axis of the future
embryo. Consequently, in most embryos, descendants of 2-cell blastomeres
contribute more cells to either the embryonic or abembryonic parts of the
blastocyst (Gardner, 2001;
Piotrowska et al., 2001;
Fujimori et al., 2003;
Plusa et al., 2005a).
Subsequently, it was suggested that this spatial distribution of the progeny
of 2-cell blastomeres depends upon separation of the animal and vegetal parts
of the zygote by second cleavage divisions
(Piotrowska-Nitsche and Zernicka-Goetz,
2005). This model is further supported by the discovery that the
degree of pluripotency differs significantly between blastomeres already at
the 4-cell stage and depends upon whether they inherit predominantly animal,
vegetal, or components of both poles of the zygote
(Piotrowska-Nitsche et al.,
2005). These differences in pluripotency appear to depend on the
extent of particular epigenetic modifications that affect development of
pluripotency (Torres-Padilla et al.,
2007). It is implicit to this second model that the early
differences between blastomeres are not determinative, but show plasticity and
can be reprogrammed if development is perturbed
(Zernicka-Goetz, 2006). Thus,
existence of such early differences between cells of the mouse embryo is
entirely compatible with the regulative nature of development.
If the route taken by each cell to their destinations could be analysed,
this should advance our understanding of how the blastocyst develops and
provide a direct method of detecting developmental regularities. A recent
study using recordings of the cell lineage in the mouse embryo documented the
proximity of cells with a shared clonal origin, the degree of asynchrony of
rounds of cell divisions and the movement of nuclei
(Kurotaki et al., 2007). The
remaining gaps in knowledge concern the relationship of cell lineage to inside
and outside positions of cells in the morulae and blastocyst and how this
might be affected by different patterns of early cleavage divisions. To
address this we have undertaken a complete analysis of all cell origins and
fates in relation to orientations of all cell divisions to ask whether cells
are allocated at random to the different blastocyst regions (embryonic and
abembryonic) and lineages (ICM and TE), or whether there are some
regularities, i.e. a pattern, to their allocation. Are the orientations of the
second cleavage divisions predictive of how pattern develops in relation to
the embryonic-abembryonic axis? Do successive cleavage divisions influence the
subsequent allocation of cells to ICM and TE in particular sectors of the
blastocyst? Finally, exactly how does the embryonic-abembryonic axis relate to
the spatio-temporal sequence of symmetric and asymmetric cell divisions?
To obtain a complete and precise dataset of 3-dimensional (3D) coordinates of all cells, their lineages and the orientation of all their divisions from the 2-cell to the blastocyst, we have developed time-lapse microscopy on multiple focal planes extending over this 3-day period. This non-invasive method showed that in the significant majority of embryos, the descendants of individual blastomeres give rise to distinct regions of the blastocyst. The 3D-lineage analysis revealed that there is a spatial and temporal relationship between symmetric and asymmetric divisions and demonstrated the way this contributes to patterning of the embryo and generation of the ICM and TE. Moreover, it indicated that the frequency of symmetric/asymmetric divisions of a blastomere correlates with its origin in relation to the AV axis of the zygote. Finally, it provided evidence that symmetric divisions anticipate the site of blastocyst cavity formation and so the orientation of the embryonic-abembryonic axis.
|
MATERIALS AND METHODS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Two-cell embryos were collected in M2 medium and then
cultured in KSOM (Piotrowska-Nitsche et
al., 2005). Time-lapse imaging was performed using a Zeiss
Axiovert microscope, Hamamatsu camera and Kinetic-Imaging software.
Fluorescence and DIC z-stacks were collected every 15 minutes, on 15
different planes for each time point, from 2-cell to blastocyst stage. The
PICT files obtained were converted to TIF using VisBio (LOCI, WI) and ImageJ
(NIH, Bethesda, MD).
Analysis of 4D movies
All cells were followed manually using SIMI Biocell software
(Schnabel et al., 1997).
Three-dimensional coordinates of nuclei were saved on average every two to
three frames and analysed as described in Results. All cell tracing was
carried out blindly, before assigning embryos to subgroups, and was
cross-checked by two researchers.
Cell divisions were classified as symmetric or asymmetric for all 8- and 16-cell blastomeres by scoring the position of daughter cells relative to the embryo surface one frame after and one before the next division in both DIC and fluorescence (see also Results). The timing of development was assessed as the period between successive second to fifth cleavages.
