Aurora A and cortical flows promote polarization and cytokinesis by inducing asymmetric ECT-2 accumulation

Version of Record

Accepted for publication after peer review and revision.

Download
Cite
Share
CommentOpen annotations (there are currently 0 annotations on this page).

Version of Record published: December 29, 2022 (This version)
Accepted Manuscript published: December 19, 2022 (Go to version)
Accepted: December 12, 2022
Received: October 6, 2022
Preprint posted: May 25, 2022 (Go to version)

1. Of interest
Rab12 is a regulator of LRRK2 and its activation by damaged lysosomes

Xiang Wang, Vitaliy V Bondar ... Anastasia G Henry

Short Report Dec 8, 2023
Further reading

Abstract
Editor's evaluation
Introduction
Results
Discussion
Materials and methods
Data availability
References
Decision letter
Author response
Article and author information
Metrics

Abstract

In the early Caenorhabditis elegans embryo, cell polarization and cytokinesis are interrelated yet distinct processes. Here, we sought to understand a poorly understood aspect of cleavage furrow positioning. Early C. elegans embryos deficient in the cytokinetic regulator centralspindlin form furrows, due to an inhibitory activity that depends on aster positioning relative to the polar cortices. Here, we show polar relaxation is associated with depletion of cortical ECT-2, a RhoGEF, specifically at the posterior cortex. Asymmetric ECT-2 accumulation requires intact centrosomes, Aurora A (AIR-1), and myosin-dependent cortical flows. Within a localization competent ECT-2 fragment, we identified three putative phospho-acceptor sites in the PH domain of ECT-2 that render ECT-2 responsive to inhibition by AIR-1. During both polarization and cytokinesis, our results suggest that centrosomal AIR-1 breaks symmetry via ECT-2 phosphorylation; this local inhibition of ECT-2 is amplified by myosin-driven flows that generate regional ECT-2 asymmetry. Together, these mechanisms cooperate to induce polarized assembly of cortical myosin, contributing to both embryo polarization and cytokinesis.

Editor's evaluation

Cytokinesis in animals involves a contractile actomyosin ring, which generates the forces needed for cell division. A key factor controlling actomyosin ring function is a protein called ECT2, which is a regulator of the signalling protein RhoA GTPase. How ECT2 gets to the correct cellular location and how it executes its functions remain a mystery, despite extensive work. This work through a set of beautiful and thorough experiments establishes the mechanism of ECT2 intracellular distribution, which involves integration of spatial and biochemical signals all contributing to the fidelity of cell division.

https://doi.org/10.7554/eLife.83992.sa0

Introduction

The division of a single animal cell into two daughter cells is facilitated by an actomyosin-based contractile ring that generates the force that constricts the cell membrane and generates the cleavage furrow (Basant and Glotzer, 2018). Furrow positioning is dictated by the position of the mitotic spindle (Rappaport, 1985). Work in Caenorhabditis elegans has shown that distinct parts of the spindle, namely the spindle midzone and the asters, are each sufficient to induce furrowing (Dechant and Glotzer, 2003; Bringmann and Hyman, 2005). The active form of the small GTPase RhoA (C.e. RHO-1) is necessary for cleavage furrow induction in these and other contexts (Kishi et al., 1993; Jantsch-Plunger et al., 2000; Zhou and Zheng, 2013). An experimentally induced equatorial zone of active RhoA suffices to induce a furrow in cultured human cells (Wagner and Glotzer, 2016). Despite progress in our understanding of cytokinesis, gaps remain in our understanding of the mechanism by which asters spatially regulate actomyosin contractility.

The guanine nucleotide exchange factor (GEF) Ect2 (C.e. ECT-2) is the primary activator of RhoA during cytokinesis (Tatsumoto et al., 1999). This Dbl-homology family member is regulated by numerous factors. Prominent among these is the centralspindlin complex which, despite containing a subunit with a RhoGAP domain, plays a significant role in RhoA activation during cytokinesis. Centralspindlin both induces relief of ECT-2 auto-inhibition and promotes Ect2 recruitment to the cortex (Yüce et al., 2005; Zhang and Glotzer, 2015; Niiya et al., 2005; Zhao and Fang, 2005; Chen et al., 2019; Su et al., 2011; Lee et al., 2018). Anti-parallel, bundled microtubule plus ends, localized kinases, and a phosphatase conspire to promote centralspindlin-dependent RhoA activation at the cell equator, midway between the two spindle poles (Wolfe et al., 2009; Burkard et al., 2009; Yüce et al., 2005; Basant et al., 2015; Hertz et al., 2016; Nishimura and Yonemura, 2006).

Despite the established role of the spindle midzone in cytokinesis in diverse metazoa, furrow formation can occur in its absence in some cell types. This is well characterized in C. elegans embryos where the centralspindlin-dependent and -independent furrows have been separated genetically (Dechant and Glotzer, 2003; Werner et al., 2007). Analogous phenomena operate in cultured human cells (Chen et al., 2021; Murthy and Wadsworth, 2008; Ramkumar et al., 2021; Rodrigues et al., 2015). In early C. elegans embryos, centralspindlin functions in parallel with NOP-1 to promote ECT-2-dependent RHO-1 activation (Morton et al., 2012; Tse et al., 2012; Fievet et al., 2012). Though it is non-essential, NOP-1 can promote formation of an incomplete furrow (Tse et al., 2012). In certain circumstances, NOP-1 suffices to induce complete furrow ingression (Canman et al., 2008; Loria et al., 2012; Zhang and Glotzer, 2015).

In addition to being genetically separable, the centralspindlin- and NOP-1-dependent pathways can be spatially separated. Whereas the position of the centralspindlin-dependent furrow correlates with the spindle midzone (Werner et al., 2007), the position of the NOP-1 dependent furrow anti-correlates with the position of the spindle asters (Hird and White, 1993; Dechant and Glotzer, 2003; Werner et al., 2007; Tse et al., 2012). Specifically, spindle asters are associated with inhibition of contractility. This inhibitory activity has generally been ascribed to the astral microtubules and is often referred to as polar relaxation/astral relaxation (Wolpert, 1960; White and Borisy, 1983; Dechant and Glotzer, 2003; Werner et al., 2007; Mangal et al., 2018; Chapa-Y-Lazo et al., 2020). In unperturbed cells, centralspindlin-dependent and -independent furrows align during cytokinesis.

