SEARCH

SEARCH BY CITATION

Keywords:

  • cell movements;
  • metaphase plate orientation;
  • axial differentiation

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Formation of the mammalian primitive streak appears to rely on cell proliferation to a minor extent only, but compensating cell movements have not yet been directly observed. This study analyses individual cell migration and proliferation simultaneously, using multiphoton and differential interference contrast time-lapse microscopy of late pregastrulation rabbit blastocysts. Epiblast cells in the posterior gastrula extension area accumulated medially and displayed complex planar movements including U-turns and a novel type of processional cell movement. In the same area metaphase plates tended to be aligned parallel to the anterior–posterior axis, and statistical analysis showed that rotations of metaphase plates causing preferred orientation were near-complete 8 min before anaphase onset; in some cases, rotations were strikingly rapid, achieving up to 45° per min. The mammalian primitive streak appears to be formed initially with its typically minimal anteroposterior elongation by a combination of oriented cell divisions with dedicated planar cell movements. Developmental Dynamics 240:1905–1916, 2011. © 2011 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Cell proliferation and cell movement are fundamental prerequisites for the early embryonic development of multicellular organisms. During vertebrate gastrulation, which brings about the formation of the body axes and the differentiation of germ layers, they can be considered as two antipodes of cellular activity which start and maintain morphogenesis before the formation of tissues with specialized fate or function. Elegant studies in teleost fish (Montero et al.,2005; Olivier et al.,2010) recently confirmed basic concepts of morphogenesis which had been formulated previously in amphibians (Vogt,1925; Keller,2002) and, most importantly, they identified molecular factors that regulate this process (Ulrich et al.,2003,2005). In comparison, the analogous development in amniote embryos is less well understood, which is partly due to the reduced accessibility of equivalent stages in the egg or intrauterine environment. However, in analogy to the amphibian blastoporus and teleost marginal zone, the primitive streak has a particularly complex function as the first irreversible axial structure in amniotes: It is about to launch into both germ layer formation and vigorous growth acceleration of the embryo. While the dynamics of primitive streak formation in reptiles are still largely unresolved (cf. Gilland and Burke,2004; Coolen et al.,2008), the clear picture known for avian primitive streak morphogenesis (Voiculescu et al.,2007) cannot be extrapolated to the mammalian situation because the starting points for primitive streak formation differ fundamentally: The chick embryo has a pre-existing cell density at the posterior marginal zone, also known as Koller's sickle (Koller,1882; Stern,1990; Callebaut and Van Nueten,1994); accordingly, cell movements associated with epithelial cell intercalation are highly responsible for rearrangement of an existing cell mass (Voiculescu et al.,2007) and hence appear to have a role predominant to that of cell proliferation (Sanders et al.,1993). Contrastingly, in the mammalian embryo the primitive streak forms in an area with a relative lack of cells compared with the rest of the embryo, namely in the so-called posterior gastrula extension (PGE) area (rabbit: Viebahn et al.,2002; pig: Hassoun et al.,2009; mouse: s. Fig. 1A in Downs and Davies,1993; human: s. Fig. 18 in Luckett,1978); as cell proliferation is not a prominent feature in the PGE area, cell migration from the anterior two-thirds of the embryonic disc instead compensates for the lack of cells in the early phase of primitive streak formation, and proliferation sets in only after the primitive streak has formed (Viebahn et al.,2002). Therefore, the net balance between migration and proliferation may tip toward migration in both amniote groups, but the paths taken by individual cells are still likely to differ in a crowded (aves) vs. a rarefied (mammalia) situation.

In addition to cell movement, the orientation of dividing cells can have a defined direction relative to the embryonic axes and may, therefore, have an important function in axial elongation (Hydra: Shimizu et al.,1995) or in determining organ shape (Drosophila wings: Strutt,2005; mouse kidney tubules: Fischer et al.,2006; vertebrate limb buds: Wyngaarden et al.,2010). Indeed, during gastrulation and at the beginning of neurulation in zebrafish, cell divisions in the epiblast are highly aligned (Concha and Adams,1998; Gong et al.,2004) such that orientation of the mitotic spindles is parallel to the long axis. In chick embryos, the mitotic spindles are preferentially orientated parallel to the long axis of the embryo at early streak stages (Wei and Mikawa,2000), during early neurulation (Sausedo et al.,1997) and during notochord extension (Sausedo and Schoenwolf,1993), thus possibly contributing to axial elongation (see however, Bodenstein and Stern,2005). But no reports appear to exist on oriented cell division during the formation of the primitive streak in the mammalian embryo although the marked differences in overall cell numbers and the topography of cell densities between mammals and birds mentioned above make direct observations in mammals a necessity.

While cell movements can be readily observed live in the chick embryo (Wei and Mikawa,2000; Cui et al.,2005; Voiculescu et al.,2007), a similar approach is more difficult in mammals due to their intra-uterine development and, in the case of rodents, due to a complex morphology, e.g., the cup-shaped egg cylinder during gastrulation (Tam and Gad,2004). Cell movements in the mouse embryo are consequently defined for the cells in the anterior visceral endoderm (Perea-Gomez et al.,2001; Srinivas et al.,2004; Kwon et al.,2008; Burtscher and Lickert,2009) and for the intra-embryonic tissues such as the epiblast only after the onset of gastrulation (Nakatsuji et al.,1986; Lawson et al.,1991; Yamanaka et al.,2007; Yen et al.,2009). Late implantation and the flat embryonic disc of the rabbit allowed indirect demonstration of extensive planar cell movements of the epiblast layer by using the live marker DiI injected between epiblast and zona pellucida at the late pregastrulation stage (Viebahn et al.,2002). However, given the negligible quantitative effect of local proliferation in the primitive streak forming area of the rabbit (Viebahn et al.,2002), the question of individual cell behavior before primitive streak formation arises, if only to find mechanisms which regulate this cell behavior at a qualitative level. The current study, therefore, intends to analyze cell movement and cell division at the onset of gastrulation simultaneously, using multiphoton time-lapse imaging of rabbit blastocysts developing in vitro. The results may answer questions as to how epiblast cells move to form the primitive streak in a mammotypical flat embryonic disc and whether proliferation contributes to primitive streak formation by orienting metaphase plates specifically according to the pre-existing anterior–posterior (A-P) axis.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Morphological Development in Culture

Embryos about to be cultured were at stage 2, i.e., at the developmental stage immediately before the formation of the primitive streak; they revealed their A-P polarization in two specific landmarks: (1) the sharp anterior contour of the anterior marginal crescent (AMC) and (2) the reduced cellular density of the PGE (cf. Fig. 1A), where the primitive streak is about to be formed. The oil drop used as constant mark of the posterior pole during culture appeared as a bright round spot under semi-dark field illumination (Fig. 1B). After a recording session and following overnight culture (16 hr), embryonic discs had developed to stage 3 with a typical increased cellular density at the anterior periphery of the AMC and a distinct primitive streak in the posterior half of the embryonic disc (Fig. 1C).

thumbnail image

Figure 1. Dorsal views of the rabbit embryonic discs. A: Darkfield illumination of a living embryonic disc at stage 2 showing the anterior marginal crescent (AMC) and the posterior gastrula extension (PGE); the dotted line marks the anterior border of the PGE area. B: Semi-dark field illumination of an embryonic disc at stage 2 immediately following injection of an oil bead (yellow arrow) between the zona pellucida and the epiblast at the posterior pole. C: Dark field illumination of an embryonic disc developed to stage 3 after overnight culture. D: DAPI-stained epiblast cell nuclei at high magnification. E: DAPI-stained hypoblast cell nuclei at high magnification. Yellow asterisks indicate the center of nuclei. Scale bar = 50 μm in A,C, 32 μm in B, 0.25 μm in D,E.