To describe the relative position of blastomeres in the 2- to 4-cell cleavage, we measured the angle between their apposing planes 15 minutes after division of the second blastomere using SIMI Biocell. We rotated the 3D representation of the embryo to look laterally at the axis defined by the daughters of the first 2- to 4-cell division and read the angle between this axis and that of the second 2- to 4-cell division (Fig. 4A).
We calculated the distance between polar body (PB) and the centre of the
two daughter cells during division in pixels using SIMI Biocell
(Fig. 4B). Descendants of
meridionally dividing blastomeres (M) were positioned equidistantly from the
PB. After equatorial/oblique division (E), only one of the two daughter cells
touched the PB and the distances between daughters and the PB differed by
approximately one cell diameter (
25-35 pixels).
From the 80 cavitated embryos, we analysed 66. Embryos were excluded because M and E divisions occurred synchronously, or the PB did not stay attached before the second cleavage, or the movie ended before cavitation.
|
RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The 8-cell stage is pivotal in the development of the mouse embryo as asymmetric divisions start from this stage, generating the first inside cells. Thus, our analysis first focused upon evaluating the spatial contribution of the four clonal descendants of each of the 8-cell blastomeres up to the 32-cell blastocyst. To achieve this, a coordinate of each 8-cell blastomere clone was assigned by calculating its centre of gravity using ImageJ (Fig. 2A-D). These were mapped with respect to the embryonic-abembryonic axis of the blastocyst. To do this, the 3D positional information obtained from the SIMI Biocell analysis was used to rotate the 3D representations of the blastocysts, so that their embryonic parts faced towards the left, placing the embryonic-abembryonic boundary in direct line of sight. We also calculated the coordinate of the centre point of the embryonic-abembryonic boundary and used this to align a tracing of the cavity. All this ensured that each embryo was identically aligned in 3D space. The orientated 3D representations were then projected onto 2D. This allowed the centre of gravity of each 8-cell clone to be accurately positioned in the embryonic or the abembryonic part of the embryo, the abembryonic part being the region around the cavity. If the centres of gravity lay upon the projected region occupied by the cavity then they were considered as abembryonic because at least half of the clone is positioned at the cavity or at the border of the cavity (Fig. 2E-H).
|
2 test; P<0.001). The group showing the
embryonic/abembryonic pattern significantly predominated. This analysis
assumes a null hypothesis in which any embryo would have an equal chance of
falling into any of these three categories. We specifically chose to do this
to exemplify a `worst-case scenario'. It could be argued that the null
hypothesis would predict each possible arrangement of eight-cell clones as
equally likely, and for the random (mixed) arrangement to vastly outnumber the
patterned and `half-half' embryos. In such a case, the number of patterned
embryos observed would have even greater significance. In this group, each
2-cell blastomere tends to contribute to either the embryonic or abembryonic
part of the blastocyst, rather than evenly to both, which shows that these
patterns reflect the lineage history of the 8-cell clones. We next considered whether the generation of blastocysts represented in the three different groups could have been due to a particular positioning of cells before the blastocyst cavity had formed or to the reorganisation of cells during the process of cavitation. To address this question, we analysed the 3D positioning of 8-cell clones in 32-cell embryos at two stages: just before and immediately following cavitation. This revealed that cavitation in general does not alter the relative spatial arrangement of clones, but clones stretch to `accommodate' the cavity (see Figs S1-S11 in the supplementary material). If at all, we observed only a slight displacement of clones as the cavity expanded.
The finding of predominance of the embryonic/abembryonic pattern indicates that in a significant number of embryos the future cavity will be surrounded predominantly by the descendents of one of the 2-cell blastomeres (Fig. 3). This suggests that the positioning of the blastocyst cavity and, therefore, the orientation of the embryonic-abembryonic axis, is biased by earlier developmental events in a significant majority of mouse embryos.
Relationship of the spatial arrangements of clones to the type of second cleavage
As the embryonic/abembryonic pattern was present in a significant majority
(61%), but not all, embryos, we next wondered whether the frequency with which
it develops might depend upon earlier division orientations, which we have
previously found to influence developmental outcomes
(Piotrowska-Nitsche et al.,
2005; Piotrowska-Nitsche and
Zernicka-Goetz, 2005;
Torres-Padilla et al., 2007).