Regional suppression of cortical contractility also occurs during polarization of the C. elegans zygote. After fertilization, the sperm-derived centrosome serves as a symmetry-breaking cue that inhibits cortical contractility (Goldstein and Hird, 1996; Cowan and Hyman, 2004). This reduction in cortical contractility induces cortical flows that advect PAR proteins, such as PAR-3 and PAR-6, to the nascent anterior of the embryo (Munro et al., 2004). Centrosomes locally concentrate Aurora A kinase (C.e. AIR-1), which is required for centrosome-directed displacement of the RhoGEF ECT-2 from the posterior cortex during polarization (Motegi and Sugimoto, 2006; Zhao et al., 2019; Klinkert et al., 2019; Kapoor and Kotak, 2019). However, the mechanism of action of Aurora A is not known, in particular it is not known whether Aurora A regulates ECT-2 during cytokinesis.

To address these questions we quantitatively examined ECT-2 distribution during polarization and cytokinesis in early C. elegans embryos. We show that ECT-2 becomes asymmetrically localized coincident with the early stages of polarity establishment, as it becomes depleted posteriorly and enriched anteriorly. The degree of ECT-2 asymmetry attenuates somewhat as the embryo progresses through mitosis and then increases again during anaphase, under the influence of the asymmetric spindle. As during polarization, ECT-2 asymmetry during anaphase requires intact centrosomes and AIR-1. A C-terminal fragment of ECT-2 largely recapitulates the dynamic localization of ECT-2 during early embryogenesis. A set of putative Aurora A sites have been identified in this fragment that are required for its displacement by centrosomes and Aurora A. While phosphorylation of ECT-2 by Aurora A serves as a symmetry breaking cue, ECT-2 asymmetry also requires myosin-dependent cortical flows. Our results suggest that Aurora A-triggered phosphorylation of ECT-2 initiates cortical flows which redistribute actomyosin and associated factors, including ECT-2, thereby regulating polarization and cytokinesis.

Results

Cortex-associated ECT-2 is asymmetric during cytokinesis

To dissect the mechanism by which spindle asters regulate contractility, we first examined the distribution of endogenous ECT-2 during early embryogenesis. Displacement of ECT-2 from the posterior cortex during polarization has been previously described in studies utilizing an ECT-2:GFP transgene. Collectively, these works revealed that centrosomes and/or their associated microtubules induce displacement of ECT-2 during polarization (Motegi and Sugimoto, 2006; Kotak et al., 2016; Kapoor and Kotak, 2019). To further investigate the dynamics of ECT-2 localization, we engineered a mNeonGreen (mNG) tag at the C-terminus of endogenous ECT-2 (Dickinson et al., 2013). ECT-2::mNG is well tolerated (>99% embryos hatch).

During both polarization and cytokinesis, we quantified the levels of cortical ECT-2 along the entire perimeter of the embryo (Figure 1—figure supplement 1). For each time point, we averaged cortical ECT-2 levels in the anterior and posterior 20% of the embryo, and calculated the ratio of these values as a measure of ECT-2 asymmetry. In the early zygote, the cortical pool of ECT-2 increases and it becomes progressively more asymmetric (Figure 1B i–iii). Images acquired during polarity establishment reveal displacement of ECT-2 from a large domain in posterior cortex and a progressive increase in the anterior cortex resulting in highly asymmetric cortical ECT-2 (Figure 1A, –480 s [relative to anaphase onset], Figure 1Bi, ii,vi). The size of the posterior domain where ECT-2 is displaced, that is, boundary length, represents ~35% of the embryo, measured at the time point of maximal ECT-2 asymmetry. ECT-2 accumulation on the anterior cortex is accompanied by an increase in cortical myosin accumulation (Figure 1—figure supplement 2), cortical contractility and pseudocleavage furrow formation, and transient elongation of the embryo (Figure 1B ii, v). The pseudocleavage furrow forms at the boundary between the myosin-enriched anterior and the myosin-depleted posterior of the embryo; myosin does not solely accumulate at the furrow.

Figure 1 with 4 supplements see all

Download asset Open asset

Cortical ECT-2 becomes asymmetric during polarization and cytokinesis.

(A) ECT-2 accumulation is asymmetric during polarization. Selected time points from a representative embryo expressing ECT-2:mNeonGreen (mNG) and mCherry:Tubulin (insets) during polarity establishment. Time zero represents anaphase onset. Blue circles indicate pronuclei and red asterisks label the approximate locations of the centrosomes. All scale bars represent 10 µm. (B) Quantification of (i) average anterior:posterior ratio of anterior and posterior cortical accumulation, (ii) average anterior cortical accumulation, (iii) average posterior cortical accumulation, (iv) average cortical accumulation over the entire cortex, and (v) normalized perimeter of embryos over time (s). n=8; solid line is the average of all embryos measured and the shaded ribbon represents the 95% confidence interval (95 CI). (vi) Box plot of boundary length. Boundary length is the fraction of the embryo cortex with normalized ECT-2 levels below threshold calculated at the time point of the maximum anterior:posterior ratio of ECT-2 for each embryo (see Materials and methods for details). Source data is available (Figure 1—source data 1). (C) ECT-2 accumulation is asymmetric during cytokinesis. A representative embryo expressing endogenously tagged ECT-2:mNeonGreen (mNG) at the indicated time points during cytokinesis beginning at anaphase onset (0 s). Red asterisks label the approximate locations of the centrosomes. (D) Quantification of the (i) average anterior:posterior ratio, (ii) average anterior cortical accumulation, (iii) average posterior cortical accumulation, and (iv) average cortical accumulation. n=11; solid line represents the average of all embryos and the shaded ribbon represents the 95% confidence interval (95 CI). (v) Box plot of boundary length at the most asymmetric time point of ECT-2 for each embryo. Source data is available (Figure 1—source data 1).