Download figure to PowerPoint

Epiblast and hypoblast, the two cell layers present at the late preprimitive streak stage, were distinguishable by their nuclear form, size and relative number when focusing on either of both layers using multiphoton microscopy of the DAPI (4′,6-diamidine-2-phenylidole-dihydrochloride) -stained blastocysts: Close to the zona pellucida surface of the embryonic disc, nuclei are oval, smaller and approximately three times as numerous as compared to those found in the layer near the blastocyst cavity, which are characterized by their large and round shapes (cf. Fig. 1D,E). Accordingly, the cells with the small nuclei were assigned to the epiblast, while the larger nuclei were assigned to hypoblast cells. For the purpose of this study, only the cells in the epiblast were considered, whereby individual cells had to be followed through several focal planes (see the Experimental Procedures section) during the period of time-lapse recording.

Cell Movement

Using multiphoton time-lapse microscopy with a time resolution of 5 min and an overall duration of 2 hr, several types of cell movement within the PGE were observed. Observation periods started either approximately 4 or 2 hr after, or immediately following, the injection of the positional oil drop marker and DAPI staining (s. Experimental Procedures). In this way, a 6-hr period of the early primitive streak forming process was accessible for direct observation. Traces of individually marked nuclei revealed overall symmetrical movements with respect to the A-P axis on both sides of the embryonic disc (see Fig. 2A–C and Supp. Movie S1), while the position of the embryo—as judged from the still pictures showing the oil drop marker at the beginning and at the end of the recording session—remained stationary. The trajectories of lateral cells were generally longer than those close to the midline. Thus, the more lateral cells performed so-called “L-turns” (Fig. 2C), i.e., a posterior movement followed by a movement from either side toward the midline. In contrast, many complex posterior–anterior cell movements near the midline were characterized as “U-turns” whose absolute movement is toward the presumed area of the primitive streak. At the end of their U-turns, some of the medial cells disappeared from the plane of focus toward the hypoblast layer.

thumbnail image

Figure 2. Planar epiblast cell movement in the posterior half of a late pre-streak embryonic disc. A–C: Three single frames from the time-lapse series (Supp. Movie S1) taken at intervals of 45 min. Red tracks mark the paths of individual epiblast cells, red dots show the position of a given cell at the time the current frame was taken. The red box in the schematic drawing of the upper left-hand corner indicates the position of the region shown in (A–C), the arrow being the projection of the stage 3 primitive streak position onto the posterior half of the stage 2 embryonic disc. In the upper right-hand corner two types of movements occurring in the posterior half of the primitive streak are shown at high magnification (U, U-turns; L, L-turns). The position of these paths is marked with white boxes in (C). D–F: processional cell movement (PCM) in the area of the prospective primitive streak taken from the same frames as shown in (A–C); the position of the selected area is indicated by a yellow box in (C). Three adjacent cells (marked “a” with a red dot and “b” and “c” both with yellow dots) are connected by lines; “b” and “c” are connected by a dotted line to highlight how cell “a” performs a procession-like movement past its two immediate neighbors. G: Vector diagram superimposed onto the frame shown in (A) and depicting the position, frequency, direction and length of PCM in the posterior gastrula extension (PGE). The points of the arrows mark the end position of processional cells in this recording period. Scale bar = 50 μm in A–C,G, 5 μm in D–F.

Download figure to PowerPoint

Detailed tracing of individual neighboring cells near to, and in, the area of the presumptive primitive streak regularly revealed a new type of cell movement: In a group of three adjacent cells one cell (labeled “a” in Fig. 2D–F) distinctly changed its position posteromedially during the period of observation (90 min in this case), while its two immediate neighbors (“b” and “c”) moved apart and converged again after cell “a” had passed between them. The movement of cell “a” across the virtual line connecting cells “b” and “c” goes further than during conventional mediolateral intercalation: Here, three cells may start to change their relative positions in a similar way but passing between cells has not been described so far (cf. Voiculescu et al.,2007). As the passing of cell “a” is reminiscent of a procession-like step on the dance floor (a metaphor which had been used before with the figure of “polonaise” and “double-whirl” movements, cf. Gräper,1929, and Wetzel,1929) this movement is described here as “processional cell movement” (PCM). This new type of movement occurred mainly near the center of the PGE, i.e., in the area of the emerging primitive streak, and was less frequently observed in lateral parts of the PGE (s. Fig. 2G and the distribution of PCM cases in Supp. Movie S2). Net movement of processional cells was posteromedial when observed in the lateral part of the PGE and predominantly medial when observed near the center of the PGE (cf. orientation of vectors in Fig. 2G).

As shapes and boundaries of individual epiblast cells during PCM were clearly visible in differential interfering contrast (DIC) movies, too (s. Fig. 3A–F and Supp. Movie S3), the complete sequence of PCM could be observed to consist of the following steps: (1) neighboring cells “b” and “c” loosen their broad and close contact (cf. bracket in Fig. 3B) completely, (2) let the processional cell “a” pass between them, and (3) re-establish close contact, again. Of interest, the necessary protrusion of a leading edge of the processional cell “a”, including the formation of a new contact zone (cf. bracket in Fig. 3C) with a fourth cell lying ahead on the processional path, was accomplished within less than 5 min (cf. Fig. 3B,C).

thumbnail image

Figure 3. Changing cell shapes during processional cell movement (PCM) as seen with differential differential interfering contrast (DIC) optics in the area of presumptive primitive streak (A–F). Six frames taken at different intervals highlight how neighboring cells “b” and “c” are intimately connected before (cf. A and bracket in B) and after (E) processional cell “a” passes or has passed between them. The short interval between the frames shown in (B) and (C) indicates the relative speed at which cell “a” protrudes its leading edge past cells “b” and “c” to make contact (bracket in C) with its “new” neighbor on the other side of cells “b” and “c.” Schematic drawing in (A) indicates position of frames within the embryonic disc. Scale bar = 10 μm.

