The transition from meiotic to mitotic spindle assembly is gradual during early mammalian development

Abstract

The transition from meiosis to mitosis, classically defined by fertilization, is a fundamental process in development. However, its mechanism remains largely unexplored. In this paper, we report a surprising gradual transition from meiosis to mitosis over the first eight divisions of the mouse embryo. The first cleavages still largely share the mechanism of spindle formation with meiosis, during which the spindle is self-assembled from randomly distributed microtubule-organizing centers (MTOCs) without centrioles, because of the concerted activity of dynein and kinesin-5. During preimplantation development, the number of cellular MTOCs progressively decreased, the spindle pole gradually became more focused, and spindle length progressively scaled down with cell size. The typical mitotic spindle with centrin-, odf2-, kinesin-12-, and CP110-positive centrosomes was established only in the blastocyst. Overall, the transition from meiosis to mitosis progresses gradually throughout the preimplantation stage in the mouse embryo, thus providing a unique system to study the mechanism of centrosome biogenesis in vivo.

Figure 1.

Stochastic MTOC assembly leads to formation of the multipolar spindle followed by progressive clustering into a barrel-shaped spindle. (A–I) Immunofluorescence staining of the mouse MII oocyte (A) and zygotes (B–I) fixed at consecutive stages of development: early interphase (18 h after hCG), in which the fertilization cone (dotted line) forms in response to sperm entry (B); mid (C)- and late (D) interphase (21 and 25 h after hCG, respectively); prophase (E); early (F), mid (G)-, and late (H) prometaphase; and metaphase (I; 28 h after hCG). Single-section images (C–H) or z-projected images of confocal sections (A, B, and I) show microtubules, pericentrin, and DNA. In B, arrowhead marks male chromatin delivered by the sperm; asterisk marks the second meiotic spindle. Note the absence of MTOC enhancement in the fertilization cone. Arrows in D and E mark MTOCs on the pronuclear surface. Upon NEBD, a multipolar spindle forms with no major axis (early prometaphase; F). Arrowheads in F and G mark the multipoles. A few major axes become visible in midprometaphase (G), consolidating into a single major axis with minor additional axes in late prometaphase (asterisks; H) and eventually forming a barrel-shaped spindle with pericentrin localized on two ring-shaped poles (metaphase; I). (J) Live imaging of mouse zygotes during the first division at prophase (left), prometaphase (middle), and metaphase (right). Z-projected images of confocal sections (3 µm thick) show microtubules (EGFP-MAP4; gray) and DNA (H2B-mRFP1). Circles and arrows mark MTOCs and the multipoles, respectively. Bars, 10 µm. Time is given in hours and minutes after NEBD.

Figure 2.

Dynein is essential for maturation of MTOCs and spindle, whereas kinesin-5 is required for spindle bipolarization. (A and B) Live imaging of mouse zygotes during the first division under inhibition of dynein by P150-CC1 (A) or of kinesin-5 by monastrol (B) at prophase (left), early (middle), and late prometaphase (right), respectively. Z-projected images of confocal sections (3 µm thick) show microtubules (EGFP-MAP4; gray). Time indicates hours and minutes after NEBD. Note that the monopolar spindle is formed in the kinesin-5–inhibited zygote (asterisk in B; right). (C) Distance of each cytoplasmic MTOC from the nucleus measured by tracking MTOCs in control, dynein-, and kinesin-5–inhibited zygotes (n = 4, 1, and 3 embryos, respectively) and plotted against time after NEBD. (D) Speed of each MTOC movement toward the nucleus in relation to the initial distance of the MTOC from the nucleus in control, dynein-, and kinesin-5–inhibited embryos. Insets show representative tracks in zygotes. (E) Whisker box plot of speed of MTOC movement toward the nucleus in control (n = 48 tracks derived from four zygotes), dynein-inhibited (n = 38 tracks derived from three zygotes), and kinesin-5–inhibited (n = 32 tracks derived from three zygotes) embryos. MTOC movement is significantly slower in dynein-inhibited embryos and faster in kinesin-5–inhibited embryos than in controls (P < 0.05, Welch two-sample t test). The lines near the middle of the boxes represent the median (50th percentile). The bottom and top of the boxes are the 25th and 75th percentile, respectively. The whiskers extend to the most extreme data point, which is no more than 1.5 times the percentile range of the box. The dots represent the extreme data point extending out of the 1.5 times percentile range of the box. (F) Summary of the potential mechanism leading to acentrosomal spindle assembly in the mouse zygote before (left) and after (right) nuclear envelope breakdown (NEBD). Bars, 10 µm.

Figure 3.