Figure 1—source data 1 This source data table consists of the measurements of ECT-2 accumulation as a function of time. Specifically, for each embryo at each time point, the table includes the average accumulation of ECT-2 in the anterior and posterior 20% of the embryo normalized to the cytoplasmic level and the ratio of these values [AntRatio, PostRatio,AntPostRatio]. The graphs in the figure present the average of these values for the set of embryos for each treatment at each time point and plotted with a 95% confidence intervals. The treatments are described in the respective figure legend.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig1-data1-v2.zip
Download elife-83992-fig1-data1-v2.zip

Following pronuclei meeting and centration, overall cortical ECT-2 levels fall and the embryo partially rounds up (Figure 1B iv, v). ECT-2::mNG remains enriched on the anterior cortex during prophase and mitosis, and it remains detectable and slightly increases on the posterior cortex (Figure 1A, –240 s, Figure 1B, i-vi). Thus, as a consequence of its prior polarization, ECT-2 is asymmetrically localized before the onset of cytokinesis. During anaphase, ECT-2 again becomes preferentially displaced from the posterior ~35% of the embryo, becoming more asymmetric (Figure 1C, 60 s, Figure 1D, i-iii). As furrow ingression begins, ECT-2 is not significantly enriched on the posterior cortex above its level in the cytoplasm (Figure 1C 100 s). A membrane probe, mCherry-PH, which detects PIP2, also exhibits anterior enrichment, though to a lesser degree than ECT-2::mNG and its levels remain stable as anaphase progresses (Figure 1—figure supplement 3; Hirani et al., 2019). Regional displacement of ECT-2 repeats in subsequent cell cycles; the cortical pool of ECT-2 decreases each cell cycle near spindle poles at anaphase and recovers in the subsequent cell cycles (Figure 1—figure supplement 4). Thus, during both polarization and cytokinesis, cortical ECT-2 is highly asymmetric.

ECT-2 distribution during cytokinesis is regulated by centrosomes

Microtubule-rich asters suppress contractility in the polar regions during cytokinesis (Werner et al., 2007). During spindle elongation, a correlation between spindle rocking and ECT-2 displacement is apparent: as the spindle pole more closely approaches one region of the cortex, the ECT-2 on the cortex in that region is preferentially displaced following a short (~10 s) delay (Figure 2A).

Figure 2 with 2 supplements see all

Download asset Open asset

ECT-2 asymmetry is responsive to the position of the spindle.

(A) A close-up view of the localization of endogenously tagged ECT-2 and tubulin during first cleavage. Images from a time-lapse series show the distribution of ECT-2:mNG and mCherry:Tubulin. Composite images are shown: ECT-2 shown in using the displayed lookup table (LUT); the tubulin signal is displayed with inverted grayscale. In the overlays, the tubulin image is advanced by 10 s relative to ECT-2. Arrows highlight two sites where local accumulation of ECT-2 declines ~10 s after the spindle approaches, as indicated by the color change in the arrows from green to red. (B) ECT-2:mNG localization in embryos under control [L4440 + DMSO] or experimental [*tba-2+tbb-2*(RNAi) + 50 µg/mL nocodazole] conditions. The images are stills from a time lapse showing ECT-2:mNG localization while the insets are mCherry:Tubulin images acquired concurrently. Asterisks label the approximate locations of the asters. (i–iii) The graphs measure the anterior:posterior accumulation ratio, anterior accumulation, and posterior accumulation of ECT-2 relative to the cytoplasm over time starting at NEBD with 95% confidence interval (CI) labeled as shaded ribbons. (iv) The box plot graphs posterior boundary length of ECT-2 at the maximum polarized time point while (v) the line graph plots average accumulation at the same time point as a function of cortical position relative to the anterior of the embryo. For this and all subsequent figures, asterisks on graphs indicate statistically significant differences between wild-type and experimental conditions at each indicated time point (*p<0.05, **p<0.01,***p<0.001, ****p<0.0001). Where multiple comparisons are present, the asterisk color reflects the experimental condition compared to wild-type. All scale bars represent 10 µm. Source data is available (Figure 2—source data 1). (C) ECT-2 localization in an *spd-5*(RNAi) and wild-type embryos beginning at NEBD. The insets show images of mCherry:Tubulin acquired at the same time points. The red asterisks label the approximate locations of the asters while the red triangles label the approximate locations of the compromised asters. The graphs measure the anterior:posterior accumulation ratio, anterior accumulation, and posterior accumulation of ECT-2 relative to the cytoplasm over time. Source data is available (Figure 2—source data 1). (D) ECT-2 localization in a *par-3*(RNAi) and wild-type embryos with images starting at anaphase onset. Red asterisks label the approximate locations of the asters. The graphs measure the anterior:posterior accumulation ratio, anterior accumulation, and posterior accumulation of ECT-2 relative to the cytoplasm over time. Source data is available (Figure 2—source data 1).

Figure 2—source data 1 The source data is as described in legend to Figure 1—source data 1.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig2-data1-v2.zip
Download elife-83992-fig2-data1-v2.zip

These results suggest that cortical levels of ECT-2 are modulated by the aster, though they do not distinguish between the role of the centrosome or the astral array of microtubules. To determine whether microtubules are required, embryos were depleted of both alpha- and beta-tubulin and treated with nocodazole to depolymerize microtubules to the greatest extent possible. Despite significant depletion of tubulin and near-complete depolymerization of microtubules (Figure 2B, insets), we observed strong displacement of ECT-2 from a broad region of the posterior cortex and the initiation of cortical contractility following mitotic exit (Figure 2B). In these embryos, the centrosomes could be identified by the residual signal from mCherry-tubulin, and the posterior displacement was centered around the position of the centrosome. ECT-2 becomes hyper-asymmetric in these embryos as compared to control embryos and the depleted zone comprises a larger region of the embryo (Figure 2B), as is also observed in embryos with a diminished, posteriorly positioned spindle (see below).

To further examine the role of the centrosome in ECT-2 asymmetry, we depleted the centrosomal scaffold protein SPD-5. Embryos depleted of SPD-5 lack organized arrays of microtubules, though some microtubules assemble in a poorly organized manner (Hamill et al., 2002; Pelletier et al., 2004). ECT-2 asymmetry is reduced in these embryos both during polarization, as previously reported (Kapoor and Kotak, 2019), and during cytokinesis; in particular, significant ECT-2 remains on the posterior cortex (Figure 2C iii). Collectively, these studies suggest that ECT-2 asymmetry during anaphase is centrosome-directed.

