| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research Article |
gorzata Waksmundzka 
niewska 2
Department of Embryology, Institute of Zoology, Warsaw University, Warsaw 02–096, Poland
ABSTRACT
The second cleavage of the mouse embryo is asynchronous. Some recent investigators have proposed that the sequence of division of blastomeres in two-cell embryos may predict the ultimate location of the descendants of these blastomeres within the blastocyst. To verify this model, we tracked the cells derived from two-cell stage blastomeres using tetramethylrhodamine-conjugated dextran as a lineage tracer. In the first variant of the experiment, we labeled one of two blastomeres in two-cell embryos and subsequently recorded which blastomere cleaved first. In the second variant of the experiment, fluorescent dextran was injected at the three-cell stage into the blastomere that had not yet cleaved. Subsequently, the fate of the progeny of labeled and unlabeled blastomeres was followed up to the blastocyst stage. Our results suggest that allocation of cells into the embryonic and abembryonic parts of the blastocyst is not determined by the order of cleavage of the first two blastomeres.
developmental biology, early development, embryo
INTRODUCTION
At the end of preimplantation development (i.e., at the blastocyst stage), the mammalian embryo contains two types of cells, the inner cell mass (ICM), which will form the fetus and parts of the placenta, and the trophectoderm cells, which will form much of the placenta but will not contribute to the fetus. The blastocyst displays axial polarity, known as the embryonic-abembryonic (Em-Ab) axis. Asymmetry of the blastocyst results from the eccentric location of the ICM, which is situated at one pole of the Em-Ab axis and, together with polar trophectoderm, forms an embryonic region of the blastocyst. The Em-Ab polarity of the blastocyst corresponds directly to the dorsoventral polarization of the postimplantation conceptus, with the embryonic pole of the blastocyst indicating orientation of the future dorsal side of the fetus [1, 2]. Until recently, it was believed that the establishment of the axes of a mammalian conceptus does not depend on the cleavage pattern of the preimplantation embryo. It was postulated that cell specification which takes place during the formation of the blastocyst is controlled by reciprocal cell communication in the compact embryo, which has already undergone several mitotic divisions [3]. However, it was recently suggested that the Em-Ab axis of the mouse blastocyst depends on the early cleavage pattern and, in most cases, is perpendicular to the plane of the first mitotic division of the zygote [4–9]. As a consequence, it was proposed that cells derived from blastomeres of two-cell embryos have different developmental fates: the progeny of one blastomere forms predominantly the ICM and polar trophectoderm, while the progeny of the other blastomere contributes primarily to the mural trophectoderm [6, 10]. However, several other studies, including the present study, failed to confirm such differential contribution of two-cell blastomeres to the separate cell lineages of the blastocyst [11–15].
The second cleavage of the mouse embryo is typically asynchronous. It has been proposed that the sequence of division of blastomeres of two-cell embryos may predict the final positioning of their descendant cells within the blastocyst. It was reported that the blastomere that divides first is the one that contributes preferentially to the embryonic part of the blastocyst [6, 9]. Moreover, it has been suggested that the fate of each two-cell blastomere is determined during the first cleavage, because the blastomere that inherits the sperm entry point has a strong tendency to divide earlier than the other blastomere [7]; however, this was later questioned by Motosugi et al. [13]. To examine whether the order of division of two-cell blastomeres determines the allocation of their descendants in the blastocyst, we performed a lineage tracing experiment injecting one blastomere with fluorochrome-conjugated dextran. In variant I of the experiment, we labeled one of two blastomeres in two-cell embryos and subsequently recorded which blastomere cleaved first. In the variant II of the experiment, fluorescent dextran was injected at the three-cell stage into the lagging blastomere. Subsequently, the fate of the progeny of labeled and unlabeled blastomeres was followed up to the blastocyst stage. Our results suggest that allocation of cells into the embryonic and abembryonic parts of the blastocyst does not depend on the order of cleavage of the first two blastomeres.
MATERIALS AND METHODS
All animal investigations performed in this study were approved by Local Ethic Committee No. 1 in Warsaw, Poland, according to the European Union Council Directive 86/609/Eec of November 24, 1986, on the approximation of laws, regulations, and administrative provisions of the member states regarding the protection of animals used for experimental and other scientific purposes [16, 17]. All animals were raised on the premises.
