PLK1- and PLK4-Mediated Asymmetric Mitotic Centrosome Size and Positioning in the Early Zebrafish Embryo

Abstract

Factors that regulate mitotic spindle positioning remain unclear within the confines of extremely large embryonic cells, such as the early divisions of the vertebrate embryo, Danio rerio (zebrafish). We find that the mitotic centrosome, a structure that assembles the mitotic spindle [1], is notably large in the zebrafish embryo (246.44 ± 11.93 μm² in a 126.86 ± 0.35 μm diameter cell) compared to a C. elegans embryo (5.78 ± 0.18 μm² in a 55.83 ± 1.04 μm diameter cell). During embryonic cell divisions, cell size changes rapidly in both C. elegans and zebrafish [2, 3], where mitotic centrosome area scales more closely with changes in cell size compared to changes in spindle length. Embryonic zebrafish spindles contain asymmetrically sized mitotic centrosomes (2.14 ± 0.13-fold difference between the two), with the larger mitotic centrosome placed toward the embryo center in a polo-like kinase (PLK) 1- and PLK4-dependent manner. We propose a model in which uniquely large zebrafish embryonic centrosomes direct spindle placement within disproportionately large cells.

Keywords: C. elegans; PCM; PLK1; PLK4; centrosome; embryo; microtubules; mitosis; spindle; zebrafish.

Figure 1.. Mitotic centrosome area scales with cell length during embryonic cell divisions in C. elegans and zebrafish.

(A) Maximum confocal projection of a C. elegans embryo at 1- and 3-cell stages. DNA and cell membrane (H2B::mCherry and PH::mCherry, inverted grayscale) and microtubules (α-tubulin::GFP, magenta) shown. Model depicting position of metaphase spindle within embryo. Bar, 10µm. (B) Three-dimensional rendering of zebrafish embryo at the 8- and 16-cell stage. Microtubule marker (EMTB-3xGFP) in grayscale. Model depicting position of spindle within embryo. Bar, 250µm. (C) Bar graphs depicting spindle length (orange) and cell length along spindle axes (gray) during C. elegans (left, n>14 embryos) and zebrafish development (right, n>3 embryos). Mean ± SEM shown. One-way ANOVA, p<0.0001 (****) for spindle and cell length in both C. elegans and zebrafish. Inset graph depicts magnified spindle length data. (D-E) Representative images of metaphase cell at the 1-cell (left) and 3-cell stage (right) in a C. elegans embryo (D), and at the 8-cell (top) and 16-cell stage (bottom) in a zebrafish embryo (E). DNA and γ-tubulin in white, DNA marked by blue arrowhead. Magnified mitotic centrosomes in insets on right. Bar, 15µm. (F) Violin plot depicting two-dimensional centrosome area (µm²) at the 1-, 2-, 3-, and 6-cell stage in C. elegans (left, n>38 centrosomes), and at the 8-, 16-, and 512-cell stage in zebrafish (right, n>28 centrosomes). One-way ANOVA, p<0.0001 (****) for both C. elegans and zebrafish. Inset depicts magnified 512-cell stage data. (G) Violin plot depicting cell length (n>14), spindle length (n>14), and mitotic centrosome area for C. elegans at the 1- and 2-cell stage (n>42 centrosomes, left), and zebrafish at the 8- and 16-cell stage (n>147 centrosomes, right). Values normalized to mean of earliest developmental stage (1-cell for C. elegans, 8-cell for zebrafish), dashed line at value of 1. See panels 1C and 1F for raw values prior to normalization. (H) Scaled model depicting cell (green), spindle (orange), and mitotic centrosome (purple) during the 1- and 2-cell stage in C. elegans embryos, and the 8- and 16-cell stage in zebrafish embryos. Percentages listed at the 2-cell and 16-cell stage refer to the percent decrease in value compared to the previous developmental stage. Bar, 20µm. For violin plots: Plot boundaries depict minimum and maximum, 25^th and 75^th quartiles represented by thin black line, median represented by thick black line. For all graphs: detailed statistical analysis in Table S1. See also Figure S1 and Video S1.

Figure 2.. Centrosomes in early zebrafish development.

(A-B) Three-dimensional rendering of mitotic spindle positioning during early embryonic divisions in a zebrafish embryo using EMTB-3xGFP (microtubules, A) and centrin-GFP (centrosome, B). Microtubules shown in depth-coded z-stack such that z-slices closest to the embryo yolk are colored red and z-slices furthest from the yolk are colored blue (A). Centrin-GFP (inverted grayscale, B) shown at the 8- and 16-cell stage. Cell highlighted by dashed box magnified in (E). Bar, 100µm. (C) Model depicting the placement of mitotic spindles within embryonic zebrafish cells from the 1-cell stage to the 16-cell stage. Cells are viewed from top of cell mass with yolk placed below (XY view). Mitotic centrosomes (purple) and metaphase plate (blue) shown. (D-E) Stills from timelapse of a cell division in EMTB-3xGFP transgenic embryo (microtubules, inverted grayscale, D) and a centrin-GFP embryo (centrosome, Fire LUT, E, insert from B). Mitotic stages denoted. Insets below centrin-GFP timelapse depict magnified poles denoted by white (top) and cyan (bottom) asterisks. (F) Violin plot depicting normalized mitotic centrosome area at 16-cell stage during prophase/prometaphase, metaphase, and anaphase. Values normalized to the mean mitotic centrosome area at prophase/prometaphase. One-way ANOVA, p<0.0310 (*). n>88 mitotic centrosomes measured. Plot boundaries depict minimum and maximum, 25^th and 75^th quartiles represented by thin black line, median represented by thick black line. Detailed statistical analysis in Table S2. (G) Maximum confocal projection of a single mitotic spindle with γ-tubulin (magenta), centrin-GFP (cyan), and DNA (DAPI, blue) shown. Insets below depict mitotic centrosome denoted by yellow box. Bar, 10µm. (H) Maximum confocal projection of a single mitotic spindle with PLK1 (magenta), DNA (DAPI, blue), and microtubules (white) shown. Insets below depict mitotic centrosome denoted by yellow box. Bar, 10µm. Detailed statistical analysis in Table S1. See also Figure S2 and Video S2–4.