Download figure to PowerPoint

In a quantitative analysis, all cases of neighbor cell changes observed between any three cells of which the central cell “a” showed posteromedial or medial translocation were recorded by “dynamic triangulation” (see Supp. Movies S2, S3, and S4) within every movie and subsequently divided into four classes as follows (Supp. Fig. S1): (1) conventional intercalation, whereby the triangular connection between the centers of neighboring cells changes from a more-or-less even shape to an elongated shape and junctions between these cells change directionally as described previously (Bertet et al.,2004; Zallen and Blankenship,2008), (2) “grade 1” PCM, whereby the processional cell passes the connecting line between its neighbors but the latter two cells do not (yet) come to lie together again (here, the original even-shaped triangle “collapses” into a single line and then changes into an inverted elongated shape), (3) “grade 2” PCM, which corresponds to complete PCM and is defined by the emergence of a completely inverted even-shaped triangle at the end of neighbor change movement, and (4) “potential end phase” of PCM, which is defined by the change from an elongated triangle to an even shaped triangle (Supp. Fig. S1 is available online). The total number of cases within these classes was compared by calculating different ratios (R1, R2 and R3) between the numbers of occurrence of conventional intercalation and that of different PCM subclasses (Suppl. Table S1). Ratios indicated that PCM was two to three times as frequent as conventional cell intercalation (cf. R2 and R3 in Supp. Table S1). For instance, the ratio R2 between the number of occurrences of conventional intercalation to that of “grade 1” and “grade 2” PCM was counted across all embryos to be 18:37, which corresponds to a P value of P = 0.0145.

Orientation of the Late Metaphase Plates

In time-lapse movies of DAPI-stained blastocysts, metaphase plates were visible as small lines in the equatorial plane of cells about to enter anaphase. As the angle of cell division became rapidly obscured soon after anaphase, the orientations of the metaphase plates were determined in the last image taken before the anaphase, i.e., 5 min at most before anaphase onset which was considered to coincide with the appearance of two parallel lines in the following image (cf. Fig. 4E,F). The angles between these late metaphase plates and the A-P axis, which was defined by the position of the AMC, were measured in 351 cells from six embryos and plotted in a rose diagram (Fig. 5A); most of these angles lay in a range around zero (between −45° and 45°) with a maximum of cases near 25°. The angular variance was 0.8 with P value = 0.0036 compared with a uniform distribution; the average angle enclosed between late metaphase plates and the A-P axis was 8.8° (Supp. Table S2). Although the angles scattered broadly, statistical analysis revealed that an orientation in a range of angles parallel to the A-P axis was the preferred orientation.

thumbnail image

Figure 4. Late metaphase plate angle definition. A–F: Single frames from a DAPI (4′,6-diamidine-2-phenylidole-dihydrochloride) time-lapse movie showing the change in orientation of a metaphase plate in the posterior gastrula extension (PGE) area in 5-min time intervals 25 min before anaphase. The outlines of the DAPI labeled DNA package (nuclei or metaphase chromosomes) of the dividing cell are marked by a yellow line and the angle of the metaphase plate with anterior–posterior (A-P) axis (solid red line) is indicated by a dotted red line. The red dot in the inset in (A) indicates the position of the dividing cell in the embryonic disc. Scale bar = 20 μm.

Download figure to PowerPoint

thumbnail image

Figure 5. Statistical analysis of late metaphase plate orientation. A: Rose diagram of 351 metaphase plate angles (taken from six embryos) grouped in intervals of 10°. The metaphase plates are taken from all areas of the embryonic disc. The anterior–posterior (A-P) axis is set to correspond with 0°, each angle α is shown together with its opposite angle (α + 180°) to produce a 360° image mirrored at the line connecting the +90° and −90° values. B: Scatter plot of late metaphase plate angles against their position along the A-P axis. Positive values on the x-axis represent metaphase plate positions in the posterior gastrula extension (PGE) area while negative values correspond to positions in the anterior part of the embryonic disc (non-PGE area). The y-axis denotes the angles of the metaphase plate. Colors represent the smoothened density of the data points, i.e., darker colors correspond to higher density. C: Rose diagram of 229 metaphase plate angles taken from the non-PGE area. D: Rose diagram of 122 metaphase plate angles taken from PGE area.

Download figure to PowerPoint

Alignment of late metaphase plate positions from six embryos revealed that there was correspondence between the angles of the metaphase plates and their position along the A-P axis (Fig. 5B). Indeed, visual inspection also showed that the angles were mostly smaller in the PGE area than those in the anterior part of the embryonic disc, referred to here as “non-PGE area.” In fact, the orientation preference of the metaphase plates in the PGE area was found to be statistically significant while the metaphase plates from the non-PGE area were not shown to have a preferred orientation (cf. Fig. 5C,D). The angular variance of the former group differed significantly from that of a uniform distribution while that of the latter did not (Supp. Table S2). Moreover, the plot of the positions of the metaphase plates along the mediolateral axis against their angles with the A-P axis showed that metaphase plates with small angles accumulated near to the center of the embryonic disc (Fig. 6).

thumbnail image

Figure 6. Scatter plot of late metaphase plate angles against their position along the mediolateral axis. The plot was constructed analogously to the one of Figure 5B using data from whole embryonic discs and six embryos. The position along the mediolateral axis was mapped to the x-axis. The x-values represent the distance of the position of metaphase plates from the midline. The y-axis and the color coding are the same as those in Figure 5B.

Download figure to PowerPoint

Metaphase Plate Rotation

During observation periods in which a particular mitosis could be continuously recorded using multiphoton microscopy of DAPI-stained blastocysts, the orientation of individual metaphase plates varied markedly and the movies showed that this was due to isolated rotation of metaphase plates rather than to curved migration of cells in metaphase (cf. Fig. 4C,E). Overt rotations occurred most frequently during the early metaphase whereas during the late metaphase, i.e., approximately 10 min before anaphase, no major rotation movements were detected. Instead, many rotations showed switches between clockwise (cf. Fig. 4C,D) and counter-clockwise rotation (cf. Fig. 4E,F), similar to an oscillating movement.

Because the low time resolution that had to be used for the DAPI movies was too coarse to study small rotatory movements or the exact angular speed, time-lapse DIC microscopy (Nomarski contrast) was performed at intervals of 40 sec. Here, epiblast and hypoblast were distinguishable by the cell size and the intercellular space in both layers, e.g., the cells of the epiblast are smaller and the space between adjacent epiblast cells is narrower than in the hypoblast (not shown). In these DIC movies, metaphase plates appeared as a bulging chromosome mass forming a broad bar that changed orientation during metaphase (s. Fig. 7A–L and Supp. Movie S5). For most metaphase plates observed in these DIC movies, the rotation was a relatively slow process that took place throughout early metaphase (30–10 min before anaphase). However, the metaphase plates of a few mitotic cells turned rapidly within a 120-sec period during late metaphase (cf. Fig. 7D,F), a movement which also included small oscillating movements described above (for clockwise rotation cf. Fig. 7A and B, B and C, F and G; for counter-clockwise rotation cf. Fig. 7C and D, D and E, I and J); the total resulting angle of these fast rotations was regularly more than 90° (Fig. 7K).

thumbnail image

Figure 7. Rotation of metaphase plates. A–L: Single frames from a differential interfering contrast (DIC) time-lapse movie (Supp. Movie S5) showing a dividing cell and the rapid rotation of its metaphase plate at 40-sec intervals; the position of this cell in the embryonic disc is indicated by the red dot in the inset in (A). A 92° rotation (arrow in K) of the metaphase plate can be seen when comparing the position of the metaphase plate extremities (yellow and blue dots) at the start of this recording (A) and immediately prior (K) to anaphase (L). Scale bar = 20 μm. M: Average absolute rotation within a 40-sec period (y-axis) at each time point before onset of anaphase (x-axis). N: Rao P values of 10 mitotic cells taken from differential interfering contrast (DIC) time-lapse movies of two embryos (y-axis) at each time point before onset of anaphase (x-axis). The P values were corrected for False Discovery Rate using the Benjamini-Hochberg procedure.