The evidence presented thus far indicates that centrosomes are the primary source of an activity that induces ECT-2 displacement during anaphase. Although the embryo contains two centrosomes at the time of the first cell division, one in the anterior, the other in the posterior, ECT-2 is preferentially displaced from the posterior cortex. Two, non-mutually exclusive explanations could underlie these differential responses. First, the centrosomes could be functionally distinct as a consequence of the disparate ages and maturity of the embedded centrioles – and assuming that the centrosomes undergo age-dependent stereotyped anterior-posterior positioning in the early embryo (Yamashita et al., 2007). Alternatively, the two centrosomes may be functionally equivalent, but, due to the underlying polarity of the embryo, the posterior centrosome is more closely opposed to the posterior cortex. To test the second model, we depleted embryos of PAR-3, to disrupt PAR protein polarity, allowing the centrosomes to be situated equidistant to the cortices on the anterior and posterior regions of the embryo (Etemad-Moghadam et al., 1995). Upon anaphase onset, embryos depleted of PAR-3 exhibit displacement of ECT-2 from both the anterior and posterior domains (Figure 2D), indicating that both centrosomes are competent to inhibit ECT-2 accumulation. Likewise, ECT-2 asymmetry during cytokinesis is reduced in embryos depleted of PAR-2 (Figure 2—figure supplement 1). To modulate spindle positioning without disrupting polarity, we depleted the Gα proteins that function redundantly to promote anaphase spindle elongation (Gotta and Ahringer, 2001). In Gα-depleted embryos, ECT-2 accumulation is asymmetric during M-phase, but upon mitotic exit, ECT-2 dissociates slowly from the anterior cortex, as in the wild-type, yet its dissociation from the posterior cortex is significantly attenuated (Figure 2—figure supplement 2), due to the reduced spindle elongation toward the posterior cortex. As a consequence, rather than ECT-2 accumulation becoming more asymmetric as anaphase progresses as in the wild-type, it becomes nearly symmetric. Collectively, these data indicate that, during cytokinesis, displacement of ECT-2 from the cortex is highly sensitive to centrosome-cortex separation.

Cortical flows contribute to asymmetric ECT-2 during polarization and cytokinesis

During polarization, the sperm-derived centrosomes that trigger asymmetric accumulation of ECT-2 are small, while ECT-2 is broadly displaced from the posterior region of the cortex, comprising ~35% of the embryo cortex (Figure 1, –480 s). Likewise, during cytokinesis, the zone of ECT-2 depletion is nearly as broad, and further increased when microtubules are depolymerized (Figure 2B). These observations suggest that ECT-2 asymmetry may involve processes that amplify the local centrosomal cue.

Anterior-directed, actomyosin-dependent, cortical flows of yolk granules and anterior PAR proteins are associated with both polarization and cytokinesis (Hird and White, 1993; Munro et al., 2004). To test whether cortical flows contribute to the displacement of ECT-2 from the posterior cortex during polarization, we examined ECT-2:mNG distribution in nop-1 mutant embryos which have attenuated cortical flows during polarization (Rose et al., 1995). In such embryos, ECT-2:mNG accumulates asymmetrically, suggesting basal flow rates suffice (Figure 3A).

Figure 3 with 4 supplements see all

Download asset Open asset

Cortical flows contribute to asymmetric cortical accumulation of ECT-2.

(A) Asymmetric accumulation of ECT-2 during polarization is reduced when cortical flows are reduced. Images from a time-lapse series show ECT-2 localization during polarization in a wild-type, *nmy-2*(RNAi), or *nop-1(it142*) embryo. The pronuclei are labeled with a blue circle and the asterisks label the approximate locations of the asters. Green arrowheads depict local reduction in ECT-2 as the centrosome is opposed to the posterior cortex. Graphs show the anterior:posterior accumulation ratio, anterior accumulation, and posterior accumulation of ECT-2 over time. The box plot depicts the posterior ECT-2 boundary at the time point of maximal asymmetry. Source data is available (Figure 3—source data 1). (B) Asymmetric ECT-2 accumulation is independent of its activators during cytokinesis. Images are stills from time-lapse movies depicting endogenous ECT-2 localization in wild-type, *nop-1(it142*), *cyk-4*(RNAi), and *nop-1(it142); cyk-4*(RNAi) embryos with time 0 as anaphase onset. Asterisks label the approximate locations of the asters. Graphs measure ECT-2 accumulation relative to average cytoplasm intensity. The boundary length of ECT-2 accumulation was measured at the most asymmetrical time point. Source data is available (Figure 3—source data 1). (C) The localization of endogenously tagged ECT-2 during cytokinesis is hyper-asymmetric in *zyg-9(b244*) embryos. Images from a time-lapse series show the distribution of ECT-2:mNG in either wild-type or *zyg-9(b244). zyg-9* loss of function results in a posteriorly localized small transverse spindle. Times shown are relative to NEBD which is approximately ~125 s prior to anaphase onset. Red asterisks label the approximate locations of the asters. The NMY-2:GFP images depict the time-dependent changes in cortical myosin in wild-type and *zyg-9*(RNAi) embryos. The images shown are maximum intensity projections of ~25 time points acquired over ~50 s during early anaphase. Each time point is itself a maximum intensity projection of four images separated by 1 µm in Z. Graphs measure ECT-2 accumulation relative to average cytoplasm intensity. Source data is available (Figure 3—source data 1). (D) Cortical ECT-2 exchanges rapidly. Fluorescence recovery after photobleaching (FRAP) recovery of ECT-2 at the anterior cortex during cytokinesis and on the cortex during polarization. A region of the cortex was imaged and a small ROI was bleached. Twenty-five images were taken at 100 ms intervals before the region was bleached followed by 175 images at 100 ms intervals to document the recovery. The kymograph qualitatively shows the FRAP recovery and the graph shows fitted curves of the recovery of each embryo measured. Recovery was fitted to y=a*(1-exp(-b*x)). The bold lines are the averages of the embryos at either polarization or cytokinesis as indicated (T_1/2 ~ 3 s). Source data is available (Figure 3—source data 1).