Chemicals, Media, and Conditions of In Vitro Culture of Embryos
All chemicals, unless otherwise stated, were obtained from Sigma-Aldrich (Pozna
, Poland). All manipulations of embryos were performed in medium 16 buffered with Hepes (M2) [18]. The medium used for long-term culture of embryos was potassium simplex optimization medium (KSOM; Specialty Media) [19]. Embryos were cultured at 37.5°C in 5% CO2 in air under paraffin oil in plastic dishes (Cellstar, Greiner Bio-one).
Obtaining Two-Cell and Three-Cell Mouse Embryos
Embryos were collected from the oviducts of superovulated F1 (CBA/H x C57Bl10) or F1 (C57Bl10 x CBA/H) females. Two- to four-month-old females were superovulated by i.p. injection of eCG (10 IU of Folligon; Intervet, Boxmeer, the Netherlands), followed 44–48 h later by i.p. injection of hCG (10 IU of Chorulon, Intervet). Females were then caged with F1 males, and females in which a vaginal plug was found were killed by cervical dislocation 44 h (for two-cell embryos) or 48 h (for three-cell embryos) after the injection of the second hormone. Isolated oviducts were put into M2, and embryos were released. Embryos were cultured in droplets of M2.
Injection of Fluorochrome-Conjugated Dextran into Two-Cell and Three-Cell Embryos
Embryos were microinjected with a solution of dextran (molecular weight, 70 000) conjugated with tetramethylrhodamine (Molecular Probes, Eugene, OR). Dextran solution was injected with a micropipette attached to a Leica micromanipulator connected to a Leica Labovert FS inverted microscope using an Eppendorf pneumatic microinjector type 5242 connected to the micromanipulator. The solution of dextran was prepared immediately before microinjection from stock solution. The stock solution (10 mg/ml of H2O) was diluted to a final concentration of 2 mg/ml in 20 mM PIPES buffer (pH 7.4) containing 0.14 M KCl. The volume of injected solution was approximately 2 pl.
Two types of experiments were performed. In variant I, the injection was into one blastomere of two-cell embryos that were isolated 44 h after hCG injection (Fig. 1A). Labeled embryos were subsequently cultured for 5 h in M2. Embryos were checked for cleavage every 20–30 min under an inverted microscope, and those that divided were subsequently inspected with a fluorescent microscope (Axiovert 135, Zeiss, Jena, Germany) to observe the sequence of the cleavage. Embryos were classified into the following three groups: 1) embryos in which the labeled blastomere cleaved first, 2) embryos in which the unlabeled blastomere cleaved first, and 3) embryos in which none of the blastomeres cleaved during the observation period. All embryos were further cultured for 52–74 h (100–120 h after hCG injection) in KSOM.
|
In variant II of the experiment, embryos isolated 48 h after hCG injection were cultured in M2 and examined every 15–20 min for cleavage, and three-cell embryos were selected for microinjection. The blastomere that had not yet cleaved (the slow blastomere) was injected with dextran (Fig. 1B). All embryos that survived injection were analyzed under a fluorescence microscope, and only those that were successfully injected with dextran (i.e., they fluoresced) were selected for subsequent culture. Embryos were cultured in conditions as already described for the next 52–74 h (100–120 h after hCG injection).
Observation of Embryos
Embryos were fixed at the blastocyst stage with 4% paraformaldehyde in PBS for 15 min at room temperature. After fixation, they were rinsed with PBS, and nuclei were stained with a 10 µM solution of DRAQ (Biostatus Ltd.) in PBS at 37.5°C for at least 30 min under paraffin oil in plastic dishes in which part of the bottom was removed and replaced with the cover glass. Embryos were examined with a confocal laser microscope (LSM 510, Zeiss). They were oriented with the Em-Ab axis of the blastocyst parallel to the bottom of the dish. Optical sections were taken every 1 µm to determine the distribution of dextran-labeled and dextran-unlabeled cells in the embryonic and abembryonic parts of the blastocyst. The number of cells was determined by calculating the number of nuclei in the entire blastocyst and in the embryonic and abembryonic parts.