Download figure to PowerPoint

For a chronometrical analysis of all metaphase plate rotations observed in two DIC recordings, an average absolute rotation value within single 40-sec time periods was calculated (Fig. 7M). In this way, two time periods were distinguishable: An early one, 14 to approximately 8 min before the onset of anaphase, showed higher average rotation values, whereas a second subsequent one, less than 8 min before anaphase, exhibited distinctly lower average rotations of the metaphase plates. Separate analysis of cells from the PGE and non-PGE areas revealed no differences in the rotation values of metaphase plates between these embryonic compartments (not shown).

The high time resolution of the DIC recordings revealed the exact timing of metaphase plate rotations with reference to anaphase onset. As a result, all cells from the PGE area whose metaphase plates were visible for a minimal 11-min period could be tested statistically using the Rao test against the null hypothesis of a uniform distribution. Ten cells from two embryos met these criteria and for these the distribution of metaphase plate angles was analyzed for each time-point recorded before anaphase. The plot of time before anaphase onset against P value of the Rao test resulted in one set of significant and another of nonsignificant P values (Fig. 7N): 8 min before anaphase, high significance levels for the orientation were attained, i.e., the time-point from which on the orientation preference was significant lay at least 440 sec before anaphase onset. There was a clear-cut phase of decreasing P values at around 8 min before anaphase and the last 12 time-points showed a P value of almost zero, corresponding to a preferred orientation of the metaphase plates relative to the A-P axis.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

This time-lapse videomicroscopy analysis of whole rabbit embryos cultured in vitro provided direct evidence of cellular movements during the formation of the primitive streak, the overt longitudinal body axis, in the PGE area of the mammalian embryo: The emergence of the primitive streak is characterized by a combination of (1) complex individual cell movements including a new type of cell movement referred to here as processional cell movement (PCM), and (2) mitotic spindle rotations which ensure that the metaphase plates are aligned preferably parallel to the A-P axis.

Multiphoton vs. DIC Time-Lapse Microscopy

The combination of multiphoton fluorescence microscopy with conventional DIC microscopy was instrumental for identifying the characteristics of the spatial cell behavior in the first 6 hr of primitive streak formation. Multiphoton microscopy of DAPI-stained blastocysts showed that the cells of the epiblast can be readily distinguished from those of the hypoblast by the different shape, size, and relative number of the nuclei, characteristics known from high-resolution histological studies of the same and other mammalian species including the mouse (cf. Viebahn,1999; Eakin and Behringer,2004; Tam and Gad,2004; Hassoun et al.,2009). A particular benefit of the DAPI-staining method is that almost all nuclei are visualized, which makes it possible to observe the metaphase plates throughout metaphase, whereas the DIC method enables one to observe the metaphase plates primarily at the late metaphase. However, the latter method allowed for use of shorter time intervals between consecutive images than the potentially cell damaging combination of DAPI-staining and multiphoton microscopy and was, therefore, instrumental for (1) verifying results obtained from the DAPI experiments, (2) defining cell membrane contact during PCM, and (3) defining the details and (high) speed of the mitotic spindle rotation and the exact timing of this rotation before the onset of anaphase.

Cell Movement

The two main paths taken by epiblast cells diverging from an apparently simple straight movement, the U- and L-turns (s. Fig. 2A–C), largely confirm the hypothesis of “polonaise” and “double whirl” movements in the PGE area toward the future primitive streak (Viebahn et al.,2002), movements which were originally described for the chick (Gräper,1929; Wetzel,1929). Cellular U-turns toward the posterior pole, however, were not observed in the present study possibly for two reasons: (1) Only the 6 initial hr of early primitive streak formation (stage 2) were captured in the movies of the present report and (2) the primitive streak elongates toward the anterior pole first before inverted U-turns are needed for lengthening of the primitive streak toward the posterior pole (stage 3). This posterior lengthening of the primitive streak is particularly apparent during the early stage 3 (as observed in some DIC movies, which allowed a longer exposition time than the DAPI movies; V.H., unpublished). The DiI tracings from which inverted U-turns were deduced (Viebahn et al.,2002), on the other hand, covered the complete period between stage 2 to 3 and recorded the cumulative effect of cell migration along cellular paths at the end the culture period. Lateral contraction of epiblast cells (which are rather flat in the PGE area: Viebahn et al.,1995) may also be a plausible mechanism to form the high and pseudostratified columnar epithelium near and in the primitive streak. However, structural signs of lateral contraction, i.e., a preferential narrowing of apical cell shapes along the transverse axis, are not seen in the DIC movies (cf. Supp. Movies S3, S4, and S5).

PCM, as the new type of planar cell movement found by time-lapse imaging of individual cells in the present study, is a movement based on the changes in the relative position of adjacent cells by which the “processional cell” crosses the boundary between its neighbors (see Supp. Movies S2, S3, S4 and Supp. Fig. S1B,C). In contrast, mediolateral cell intercalation observed in the epiblast of the pregastrulation chick embryo (Voiculescu et al.,2007) was based on the changes in the relative positions of non-neighboring cells in different regions, thus showing effects caused by more general cell movements. Therefore, the behavior of neighboring cells immediately before gastrulation in the rabbit requires the new definition given above and could be suitably described as a sub- (or micro-) movement of global (or macro) “polonaise” movements.

In the rabbit embryo, the major net movement leading to primitive streak formation appears not to be simply a mass “polonaise” and “double whirl” movement or conventional mediolateral cell intercalation, but a combination of mainly two variations of trajectory paths (U- and L-turns) and by PCM, the new type of cell movement (cf. Fig. 2). Both PCM and conventional intercalation are preferentially localized in the presumptive primitive streak area and therefore, they may functionally correlate with cell accumulation. However, PCM seems to be more frequent compared with conventional intercalation as ratios amount up to values of 2.9, depending on the stringency with which they are calculated (cf. Supp. Table S1). A conservative calculation (R2 = 2.1) takes only cases of proven PCM into account, i.e., “grade 1” and “grade 2” PCM, in which cell “a” has moved beyond the connecting line between its immediate neighbors “b” and “c”. This suggests already that neither conventional intercalation nor oriented epithelial contraction (s. above) but PCM may be a main mechanism, in addition to mass “polonaise” and “double whirl” macro-movements, for cell accumulation in the presumptive primitive streak forming area. In addition, the re-approaching of neighbor cells “b” and “c” during the end phase of PCM may help avoiding any elongation in the A-P plane at this early stage of primitive streak formation. Moreover, the extensive movement of a cell passing between its neighbors (“grade 2” PCM) may be a sign of the increased distance that individual cells have to move from the highly proliferative belt in the non-PGE area (cf. Fig. 7C in Viebahn et al.,2002) to the center of the “rarefied” PGE area. In the chick, in contrast, with its high cellular density in Koller's sickle and with its apparent lack of proliferation in the primitive streak forming area (Sanders et al.,1993), the additive effects of classical convergent extension movements may be necessary for (1) increasing the cellular density in the primitive streak and (2) for the transformation of the circular posterior pole of the embryonic disc into an elongated pear-shape (Voiculescu et al.,2007).

