QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 30 August 2006
doi: 10.1242/dev.02578

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Strauss, B. Articles by Papalopulu, N. Search for Related Content PubMed PubMed Citation Articles by Strauss, B. Articles by Papalopulu, N.

A default mechanism of spindle orientation based on cell shape is sufficient to generate cell fate diversity in polarised Xenopus blastomeres

Bernhard Strauss^1,2, Richard J. Adams² and Nancy Papalopulu^1,2,*,

¹ The Wellcome Trust/Cancer Research UK Gurdon Institute, Tennis Court Road, Cambridge CB2 1QN, UK.
² Department of Physiology, Development and Neuroscience, Downing Site, University of Cambridge, Cambridge CB2 3DY, UK.

View larger version (62K): [in a new window]   Fig. 1. Cells in the Xenopus blastula divide according to their shape. (A) Xenopus blastula cells have different shapes; arrows indicate cells with small apical surfaces and a long axis perpendicular to the surface. (B,C) Examples of single cells (B) and a pair of cells (C) dissected out (arrowhead indicates flat cell with a large apical surface; arrow indicates apicobasally elongated cell with a small apical surface). (D,E) Confocal image stacks of embryos stained for -tubulin (red in D, white in E) and DNA (yellow in D). A top view section and the corresponding side projection of the stack are shown. In D, the spindle is oriented parallel to the surface; in E it is perpendicular to the surface. The cell in E has a small apical surface and a spindle aligned with the long axis that is perpendicular to the surface. Such cells will undergo a fate-asymmetric division. Spindle alignment with the long axis of the cell is also observed in cells with parallel spindles (D). (F-H) Spindle orientation was visualised in vivo using tau-GFP-injected embryos and alignment was analysed at late anaphase between the 8th and the 10th division. Spindles aligned within 25° of the long axis were scored as aligned with the long axis (LA) (F,H). Alignment within 25° of the short axis was scored as aligned with the short axis (SA) (G,H). n=216 divisions in 11 embryos. Scale bars: 50 µm. (I,J) 3D reconstructions of sister cells that were identified by the remnant of the midbody (white arrow). (I) A pair of flat cells that are the product of a parallel division and (J) a flat cell and an apicobasally elongated cell, the product of an oblique division. The previous cleavage plane can be deduced from the position of the midbody remnant. In all these cells, the axis of the spindle (red line) is aligned close to the long axis of the cell. Movies 1 and 2 in the supplementary material correspond to these images. (K,L) Computational analysis of spindle alignment with the long axis in 3D reconstructions shows that the median deviation is 12° at metaphase and most spindles are within 20° of the long axis. The distribution of angles was non-random, as determined by Watson's U2n test (P<0.005). (M) Correlation of the elongation factor with spindle alignment to the long axis. (A value of 1 corresponds to a sphere.) Elongated cells show better alignment with the long axis.

View larger version (34K): [in a new window]   Fig. 2. Isolated spherical blastomeres divide differently from cells in the embryo and often bisect a small apical surface. (A) Example of a colour-coded frame from a time-lapse movie that shows the three possible spindle orientations in the embryo, quantified for the 128- to 256-cell division in 10 embryos (n=447 divisions). Par., parallel; Per., perpendicular; Obl., oblique, as defined in the Materials and methods. (B) These three types of spindle orientation occur also in isolated blastomeres at 128-cell stage, as example frames from Movies 1-7 in the supplementary material show; however, the proportion of each type of division differs. (C) Comparison of the proportions of the three division types between the embryo (white bars) and isolated blastomeres, n=151 (black bars). Oblique and perpendicular divisions have significantly different proportions in isolated cells when compared with the embryo (-square goodness of fit, P=0.001). (D) In control embryos, cells with a very small apical surface (less than or equal to one quarter of its sister) will always divide perpendicularly and the apical pigment is not divided. Arrows in D indicate cells with a small apical domain. (E,F) In isolated blastomeres (E) and in embryos that were raised in Ca2+/Mg2+-free medium (D), such small apical surfaces do become divided. (F) Comparison of frequencies with which the cleavage plane divides a small apical surface. In control embryos: 0% (n=42 divisions in time-lapse movies of 24 embryos). In embryos raised in Ca2+/Mg2+-free medium, with rounded cells: 32% (n=60 cells with small apical surface were analysed in 12 different embryos).