There are four different permutations of second cleavage division depending on
their sequence and orientation with respect to the AV axis of the zygote. A
meridional division (along the AV axis) of one 2-cell blastomere may be
followed by an equatorial division of its sister cell (ME embryos), or this
sequence might be reversed (EM embryos). These two division patterns are most
common (Gardner, 2002;
Piotrowska-Nitsche and Zernicka-Goetz,
2005). Less common are divisions in which both blastomeres divide
with the same orientation: either both meridionally (MM) or equatorially (EE).
Our previous studies showed that how the embryo divides at the 2- to 4-cell
transition significantly affects its subsequent development. Thus, for
example, embryos in which animal and vegetal components are separated in both
blastomeres (EE embryos) were compromised in their developmental potential in
relation to the other groups of embryos. Interestingly, the 4-cell blastomeres
of one of the most common groups, ME embryos, were found to differ
significantly from each other. Specifically, the blastomere that inherits
vegetal components was found to be restricted in its developmental potential
and extent of particular epigenetic modifications than other blastomeres
(Piotrowska-Nitsche et al.,
2005; Torres-Padilla et al.,
2007). Therefore, we wished to examine whether the pattern of
8-cell clones might develop in relation to the different spatio-temporal
pattern of second cleavage divisions that affects separation of animal and
vegetal components of the zygote.
|
|
=64° and 65°)
than they did in MM and EE embryos (=15° and 28°)
(Fig. 4C). Within the group of
66 embryos, there were 24 ME (36%), 22 EM (33%), 13 MM (20%) and 7 EE (11%)
embryos. Thus, embryos in which the second cleavages were perpendicular to
each other (ME and EM) were most common, which is in agreement with some
previous studies (Gardner,
2002; Piotrowska-Nitsche and
Zernicka-Goetz, 2005), but not with others
(Louvet-Vallee et al.,
2005).
When we examined the distribution of the 8-cell clones in blastocysts, we
found that the frequency of development of the embryonic/abembryonic pattern
differed depending on the second cleavage orientations. It was evident in 71%
of ME, 55% of EM, 54% of MM and 57% of EE embryos. Thus, strikingly, ME
embryos display a significant tendency to develop the embryonic/abembryonic
pattern (2 test, P=0.014;
Fig. 4D), suggesting that the
second cleavages bisecting the AV axis could influence development.
Relationships between symmetric and asymmetric divisions in generating inside and outside cells
We next asked whether the specific spatial distribution of 8-cell clones,
revealed by the above analysis, indicated any regional differences in the
generation of inside and outside cells leading to embryonic/abembryonic
pattern. Inner, pluripotent cells are generated together with outer cells
through asymmetric/differentiative divisions of some 8- and 16-cell
blastomeres, whereas symmetrically/conservatively dividing cells generate two
outside daughters (Johnson and Ziomek,
1981). Thus, to determine whether there is any relationship
between these division types, we analysed all divisions in terms of whether
they were symmetric or asymmetric at the 8- to 16-cell- and 16- to 32-cell
transitions and measured all cell cycle lengths. To determine the division
orientation, we scored the position (inner or outer) of daughter cells both
immediately after their mitotic division and also at the end of their cell
cycle, to check whether cells had changed their position. We found that in
most cases (95.1%, n=1578), they did not change their position and so
cells scored as inner contributed to ICM.
|
). To
explore this further, we focused on analysing each division undertaken by
individual 8-cell blastomeres.
The two daughters of a symmetric division can subsequently divide either
asymmetrically or symmetrically with equal probability, if at random. Thus,
the possible division permutations of the daughters of symmetric divisions
are: SS (both symmetric), AA (both asymmetric) and AS/SA (one of each, in
either sequence). The outside daughter cell of an asymmetric division can also
divide either asymmetrically or symmetrically, again with equal probability if
at random (the inside daughter divides to generate two inside cells; inner
division, I). However, we found that the type of successive division
undertaken by an outside cell from the fourth to the fifth cleavage is not
taken at random. Our analysis showed that a symmetrically dividing mother cell
most frequently produced daughters of which one undertook an asymmetric
division and the other a symmetric division (2 test,
P<0.001; Fig. 5B).