Figure 3—source data 1 This file contains the source data for Figure 3A–C is as described in legend to Figure 1—source data 1. This file also contains the source data for Figure 3D which includes the individual and average a and b coefficients used to plot the fluorescence recovery after photobleaching (FRAP) data.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig3-data1-v2.zip
Download elife-83992-fig3-data1-v2.zip

To more strongly disrupt cortical flows during polarization, we depleted non-muscle myosin, NMY-2, in embryos expressing ECT-2:mNG (Munro et al., 2004). In embryos with reduced myosin-dependent flows, ECT-2:mNG asymmetry is highly attenuated, though a local inhibitory effect on ECT-2:mNG accumulation in the immediate vicinity of the centrosome is detectable; these results are consistent with an earlier report (Motegi and Sugimoto, 2006; Figure 3A and Figure 3—figure supplement 1). These results suggest that cortical flows amplify the local, centrosome-induced reduction in cortical ECT-2 accumulation.

During cytokinesis, ECT-2 activation requires two factors that function in parallel to promote RHO-1 activation, the centralspindlin subunit CYK-4 and NOP-1 (Tse et al., 2012). Redistribution of ECT-2 from the posterior cortex could reflect the displacement of ECT-2 in complex with either one or both of these activators. To test each of these possibilities, we examined the distribution of ECT-2 during cytokinesis in embryos in which the functions of NOP-1 and CYK-4 are reduced, either alone or in combination. ECT-2 displacement and asymmetric accumulation was not abolished by these perturbations (Figure 3B). Thus, basal levels of RHO-1 activation suffice to promote displacement of ECT-2 from the posterior cortex during anaphase.

Enhanced cortical flows have been reported during cytokinesis in embryos with attenuated astral microtubules, such as nocodazole-treated embryos (Figure 2B; Hird and White, 1993), embryos with posteriorly positioning spindles due to persistent katanin activity (Werner et al., 2007) or depletion of ZYG-9. These perturbations result in suppression of cortical contractility in the posterior cortex and enhancement of contractility in the anterior cortex. These changes induce enhanced cortical flows of myosin toward the anterior that drive formation of an ectopic furrow in the anterior at the boundary between regions of high and low myosin accumulation (Figure 3C). Posteriorly positioned spindles also induce a centralspindlin-dependent furrow coincident with the spindle midzone (Werner et al., 2007). To quantitatively characterize these flows, we assayed the movement of myosin (NMY-2) punctate in wild-type and ZYG-9-depleted embryos. As compared to control embryos, ZYG-9-depleted embryos contain more cortical myosin foci which travel more than 1 µm, and these foci also move faster and travel further (Figure 3—figure supplement 3). To test whether enhanced cortical flows influence ECT-2 localization, we examined ECT-2::mNG in zyg-9(b244ts) embryos. Under these conditions, during anaphase, ECT-2 becomes hyper-asymmetric with a stronger reduction in the posterior cortical pool of ECT-2, an increase in the anterior cortical pool, and an expansion of the posterior domain depleted of ECT-2 (Figure 3C). As expected, the two poles of the small spindle appear to suppress ECT-2 accumulation equally. We conclude that cortical flows contribute to ECT-2 asymmetry.

Polar relaxation was proposed to result from dynein-mediated stripping of cortical myosin that generates defects in the cortical network, resulting in anterior-directed flows, contributing to polar relaxation (Chapa-Y-Lazo et al., 2020). However, asymmetric ECT-2 accumulation is not abrogated by depletion the heavy chain of cytoplasmic dynein, DHC-1. To the contrary, DHC-1 depletion, which would be predicted to compromise dynein-mediated stripping of cortical myosin, causes enhanced ECT-2 asymmetry, akin to the situation in embryos in which microtubule assembly is strongly suppressed, presumably due to the close association of the centrosomes with the posterior cortex (Figure 3—figure supplement 2). Thus, not only are cortical flows required for asymmetric ECT-2, increases or decreases in cortical flows enhance or attenuate the extent of ECT-2 asymmetry, respectively.

To understand how cortical flows promote the asymmetric accumulation of ECT-2, we sought to determine whether ECT-2 itself undergoes long range flows or if ECT-2 dynamically interacts with components that undergo such flows. To that end, we performed fluorescence recovery after photobleaching (FRAP) assays on ECT-2:mNG to measure the dynamics of its cortical association. Following bleaching, ECT-2:mNG rapidly recovers with a t_1/2 of ~3 s; with no obvious difference detected between polarization and cytokinesis. We conclude that ECT-2 asymmetry is likely to reflect the dynamic association of ECT-2 with more stably associated cortical component(s) that undergo cortical flow (Figure 3D).

Although much of ECT-2 is highly dynamic, we found one context in which ECT-2 associates more stably with the cortex. In ZYG-9-deficient embryos, despite the close proximity of centrosomes, ECT-2:mNG accumulates on the furrow that forms at the posterior pole, in a centralspindlin-dependent manner (Werner et al., 2007). FRAP analysis of the posterior and anterior furrows indicates that in both cases the mobile fraction of ECT-2 recovers quickly, but only ~50% of ECT-2 on the posterior furrow recovers, whereas the anterior furrow more fully recovers (Figure 3—figure supplement 4).

AIR-1, an Aurora A kinase, regulates ECT-2 asymmetry during cytokinesis

To further examine the origin of ECT-2 asymmetry, we examined the impact of Aurora A kinase (AIR-1) on ECT-2 distribution during cytokinesis. AIR-1 has been shown to play a central role in embryo polarization (Kapoor and Kotak, 2019; Zhao et al., 2019; Klinkert et al., 2019). Although AIR-1 functions in polarity establishment, AIR-1-depleted embryos are not unpolarized. Rather, in most such embryos, due the existence of a parallel, pathway (Motegi et al., 2011), posterior PAR proteins become enriched at both the anterior and the posterior cortex, and cortical flows emanate from both poles toward the equator (Kapoor and Kotak, 2019; Zhao et al., 2019; Klinkert et al., 2019; Reich et al., 2019). As a consequence, depletion of AIR-1 by RNAi reduces, but does not abolish, ECT-2 asymmetry during cytokinesis (Figure 4A). The residual asymmetry in ECT-2 accumulation in AIR-1-depleted embryos is further reduced by co-depletion of PAR-2 (Figure 4—figure supplement 2).

Figure 4 with 2 supplements see all

Download asset Open asset

Asymmetric ECT-2 accumulation requires AIR-1.

