Author response: The recycling endosome protein Rab25 coordinates collective cell movements in the zebrafish surface epithelium

Article Figures and data Abstract Introduction Results Discussion Materials and methods Appendix 1 Data availability References Decision letter Author response Article and author information Metrics Abstract In emerging epithelial tissues, cells undergo dramatic rearrangements to promote tissue shape changes. Dividing cells remain interconnected via transient cytokinetic bridges. Bridges are cleaved during abscission and currently, the consequences of disrupting abscission in developing epithelia are not well understood. We show that the Rab GTPase Rab25 localizes near cytokinetic midbodies and likely coordinates abscission through endomembrane trafficking in the epithelium of the zebrafish gastrula during epiboly. In maternal-zygotic Rab25a and Rab25b mutant embryos, morphogenic activity tears open persistent apical cytokinetic bridges that failed to undergo timely abscission. Cytokinesis defects result in anisotropic cell morphologies that are associated with a reduction of contractile actomyosin networks. This slows cell rearrangements and alters the viscoelastic responses of the tissue, all of which likely contribute to delayed epiboly. We present a model in which Rab25 trafficking coordinates cytokinetic bridge abscission and cortical actin density, impacting local cell shape changes and tissue-scale forces. Introduction During metazoan development, epithelial cells are often organized into cohesive sheets that undergo large-scale cellular movements. In proliferative tissues, cell division is fundamental to the establishment of normal tissue architecture (Gibson et al., 2006). Cell division also poses a challenge for developing epithelia, as dividing cells must maintain cell-cell contacts while simultaneously coordinating shape changes with surrounding cells (Higashi et al., 2016). In the zebrafish embryo, the single-cell thick surface epithelium, the enveloping layer (EVL), undergoes rounds of proliferation during epiboly, a gastrulation movement describing the expansion and spreading of a tissue (Campinho et al., 2013). Thus, the zebrafish gastrula is an ideal system to investigate epithelial morphogenesis during gastrulation. The multilayered zebrafish blastoderm is positioned on top of a large yolk cell. During epiboly, the blastoderm spreads vegetally to completely cover the yolk cell (reviewed in Bruce and Heisenberg, 2020). Epiboly initiates at dome stage when EVL spreading is triggered by a reduction in tissue surface tension, that enables vegetal directed tissue expansion (Morita et al., 2017). EVL cells thin apicobasally as the apical surface area expands, with cell divisions preferentially oriented along the animal-vegetal axis (Campinho et al., 2013). EVL proliferation slows as the blastoderm covers half of the yolk cell at 50% epiboly (Campinho et al., 2013). At 6hpf, a contractile actomyosin ring assembles in the region of the yolk cell adjacent to the blastoderm, the yolk syncytial layer (YSL) (Behrndt et al., 2012). Contractile forces generated by the ring tow the tightly attached EVL vegetally. During the later stages of epiboly, as EVL cells continue to elongate and spread, EVL tension is highest, particularly in marginal regions closest to the yolk cell (Campinho et al., 2013). A subset of marginal EVL cells actively change shape and rearrange to leave the margin, enabling the tissue circumference to narrow and eventually close at the vegetal pole (Köppen et al., 2006). Overall, EVL morphogenesis involves coordination between local cell shape changes, rearrangements and tissue-scale forces. Cell divisions play a critical and conserved role during epithelial morphogenesis in multiple systems (Campinho et al., 2013; Firmino et al., 2016; Lau et al., 2015; Wyngaarden et al., 2010). Following division, daughter cells either share a common interface or cell rearrangements occur (Firmino et al., 2016; Gibson et al., 2006). Both events can be viewed as multicellular processes as cell division impacts both the dividing cell and adjacent neighbors. For example, during the formation of de novo cell contacts between newly formed daughter cells, neighboring cells resist contractile forces to maintain shape as shown in the Xenopus