Thus, cells derived from a symmetrically dividing mother have an equal chance
to divide symmetrically or asymmetrically. However, when the sequence of
division of the daughters is considered, the first dividing daughter was more
likely to divide symmetrically and the second asymmetrically
(P=0.028; Fig. 5B).
Additionally, an asymmetrically dividing mother cell most frequently gave rise
to an outside daughter that divided symmetrically (2 test,
P<0.001; Fig. 5B).
This suggests that, in general, when we consider all embryos together, there
is a tendency for a compensatory relationship between symmetric and asymmetric
divisions in sequential cell divisions that might be important in regulating
the number of inside versus outside cells.
The specific division orientation of a blastomere might be affected by its
age or by whether it divides earlier or later than its neighbours, as
previously suggested (Garbutt et al.,
1987). However, our dataset indicated that there were no
significant differences in cell cycle lengths between symmetrically and
asymmetrically dividing cells (Table
1). It also showed that blastomeres undertaking earlier or later
fourth cleavage divisions had no tendency to divide symmetrically rather than
asymmetrically (2 test, P=0.89). The same held true
for 16-cell blastomeres when they undertook the fifth cleavage round. Thus, it
appears that neither age nor division order are likely to be factors
determining the kind of division cells undertaken. We also measured and
analysed cell cycle lengths of inside versus outside cells in the 66 embryos
studied. This showed that once positioned, inner cells have a significantly
longer cell cycle than outer cells (54 minutes longer; Student's
t-test to compare average cell cycle lengths, P<0.001;
Table 2).
|
|
;
Kelly et al., 1978;
Graham and Deussen, 1978;
Surani and Barton, 1984) and
contribute preferentially to the embryonic rather than abembryonic blastocyst
region (Piotrowska et al.,
2001). We found, however, that each of the 2-cell blastomeres
contributed on average an equal number of cells to the ICM at the 32-cell
stage, and the proportion of inside cells generated by the earlier-dividing
blastomere was insignificantly higher (52% versus 48% from earlier or later
dividing, respectively; Student's t-test, P=0.14; see Table
S1 in the supplementary material). The TE also comprised equal proportions of
cells derived from each of the 2-cell blastomeres (50% from both). Likewise,
we saw no tendency for the earlier-dividing 2-cell blastomere to contribute
more cells to the embryonic part of the blastocyst, consistent with studies
that reported embryonic-abembryonic pattern irrespective of cell division
order (Fujimori et al., 2003).
Determination of cell cycle lengths of all cells also allowed us to ask
whether a tendency to divide earlier is inherited by the progeny of an
earlier-dividing cell. We found no significant differences in cell cycle
length between progeny of either 2-cell blastomere
(Table 3). The differences in
the spatial contributions made by the earlier- versus later-dividing 2-cell
blastomere in previous studies (Barlow et
al., 1972; Kelly et al.,
1978; Graham and Deussen,
1978; Surani and Barton,
1984; Piotrowska et al.,
2001) might be due to labelling and/or repeated manipulation of
embryos, which influenced the behaviour of cells in ways that did not occur
when using the non-invasive time-lapse methods of the present approach.
|
|
; Torres-Padilla et al.,
2007). Moreover, it suggests that mouse embryos cannot be entirely
symmetric, but show some polarity along their AV axis that influences the
pattern of symmetric and asymmetric cell divisions.
Patterns of symmetric and asymmetric divisions anticipate the orientation of the embryonic-abembryonic axis
The positioning of the blastocyst cavity defines the orientation of the
embryonic-abembryonic axis. Our analyses of the spatial arrangement of 8-cell
clones in the blastocyst revealed that the majority of mouse embryos develop
with pattern with respect to the embryonic-abembryonic axis. This raised the
possibility that the orientation of this axis could be predicted by a
different spatio-temporal sequence of symmetric versus asymmetric divisions at
the future embryonic and abembryonic poles. To address this, we followed the
generation of the inner and outer cell populations and asked whether two
features of the blastocyst, the dovetailed 1/8 clone and the cavity, might
form with respect to a particular distribution of asymmetric versus symmetric
cell divisions.