RESULTS
Effect of Injection of Dextran Solution on Development of Embryos
First, we analyzed the sequence of division of blastomeres in two-cell embryos that were injected with fluorescent dextran into one blastomere (Fig. 1A). In 57% (52/92) of embryos in variant I, the unlabeled blastomere divided earlier. However, this difference was not statistically significant (P = 0.37, chi-square test).
Second, we examined the success rate of development of microinjected embryos to the blastocyst stage (Table 1). The numbers of embryos that reached the blastocyst stage were not significantly different between the group of embryos in which one of the blastomeres was labeled at the two-cell stage and the group of embryos in which one of the blastomeres was labeled at the three-cell stage. However, there was a significant difference between the numbers of injected and control (noninjected) embryos that reached the blastocyst stage of development (approximately 48% and 76%, respectively).
|
Contribution of Cells Derived from Dextran-Labeled Blastomeres in Blastocysts
Of 196 embryos in which one of the blastomeres was labeled with fluorescent dextran at the two-cell or three-cell stage, 94 reached the blastocyst stage (Table 1, variants I and II together). However, in 23 blastocysts (24%), we were unable to determine the exact number and distribution of cells derived from the microinjected blastomere. In most cases, this failure was due to the expansion of blastocysts. Expanded blastocysts have a high number of cells, and their ICMs are flattened against polar trophectoderm; therefore, the border zone of the ICM is difficult to locate. Therefore, we analyzed the distribution of labeled cells in the remaining 71 blastocysts (42 blastocysts from variant I and 29 blastocysts from variant II). Each blastocyst was assigned to the young group or to the expanded group according to the size of the blastocyst cavity (Fig. 2). The blastocyst was considered young if the diameter of the cavity was no larger than two thirds the diameter of the embryo (Fig. 3A and supplemental Fig. S1 available online at http://www.biolreprod.org). The blastocyst was considered expanded if the diameter of the cavity was larger than two thirds the diameter of the embryo (Fig. 3B and supplemental Fig. S2 available online at http://www.biolreprod.org). The border zone between the embryonic and abembryonic parts was defined differently for young and expanded blastocysts. For young blastocysts, the border zone separates the ICM and polar trophectoderm (the embryonic part) from the mural trophectoderm (the abembryonic part), but the border zone (containing one layer of superficial cells in the ICM) belongs to the abembryonic part [6]. For expanded blastocysts, the entire ICM and polar trophectoderm belong to the embryonic part (Fig. 2). For each blastocyst, we determined the total number of cells, the number of cells derived from the labeled blastomere, and the distribution pattern of labeled and unlabeled cells within the embryonic and abembryonic parts.
|
|
Most of the blastocysts (88% [37/42]) that developed from embryos labeled at the two-cell stage (variant I) had more cells derived from the unlabeled blastomere than from its labeled counterpart (data not shown). In variant II, dextran was injected at the three-cell stage into the blastomere in which cleavage was delayed. In 79% (23/29) of blastocysts in this group, more than 50% of cells were derived from the unlabeled blastomere (data not shown).
Order of Cleavage of Two-Cell Stage Blastomeres and Contribution of Their Progeny in Blastocysts
In variant I, we labeled one of the blastomeres in two-cell embryos with fluorescent dextran and observed which blastomere (labeled or unlabeled) divided first. In variant II, we observed the order of cleavage of two-cell stage blastomeres and then microinjected the lagging blastomere. This allowed us to track the distribution of the descendants of the blastomere that cleaved first (the fast blastomere) and of the lagging blastomere (the slow blastomere) within the blastocyst.
For each blastocyst, we determined the numbers of cells derived from the fast blastomere and from the slow blastomeres in the embryonic part and in the abembryonic part. For young blastocysts, the allocation of cells derived from two-cell stage blastomeres into the border zone was also examined. The distribution of cells was scored separately for blastocysts from variant I and variant II (Table 2 and supplemental Tables S1 and S2 available online at http://www.biolreprod.org). We arbitrarily decided that, when one type of cells constituted more than 57.5% of the entire blastocyst or its parts, this type of cell was classified as dominating. When the proportion of fast and slow cells in the blastocyst was in the range of 42.5%–57.5%, this blastocyst was classified as being composed of equal number of both cell types.