gastrula and Drosophila pupal notum (Herszterg et al., 2013; Higashi et al., 2016). Cell divisions also impact tissue scale forces. Oriented cell divisions have well-characterized roles in both propagating forces and dissipating anisotropic tensions (reviewed in Godard and Heisenberg, 2019). During zebrafish epiboly, oriented cell divisions relieve stresses in the EVL to facilitate tissue spreading (Campinho et al., 2013). More recently, cell divisions have been shown to contribute to the viscoelastic properties of cells. For example, the deep cells in the central region of the zebrafish blastoderm undergo a rapid decrease in tissue viscosity following mitotic rounding that is important for initiating epiboly (Petridou et al., 2019). Similarly, inhibiting cell division in amniote gastrulation resulted in high tissue viscosity, slowing cell Polonaise movements (Saadaoui et al., 2020). While our understanding of the effects of cell divisions on tissue development has advanced considerably, the role of the terminal step of cytokinesis, intercellular bridge abscission has not been examined in detail. Following contraction of the cytokinetic ring, sister cells remain connected via intercellular bridges (Burgos and Fawcett, 1955). Actomyosin and microtubule networks constrict the bridges to 1–2 µM (Mierzwa and Gerlich, 2014) and subsequently, both cytoskeletal networks are removed from bridges which is followed by bridge cleavage adjacent to the cytokinetic midbody (Connell et al., 2009; Frémont et al., 2017; Mierzwa and Gerlich, 2014). The timing of bridge abscission can affect many biological processes. When abscission is delayed in cancer cell culture models, neighboring cell division events can tear open adjacent cytokinetic bridges resulting in the formation of binucleate cells (Dambournet et al., 2011). In mouse ES cells, abscission impacts cell fate by acting as a switch for pluripotency exit (Chaigne et al., 2020). Lastly, cytokinetic bridges act as landmarks for lumen development in both cell culture and zebrafish, with the timing of abscission critical for successful lumen expansion (Rathbun et al., 2020; Willenborg et al., 2011). More specifically, in Kupffer's vesicle, the zebrafish left-right laterality organ, premature cytokinetic bridge cutting via laser ablation disrupted lumen morphogenesis (Rathbun et al., 2020). Open questions include what happens to cells if they remain interconnected by cytokinetic bridges during tissue morphogenesis and do cytokinesis failures hinder cell rearrangements or disrupt tissue-scale forces? Furthermore, what subcellular pathways are required for abscission during embryonic development? Rab proteins are the largest family of small GTPases and are important for intracellular membrane trafficking (Hutagalung and Novick, 2011; Stenmark, 2009). Endocytic-membrane recycling is critical for both cell division and cytokinetic abscission. In cell culture, Rab11-positive recycling endosomes coordinate cell cycle progression and mitotic spindle positioning (Hehnly and Doxsey, 2014). Recycled membrane is trafficked to the midbody and impaired Rab11 or Rab35 endocytic-recycling pathways result in delayed or failed abscission (Dambournet et al., 2011; Frémont and Echard, 2018; Frémont et al., 2017; Kouranti et al., 2006). For example, disrupting Rab11 directed trafficking perturbs Kupffer's vesicle morphogenesis during embryonic development in the zebrafish (Rathbun et al., 2020). Here, we characterize Rab25, an epithelial specific membrane recycling protein that is a member of the Rab11 subfamily (Goldenring et al., 1993; Mitra et al., 2017). Rab25 has been implicated in cancer cell metastasis (Caswell et al., 2007; Mitra et al., 2017), but its role in embryonic tissue morphogenesis has not been examined. We demonstrate that maternal-zygotic Rab25a and Rab25b mutant embryos exhibit epithelial spreading delays. Here, we propose an EVL trafficking defect in mutant embryos resulted in delayed or failed intercellular bridge abscission during cytokinesis which slowed epiboly movements. Failure of timely bridge scission resulted in the progressive formation of multinucleate cells which remained mitotically active, resulting in heterogenous cell sizes with anisotropic cell shapes and spatial arrangements in the EVL. Abnormal EVL cells exhibited an