A functional prerequisite for PCM is planar polarization of the cells involved which may be conveyed by the signals of the PCP pathway (Zallen,2007). Initially, attempts at determining the distribution of Vangl2 or Prickle1, two members of the PCP pathway expressed in or near the primitive streak of chick (cf. Zhang and Levin,2009) and mouse (Suriben,2009), in the rabbit embryonic disc have so far been unsuccessful (V.H., unpublished observations). However, global molecular factors controlling PCP and cell migration such as Wnt5a or Wnt11 (Gong et al.,2004; Ulrich et al.,2005) and Dkk1, a regulatory factor acting from a distance (Caneparo et al.,2007) and suitably expressed in the anterior marginal crescent (Idkowiak et al.,2004), can now be tested in a mammalian system on the basis of individual cell behavior described in the present report. In a different context, the fact that single cells pass their neighboring cells correlates with cells individually labeled for hyaluronan (V.H., unpublished observations), which has been shown to be instrumental in cell migration for tail regeneration (Contreras et al.,2009) and may be a further key factor in determining when cells move as singletons and when as a group (Arboleda-Estudillo et al.,2010).

Analysis of Metaphase Plate Angles

In comparison with previous studies measuring the orientation of mitotic spindles rather than the orientation of metaphase plates (Wyngaarden et al.,2010), the DAPI-staining approach taken in the present study allows for high precision measurements because the metaphase plates can be perceived as lines, rather than deducing orientation from lines through estimated cell centers. Taking into account the circular nature of the data, circular statistical representations were used. In principle, angles could also be mapped on to a linear scale and subsequently analyzed by linear statistics. However, a linear scale could divide two neighboring angle values to opposite ends of an interval; the variance would then depend on the cutting point which would lead to undesirable ambiguity; this ambiguity disappears through use of circular representation of statistical values.

The resulting rose diagrams do not always exhibit clear-cut peaks (Fig. 5); hence, the necessity to apply suitable statistics designed to yield precise information on preferred angles. However, the fact that statistical analyses of the orientation performed in this study show a less pronounced peak than those of other studies (e.g., Concha and Adams,1998) could possibly be explained by the more complex cell movements in the rabbit (as compared to other model organisms) or by the relative uncertainty in determining the exact orientation of the A-P axis before emergence of the primitive streak: Obviously, this is particularly important in the mammalian embryo where the A-P axis as the reference for angle measurement before the appearance of the primitive streak has to be deduced from the estimated position of median AMC or PGE centers, i.e., the A-P axis cannot always be determined with the same degree of confidence as the axis of the primitive streak. This slight uncertainty in the position of the A-P axis may account for both the rather broad scattering of values (Fig. 5D) and the deviation (by approximately 8°) of the preferred metaphase plate orientation from a presumed “ideal” orientation exactly parallel to the A-P axis. However, these results may be taken at least as a first indication of oriented cell division before gastrulation in the mammalian embryo.

The first scatter plot (Fig. 5B) shows a clear pattern reflected by the data analysis, namely that the angles of the metaphase plates in the PGE are smaller (preferred parallel orientation) than those in the non-PGE area; this is also reflected by the markedly reduced number of large angles (i.e., white areas) in the right-hand corners of the plot (Fig. 5B). In contrast, the second plot (Fig. 6) is more difficult to interpret. One of the reasons for this lies in the difficulty of ascertaining the location of the cells which are to form the primitive streak in the subsequent phase of development. Choosing a conservative estimate for the width of their location leads to the volume of data being too small for statistical analysis. However, the visual impression is that the small angles accumulate around the center of the plot, which would indicate that small angles tend to occur near the A-P axis.

Oriented Cell Division

In chick and zebrafish, the preferred orientation of the metaphase plates is perpendicular to the A-P axis (chick: Wei and Mikawa,2000; zebrafish: Concha and Adams,1998) and later, during neurulation, the orientation changes to a parallel direction (Concha and Adams,1998). However, the orientation of cell divisions in the chick primitive streak forming area (Wei and Mikawa,2000) appears to be in contrast with the present report; but its functional significance for primitive streak formation had earlier been questioned on the basis of a mathematical model (Bodenstein and Stern,2005). Differences in spindle orientation between gastrulating chick and rabbit embryos may come as no surprise because of the marked disparity in overall size, cell numbers and the topography between the two species at this stage. The preferred orientation of the metaphase plates approximately parallel to the A-P axis, as observed in this study, seems to be advantageous for achieving a high cell density in the region of the presumptive primitive streak, similar to the effect of PCM in this area: Daughter cells divide along the mediolateral axis in an area in which neither A-P elongation (because of the early stage) nor mediolateral widening of the embryonic disc was detected, i.e., the dividing cells accumulate in the area of the primitive streak without deformation (e.g., widening) of the embryonic disc. This is supported by the mechanical effect which the lateral embryonic disc borders may have on tissue flow as cells arriving at these borders through oriented cell division have to move toward the midline because the borders limit lateral migration. In addition, movements of lateral cells (s. above L-turns) toward the presumptive primitive streak coincide with the movements of lateral daughter cells to increase this net medial cellular movement.

Metaphase Plate Rotation

Planar rotation before anaphase onset has previously been described as a mechanism for achieving defined orientation; in fact, misorientation of the mitotic spindles in normal rat kidney cells is corrected by direct rotation of the spindles (O'Connell and Wang,2000). More closely related to axis formation, the rotation of mitotic spindles establishes a preferred orientation of cell divisions in the closing Xenopus neural tube: metaphase plates rotate rapidly approximately 1.5 min before anaphase with an angular speed of approximately 60° per min (Kieserman and Wallingford,2009). Rapid rotations can also be observed in the pregastrulation embryonic disc of the rabbit (Fig. 7A–L) but they occurred within a larger time window (an 8-min period) before anaphase and were somewhat slower (up to 45° per min). As most metaphase plates were already oriented parallel to the A-P axis before the 8-min time window, these late and rapid rotations may be seen as “corrective” movements. However, the functional significance of these sudden rotations and of the slight oscillations (s. Supp. Movie S5), for that matter, remains to be determined by future studies which would also analyze the fate of cells that arise from rapidly rotating metaphase plates.