Analysis of the spatial allocation of cells within the dovetailed clone
showed that it comprised on average 53% embryonic cells (28% ICM, 25% polar
TE), 31% boundary cells (16% inner surface, 15% boundary TE) and 15%
abembryonic cells (mural TE). By contrast, the contributions made by the other
three clones from the same 2-cell blastomere were: 12% embryonic cells (8%
ICM, 4% polar TE), 35% boundary cells (24% inner surface, 11% boundary TE) and
53% abembryonic cells. In agreement with this, analysis of the cell division
patterns revealed that the dovetailed clone arose after significantly more
asymmetric divisions during the fourth cleavage than undertaken by the other
three clones derived from the same 2-cell blastomere
(Table 4; 2 to
compare frequencies between clones, P=0.038). Furthermore, in the
subsequent fifth cleavage, inner divisions were significantly more prevalent
in the dovetailed region than in its three sister clones
(Table 4; P=0.016).
This indicates that specific patterns of symmetric and asymmetric divisions
might influence how individual blastomeres contribute to particular blastocyst
regions.
|
|
|
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Development of embryonic/abembryonic pattern
This non-invasive 3D lineage analysis of all cells, their origins,
behaviour and their final positioning at the blastocyst stage, provides
support for cell-labelling lineage studies that indicated a non-random
allocation of the progeny of 2-cell and 4-cell blastomeres in the majority of
mouse embryos (Gardner, 2001;
Piotrowska et al., 2001;
Fujimori et al., 2003;
Piotrowska-Nitsche and Zernicka-Goetz,
2005). The embryonic/abembryonic pattern observed in the current
study is one in which the 8-cell clones in 61% of embryos come to occupy
distinct regions in relation to the embryonic-abembryonic axis. Four 8-cell
clones originating from one 2-cell blastomere are mainly positioned in the
embryonic part of the embryo. Of the four clones originating from the other
2-cell blastomere, three are positioned in the abembryonic part and one
dovetailed clone crosses more into the embryonic part and as such leads to a
tilt between the embryonic-abembryonic boundary and the boundary between
descendants of the 2-cell blastomeres. Thus, the tilt is the consequence of
predominant asymmetric divisions that position the dovetailed clone
(Table 4,
Fig. 2H). This finding alone
could account for why alternative models have been proposed to explain the
clonal distribution of cells in the blastocyst. This tilt was interpreted by
some authors as evidence of random cell arrangement and mixing
(Alarcon and Marikawa, 2003;
Chroscicka et al., 2004;
Motosugi et al., 2005), and
not, as shown here, as an actual part of the embryonic/abembryonic
pattern.
Development in relation to the animal-vegetal axis
The present lineage-tracing analysis also gives insight into another
question under debate: is the mouse embryo entirely symmetric or not? The
current study suggests that the extent of development of the
embryonic/abembryonic pattern depends on how the embryo divides with respect
to the AV axis of the zygote. The first zygotic cleavage usually occurs along
the AV axis (Plusa et al.,
2005a). Only in the second cleavage rounds do equatorial
divisions, separating animal and vegetal parts, become significant
(Gardner, 2002).
Embryonic/abembryonic pattern significantly predominates in embryos in which
the animal and vegetal cells are separated by the later second cleavage
division (ME embryos). This is consistent with earlier studies indicating that
cells inheriting either the animal, vegetal, or both poles of the zygote have
different properties (Piotrowska-Nitsche
and Zernicka-Goetz, 2005;
Piotrowska-Nitsche et al.,
2005; Torres-Padilla et al.,
2007). This argues for zygote organisation (pre-pattern) and its
AV polarity having influence upon the development of mouse embryos and argues
against the view that mouse embryos are entirely symmetrical without any
pre-pattern (Hiiragi and Solter,
2004; Louvet-Vallee et al.,
2005). This inconsistency might be because the authors expressing
the latter view could not examine the possible influence of AV axis because
the marker of this axis (the PB) did not stay attached to the embryos they
studied (Hiiragi and Solter,
2004). Without any marker, it is not possible to determine whether
AV axial information affects development and the development of differences
among blastomeres might be classified as stochastic
(Dietrich and Hiiragi, 2007).
There is evidence indicating that development of the mouse embryo is
influenced by whether cells divide along or perpendicular to the AV axis.
Since the patterns of such divisions differ between embryos, it is essential
to classify them accordingly to recognise the extent of, and possible reasons
behind, development of blastocyst patterning. Although the
embryonic/abembryonic pattern clearly predominates in ME embryos, it is also
seen in others. It is possible that this also reflects the way in which the AV
axis is partitioned in these embryos as a result of variability in the
orientation of cleavage divisions. Taken together, our results suggest that
developmental properties are polarised in the zygote. Because cleavage
divisions partition the zygote in different ways, embryos differ from each
other and so cannot all have a fixed relationship between lineage and
fate.