|
We observed that most blastocysts (62% [26/42]) analyzed in variant I contained more cells derived from the fast blastomeres (supplemental Table S1 available online at http://www.biolreprod.org). In the group of 10 young blastocysts, all embryos had more fast cells than slow cells (on average, 25 versus 11 cells) (Table 2). However, among 32 expanded blastocysts from variant I, the numbers of blastocysts in which slow cells and fast cells dominated were equal. On average, blastocysts in this group contained 25 fast cells and 26 slow cells. In all young blastocysts from variant I, the border zone was composed mostly of cells derived from the fast blastomere. On average, the border zone was composed of six fast cells and three slow cells (Table 2 and supplemental Table S1 available online at http://www.biolreprod.org).
Most blastocysts (79%) obtained in variant II (6 of 8 young blastocysts and 17 of 21 expanded blastocysts) were composed predominantly of the progeny of fast two-cell blastomeres (supplemental Table S2 available online at http://www.biolreprod.org). On average, young blastocysts contained 20 fast cells and 15 slow cells, and expanded blastocysts contained 30 fast cells and 18 slow cells (Table 2). In the border zones, fast cells dominated in three young blastocysts, slow cells dominated in another three young blastocysts, and equal numbers of cells were derived from both blastomeres in the remaining two young blastocysts (supplemental Table S2 available online at http://www.biolreprod.org). On average, there were four fast cells and four slow cells in the border zones of blastocysts obtained in variant II (Table 2 and supplemental Table S2 available online at http://www.biolreprod.org).
Order of Cleavage of Two-Cell Stage Blastomeres and Distribution of Their Progeny to Embryonic and Abembryonic Parts of Blastocysts
In 45% (19/42) of blastocysts (young and expanded together) from variant I, the embryonic parts contained approximately the same proportions of slow and fast cells (Table 2 and supplemental Tables S1 and S2 available online at http://www.biolreprod.org). The frequencies of blastocysts with embryonic parts in which slow and fast cells dominated were almost equal (29% [12/42] and 26% [11/42], respectively). On average, the embryonic part of the blastocysts in this group contained eight fast cells and seven slow cells (Table 2 and supplemental Table S1 available online at http://www.biolreprod.org).
In variant I blastocysts, slow cells and fast cells equally contributed to the abembryonic parts. In this group of blastocysts, the abembryonic part contained on average 16 fast cells and 16 slow cells (Table 2 and supplemental Table S1 available online at http://www.biolreprod.org).
In 45% (13/29) of variant II blastocysts, the embryonic parts contained equal proportions of slow and fast cells (Table 2 and supplemental Table S2 available online at http://www.biolreprod.org). Blastocysts with embryonic parts composed mostly of slow cells (31% [9/29]) were observed as often as blastocysts with embryonic parts composed mostly of fast cells (24% [7/29]). In this group, the embryonic part of the blastocyst contained on average 9 fast cells and 10 slow cells.
In 66% (19/29) of variant II blastocysts, fast cells dominated the abembryonic parts. In 28% (8/29) of blastocysts, equal proportions of fast and slow cells were observed in the abembryonic parts (Table 2 and supplemental Table S2 available online at http://www.biolreprod.org). Although slow cells dominated in only 7% (2/29) of the abembryonic parts of blastocysts, the proportion of slow cells was never greater than 67.5%. On average, blastocysts in this group contained 18 fast cells and 8 slow cells in their abembryonic part.
DISCUSSION
Cleavages of the mouse embryo are asynchronous. Usually during the second cleavage, one two-cell stage blastomere cleaves earlier than its counterpart, and the three-cell embryo forms. In this study, we investigated whether there is a relationship between the order of division of two-cell blastomeres and the pattern of cell allocation in the mouse blastocyst. Our objective was to verify the recently proposed hypothesis [6] that, in contrast to the previous suggestions [20, 21], blastomeres at the two-cell stage already have a predictable fate. According to this hypothesis, the earlier dividing blastomere in a two-cell embryo has a strong tendency to form the embryonic part of the blastocyst, and the blastomere that cleaves as the second blastomere forms the abembryonic part (reviewed by Zernicka-Goetz [10, 22]). We performed lineage tracing of the descendants of two-cell fast and slow blastomeres and found that cells derived from fast-dividing and slow-dividing blastomeres do not have a specific fate and that their allocation does not delineate the Em-Ab polarity of the blastocyst.