Conclusions

The live recordings of this study suggest that both oriented cell division and a dedicated type of cell movement (here referred to as processional cell movement or PCM) act in concert to form the early mammalian primitive streak by cell accumulation with minimal anteroposterior elongation. However, analysis of planar cell behavior still needs to be extended to the whole embryonic disc and throughout the entire period of primitive streak formation to record a higher proportion of complete PCM and to take specific cell behavior (e.g., tandem movements, V.H., unpublished observations) prior and following PCM into account. Similar recordings may also be carried out in other mammalian model organisms in which axial differentiation is apparent in a planar embryonic disc before primitive streak formation, such as in the tammar wallaby (Hickford et al.,2008). This may then also reveal a possible functional correlation between PCM and oriented cell division during primitive streak formation. Determining the fate of oriented cell divisions and of processional cells and the molecular mechanisms causing their accompanying cellular deformation (cf. Lecuit and Lenne,2007; Krieg et al.,2008; Mammoto and Ingber,2010) may lead an informed assessment as to whether PCM or oriented cell division is the major factor for creating the cell density of the mammalian primitive streak. However, this study already suggests that analyzing cell proliferation and cell movement simultaneously during mammalian gastrulation may establish new regulatory mechanisms, or may indicate how molecular factors act at the cellular level to control axial differentiation in the mammalian embryo (cf. Arnold and Robertson,2009). In addition, questions can now be asked as to where and when axial information is generated in species with a mammotypical flat embryonic disc and how these factors control complex individual cell behavior leading to the overt, irreversible axial differentiation during gastrulation.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Rabbit Blastocysts

Naturally mated New Zealand White rabbits (Lammers, Euskirchen, Germany) were superovulated by single injections of pregnant mare serum gonadotropin (100 IE intramuscularly; Intergonan, Intervet, Unterschleiβheim, Germany) and human choriogonadotropin (180 IE intramuscularly; Predalon, Organon, Oberschleiβheim, Germany) 72 hr in advance and immediately before mating, respectively. Animals were killed by means of administration of an overdose (320 mg) of Narcoren (Merial, Hallbergmoos, Germany) intravenously. At 6.2 days postcoitum (dpc) blastocysts were flushed from uteri using warm (37°C) phosphate buffered saline (PBS) and then washed in warm PBS twice to remove blood and tissue residue. Blastocysts were then transferred to warm HAM's F10 culture medium (Biochrom AG, Berlin, Germany), supplemented with 20% fetal calf serum, 50 IU penicillin and 50 μg/ml streptomycin (both Biochrom, Berlin) and kept in an incubator at 37°C under 5% CO2 until used for time-lapsing microscopy (see below). After recording, blastocysts were cultivated at the same conditions up to 18 hr to reach the next stage of development (stage 3) and only those that had developed to the stage of gastrulation were taken for this study. Embryos were staged before, and upon completion of, time-lapse recording as well as at the end of the culture period using dark field optics (Leica MZ 16 Stereomicroscope, Leica Microsystems, Wetzlar, Germany). Dark field photographs were taken using a SPOT Insight wide-field 4 MP CCD color digital camera (Visitron Systems GmbH, Puchheim, Germany).

Time-Lapse Recording Using Multiphoton Microscopy

Intact blastocysts (6.2 dpc) were injected with the lipophilic carbocyanine DiI dissolved in germ oil using the CellTram vario system (Eppendorf, Hamburg, Germany). The microinjection of an oil bead between the zona pellucida and the embryonic disc was used as a marker to find the optimal orientation of the blastocyst on the stage of the confocal laser microscope and to verify that the embryo does not move as a whole during a recording session. Microinjection was carried out using a borosilicate glass capillary micropipette (1.0 mm O.D. × 0.78 mm ID, HARVARD apparatus) pulled in a Flaming/Brown Micropipette Puller (Model P-97, Sutter Instrument Co., Science Products GmbH, Hofheim, Germany). The blastocysts with the oil mark were photographed before and after time-lapse recording with a semi-dark field illumination (Stemi SV 11, Zeiss, Germany). After injection, blastocysts were stained with the vital DNA dye DAPI (Sigma, Muenchen, Germany) at a concentration of 1μg/ml in equilibrated Ham's F10 medium for 1 hr at 37°C and 5% CO2 (Reupke et al.,2009). Injected and stained blastocysts were positioned in a glass bottom dish in such a way that the embryonic disc faced the glass bottom and the region of interest of the disc lay in the center of the microscope lens. To avoid displacement, the blastocyst was placed on a metal ring with a diameter slightly smaller than the diameter of the blastocyst and the dish was filled with Ham's F10 medium to a level just below the abembryonic pole of the blastocyst. This was carried out immediately before the start of recording in a chamber mounted onto an inverse True Confocal Scanner Leica TCS SP2 microscope (Leica Microsystems, Wetzlar, Germany). Using the multiphoton effect of a tunable Ti:Sapphire Chameleon laser (Coherent, Dieburg, Germany) controlled by an electro-optic modulator (LINOS Photonics GmbH & Co. KG, Göttingen, Germany), stained blastocysts were visualized. One image as the average of three single scans was recorded every 5 min. Recordings were limited to a maximum of 2 hr because blastocyst expansion lead to tissue movement in the vertical plane and this gradually moved the embryonic disc out of the working range of lens during the time of the time-lapse recording; to follow individual cells while they were moved co-axially by this general blastocyst expansion, pictures were taken at every time point in four to six optical focal planes at intervals of 10 μm, depending on the size, thickness and co-axial position of the embryonic disc. To assure the position and overall morphology of the embryonic disc at the beginning and at the end of each time-lapse recording single transmission contrast images were taken using an Ar-laser. Tracking of cell movement was carried out in eight embryos using a ×20 lens and, therefore, in an area measuring 350 × 250 μm, which corresponds roughly to the size of the PGE area.

Time-Lapse Recording Using Differential Interference Contrast (DIC)

Images were acquired at an interval of 40 sec using an Axiovert 200M microscope equipped with an Axiocam MRm camera (both Zeiss, Jena, Germany) by means of Nomarski contrast and the AxioVison software. If required, image acquisition was briefly interrupted to assure optimal positioning of the area of interest in the depth of the focal plane; this lead to image sequences of variable length. Analysis with the DIC optics was carried out in 10 embryos using a ×40 lens with an observation of segment of 62 × 47 μm corresponding to approximately one quarter of the PGE area.

Image Analysis

Every single stack of images was projected onto a single plane using Leica Confocal Software to obtain a two-dimensional (2D) projection image from the 3D image stack along the orthogonal axis. Then, the projection images were assembled using ImageJ software (Wayne Rasband, National Institutes of Health, USA, http://rsbweb.nih.gov/ij/) and a movie was made from all projected images (Abramoff et al.,2004). Individual cell nuclei were tracked using plug-in MTrackJ. For each mitotic cell of the DAPI time-lapse movies, the orientation of the metaphase plates was marked with a line in the last image before the anaphase. In the DIC movies the orientation of the metaphase plates was indicated at every time point of the recording using ImageJ. The angles between the metaphase plate and the A-P axis were determined in such a way that 0° resp. 180° designates parallel lines to the A-P axis. Positive values describe a counter-clockwise deviation from the A-P axis. Because the lines determining the orientation of the metaphase plates have no distinguishable ends, angles differing by 180° were considered to be identical.