(A) AIR-1 promotes asymmetric cortical accumulation of ECT-2. Images from a time-lapse series show the localization of endogenously tagged ECT-2 during cytokinesis in either wild-type or embryos depleted of Aurora A kinase, AIR-1, by RNAi. The red asterisks mark the approximate location of the asters while the red triangles mark the center of the unorganized asters. Quantitation of the accumulation of cortical ECT-2:mNG. Note that NEBD is delayed in AIR-1-depleted embryos resulting in contractility at an earlier time point (Hachet et al., 2007; Portier et al., 2007). Source data is available (Figure 4—source data 1). (B) TPXL-1 contributes to inhibition of ECT-2 accumulation. Images from a time-lapse series show the accumulation of ECT-2:mNG in either a wild-type or *tpxl-1*(RNAi) embryo. Graphs quantify the cortical accumulation of ECT-2 relative to cytoplasm. Source data is available (Figure 4—source data 1). (C) AIR-1, but not TPXL-1, is required for inhibition of ECT-2 accumulation. ECT-2 was imaged in a *zyg-9(b244*) background in control embryos or embryos depleted of AIR-1 or TPXL-1. Images are shown from a time-lapse acquisition of a representative embryo with NEBD set as time 0. The graphs depict ECT-2 accumulation relative to the cytoplasm and the box plot quantifies the boundary length of ECT-2 inhibition. Note that NEBD is delayed in AIR-1-depleted embryos resulting in contractility at an earlier time point. Source data is available (Figure 4—source data 1). (D) ECT-2 cortical accumulation is reduced when AIR-1 activity is increased by *saps-1*(RNAi). Images from a time-lapse series show the accumulation of ECT-2:mNG in either a wild-type or *saps-1*(RNAi) embryo with anaphase onset as time 0. Quantitation shows ECT-2 accumulation over time. Source data is available (Figure 4—source data 1).

Figure 4—source data 1 The source data is as described in legend to Figure 1—source data 1.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig4-data1-v2.zip
Download elife-83992-fig4-data1-v2.zip

The reduction in cytokinetic ECT-2 asymmetry upon depletion of AIR-1 may reflect a direct role of AIR-1 in ECT-2 localization or it may result from the small spindles that assemble in such embryos (Hannak et al., 2001), which, in turn, cause the centrosomes to lie more distal from the cortex. To distinguish between these possibilities, we depleted AIR-1 in embryos defective for ZYG-9 function. Although the spindle assembles close to the posterior pole, it does not induce the pronounced ECT-2 asymmetry observed in the presence of AIR-1 (Figure 4C). We conclude that AIR-1 promotes displacement of ECT-2 from the cortex during anaphase.

Given that AIR-1 has multiple functions in centrosome maturation, microtubule stability, and embryo polarization, the effect of AIR-1 depletion on ECT-2 accumulation might result from indirect effects at an earlier stage of the cell cycle (Hannak et al., 2001; Zhao et al., 2019). To rule out these possibilities, we used a chemical inhibitor of AIR-1, MLN8237 (Sumiyoshi et al., 2015). Treatment of embryos with MLN8237 at metaphase resulted in a rapid enhancement in cortical ECT-2 accumulation during cytokinesis, indicating that AIR-1 functions during anaphase to promote ECT-2 displacement from the cortex (Figure 4—figure supplement 1).

Some, but not all, of AIR-1 functions involve its binding partner, TPXL-1, an ortholog of Tpx2 (Ozlü et al., 2005). Tpx2 is a direct activator of Aurora A kinase activity (Zorba et al., 2014). Furthermore, TPXL-1 prominently decorates astral microtubules (Ozlü et al., 2005), which are positioned where they could conceivably modulate cortical ECT-2 accumulation. Indeed, TPXL-1 has been proposed to promote the displacement of contractile proteins from the anterior cortex of the early C. elegans embryo (Mangal et al., 2018). To investigate whether an AIR-1/TPXL-1 complex plays a role in ECT-2 displacement during cytokinesis, we depleted TPXL-1 in embryos expressing ECT-2::mNG. Overall cortical ECT-2 accumulation is enhanced by TPXL-1 depletion, though the degree of ECT-2 asymmetry is unaffected (Figure 4B). TPXL-1 depletion results in a number of phenotypes, including reduced spindle length during metaphase (Lewellyn et al., 2010), which, like AIR-1 depletion, might indirectly impact ECT-2 accumulation by increasing the distance between spindle pole and the cortex. To distinguish between these possibilities, we examined ECT-2 localization in embryos deficient in both ZYG-9 and TPXL-1 so that the attenuated spindle assembles close to the embryo posterior. Under these conditions, we observed robust depletion of ECT-2 at the posterior pole in zyg-9(b244) embryos depleted of TPXL-1, but not AIR-1 (Figure 4C). We conclude that while AIR-1 is a major regulator of the asymmetric accumulation of ECT-2, the TPXL-1/AIR-1 complex does not play a central role in this process.

To further explore the function of AIR-1 in ECT-2 regulation, we sought to enhance AIR-1 activity. To that end, we depleted PPH-6 or SAPS-1, the two major subunits of the protein phosphatase 6 (Afshar et al., 2010). PPH-6 interacts with AIR-1 and is implicated in dephosphorylation of the activation loop of AIR-1 (Kotak et al., 2016). Prior work revealed that depletion of either PPH-6 or SAPS-1 blocks pseudocleavage, reduces the accumulation of cortical myosin, and attenuates the forces that drive spindle elongation (Afshar et al., 2010) and more recent work demonstrates that AIR-1 is epistatic to PPH-6 (Kotak et al., 2016). As expected for a negative regulator of AIR-1, SAPS-1 depletion results in an overall reduction in cortical ECT-2 (Figure 4D). Nevertheless, the cortical pool of ECT-2 remains somewhat asymmetric in SAPS-1-depleted embryos. These results indicate that AIR-1 activity inhibits cortical association of ECT-2.