Microinjection of fluorochrome-conjugated dextran is an invasive method of cell labeling, and possible adverse effects have to be considered when interpreting results of the experiments. In a previous study, injection of fluorochrome-conjugated dextran into one-cell embryos slowed down cleavages and resulted in blastocysts with cell numbers lower than those in the controls [11]. In the present study, cells derived from labeled cells cleaved at a slower rate than cells derived from unlabeled blastomeres. This resulted in blastocysts in which unlabeled cells predominated. However, in embryos that were injected into one blastomere at the two-cell stage (variant I), there was no difference in the timing of divisions between labeled and unlabeled blastomeres during the period when division from the two-cell stage to the four-cell stage was taking place. This suggests that divisions between the two-cell stage and the four-cell stage occur at a normal pace in labeled and unlabeled blastomeres and that the deceleration of divisions occurs later in development. However, we cannot exclude the possibility that in variant I (because of microinjection) the order of division of two-cell blastomeres to the four-cell stage was reversed and that the blastomere that would normally (if not manipulated) cleave as the first blastomere cleaved as the second blastomere.
In variant II, we labeled the blastomeres that already had a tendency to cleave later than their sister blastomeres. In this case, the natural order of cleavage of two-cell blastomeres was not perturbed, but subsequent divisions of cells descending from labeled slow blastomeres could have been retarded. This explains why in both experimental variants the blastocysts had more cells derived from unlabeled blastomeres than from labeled blastomeres.
In another study in which two-cell blastomeres were labeled with cell membrane markers, the proportion of cells derived from both blastomeres in blastocysts was close to 1 (1:1.17) [6]. In the present study, the proportion of cells derived from labeled and unlabeled blastomeres was 0.47:1 in blastocysts in variant I and 0.61:1 in variant II (data not shown). Therefore, in our study the microinjection of labeled dextran caused abnormalities in the timing of cleavage.
However, our experiments allowed us to establish that the descendants of both of the first two blastomeres (regardless of whether the labeled blastomere was the slow blastomere or the fast blastomere) populated the embryonic and abembryonic parts of the blastocysts. The results of variant II provided additional insight into the issue of the contributions of two-cell blastomeres to the embryonic and abembryonic parts of the blastocysts. In variant II, the blastocysts were derived from two-cell embryos in which the natural difference of the timing of cleavage between the fast and slow blastomeres was experimentally increased by the injection of the slow blastomere. However, cells derived from the slow blastomere efficiently contributed to the embryonic part of the blastocyst. Despite the fact that in variant II slow two-cell stage blastomeres were slowed down because of microinjection of the dextran solution, their descendants comprised on average 51% of cells in the embryonic part of the blastocyst (Table 2). Therefore, we conclude that the difference in the cell cycle length between the first two blastomeres of the mouse embryo is not a factor that independently directs descendant cells into different pathways of development.
Findings suggest that the polarity of the two-cell embryo forecasts the Em-Ab axis of the blastocyst and that in the two-cell embryo each blastomere already has a distinct fate. A study by Gardner [5]—in which the plane of the first cleavage was marked with droplets of oil injected into zona pellucida or in which two-cell embryos were immobilized in alginate gel—demonstrated that the plane of the first cleavage dictates the polarity of the blastocyst (i.e., the Em-Ab axis of the blastocyst is always orthogonal to the plane of the first cleavage). In another study, fluorescent membrane-soluble markers were used to demonstrate that cells derived from two-cell blastomeres had a strong preference to locate in the embryonic part or in the abembryonic part of a blastocyst [6]. A well-defined clonal border was observed between the progeny of the first two blastomeres; this border reflected the plane of the first cleavage and was located orthogonally to the Em-Ab axis [6]. These observations were independently confirmed when blastomeres were labeled by injecting Cre recombinase to locally activate a reporter transgene, and subsequently the distribution of the descendants of two-cell blastomeres in blastocysts was analyzed [8]. It is also suggested at the two-cell stage that the blastomere that inherited the sperm entry point begins its next cleavage earlier than its counterpart [7]. This has led to the conclusion that the polarity of the blastocyst is already established at fertilization [10]. However, other experiments demonstrate highly regulative capacities of mammalian preimplantation embryos [3, 23–25].