overall reduction in cortical actin density during epiboly. Sparse actomyosin networks were associated with uncoordinated cell behaviors, balanced tensions, and altered viscoelastic responses during epiboly. Results Fluorescently tagged Rab25 localizes near the plasma membrane, centrosomes and cytokinetic midbody In zebrafish, rab25a and rab25b transcripts are maternally deposited and their levels increase during gastrulation (https://www.ebi.ac.uk/gxa/home). Whole-mount in situ hybridization revealed ubiquitous expression of both genes in the blastoderm prior to epiboly initiation, followed by EVL restricted expression during epiboly (Figure 1A). No expression was detected using sense control probes (Figure 1—figure supplement 1A). EVL expression is consistent with reported epithelial-specific function of Rab25 in mammals (Goldenring et al., 1993). Figure 1 with 1 supplement see all Download asset Open asset rab25a and rab25b expression pattern and subcellular localization. (A) Bright field images of whole-mount in situ hybridizations for rab25a (top row) and rab25b (bottom row), in WT embryos; lateral views with animal pole positioned to the top. (A') Section of a WT embryo showing rab25a expression restricted to the EVL (arrow, top row far right panel); transcripts absent from the deep cells (DEL) and yolk cell (YC). (B) Confocal z-projections of Venus-Rab25a and eGfp-Rab25b subcellular localization in WT embryos; red arrowheads denote perinuclear aggregates. Scale bar 20 μm. (C) Confocal z-projections of stills from time-lapse movies of transgenic Tubulin-GFP (green) embryos expressing mCherry-Rab25b (magenta). Arrowheads denote co-localization of mCherry-Rab25b at centrosomes. Scale bar 20 μm. (D) Confocal z-projections of eGfp-Rab25b (white) and mCherry-Mklp1 (teal) localization in WT embryos; box highlights enrichment of eGfp-Rab25b adjacent to the midbody. Scale bar 20 μm. To examine the dynamic subcellular distribution of Rab25a and Rab25b in the EVL, N-terminally fluorescently tagged Rab25a and Rab25b constructs were expressed in WT embryos by RNA injection at the one-cell stage. By live confocal microscopy, Venus-Rab25a and eGfp-Rab25b were observed at the plasma membrane and in motile cytoplasmic puncta (Figure 1B; Video 1). Between dome and 50% epiboly, when the EVL is most proliferative, Venus-Rab25a and eGfp-Rab25b appeared to localize near centrosomes during mitosis and then moved toward the opposing poles of dividing cells (Figure 1B, arrowheads; Video 1). Injection of mcherry-rab25b mRNA into transgenic animals expressing Tubulin-Gfp (Fei et al., 2019) confirmed that mCherry-Rab25b localized near centrosomes and dissipated following cell division (Figure 1C; arrows). During cytokinetic abscission, co-expression of eGfp-Rab25b with the midbody marker mCherry-Mklp1 demonstrated that eGfp-Rab25b dynamically localized within intercellular bridges adjacent to the midbody (Figure 1D), the sites of bridge scission. Video 1 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg N-terminally tagged-Rab25 constructs dynamics during cell proliferation and early epiboly. WT embryo expressing Venus-Rab25a (left panel) and eGfp-Rab25b (right panel) beginning at 30% epiboly; Scale bar 20 μm. Proliferation in the EVL largely subsides during late epiboly progression stages (Campinho et al., 2013). Consequently, both eGfp-Rab25b and Venus-Rab25a were primarily distributed near plasma membrane regions between 50 and 80% epiboly, as well as in small dynamic cytoplasmic puncta (Video 2). Notably, as marginal EVL cells intercalated into submarginal zones, Rab25 constructs became enriched at tricellular vertices contacting the yolk cell (Video 3). Overall, the redistribution of Rab25 constructs to centrosomes and the midbody are suggestive of a role in cell division. While Rab25 has not previously been shown to regulate these processes, the closely related Rab11 has well known functions in mitosis and cytokinesis (Hehnly and Doxsey, 2014; Rathbun et al., 2020). Furthermore, recruitment of fluorescent-Rab25 to cell vertices is similar to Rab11 distribution in the Xenopus laevis neuroepithelium (Ossipova et al., 2014), implicating Rab25 in cell shape changes during epiboly. Video 2 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg eGfp-Rab25b live distribution at the plasma