Progressive transition from acentrosomal to centrosomal spindle formation during mouse preimplantation development. (A and B) Live imaging of mouse embryos during the second division (A) and the fourth division (B) at prophase, prometaphase, and metaphase. Note the small microtubule asters (presumably MTOCs; yellow arrowheads) in two-cell and eight-cell stage embryos before NEBD. Red arrowheads mark multipoles of the spindle at prometaphase. Z-projected images of 3-µm confocal sections showing microtubules (EGFP-MAP4; gray) and DNA (H2B-mRFP1). Time is shown in hours and minutes after hCG. (C–E) Immunostaining of embryos fixed during the second division (showing one of the two cells; C), the fourth division (D), and at 113 h after hCG (E4.5; E) at prophase, prometaphase, and metaphase and stained for DNA, microtubules, and pericentrin. In prometaphase at the two-cell and eight-cell stages, several small MTOCs are visible (arrowheads), whereas only two bright MTOCs (arrows) before and after NEBD are seen at E4.5. All pictures are projected images of 0.38-µm stacks. Insets in E represent a zoom of the boxes. Bars: (A–E, main images) 10 µm; (E, insets) 5 µm.

Figure 4.

Emergence of the centrin- and CP110-positive centrosome and the kinesin-12–positive spindle in the E3.5 blastocyst. (A) Immunostaining of embryos fixed at the 16-cell stage, E3.5 (between 32- and 64-cell stages), and E4.5 (with >128 cells) and stained for DNA (blue), centrin, and pericentrin. At the 16-cell stage, all pericentrin-positive MTOCs are negative for centrin. At E3.5, some MTOCs are positive for pericentrin and negative for centrin (white arrowheads), whereas others are positive for both (yellow arrowheads). At E4.5, one or two dots positive both for pericentrin and centrin are visible in each cell (yellow arrowheads). Note that the laser intensity for centrin detection was enhanced in 16-cell stage and E3.5 embryos, resulting in the enhanced signal in the cytoplasm. Because single-section images of confocal microscopy are shown, the centrin and pericentrin signal of other cells are out of focus. Bar, 10 µm. (B, left) Immunostaining of mouse embryos fixed at the 16-cell stage, E3.5, and E4.5 and stained for DNA (blue), CP110, and pericentrin. At the 16-cell stage, MTOCs are negative for CP110. At E3.5, only some MTOCs are positive for CP110 (yellow arrowheads; white arrowheads mark those negative for CP110), whereas at E4.5, all MTOCs are positive. (right) Immunostaining of mouse embryos fixed at the 16-cell stage, E3.5, and E4.5 and stained for DNA (blue), kinesin-12, pericentrin, and microtubules. At the 16-cell stage, the spindle is negative for kinesin-12 (white arrowheads). At E3.5, only some spindles are positive for kinesin-12 (yellow arrowheads), whereas at E4.5, most of the spindles are positive (yellow arrowheads). Z-projected sections of confocal images. Bars, 5 µm. (C) The fraction of centrin-positive (yellow) and -negative (red) MTOCs, CP110-positive (green) and -negative (blue) MTOCs, and of kinesin-12–positive (orange) and –negative (violet) spindles at the 16-cell, E3.5, and E4.5 stages.

Figure 5.

Progressive change in MTOCs and the time required for division. (A) Progressive change in diameter of MTOCs shortly before NEBD in the embryos at consecutive developmental stages. (B) Whisker box plot of the number of MTOCs per cell at one-cell, two-cell, eight-cell, E3.5, and E4.5 stages, shortly before NEBD. The lines near the middle of the boxes represent the median (50th percentile). The bottom and top of the boxes are the 25th and 75th percentile, respectively. The end of the bottom whisker is the 5th percentile, and the end of the top whisker is the 95 percentile. (C) Duration of cell division (from NEBD to the beginning of anaphase) at consecutive stages of preimplantation development. Note that from the eight-cell stage on, the duration of cell division (∼20 min or less) becomes only twice as long as the time interval of recording (every 10 min); thus, the data cannot be as precise (indicated by a broken line) as those for the earlier stages. In A and C, the vertical bars indicate the range of values.

Figure 6.

Gradual change in spindle characteristics and establishment of the spindle length regulation during mouse preimplantation development. (A) Progressive change in radius and diameter of the spindle in the embryos at consecutive developmental stages. Insets show a representative image of the spindle and its pole at each stage. Bar, 10 µm. (B, left) Change in the mean size of the spindle and of the cell in the embryos at consecutive developmental stages. (right) Cell diameter and spindle length are plotted as colored circles for individual embryos at different developmental stages, illustrating that their ratio (slope of the black line) remains constant from the fourth to eighth division. Those for experimentally micromanipulated embryos are shown as colored crosses. (C, left) Experimentally micromanipulated zygotes in which two thirds (top) or half (bottom) of the cytoplasm was removed. Note that two pronuclei are visible (yellow arrowheads) after cytoplasm removal. Metaphase spindle and measurement of its size and cell size by live imaging of the micromanipulated embryos during the subsequent divisions. All pictures are projected images of 4.5-µm stacks. Bars, 20 µm. In A and B, the vertical bars indicate the range of values.

Figure 7.

Gradual transition from meiotic to mitotic spindle assembly throughout the preimplantation stage in the mouse embryo. Summary of the progressive transition from meiosis to mitosis throughout mouse preimplantation development. See Discussion for details.