The C-terminus of ECT-2 recapitulates the asymmetric distribution of ECT-2

To gain insight into the mechanism by which AIR-1 regulates ECT-2 accumulation, we sought to identify a region of ECT-2 that is sufficient to recapitulate the asymmetric accumulation of ECT-2. As previously shown, a C-terminal fragment of ECT-2 that extends from the N-terminus of the DH domain to the C-terminus exhibits cortical accumulation (Chan and Nance, 2013). Further deletion analysis indicates that cortical accumulation is independent of the majority of the DH domain, but required both the PH domain and a portion of the region C-terminal to the PH domain, but not its entirety (ECT-2^C) (Figure 5A). This fragment can localize to the cortex of embryos depleted of endogenous ECT-2 (Figure 5—figure supplement 1); although it does not accumulate asymmetrically under these conditions. The accumulation pattern of ECT-2^C is similar to that of endogenous ECT-2 during both polarization and cytokinesis (Figure 5D and E). This ECT-2 fragment is also recruited to the posterior, centralspindlin-dependent furrow that forms in ZYG-9-depleted embryos (not shown). AIR-1 depletion resulted in a reduction in the degree of its asymmetric accumulation (Figure 5—figure supplement 2). These data indicate that this C-terminal fragment of ECT-2 is sufficient to respond to the inhibitory signal from the centrosomes, and that the N-terminal regulatory domain and the DH domain of ECT-2 are dispensable for asymmetric ECT-2, and its recruitment by centralspindlin.

Figure 5 with 2 supplements see all

Download asset Open asset

Phosphorylation of ECT-2 regulates its asymmetric cortical accumulation.

(A) Schematic of GFP:ECT-2 truncations tested in this study and summary of their localization patterns. (B) Schematic of ECT-2^C with locations of mutated residues. Residues in the loop in the PH domain are colored purple. (C) Alpha Fold prediction of the ECT-2 PH domain prediction (AF-Q9U364-F1) aligned with the crystal structure of the PH domain of Bcr-Abl (PDB 5OC7). (D) Stills from representative time series of GFP:ECT-2^C either wild-type, 3A, or 6A mutants and the quantitation of accumulation of each transgene with NEBD as time 0. Source data is available (Figure 5—source data 1). (E) Asymmetric accumulation of GFP:ECT-2^C-3A is lost when cortical flows are reduced. Images shown are from time-lapse acquisitions of either control or NMY-2-depleted embryos from transgenic lines expressing either GFP:ECT-2^C or GFP:ECT-2^C-3A. Graphs (i) – (**iii**) quantify GFP:ECT-2^C wild-type accumulation over time and graphs (iv)–(vi) quantify GFP:ECT-2^C-3A accumulation over time, both with t=0 set as NEBD. Green arrowheads depict local reduction in ECT-2 as the centrosome is opposed to the posterior cortex. Source data is available (Figure 5—source data 1). (F) GFP:ECT-2^C-3A is more refractory than GFP:ECT-2^C-WT to GBP:mCherry:AIR-1 during cytokinesis. Panels show stills from time series of embryos co-expressing either GFP:ECT-2^C-WT or GFP:ECT-2^C-3A with GBP:mCherry:AIR-1. The graphs reflect GFP:ECT-2^C cortical accumulation over time starting at anaphase onset. ECT-2 accumulation on the spindle was excluded from measurements. Source data is available (Figure 5—source data 1).

Figure 5—source data 1 The source data is as described in legend to Figure 1—source data 1.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig5-data1-v2.zip
Download elife-83992-fig5-data1-v2.zip

Putative phosphorylation sites in the PH domain impact cortical accumulation of ECT-2

As AIR-1 is required for displacement of ECT-2 from the posterior cortex, we examined whether ECT-2 contains putative AIR-1 phosphorylation sites that could regulate its association with the cortex. Using a minimal Aurora consensus site [(K/R)|(K/R)(S/T)] (i.e. at least one basic residue, one to two residues N-terminal to an S/T residue) (Meraldi et al., 2004), we identified five putative sites in ECT-2^C. Two putative sites were located in an Alpha Fold predicted loop in the PH domain (Jumper et al., 2021), as was an additional serine residue with a near fit to the consensus sequence (RHAS⁶⁴³) (Figure 5B and C). We generated GFP-tagged transgenes in which either all six sites or the three sites in the PH domain loop were mutated to alanine, a non-phosphorylatable residue. As compared to ECT-2^C, both variants with alanine substitutions exhibited increased cortical accumulation (Figure 5D and E). Nevertheless, these variants accumulated in an asymmetric manner. ECT-2^C asymmetry temporally correlated with anteriorly directed cortical flows (Figure 5D and E), raising the possibility that asymmetric accumulation of endogenous ECT-2 drives flows that cause asymmetry of the transgene, irrespective of its phosphorylation status.

To test whether phosphorylation regulates the displacement of ECT-2 derivatives during polarization, we depleted non-muscle myosin, NMY-2, in embryos expressing GFP:ECT-2^C or GFP:ECT-2^C_3A. In the absence of myosin-dependent flows, the degree of GFP:ECT-2^C asymmetry is dramatically attenuated (Figure 5E). In 7/9 embryos, however, a local inhibitory effect in the immediate vicinity of the centrosome is detectable. In contrast, under these conditions, only 1/10 ECT-2^C_3A embryos exhibit significant displacement near the centrosome. These results suggest that centrosome-associated AIR-1 locally inhibits cortical ECT-2 accumulation.

To test whether ECT-2 displacement from the cortex can be triggered by Aurora A, we co-expressed a derivative of AIR-1 tagged with the GFP-binding domain (GBP::mCherry::AIR-1) with fusions of GFP:ECT-2^C and GFP:ECT-2^C_3A (Klinkert et al., 2019). Upon anaphase onset, GBP::mCherry::AIR-1 induces displacement of GFP:ECT-2^C (Figure 5F). Importantly, GFP:ECT-2^C_3A is refractory to GBP::mCherry::AIR-1-induced displacement (Figure 5F). These results support the hypothesis that AIR-1 directly modulates ECT-2 localization by phosphorylation of residues within a loop in the PH domain.

AIR-1 regulates centralspindlin-dependent furrowing

The AIR-1 pathway has been shown to impact centralspindlin-independent furrowing. For example, embryos deficient in centralspindlin subunits form ingressing, NOP-1-dependent furrows during cytokinesis which fail to form upon depletion of SAPS-1/PPH-6, a complex that negatively regulates AIR-1 (Afshar et al., 2010). Conversely, depletion of AIR-1 results in ectopic furrows during pseudocleavage (Kapoor and Kotak, 2019) and cytokinesis (Figure 4A). NOP-1, the upstream-most factor in the centralspindlin-independent pathway (Tse et al., 2012), is required for these ectopic furrows (Kapoor and Kotak, 2019).