membrane and cell vertices during late epiboly. WT embryo expressing eGfp-Rab25b beginning at 6hpf; Scale bar 20 μm. Video 3 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg Venus-Rab25a is dynamically recruited to vertices of rearranging marginal cells. WT embryo expressing Venus-Rab25a beginning at 7hpf; Scale bar 20 μm. Rab25a and Rab25b are required for normal epiboly movements To explore the functions of Rab25a and Rab25b, CRISPR/Cas9 gene editing was used to generate maternal-zygotic (MZ) mutant lines. Guide RNAs were designed to target exon two which encodes the GTPase domain, the functional domain of Rab proteins (Mitra et al., 2017). We characterized two rab25a mutant alleles from two founder fish, a 13-base pair (bp) deletion (2.3) and 29 bp insertion (4). Each allele contained a premature stop codon that disrupted the GTPase domain (see Materials and methods). An 18 bp deletion was generated in rab25b which produced an in-frame mutation and deleted a portion of the GTPase domain (see Materials and methods). Quantitative PCR analysis of transcript levels at shield stage showed undetectable levels of rab25a transcripts and reduced rab25b transcripts in MZrab25a and MZrab25b embryos, respectively (Figure 2—figure supplement 1A). Nonsense-mediated decay can result in genetic compensation of similar sequences hence we examined the levels of rab25a, rab25b and rab11a in mutant embryos (El-Brolosy and Stainier, 2017; Rossi et al., 2015). MZrab25a embryos had elevated levels of rab25b transcript, whereas MZrab25b mutants did not have elevated levels of rab25a transcript. rab11a transcript levels were similar to WT in both mutants. Thus, MZrab25a mutants appeared to exhibit genetic compensation in the form of transcriptional adaption of rab25b, whereas MZrab25b mutants did not. Double MZrab25a/rab25b embryos exhibited phenotypes similar to those of MZrab25b single mutants (Figure 2—figure supplement 1B) with moderate increases in the severity of the EVL defects compared to either MZrab25a or MZrab25b single mutants. These results support our hypothesis that genetic compensation by rab25b partially rescues MZrab25a mutant embryos, producing a milder phenotype than MZrab25b or MZrab25a/rab25b mutants. To examine the mutant phenotypes, WT, MZrab25a, and MZrab25b embryos were time-matched and examined live by light microscopy. A subset of MZrab25a (36/100) and MZrab25b (30/100) embryos exhibited abnormal blastoderm morphology during early blastula stages (Figure 2—figure supplement 1D). Notably, the majority of MZrab25a (20/36) and MZrab25b (20/30) mutant embryos with early blastoderm defects recovered by 4.3hpf and MZrab25a and MZrab25b embryos initiated epiboly on time (Figure 2A, 4.3hpf, dome). Following this, MZrab25a and MZrab25b mutants exhibited strong epiboly delays that worsened over time, with the greatest delays observed at late epiboly stages (Figure 2A, 6-9hpf, Figure 2B). The epiboly delay in MZrab25b embryos could be rescued by a transgenic construct expressing full length rab25b under the control of a ß-actin promoter, indicating that the phenotype results from the loss of Rab25b (Figure 2—figure supplement 1C). Figure 2 with 2 supplements see all Download asset Open asset Epithelial spreading delays and heterogenous cell morphology, size and spatial arrangements in MZrab25a and MZrab25b embryos. (A) Time-matched bright field images of lateral views of WT, MZrab25a and MZrab25b embryos during epiboly. Arrowheads indicate blastoderm margin, asterisks denote embryonic organizer (shield). (B) Quantification of epiboly progression after 8hpf in: WT (n = 80), MZrab25a (2.3, n = 87), MZrab25a (4,n = 28); WT (n = 97), Mrab25a (2.3,n = 73), MZrab25a (4,n = 49); WT (n = 29), MZrab25b (n = 21), Mrab25b (n = 18). Means: SEM; Two-Way ANOVA; ***p<0.001, ****p<0.0001.