Frequency of asymmetric versus symmetric divisions and the embryonic-abembryonic axis
Examination of the spatial and temporal patterns of symmetric and
asymmetric divisions at the 8- to 16-cell and 16- to 32-cell transitions also
allowed us to address another question under current debate: does the
embryonic-abembryonic axis become oriented at random or could its orientation
be predicted by earlier developmental events? We found that the blastocyst
cavity has a significant tendency to develop where symmetric divisions
predominate. This might suggest that junctions between inner and outer cells
are weaker adjacent to symmetrically dividing cells, thus facilitating the
cavity formation at that site. One possible explanation for this might be the
absence of mid-bodies between inner and outer cells as a consequence of their
division history (Plusa et al.,
2005b). Interestingly, the formation of the cavity within a region
of symmetric divisions was again particularly evident in ME embryos. Thus, it
appears that positioning of the cavity, and so the orientation of
embryonic-abembryonic axis, is influenced by a pattern of earlier cell
divisions.
What determines whether divisions are symmetric or asymmetric? One
possibility is blastomere age or division order, as suggested previously
(Garbutt et al., 1987).
However, our lineage-tracing analysis did not reveal any significant
correlation between cell cycle lengths or order of divisions and division
orientation. Thus, how cell division orientation is determined remains
unclear, but the finding that vegetally-derived cells take preferentially
symmetric divisions might help to shed light on this process in the
future.
Does shape influence patterning?
The contemporaneous lineage study of Kurotaki et al.
(Kurotaki et al., 2007)
suggests that the orientation of the embryonic-abembryonic axis develops in
response to the shape of the zona pellucida. This contrasts with another
recent study showing that the embryonic-abembryonic axis of the mouse
blastocyst is prepatterned and develops independently of the zona pellucida
(Gardner, 2007). To address
this discrepancy we analysed the embryos from our study using the approach of
Kurotaki et al. by measuring the angle between the 2-cell boundary and the
embryonic-abembryonic axis. We confirm that the 2-cell embryo is oriented
along the long axis of the zona in most (85%) cases. However, in only 35% of
embryos was the angle between the 2-cell boundary and the
embryonic-abembryonic axis more than 70°, in contrast to Kurotaki et al.
(Kurotaki et al., 2007) who
found this relationship in 64% of embryos. Thus, in embryos analysed in the
present 4D lineage-tracing study, the zona pellucida does not appear to have a
role in patterning.
This is not to say that shape cannot influence patterning. It has been
demonstrated previously by us and others that the shape of the embryo could
influence development (Gray et al.,
2004; Plusa et al.,
2005a; Gardner and Davies,
2002). Thus, in experimentally elongated embryos, cells tend to
divide through their short axis and hence, if indeed embryos were to adopt the
shape of a considerably elongated zona, this might affect division
orientation. The time-lapse studies we carried out here indicate that
blastomeres were not significantly restrained by the zona from compaction up
to cavitation; a gap of a few microns separated the cells from the zona. Thus,
we cannot account for the response of the embryos to the zona shape in the
study of Kurotaki et al. (Kurotaki et al.,
2007). It should be noted, however, that Kurotaki et al. did not
examine cell division orientations either at the early cleavage stages, with
respect to the AV axis and each other, or at the later stages when asymmetric
divisions separate inside from outside cells. In the absence of this
information, any relationship between lineages and their division patterns and
the orientation of the embryonic-abembryonic axis would be difficult to
find.
Multiple ways to build a blastocyst: developmental safeguards?
Our time-lapse studies indicate that there might be more than one way in
which embryos develop into blastocysts. The embryonic/abembryonic pattern is
seen in the significant majority of embryos, but is not exclusive. This is
perhaps unsurprising, given that mouse development is variable in other
aspects. On the level of single cells, pattern is not invariant; for example,
clone #4 does not always arise from the same `progenitor/mother' cell. Thus,
as with the assembly of any complex structure, in some cases different
construction techniques may be applied such that the end-point can be achieved
by following different paths. It will be difficult to test whether patterned
embryos have any developmental advantage over non-patterned, because this
necessitates assessing the developmental success of different embryo types in
the same mother. However, embryos in which animal and vegetal parts are
separated in both 2-cell blastomeres have significantly reduced viability
(Pitorowska-Nitsche and Zernicka-Goetz, 2005), suggesting that some
`developmental routes' might be more favourable than others.
