The maternal-effect gene cellular island encodes aurora B kinase and is essential for furrow formation in the early zebrafish embryo

Abstract

Females homozygous for a mutation in cellular island (cei) produce embryos with defects in cytokinesis during early development. Analysis of the cytoskeletal events associated with furrow formation reveal that these defects include a general delay in furrow initiation as well as a complete failure to form furrow-associated structures in distal regions of the blastodisc. A linkage mapping-based candidate gene approach, including transgenic rescue, shows that cei encodes the zebrafish Aurora B kinase homologue. Genetic complementation analysis between the cei mutation and aurB zygotic lethal mutations corroborate gene assignment and reveal a complex nature of the maternal-effect cei allele, which appears to preferentially affect a function important for cytokinesis in the early blastomeres. Surprisingly, in cei mutant embryos a short yet otherwise normal furrow forms in the center of the blastodisc. Furrow formation is absent throughout the width of the blastodisc in cei mutant embryos additionally mutant for futile cycle, which lack a spindle apparatus, showing that the residual furrow signal present in cei mutants is derived from the mitotic spindle. Our analysis suggests that partially redundant signals derived from the spindle and astral apparatus mediate furrow formation in medial and distal regions of the early embryonic blastomeres, respectively, possibly as a spatial specialization to achieve furrow formation in these large cells. In addition, our data also suggest a role for Cei/AurB function in the reorganization of the furrow-associated microtubules in both early cleavage- and somite-stage embryos. In accordance with the requirement for cei/aurB in furrow induction in the early cleavage embryo, germ plasm recruitment to the forming furrow is also affected in embryos lacking normal cei/aurB function.

Figure 1. Embryos from cei homozygous females exhibit defects in cytokinesis.

Cytokinesis defects in live (A–H, O–R) and fixed (I–N) wild-type (A–D, I–K) and cei mutant (E–H, L–R) embryos. (A–H) Side views of live wild-type (A–D) and maternally mutant cei (E–H) embryos at time points equivalent to (in wild-type embryos) the 2-cell (A,E), 8-cell (B,F), 64-cell (C,G) and 1,000-cell (D,H) stages. Maternally mutant cei embryos exhibit a rudimentary furrow (arrowhead in E) and form syncytial embryos (H). (I–N) Animal views of fixed wild-type (I–K) and maternally mutant cei (L–N) embryos, labeled with an antibody against ß-catenin, a component of cell adhesion junctions present in mature furrows (I,L) and the DNA dye DAPI (J,M). Merged images shown in (K,N). (O–R) Side views of live embryos at the 1,000 cell stage, showing the phenotypic range of maternally mutant cei embryos. Embryos exhibit various degrees of aberrant syncytium formation. Images are representative for the categories presented in Table 1: C4 (most severe – O), C3 (P), C2 (Q), C1 (least severe – R). Arrowheads show the limits of a single cellularized region that typically sits atop the syncytial region.

Figure 2. DNA segregation defects in embryos with reduced cei/aurB function.

Animal views of fixed embryos labeled with DAPI. (A–B) Wild-type (A) and maternally mutant cei (B) embryos at 60 min p.f. (C–D) Solvent- (C) and ZM2- (D) treated embryos at 75 min p.f. Arrowheads indicate DNA bridges.

Figure 3. cei mutant embryos have defects in the induction of furrow associated structures.

(A–H) Animal views of fixed wild-type (A–D) and mutant (E–H) embryos synchronized by in vitro fertilization and fixed at the indicated time points. Initiation of furrow formation (arrowhead in B′) is delayed in mutant embryos (F′), and furrow associated structures form only in medial region in the mutant (arrowheads in G′,H′). (I,J) Fixed wild-type (I) and mutant (J) embryos labeled with fluorescent phalloidin, showing the formation of a truncated furrow in the center of the blastodisc (arrowhead in J′).