(N = 3). (C) Confocal z-projections of time-matched lateral views of WT, MZrab25a and MZrab25b embryos at 8hpf stained with phalloidin and corresponding apical surface area heat maps. Cooler colors represent smaller areas, warmer colors represent larger areas. Yellow boxes indicate cells with reduced apices. Red arrows denote cells with increased apical surface areas. White arrows indicate curved cell junctions. Scale bar 100 μm. (D) Frequency distribution of apical surface areas of WT (n = 817, N = 8), MZrab25a 2.3 (n = 651, N = 14), MZrab25a 4 (n = 654, N = 15) and MZrab25b (n = 503, N = 15) embryos at 6hpf. (E) EVL Cell number in WT (n = 8), MZrab25b (n = 8), MZrab25a 2.3 (n = 9) and MZrab25a 4 (n = 7) embryos at 6hpf. Means: SEM; Two-Way ANOVA; ***p<0.001. (F) Frequency distributions of EVL cellular contacts number at 6hpf in WT (n = 817, N = 8), MZrab25a 2.3 (n = 651, N = 14), MZrab25a 4 (n = 654, N = 15) and MZrab25b (n = 503, N = 15). (G) Circularity quantifications for WT and MZrab25b embryos during epiboly. 30% epiboly: WT (n = 7), Mrab25b (n=7). Shield: WT (n=14), Mrab25b (n=10). 80% epiboly: WT (n=8), Mrab25b (n=9). Means: SEM: One-way ANOVA, ****p<0.0001. The mutant phenotypes apparently did not result from a general developmental delay, as the zebrafish organizer formed at the same time as time-matched WT embryos (Figure 2A, 6hpf, white asterisk). During epiboly, deep cells are positioned behind the leading edge of the EVL and do not move past the EVL margin (reviewed in Bruce and Heisenberg, 2020). Examination of deep cell and EVL epiboly revealed they were equally delayed, suggesting the blastoderm delay could be the result of an EVL-specific defect (Figure 2—figure supplement 1E). Overall, we found that MZrab25b phenotypes were more severe than either of the MZrab25a alleles. Despite the strong epiboly delay, most MZrab25a (87/100) and MZrab25b (99/100) embryos completed gastrulation and survived to adulthood. We examined patterning and germ layer specification by in situ hybridization. EVL differentiation is required for normal epiboly (Fukazawa et al., 2010), and we found that expression of the EVL marker keratin4 (krt4) was similar in mutant and WT embryos (Figure 2—figure supplement 1F). Analysis of the mesoderm marker, goosecoid, and the endodermal marker sox17, also showed normal expression patterns (Figure 2—figure supplement 1F). Occasionally, sox17 staining showed that the dorsal forerunner cluster was disorganized in MZrab25a and MZrab25b embryos (Figure 2—figure supplement 1F). Overall, germ layer specification appeared to be largely normally in mutant embryos. Together these data suggested that Rab25a and Rab25b may have a specific role in epiboly during zebrafish gastrulation. Cell shape and rearrangements indicative of epithelial defects in MZrab25a and MZrab25b embryos The EVL restricted expression of rab25a and rab25b pointed to a primary defect in the EVL of Rab25 mutant embryos, consistent with the epithelial-specific role of Rab25 in mammals (Goldenring et al., 1993; Jeong et al., 2019). with an EVL analysis of phalloidin stained Rab25 mutant embryos revealed cellular size and shape compared to (Figure also revealed that yolk cell actin networks were in MZrab25a and MZrab25b embryos, suggesting yolk cell defects may contribute to the mutant phenotypes (Figure of the yolk cell can be the result of development and can impact the of epiboly et al., 2016). Thus, we examined the cortical actin networks of stage mutant embryos by phalloidin staining to While WT and MZrab25a yolk cell actin networks were a small number of MZrab25b embryos localized yolk cell cortical actin (Figure 2—figure supplement Thus, while we cannot the that actin defects in the to the yolk cell phenotypes, that they produced the defects in mutant EVL cells. In support of this, EVL cell morphology was normal at the of epiboly following cleavage stages in all mutant embryos examined (Figure 2—figure supplement By shield mutant EVL cells exhibited a of defects compared to (Figure 2—figure supplement Furthermore, these defects the greatest epiboly delays which between suggesting a between the EVL defects and the epiboly delays. Lastly, actin in the yolk cell of embryos failed to EVL defects that those in Rab25 mutant embryos (Figure 2—figure supplement all these that the EVL phenotypes in Rab25 mutant embryos are likely to be cell we our analysis on the EVL of MZrab25a and MZrab25b mutants during epiboly. embryos at epiboly were used to the EVL This stage was because that was far epiboly in most mutants by 8hpf and was also when embryos exhibited the largest delays (Figure EVL surface area heat and distributions were generated to the surface areas of cells in phalloidin stained mutant and WT embryos. MZrab25a and MZrab25b embryos exhibited heterogenous cell surface areas (Figure In WT, EVL cells had apical surface areas between and while the distribution of EVL surface areas in MZrab25a and MZrab25b mutant embryos was indicating both smaller and larger cell (Figure of apical surface area in mutants was associated with anisotropic cell shapes and disorganized spatial While WT cells were or as MZrab25a and MZrab25b cells exhibited cell shape and abnormal of (Figure This was by and cell-cell with larger indicating increased cell shape (Figure with increased apical surface area to have increased (Figure red whereas cells with reduced to be (Figure Notably, normally cells in mutants still anisotropic cell shapes and cell contacts (Figure white arrows). to WT embryos, the EVL defects were associated with reduced cell number in MZrab25b but not MZrab25a embryos (Figure Cell size is reported to scale with number et al., 2017). Thus, we examined mutant cells with large apical surface areas were by WT, MZrab25a and MZrab25b embryos with and the marker cells in MZrab25a and MZrab25b embryos were or multinucleate (Figure cell size with number (Figure Overall, multinucleate large cells suggested cytokinesis failures in MZrab25a and MZrab25b embryos (Rathbun et al., consistent with the of EVL defects following epiboly initiation, when the EVL is most proliferative (Campinho et al., 2013). Figure 3 Download asset Open asset cytokinetic bridges and delayed abscission in MZrab25b embryos. (A) Confocal z-projections of WT and MZrab25b embryos at 50% epiboly expressing (green) and (magenta). positioned Scale bar μm. (B) surface area quantifications of WT (n = N = MZrab25a binucleate (n = and MZrab25b = and multinucleate (n = at 80% epiboly. SEM; ANOVA; of stills from confocal of WT and MZrab25b embryos for and plasma membrane at dome white WT arrows bridge in MZrab25b cell. Red arrowheads indicate cytokinetic bridge open by neighboring morphogenic Scale bar 20 μm. (D) bridge in WT (n = N = and MZrab25b (n = N = embryos during epiboly in EVL marginal SEM; (E) bridge length following formation in WT (n = = and MZrab25b (n = = embryos during epiboly in EVL marginal SEM; abscission defects during epiboly in MZrab25b mutants of eGfp-Rab25b near the cytokinetic midbody and the of large EVL cells in MZrab25a and MZrab25b embryos that Rab25 may function in To investigate abscission defects in Rab25 we on MZrab25b embryos because they exhibited more severe defects than MZrab25a embryos. bridges were characterized by and membrane in WT and mutant embryos. Following mitosis in WT cells, were organized into apical cytokinetic bridges (Figure white Video cytokinetic bridges a of bridge and over (Figure = bridges then remained connected for an in which bridge length did not change abscission (Figure = Video 4 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg WT EVL WT embryo with and mRNA and (green) beginning at dome stage. Scale bar 20 μm. During cytokinesis in MZrab25b embryos, the plasma membrane at the cleavage similar to WT embryos (Figure Following two of abscission defects were The during the of bridge formation when bridge in a small of mutant cells during cytokinesis leading to the formation of binucleate cells (Figure Video In these MZrab25b embryos expressing the midbody marker showed a of mCherry-Mklp1 in cytokinetic which bridge (Video WT (left MZrab25b (right Video Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg bridge in Mrab25b Mrab25b embryo with and mRNA and (green) beginning at dome stage. Scale bar 20 μm. Video Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg formation of the cytokinetic midbody in MZrab25b embryos. expressing mCherry-Mklp1 (green) beginning at dome stage showing formation of midbody in of Mrab25b embryo expressing and mCherry-Mklp1 of the cytokinetic midbody which bridge Scale bar 20 μm. The defect later during the abscission Despite following their formation (Figure cytokinetic bridges in these appeared to and narrow normally (Figure red In WT embryos, daughter cells remained interconnected for bridge scission (n = while in bridges as suggesting abscission either failed or was delayed (Figure n = 15). While bridges were still present at the of the most intercellular bridges appeared to be open by in adjacent EVL as mitosis or cell resulting in (Figure red Video

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.