Figure 3.. Mitotic centrosomes are asymmetric in size.

(A) Left, model depicting a C. elegans embryo at the 3-cell stage with location of mitotic cell denoted by spindle illustration. Right, maximum confocal projection of a 3-cell stage embryo. Inset, depicts mitotic spindle shown on right. γ-tubulin and nuclear marker (grayscale) shown, γ-tubulin-positive mitotic centrosomes denoted by magenta arrowheads. Note that spindle inset has been rotated to a vertical orientation for comparison to (C). Bar, 10µm. (B) Violin plot depicting C. elegans mitotic centrosome area at the 1-, 3-, and 6-cell stage binned by size (larger or smaller). One-way ANOVA, p=0.0442 (*), p=0.0226 (*), and p=0.0017 (**), respectively. (C) Model depicting a 16-cell embryo with maximum confocal projections of a representative cell (denoted with purple box). Fixed 16-cell metaphase embryo immunostained for γ-tubulin (magenta/inverted grayscale) and DNA (DAPI, blue). Embryonic midline denoted with orange dashed line. Bar, 10µm. (D) Violin plot depicting zebrafish mitotic centrosome area at the 8-, 16-, and 512-cell stage binned by size (larger or smaller). One-way ANOVA, p=0.0009 (***), p<0.0001 (****). Inset depicts magnified 512-cell stage values. (E) Violin plot depicting the ratio of larger to smaller mitotic centrosome area for 1- (n=23), 3- (n=23) and 6-cell stage C. elegans embryos (n=19 spindles) and 8- (n=73), 16- (n=172), and 512-cell stage zebrafish embryos (n=14 spindles). Dotted line represents a ratio of 1. (F) Model depicting the positioning of the asymmetric zebrafish mitotic centrosomes in relation to the embryonic midline during the 8- and 16-cell stages. The larger of the two mitotic centrosomes (purple) is placed closest to the embryonic midline (orange), providing directionality (turquoise arrow). Violin plots depicting mitotic centrosome area at the 8- (left) and 16-cell stage (right) shown with corresponding model. One-way ANOVA, p<0.0001 (****). For all violin plots: Plot boundaries depict minimum and maximum, 25^th and 75^th quartiles represented by thin black line, median represented by thick black line. Detailed statistical analysis in Table S3. See also Figure S3.

Figure 4.. PLK1 and PLK4 activity are required for asymmetric mitotic centrosome positioning.

(A) Representative images of 16-cell stage embryos during metaphase under conditions of DMSO (left), 1µM BI2536 (center), or 1µM centrinone treatment (right). Single cells denoted in embryo image magnified in inset. γ-tubulin (magenta/inverted grayscale), and DNA (DAPI, blue) shown. Model depicting mitotic centrosome positioning in embryo shown on left (cyan/correct and gold/incorrect positioning depicted with arrows). Large and smaller mitotic centrosomes not drawn to scale in model. Bar, 10 µm. (B) Bar graph depicting percentage of spindles with largest centrosome pointed towards midline under conditions of DMSO (gray), BI2536 (100nM or 1µM, blue), or centrinone (100nM or 1µM, gold) exposure. (C-D) Violin plot depicting the ratio of mitotic centrosome areas binned by size (larger-to-smaller centrosome ratio, C) or centrosome areas unbinned (D) under conditions of DMSO (gray), BI2536 (100nM or 1µM, blue), or centrinone (100nM or 1µM, gold) exposure. Metaphase centrosome areas measured from γ-tubulin signal from fixed zebrafish embryos at the 16-cell stage. One-way ANOVA with Dunnett’s multiple comparison test performed with DMSO control. (E) Model depicting the positioning of the asymmetric mitotic centrosomes in relation to the embryonic midline during the 16-cell stages under conditions of DMSO (gray), BI2536 or centrinone (purple) exposure. Mitotic centrosomes (purple), metaphase plate (blue), and embryonic midline (orange dashed line) depicted. (F) Bar graphs representing percentage of normal embryos (gray), dead (black), or with abnormal phenotypes (orange) at 2, 4, 9, and 120 hours post-fertilization (hpf) in uninjected embryos (n=271 embryos) and after DMSO vehicle control injection (n=131 embryos), 1µM BI2536 (n=200 embryos), or 1µM centrinone (n=247 embryos) injections. For all violin plots: Plot boundaries depict minimum and maximum, 25^th and 75^th quartiles represented by thin black line, median represented by thick black line. For all graphs: Detailed statistical analysis in Table S4. See also Figure S4.