In an attempt to reconcile these apparently contradictory data with the regulative behavior of the embryo, it was recently suggested that the pattern of early development is not predetermined, but rather there is a certain bias toward development of the fast blastomere into the embryonic part of the blastocyst and the slow blastomere into the abembryonic part [22]. However, other studies that used a lipophilic dye to label the plasma membranes of embryos [12, 13], as well as studies on embryos injected with fluorescent markers [11], failed to demonstrate the biased contribution of two-cell blastomeres to the embryonic or abembryonic parts of blastocysts.
The role of the sperm entry point in patterning the early embryo has also been questioned [13]. Alarcon and Marikawa [15] recently showed that when lineages of early-dividing and late-dividing two-cell blastomeres are traced using plasma membrane marker, the order of the second cleavage does not correlate with the Em-Ab polarity of the blastocyst; no relationship was found between the order of the second cleavage of isolated two-cell blastomeres and the tendency of descendant sister half embryos to form the blastocyst cavity. The present results, obtained using a different method of lineage tracing, support the view that the asynchrony of the second cleavage does not reflect the prepatterning of the mouse embryo and confirm the conclusions of Alarcon and Marikawa. Although the method that we used certainly retarded divisions of the labeled blastomere, we believe that our data analysis provides additional insight into the relationship between the order of blastomere division and cell allocation in the blastocyst. Even when asynchrony of cleavage between two-cell blastomeres was increased due to experimental conditions in our study, cells derived from the slow blastomere still were able to populate the embryonic part of the blastocyst as efficiently as cells derived from the fast blastomere.
Our results also contradict a recently proposed model of early prepatterning of the mouse embryo [26] that suggests that the maternal Cdx2 gene product is asymmetrically distributed in the mouse zygote and becomes concentrated in the late-dividing two-cell stage blastomere. In that study during subsequent development, cells derived from the blastomere that contained the transcription factor CDX2 formed the mural and polar trophectoderm of the blastocyst, and cells derived from the other blastomere formed the ICM [26]. Because this was just a single study—and none of the previous studies in which cell lineages of early mouse embryos were examined confirmed such distribution of the descendants of the first two blastomeres [4, 6, 8, 11, 13, 15, 20, 27]—this issue remains open and needs further examination.
An explanation for the various findings about pattern formation in early mouse development remains unknown. Differences in the methods of blastomere labeling [9] or differences in the mouse strains [26] have been suggested. However, these explanations are unlikely because studies using the same method of cell labeling gave different results [11, 26]. Even when the same strain of mice and the same method of cell lineage tracing were used, different investigators obtained contradictory results [6, 12, 15]. The role of early cleavages in determining the organization of the mouse blastocyst remains a subject of debate [22, 28–31], and more investigation is needed to reconcile contradictory findings. In particular, the exact timing and the molecular background of early cell specification events in mammalian development remain enigmatic.
ACKNOWLEDGMENTS
We thank Prof. A.K. Tarkowski and Dr. Maria A. Ciemerych-Litwinienko for critical reading of the manuscript and Dr. M. Kloc for excellent help during its preparation.
FOOTNOTES
1 Correspondence: Marek Maleszewski, Department of Embryology, Institute of Zoology, Warsaw University, ul. Miecznikowa 1, Warsaw 02–096, Poland. FAX: 48 22 55 41 210; maleszewski{at}biol.uw.edu.pl 
2 Current address: Institute of Genetics and Animal Breeding, Polish Academy of Science, Jastrz
biec 05–552, Poland.
Received: 5 May 2006.
First decision: 24 May 2006.
Accepted: 29 June 2006.
REFERENCES
Related articles in Biol Reprod:
This article has been cited by other articles:
|
|
 |
|
  Y. Kurotaki, K. Hatta, K. Nakao, Y.-i. Nabeshima, and T. Fujimori Blastocyst Axis Is Specified Independently of Early Cell Lineage But Aligns with the ZP Shape Science, May 4, 2007; 316(5825): 719 - 723. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