Figure 4. Defective cytoskeletal dynamics in cei/aurB mutant embryos.

(A–G) High magnification images of wild-type (A–C) and cei mutant (D–G) embryos fixed at the indicated time points and labeled with an anti-α-tubulin antibody. Images show the FMA structure, which after formation of the furrow (A) becomes enriched in the distal region of the wild-type embryo (B) and eventually disassembles (C). In maternal cei mutants, the truncated FMA forms in the center of the blastodisc (D) and neither translocates distally nor becomes disassembled (E,F). Arrowheads in (F) indicate FMA remnants corresponding to the second cleavage planes. A rare maternal cei mutant embryo with a weaker phenotype (G) showing that defects in FMA reorganization and disassembly can be observed even when the furrow encompasses the entirety of the blastodisc.

Figure 5. Maternal expression of wild-type aurora B kinase rescues the cytokinesis defects in cei mutant embryos.

Embryos from cei/cei homozygous mutant females show a strong syncytial phenotype at the 1,000 cells stage (A) and show defective accumulation of ß-catenin in mature furrows at the 8-cell stage (C). Embryos from sibling cei/cei homozygous mutant females which carry transgenic copies of a maternally-expressed wild-type zebrafish aurB gene show significant rescue of the syncytial (B) and ß-catenin furrow accumulation phenotype (D). Side views of 1,000-cell stage live embryos (A,B) and animal views of embryos fixed at the 8-cell stage and labeled with an anti-ß-catenin antibody (green) and a DNA dye (blue) (C,D). Homozygosity for cei and presence of the transgene were determined by genotyping as in Materials and Methods.

Figure 6. Zygotic cei/aurB function is essential for embryonic development.

(A–C) Side views of live 24-hour wild-type (A), hi1045 homozygous (B) and aurB morphant (C) embryos. Arrowheads in (B,C) indicate cell necrosis in the brain. (D–F) Side views of live 5-day wild-type (D) and hi1045/cei transheterozygous (E) embryos, as well as hi1045/cei transheterozygotes containing transgenic copies of the wild-type zebrafish aurB gene (F). Arrowheads indicate the swim bladder, which is inflated in viable embryos (D,F) and indicates inviability when not inflated (E).

Figure 7. Cytokinesis defects in embryos lacking zygotic aurB function.

(A–C″) Wild-type (A), aurB^hi1045 homozygote (B) and aurB morphant (C) embryos fixed at 24 hours p.f. and labeled to detect ß-catenin (green) and DNA (red). (A–C) Overview of the head region. (A′–C′) Optical section through the brain region, showing internal cells. aurB^hi1045 homozygotes and aurB morphants show a high frequency of compact nuclei which are typically arranged in pairs (some examples indicated with a white bar at their left flank), consistent with cell death after a failure to undergo proper cytokinesis. (A″–C″) Optical section through a surface layer of the same region, corresponding to the EVL or peridermal layer. In this layer, cells in aurB^hi1045 homozygotes and aurB morphants show a high frequency of multilobular nuclei (some examples indicated by asterisks), again suggestive of defects in cytokinesis. Wild-type embryos injected with control MO exhibit normal cellular and nuclear morphologies (not shown), similar to those observed in untreated wild-type embryos (A). (D–E) High magnification images of a wild-type (D) and aurB^hi1045 homozygous (E) embryos labeled to detect ß-catenin (green), microtubules (red) and DNA (blue). Mutant embryos exhibit closely apposed pairs of nuclei (asterisks) that lack an intervening adhesive membrane.

Figure 8. Subcellular localization of Cei/AurB protein.