Embryos might take slightly different routes of development depending on how components of the zygote become partitioned through patterns of early cell divisions. Pattern can reflect lineage history in embryos partitioned in particular ways along the AV axis. This raises the possibility of specific components distributed along this axis that can influence development. In embryos undergoing other cleavage patterns, such components might be partitioned in a way that gives rise to progeny with more mixed developmental properties. An ability to control differential gene expression in more than one way could reflect regulatory mechanisms in the embryo that ensure its normal development depending on which route it takes earlier on. Such redundant mechanisms are employed in biological systems to safeguard complex processes from environmental perturbations. The ability of the embryo to regulate might then mask the presence of early pattern. Indeed, the very act of experimental manipulation could bring a correction mechanism into play that triggers differential gene expression and forces development in a specific direction.
Even though there appears to be more than one route towards the development
of blastocyst, there is a considerable weight of evidence pointing to a
relationship between how the embryo divides in relation to the AV axis and
subsequent developmental processes. An appreciation that blastomeres
inheriting different parts of the zygote differ, is assisting our
understanding. For example, it has permitted the discovery of the earliest
epigenetic modification known to date that is important for cell pluripotency
(Torres-Padilla et al., 2007;
Hemberger and Dean, 2007).
Hence, discovering the rules that govern the development of patterned embryos
provides the potential for gaining greater insight into developmental
mechanisms operating at this early stage of embryogenesis.
Supplementary material
Supplementary material for this article is available at
http://dev.biologists.org/cgi/content/full/135/5/953/DC1
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alarcon, V. B. and Marikawa, Y. (2003).
Deviation of the blastocyst axis from the first cleavage plane does not affect
the quality of mouse postimplantation development. Biol.
Reprod. 69,1208
-1212.
Barlow, P., Owen, D. A. J. and Graham, C. F. (1972). DNA synthesis in the preimplantation mouse embryo. J. Embryol. Exp. Morphol. 27,431 -435.[Medline]
Chroscicka, A., Komorowski, S. and Maleszewski, M. (2004). Both blastomeres of the mouse 2-cell embryo contribute to the embryonic portion of the blastocyst. Mol. Reprod. Dev. 68,308 -312.[CrossRef][Medline]
Dietrich, J. E. and Hiiragi, T. (2007).
Stochastic patterning in the mouse preimplantation embryo.
Development 134,4219
-4231.
Fleming, T. P. (1987). A quantitative analysis of cell allocation to trophectoderm and inner cell mass in the mouse blastocyst. Dev. Biol. 119,520 -531.[CrossRef][Medline]
Fujimori, T., Kurotaki, Y., Miyazaki, J. and Nabeshima, Y.
(2003). Analysis of cell lineage in two- and four-cell mouse
embryos. Development
130,5113
-5122.
Garbutt, C. L., Johnson, M. H. and George, M. A.
(1987). When and how does cell division order influence cell
allocation to the inner cell mass of the mouse blastocyst?
Development 100,325
-332.
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]
Gardner, R. L. (2001). Specification of embryonic axes begins before cleavage in normal mouse development. Development 128,839 -847.[Abstract]
Gardner, R. L. (2002). Experimental analysis of
second cleavage in the mouse. Hum. Reprod.
17,3178
-3189.
Gardner, R. L. (2007). The axis of polarity of
the mouse blastocyst is specified before blastulation and independently of the
zona pellucida. Hum. Reprod.
22,798
-806.
Gardner, R. L. and Davies, T. J. (2002).
Trophectoderm growth and bilateral symmetry of the blastocyst in the mouse.
Hum. Reprod. 17,1839
-1845.