Statistical Analysis

For the circular variance as well as for the assessment of its significance level against uniform distribution, the Rao test for homogeneity of angular data (Rao,1967) was applied. Computation was done using the R port of the package CircStats (Berens,2009). For this purpose, all angles were multiplied by two to obtain data in the full circle range [−180°, 180°]. Circular variance and mean value, also called dispersion and polar vector, were calculated from these data. The mean value was subsequently multiplied by 0.5. The homogeneity tests were cross-checked by Monte-Carlo simulation. To obtain an intuitive picture of the order of magnitude of variation among angles, data were mapped to the interval [−90°, 90°] and the standard deviation of these linearized data was computed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

The excellent technical assistance of Kirsten Falk-Stietenroth and Irmgard Weiβ is gratefully acknowledged. We also thank Hans-Georg Sydow for his unfailing help with image processing.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information
  • Abramoff MD, Magelhaes PJ, Ram SJ. 2004. Image processing with ImageJ. Biophoton Int 11: 3642.
  • Arboleda-Estudillo Y, Krieg M, Stuhmer J, Licata NA, Muller DJ, Heisenberg CP. 2010. Movement directionality in collective migration of germ layer progenitors. Curr Biol 20: 161169.
  • Arnold SJ, Robertson EJ. 2009. Making a commitment: cell lineage allocation and axis patterning in the early mouse embryo. Nat Rev Mol Cell Biol 10: 91103.
  • Berens P. 2009. CircStat: a MATLAB toolbox for circular statistics. J Stat Softw. 31: 10.
  • Bertet C, Sulak L, Lecuit T. 2004. Myosin-dependent junction remodelling controls planar cell intercalation and axis elongation. Nature 429: 667671.
  • Bodenstein L, Stern CD. 2005. Formation of the chick primitive streak as studied in computer simulations. J Theor Biol 233: 253269.
  • Burtscher I, Lickert H. 2009. Foxa2 regulates polarity and epithelialization in the endoderm germ layer of the mouse embryo. Development 136: 10291038.
  • Callebaut M, Van Nueten E. 1994. Rauber's (Koller's) sickle: the early gastrulation organizer of the avian blastoderm. Eur J Morphol 32: 3548.
  • Caneparo L, Huang YL, Staudt N, Tada M, Ahrendt R, Kazanskaya O, Niehrs C, Houart C. 2007. Dickkopf-1 regulates gastrulation movements by coordinated modulation of Wnt/beta catenin and Wnt/PCP activities, through interaction with the Dally-like homolog Knypek. Genes Dev 21: 465480.
  • Concha ML, Adams RJ. 1998. Oriented cell divisions and cellular morphogenesis in the zebrafish gastrula and neurula: a time-lapse analysis. Development 125: 983994.
  • Contreras EG, Gaete M, Sanchez N, Carrasco H, Larrain J. 2009. Early requirement of Hyaluronan for tail regeneration in Xenopus tadpoles. Development 136: 29872996.
  • Coolen M, Nicolle D, Plouhinec JL, Gombault A, Sauka-Spengler T, Menuet A, Pieau C, Mazan S. 2008. Molecular characterization of the gastrula in the turtle Emys orbicularis: an evolutionary perspective on gastrulation. PLoS One 3: e2676.
  • Cui C, Yang X, Chuai M, Glazier JA, Weijer CJ. 2005. Analysis of tissue flow patterns during primitive streak formation in the chick embryo. Dev Biol 284: 3747.
  • Downs KM, Davies T. 1993. Staging of gastrulating mouse embryos by morphological landmarks in the dissecting microscope. Development 118: 12551266.
  • Eakin GS, Behringer RR. 2004. Diversity of germ layer and axis formation among mammals. Semin Cell Dev Biol 15: 619629.
  • Fischer E, Legue E, Doyen A, Nato F, Nicolas JF, Torres V, Yaniv M, Pontoglio M. 2006. Defective planar cell polarity in polycystic kidney disease. Nat Genet 38: 2123.
  • Gilland EH, Burke AC. 2004. Gastrulation in reptiles. In: SternC, editor. Gastrulation: from cells to embryo. New York: Cold Spring Harbor. p 205217.
  • Gong Y, Mo C, Fraser SE. 2004. Planar cell polarity signalling controls cell division orientation during zebrafish gastrulation. Nature 430: 689693.
  • Gräper L. 1929. Die Primitiventwicklung des Hühnchens nach stereokinematographischen Untersuchungen, kontrolliert durch vitale Farbmarkierung und verglichen mit der Entwicklung anderer Wirbeltiere. Arch Entwickl Mech Org : 382429.
  • Hassoun R, Schwartz P, Feistel K, Blum M, Viebahn C. 2009. Axial differentiation and early gastrulation stages of the pig embryo. Differentiation 78: 301311.
  • Hickford D, Shaw G, Renfree MB. 2008. In vitro culture of peri-gastrulation embryos of a macropodid marsupial. J Anat 212: 180191.
  • Idkowiak J, Weisheit G, Plitzner J, Viebahn C. 2004. Hypoblast controls mesoderm generation and axial patterning in the gastrulating rabbit embryo. Dev Genes Evol 214: 591605.
  • Keller R. 2002. Shaping the vertebrate body plan by polarized embryonic cell movements. Science 298: 19501954.
  • Kieserman EK, Wallingford JB. 2009. In vivo imaging reveals a role for Cdc42 in spindle positioning and planar orientation of cell divisions during vertebrate neural tube closure. J Cell Sci 122: 24812490.
  • Koller C. 1882. Untersuchungen über die Blätterbildung im Hühnerkeim. Arch Mikrosk Anat : 174211.
  • Krieg M, Arboleda-Estudillo Y, Puech PH, Käfer J, Graner F, Müller DJ, Heisenberg CP. 2008. Tensile forces govern germ-layer organization in zebrafish. Nat Cell Biol 10: 429436.
  • Kwon GS, Viotti M, Hadjantonakis AK. 2008. The endoderm of the mouse embryo arises by dynamic widespread intercalation of embryonic and extraembryonic lineages. Dev Cell 15: 509520.
  • Lawson KA, Meneses JJ, Pedersen RA. 1991. Clonal analysis of epiblast fate during germ layer formation in the mouse embryo. Development 113: 891911.
  • Lecuit T, Lenne P. 2007. Cell surface mechanics and the control of cell shape, tissue patterns and morphogenesis. Nature Rev Mol Cell Biol 8: 633644.
  • Luckett WP. 1978. Origin and differentiation of the yolk sac and extraembryonic mesoderm in presomite human and rhesus monkey embryos. Am J Anat 152: 5997.
  • Mammoto T, Ingber DE. 2010. Mechanical control of tissue and organ development. Development 137: 14071420.
  • Montero JA, Carvalho L, Wilsch-Brauninger M, Kilian B, Mustafa C, Heisenberg CP. 2005. Shield formation at the onset of zebrafish gastrulation. Development 132: 11871198.
  • Nakatsuji N, Snow MHL, Wylie CC. 1986. Cinemicrographic study of the cell movement in the primitive-streak-stage mouse embryo. J Embryol Exp Morphol 96: 99109.
  • O'Connell CB, Wang YL. 2000. Mammalian spindle orientation and position respond to changes in cell shape in a dynein-dependent fashion. Mol Biol Cell 11: 17651774.
  • Olivier N, Luengo-Oroz MA, Duloquin L, Faure E, Savy T, Veilleux I, Solinas X, Debarre D, Bourgine P, Santos A, Peyrieras N, Beaurepaire E. 2010. Cell lineage reconstruction of early zebrafish embryos using label-free nonlinear microscopy. Science 329: 967971.
  • Perea-Gomez A, Lawson KA, Rhinn M, Zakin L, Brulet P, Mazan S, Ang SL. 2001. Otx2 is required for visceral endoderm movement and for the restriction of posterior signals in the epiblast of the mouse embryo. Development 128: 753765.