(A–S) Localization of Cei/AurB protein in the early embryo. Animal views of fixed wild-type embryos labeled to detect microtubules, Cei/AurB protein and DNA. (A–P) Series of images detailing the progression of the spindle during the cell cycle as indicated. During prometaphase and metaphase, Cei/AurB protein colocalizes with the forming asters (A–D) and spindles (E–H). During anaphase (I–L), Cei protein remains localized to astral microtubules and the spindle, although localization to the latter appears to subside. During telophase (M–P), Cei/AurB protein transitions to the furrow proper, where it begins to form a ladder-like pattern of short filaments perpendicular to the plane of the furrow (arrows in N′). Panels M′–P′ are 3× magnifications of M–P. (Q–S) During furrow maturation in cytokinesis (shown in this embryo during completion of the first cell division cycle at 38 min p.f.), Cei protein further accumulates along the length of the furrow as punctate aggregates (asterisks in R′) or short filaments perpendicular to the furrow (arrows in R′). These accumulations correspond to sites of microtubule bundling (compare to Q′,S′). Panels Q′–S′ are 3× magnifications of Q–S. All described sites of Cei localization are absent in control labelings using preimmune serum (not shown). (T–AE) Colocalization of Cei/AurB protein to midbodies in fixed 15-somite (16 hours p.f.) embryos. (T–W) Field of cells in a wild-type embryo, showing midbody-like structures (arrowheads in T,U,W). (X–AA, AB–AE) Fields of cells in hi1045 homozygous mutant embryos. Midbody-like structures are present (arrowheads in X,Y,AA,AB,AC,AE) but exhibit reduced levels of Cei/AurB colocalization and often exhibit a splayed, less compact structure (asterisk in AB,AE). In some cases, DNA bridges can also be observed (arrowhead in Z).

Figure 9. Germ plasm recruitment in cei/aurB mutant embryos.

Localization of vasa mRNA by fluorescent in situ hybridization in wild-type embryos fixed at 49 min p.f. (A,B) and 65 min p.f. (C,D). In the wild-type, germ plasm becomes recruited as an elongated rod-like structure in the approximately 2/3 distal regions of the furrow (A′), then localizes as a compact structure at the distal furrow end (C′). Recruitment of germ plasm is reduced or absent in cei mutants (B′, D′). Instead, aggregated germ plasm often becomes recruited to two aggregates in the center of the blastodisc, corresponding to the edges of the truncated furrow-like structure observed in these mutants (D″). Arrowheads flank planes of cleavage.

Figure 10. Redundant signals contribute to furrow formation in the early blastomeres.

Wild-type (A–D) and cei; fue double mutant (E–H) embryos were fixed at 49 min p.f. and labeled with anti–α-tubulin antibodies (A,E; red in D,H), anti–γ-tubulin antibodies (B,F; green in D,H) and DAPI (C,G; blue in D,H). (A–D) At this stage, wild-type embryos exhibit an FMA along the span of the first cleavage plane (1^st). (E–H) cei; fue double mutant embryos lack an FMA in either medial or distal regions of the 1^st cleavage plane (compare to Figure 3D,H). Double mutants show the expected delay in furrow induction, characteristic of the cei mutation, in the less mature furrow corresponding to the 2^nd cleavage plane (arrowhead in (E); compare to Figure 3B and 3F). Double mutants also exhibit the pronuclear fusion defect characteristic of the fue mutation . Similar results can be observed in embryos until 65 min p.f. (not shown).

Figure 11. Model for the action of cei/aurB and other furrow induction signals in the early zebrafish embryo.

Top images: Cei/AurB protein associated with astral microtubule ends and midzone-derived signals constitute partially redundant signals for furrow formation in the early zygote. Both astral microtubule- and spindle-derived signals act redundantly to initiate furrow formation in the medial region of the early embryonic blastodisc. On the other hand, astral-derived signals, dependent on cei function, are essential for furrow formation in distal regions of the blastodisc. Our analysis does not rule out a role for Cei/AurB, which is also localized to the spindle, as a mediator of a spindle-derived signal. Bottom images: Furrow induction results in the formation of furrow-associated cytoskeletal structures such as the FMA (green), as well as germ plasm recruitment (blue).