Graham, C. F. and Deussen, Z. A. (1978). Features of cell lineage in preimplantation mouse development. J. Embryol. Exp. Morphol. 48,53 -72.[Medline]
Gray, D., Plusa, B., Piotrowska, K., Na, J., Tom, B., Glover, D. M. and Zernicka-Goetz, M. (2004). First cleavage of the mouse embryo responds to change in egg shape at fertilization. Curr. Biol. 14,397 -405.[CrossRef][Medline]
Hadjantonakis, A. K. and Papaioannou, V. E. (2004). Dynamic in vivo imaging and cell tracking using a histone fluorescent protein fusion in mice. BMC Biotechnol. 4, 33.[CrossRef][Medline]
Hemberger, M. and Dean, W. (2007). Epigenetic arbitration of cell fate decisions: tipping the bias. Dev. Cell 12,176 -178.[CrossRef][Medline]
Hiiragi, T. and Solter, D. (2004). First cleavage plane of the mouse egg is not predetermined but defined by the topology of the two apposing pronuclei. Nature 430,360 -364.[CrossRef][Medline]
Johnson, M. H. and Ziomek, C. A. (1981). The foundation of two distinct cell lineages within the mouse morula. Cell 24,71 -80.[CrossRef][Medline]
Kelly, S. J., Mulnard, J. G. and Graham, C. F. (1978). Cell division and cell allocation in early mouse development. J. Embryol. Exp. Morphol. 48, 37-51.[Medline]
Kurotaki, Y., Hatta, K., Nakao, K., Nabeshima, Y. and Fujimori,
T. (2007). Blastocyst axis is specified independently of
early cell lineage but aligns with the ZP shape.
Science 316,719
-723.
Louvet-Vallee, S., Vinot, S. and Maro, B. (2005). Mitotic spindles and cleavage planes are oriented randomly in the two-cell mouse embryo. Curr. Biol. 15,464 -469.[CrossRef][Medline]
Motosugi, N., Bauer, T., Polanski, Z., Solter, D. and Hiiragi,
T. (2005). Polarity of the mouse embryo is established at
blastocyst and is not prepatterned. Genes Dev.
19,1081
-1092.
Piotrowska, K. and Zernicka-Goetz, M. (2001). Role for sperm in spatial patterning of the early mouse embryo. Nature 409,517 -521.[CrossRef][Medline]
Piotrowska, K., Wianny, F., Pedersen, R. A. and Zernicka-Goetz,
M. (2001). Blastomeres arising from the first cleavage
division have distinguishable fates in normal mouse development.
Development 128,3739
-3748.
Piotrowska-Nitsche, K. and Zernicka-Goetz, M. (2005). Spatial arrangement of individual 4-cell stage blastomeres and the order in which they are generated correlate with blastocyst pattern in the mouse embryo. Mech. Dev. 122,487 -500.[CrossRef][Medline]
Piotrowska-Nitsche, K., Perea-Gomez, A., Haraguchi, S. and
Zernicka-Goetz, M. (2005). Four-cell stage mouse blastomeres
have different developmental properties. Development
132,479
-490.
Plusa, B., Hadjantonakis, A. K., Gray, D., Piotrowska-Nitsche, K., Jedrusik, A., Papaioannou, V. E., Glover, D. M. and Zernicka-Goetz, M. (2005a). The first cleavage of the mouse zygote predicts the blastocyst axis. Nature 434,391 -395.[CrossRef][Medline]
Plusa, B., Frankenberg, S., Chalmers, A., Hadjantonakis, A. K.,
Moore, C. A., Papalopulu, N., Papaioannou, V. E., Glover, D. M. and
Zernicka-Goetz, M. (2005b). Downregulation of Par3 and aPKC
function directs cells towards the ICM in the preimplantation mouse embryo.
J. Cell Sci. 118,505
-515.
Schnabel, R., Hutter, H., Moerman, D. and Schnabel, H. (1997). Assessing normal embryogenesis in Caenorhabditis elegans using a 4D microscope: variability of development and regional specification. Dev. Biol. 184,234 -265.[CrossRef][Medline]
Surani, M. A. and Barton, S. C. (1984). Spatial distribution of blastomeres is dependent on cell division order and interactions in mouse morulae. Dev. Biol. 102,335 -343.[CrossRef][Medline]
Torres-Padilla, M. E., Parfitt, D. E., Kouzarides, T. and Zernicka-Goetz, M. (2007). Histone arginine methylation regulates pluripotency in the early mouse embryo. Nature 445,214 -218.[CrossRef][Medline]
Zernicka-Goetz, M. (2006). The first cell-fate decisions in the mouse embryo: destiny is a matter of both chance and choice. Curr. Opin. Genet. Dev. 16,406 -412.[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