View larger version (33K): [in a new window]   Fig. 3. Blastula cells align the spindle with an experimentally induced long axis. (A) Round isolated blastomeres were compressed and filmed to assess the orientation of the cleavage plane with respect to the introduced long axis: parallel (spindle perpendicular to the long axis), perpendicular (spindle parallel to long axis) and oblique (orientation between these two categories). Arrowheads indicate cleavage planes. (B) Cleavage plane orientations were analysed with respect to the time that has elapsed between mechanical deformation and cytokinesis (n=100 divisions). The shorter the elapsed time between deformation and cytokinesis (0-3 minutes, 3-5 minutes), the smaller the percentage of spindles that align with the long axis. When a long axis is imposed 5-15 minutes before cytokinesis, the spindle aligns with it in 100% of cases. The correlation between division type and time, measured to the nearest half minute, was highly statistically significant (Spearman's correlation -0.647, n=100).

View larger version (73K): [in a new window]   Fig. 4. Spindle alignment takes place early in the cell cycle; large angle oscillations do not occur. (A) Different spindle orientation behaviour is shown in frames from 4D series of tau-GFP injected embryos; LA, spindle is already set up within 25° of the long axis at prophase/prometaphase; RLA, spindle rotates into the long axis; SA, spindle is already set up within 25° of the short axis; RSA, spindle rotates into the short axis (see Movie 6 in the supplementary material). (B) Quantification of spindle orientation behaviour in 216 cells from 11 different embryos. The majority of spindles are already close to the long axis when the spindle is set up. (C) Example frames of a spindle rotating into the long axis, RLA. Scale bars in A,C: 20 µm. (D) Analysis of the temporal dynamics of such rotation movements in 12 cells from three different embryos. The angular changes (y-axis) with respect to the long axis are plotted against time (x-axis). Time intervals between frames are 92 seconds in cells 1-8 and 75 seconds in cells 9-12. The broken line indicates 25°. Overall, orientation movements into the long axis occur via small angle adjustments as for example in graphs 1 and 11 (number 11 corresponds to the cell shown in C).

View larger version (24K): [in a new window]   Fig. 5. Most spindles align with the long axis at prophase. The deviation of the spindle axis from the long axis was measured in parallel (symmetrically) dividing fixed cells at different time-points during mitosis (A-C). Charts on the left show the distribution of the raw data; y-axis=0° corresponds to long axis; each triangle represents one measurement. The radial line indicates the median angle deviation. Graphs on the right show the percentage of cells in angle classes of 20°. The majority of cells have their spindles aligned within 20° of the long axis throughout the cell cycle. (D) Cells with a spindle axis perpendicular to the surface were measured at prophase (n=41 cells).

View larger version (31K): [in a new window]   Fig. 6. Astral microtubules are important for spindle alignment into the long axis. The deviation of the spindle axis from the long axis was measured in cells treated with nocodazole. (A,D) Antibody staining for -tubulin in treated (A) and control (D) cells. (B,C) A reduction of astral microtubules leads to an increase of the median deviation from the long axis compared with the controls (E,F). Eighty-five cells were analysed in ten nocodazole-treated embryos and 75 cells in 12 control embryos. The difference between the median angles is statistically significant by the Mann-Whitney test (P<0.001).

View larger version (79K): [in a new window]   Fig. 7. Astral microtubules are most abundant in prophase and anaphase/telophase. (A) Example frames from a time-lapse movie showing a cell expressing EB1-GFP. Movies were generated from projections of z-stacks with an imaging depth of about 15 µm to visualise quantitative changes in cortical astral microtubules. Cortical EB1 signal disappears around metaphase. Seven cells were analysed throughout mitosis in three different embryos. (B) Quantification of EB1 signals throughout the cell cycle of one cell. I/P, interphase/prophase; PM, prometaphase; M, metaphase; A, anaphase; T/CK, telophase/cytokinesis; y-axis, number of EB1 signals; x-axis, time (15.3 seconds between frames). (C) Astral microtubules shown in fixed cells by antibody staining against -tubulin. During prometaphase and metaphase, astral microtubules are strongly reduced; mb, midbody. Centrosomes are separated in late anaphase (arrow). Scale bars: 20 µm.