  • Rao CR. 1967. Least squares theory using an estimated dispersion matrix and its application to measurement of signals. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics. Berkeley: University of California Press. p 355372.
  • Reupke T, Püschel B, Viebahn C. 2009. Tracing and ablation of single cells in the mammalian blastocyst using fluorescent DNA staining and multiphoton laser. Histochem Cell Biol 131: 521530.
  • Sanders EJ, Varedi M, French AS. 1993. Cell proliferation in the gastrulating chick embryo: a study using BrdU incorporation and PCNA localization. Development 118: 389399.
  • Sausedo RA, Schoenwolf GC. 1993. Cell behaviors underlying notochord formation and extension in avian embryos: quantitative and immunocytochemical studies. Anat Rec 237: 5870.
  • Sausedo RA, Smith JL, Schoenwolf GC. 1997. Role of nonrandomly oriented cell division in shaping and bending of the neural plate. J Comp Neurol 381: 473488.
  • Shimizu H, Bode PM, Bode HR. 1995. Patterns of oriented cell division during the steady-state morphogenesis of the body column in hydra. Dev Dyn 204: 349357.
  • Srinivas S, Rodriguez T, Clements M, Smith JC, Beddington RS. 2004. Active cell migration drives the unilateral movements of the anterior visceral endoderm. Development 131: 11571164.
  • Stern CD. 1990. The marginal zone and its contribution to the hypoblast and primitive streak of the chick embryo. Development 109: 667682.
  • Strutt D. 2005. Organ shape: controlling oriented cell division. Curr Biol 15: R758R759.
  • Suriben R. 2009. Posterior malformations in Dact1 mutant mice arise through misregulated Vangl2 at the primitive streak. Nat Genet 41: 977985.
  • Tam PPL, Gad JM. 2004. Gastrulation in the mouse embryo. In: SternC, editor. Gastrulation: from cells to embryo. New York: Cold Spring Harbor. p 233262.
  • Ulrich F, Concha ML, Heid PJ, Voss E, Witzel S, Roehl H, Tada M, Wilson SW, Adams RJ, Soll DR, Heisenberg CP. 2003. Slb/Wnt11 controls hypoblast cell migration and morphogenesis at the onset of zebrafish gastrulation. Development 130: 53755384.
  • Ulrich F, Krieg M, Schotz EM, Link V, Castanon I, Schnabel V, Taubenberger A, Mueller D, Puech PH, Heisenberg CP. 2005. Wnt11 functions in gastrulation by controlling cell cohesion through Rab5c and E-cadherin. Dev Cell 9: 555564.
  • Viebahn C. 1999. The anterior margin of the mammalian gastrula: comparative and phylogenetic aspects of its role in axis formation and head induction. Curr Top Dev Biol 46: 63103.
  • Viebahn C, Mayer B, Hrabe de Angelis M. 1995. Signs of the principal body axes prior to primitive streak formation in the rabbit embryo. Anat Embryol (Berl) 192: 159169.
  • Viebahn C, Stortz C, Mitchell SA, Blum M. 2002. Low proliferative and high migratory activity in the area of Brachyury expressing mesoderm progenitor cells in the gastrulating rabbit embryo. Development 129: 23552365.
  • Vogt W. 1925. Gestaltungsanalyse am Amphibienkeim mit örtlicher Vitalfärbung. I. Teil. Methodik und Wirkungsweise der örtlichen Vitalfärbung mit Agar als Farbträger. Rou's Archiv Entwickl Mech : 542610.
  • Voiculescu O, Bertocchini F, Wolpert L, Keller RE, Stern CD. 2007. The amniote primitive streak is defined by epithelial cell intercalation before gastrulation. Nature 449: 10491052.
  • Wei Y, Mikawa T. 2000. Formation of the avian primitive streak from spatially restricted blastoderm: evidence for polarized cell division in the elongating streak. Development 127: 8796.
  • Wetzel R. 1929. Untersuchungen am Hühnchen. Die Entwicklung des Keims während der ersten beiden Bruttage. Arch Entwickl Mech Org : 188321.
  • Wyngaarden LA, Vogeli KM, Ciruna BG, Wells M, Hadjantonakis AK, Hopyan S. 2010. Oriented cell motility and division underlie early limb bud morphogenesis. Development 137: 25512558.
  • Yamanaka Y, Tamplin OJ, Beckers A, Gossler A, Rossant J. 2007. Live imaging and genetic analysis of mouse notochord formation reveals regional morphogenetic mechanisms. Dev Cell 13: 884896.
  • Yen WW, Williams M, Periasamy A, Conaway M, Burdsal C, Keller R, Lu X, Sutherland A. 2009. PTK7 is essential for polarized cell motility and convergent extension during mouse gastrulation. Development 136: 20392048.
  • Zallen JA. 2007. Planar polarity and tissue morphogenesis. Cell 129: 10511063.
  • Zallen JA, Blankenship J. 2008. Multicellular dynamics during epithelial elongation. Sem Cell Dev Biol 19: 263270.
  • Zhang Y, Levin M. 2009. Left-right asymmetry in the chick embryo requires core planar cell polarity protein Vangl2. Genesis 47: 719728.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
DVDY_22687_sm_suppfig1.tif1083KSupporting Figure 1 Principal neighbor cell changes along the mediolateral axis and in the vicinity of the presumptive primitive streak area, showing relative positions of the cells involved at the start (a, b, and c) and at the end (a′, b′, and c′) of their movement. The relative position of the cells in every group of cells was depicted with green (start) and magenta (end) triangles to highlight the principal change in the arrangement as a result of conventional intercalation (A) or in different processional cell movement (PCM) subclasses (B, C, D).
DVDY_22687_sm_suppmovie1.mov5328KSupporting Movie 1 Cell movements (L- and U-turns) in the posterior half of DAPI (4′,6-diamidine-2-phenylidole-dihydrochloride) -stained rabbit embryonic disc before gastrulation by means of multiphoton time-lapse recording. The still pictures shown in Figure 2 were taken from this movie. For details see legend of Figure 2A–C.
DVDY_22687_sm_suppmovie2.mov5106KSupporting Movie 2 Cellular movements in the presumptive primitive streak area (same image series as in Supp. Movie S1) demonstrating processional cell movement (PCM) semiquantitatively. Processional cells were marked “a” and their neighboring cells “b” and “c.” The relative position of the cells in every group of cells was depicted with green (start) and magenta (end) triangles to show the changes in the arrangement as a result of PCM. For details, see Figure 2D–G.
DVDY_22687_sm_suppmovie3.mov1596KSupporting Movie 3 Processional cell movement (PCM) as seen with differential interfering contrast (DIC) time-lapse recording shown in Fig. 3. The changes in the relative position of the neighboring cells were illustrated with triangles as in movie 2. For details, see legend of Supp. Movie S2 and Figure 3.
DVDY_22687_sm_suppmovie4.mov4745KSupporting Movie 4 Three triplets of neighboring cells in the presumptive primitive streak area exhibiting processional cell movement (PCM). For details, see legend of Supp. Movie S2.
DVDY_22687_sm_suppmovie5.mov3161KSupporting Movie 5 Rapid metaphase plate rotation in the posterior gastrula extension (PGE) area as seen with differential interfering contrast (DIC) time-lapse recording shown in Figure 6. For details, see legend of Figure 7.
DVDY_22687_sm_suppinfo.pdf383KSupporting Information.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.