To extend these findings, we first tested whether AIR-1 depletion bypasses the requirement for CYK-4 and NOP-1 in ECT-2 activation. To that end, we depleted both AIR-1 and CYK-4 in nop-1(it142) embryos. These triply deficient embryos fail to furrow during anaphase (100%, n=8). Next, we asked whether AIR-1 impacts centralspindlin-dependent furrowing, by depleting AIR-1 in NOP-1-deficient embryos. While 100% of embryos deficient in NOP-1 furrow to completion, ~30% embryos deficient in both AIR-1 and NOP-1 ingress with markedly slower kinetics and some fail to complete ingression (Figure 6A). These results suggest that while AIR-1 primarily impacts NOP-1-directed furrowing, it also affects centralspindlin-directed furrowing.Due to the role of AIR-1 in promoting nuclear envelope breakdown (Hachet et al., 2007; Portier et al., 2007), embryos depleted of AIR-1 appear to initiate furrowing more rapidly when NEBD is used as a reference point.

Figure 6

Download asset Open asset

AIR-1 is involved in centralspindlin-dependent furrowing.

(A) Furrowing is variable in *nop-1(it142); air-1*(RNAi) embryos. Shown are cortical maximum intensity projections 6 μm deep of NMY-2:GFP; *nop-1(it142*) embryos with or without depletion of AIR-1. Furrow width as a function of time starting at NEBD in individual embryos is shown graphically using a rolling average period of three time points. Note that NEBD is delayed in AIR-1-depleted embryos resulting in contractility at an earlier time point. Source data is available (Figure 6—source data 1). (B) ECT-2^T634E exhibits reduced cortical accumulation. Embryo images from time series of ECT-2:mNG or ECT-2^T634E:mNG expressing embryos in strains co-expressing mCherry:Tubulin. (i–iii) Graphs represent ECT-2 accumulation as a function of time with anaphase onset as time 0 s. Source data is available (Figure 6—source data 1). (C) ECT-2^T634E slows furrow ingression and affects central spindle-dependent furrowing. (i) Furrow kinetics of ECT-2:mNG or ECT-2^T634E:mNG in strains co-expressing mCherry:Tubulin. Graphs represent the extent of furrow ingression of individual embryos as a function of time starting at anaphase onset using a rolling average of three time points. (ii) Furrow kinetics of ECT-2:mNG or ECT-2^T634E:mNG in strains co-expressing mCherry:Tubulin. Graphs represent the extent of furrow ingression of individual embryos as a function of time starting at anaphase onset using a rolling average of three time points. Source data is available (Figure 6—source data 1). (F) ECT-2^T634E:mNG is more dosage sensitive than ECT-2:mNG. (i) Average total intensity of ECT-2 in embryos measured and graphed as box whisker plots for control embryos or embryos partially depleted of ECT-2. Furrow kinetics of control or partially depleted ECT-2 embryos expressing ECT-2:mNG or ECT-2^T634E:mNG co-expressing NMY-2:mKate were measured and used to determine (ii) the time between anaphase onset and furrow initiation (ingression to 90% furrow width) and (iii) the fraction of embryos that failed division. All non-depleted embryos completed division. Source data is available (Figure 6—source data 1).

Figure 6—source data 1 This source data table contains (I) the measurements of furrow width at each time point of each embryo. These data are plotted in Figure 6A and Figure 6Ci, ii. The progression of furrowing of individual embryos are graphed as a function of time. (II) The source data for Figure 6B is as described in legend to Figure 1—source data 1. (III) This source data contains the normalized ECT-2-mNG intensities of each embryo. Individual embryos are plotted as well as a box plot summarizing the data, and (IV) the data table containing the duration between anaphase onset and the time at which each embryo reached 10% furrow ingression (furrow width is 90% of initial width). Individual embryos are plotted as well as a box plot summarizing the data.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig6-data1-v2.zip
Download elife-83992-fig6-data1-v2.zip

Finally, we sought to confirm that the effects resulting from manipulating AIR-1 reflects the specific impact of this kinase on ECT-2. To that end, we generated a strain in which one of the putative phospho-acceptor residues in ECT-2, T634, was mutated to a glutamic acid. This strain could be readily maintained as a homozygote (>98% hatch rate), indicating that this allele retains significant function, although cortical accumulation of ECT-2^T634E is reduced and less asymmetric. To test whether the T634E substitution impacts ECT-2 activity, we used RNAi to partially deplete ECT-2:mNG and ECT-2^T634E:mNG. Reduction of ECT-2:mNG levels to ~20% of wild-type results in cytokinesis failure in only ~11% of embryos. Thus, in the wild-type, ECT-2 accumulates to levels in excess of that required for completion of cytokinesis. In contrast, a similar reduction of ECT-2^T634E:mNG results in cytokinesis failure in ~50% of embryos, suggesting that phosphorylation of ECT-2 on T634 attenuates ECT-2 function. Importantly, whereas ECT-2^T634E:mNG slows the rate of furrow ingression as compared to wild-type, furrow initiation is not delayed. In contrast, partial depletion of ECT-2 does not affect the rate of furrow ingression, but it delays furrow initiation. These results suggest that the defects resulting from the T634E substitution are distinct from those resulting from an overall reduction in ECT-2 function. To test whether this substitution affects centralspindlin-dependent furrowing, we depleted NOP-1 in embryos expressing ECT-2:mNG and ECT-2^T634E:mNG. Whereas ECT-2:mNG embryos depleted of NOP-1 complete cytokinesis with only a slight delay, embryos expressing ECT-2^T634E and depleted of NOP-1 exhibit highly variable furrowing behavior. While a minority of such embryos complete the first division (5/12), many embryos form furrows that ingress at variable rates and to variable extents (Figure 6Bv). We conclude that the introduction of a single phosphomimetic residue at a putative AIR-1 site in a flexible loop in the PH domain of ECT-2 impairs centralspindlin-dependent furrowing.

Discussion

During cytokinesis, cleavage furrow formation is directed by a combination of a positive signal that spatially correlates with the position of the spindle midzone, and a negative signal associated with spindle poles. The positive cue is centralspindlin-dependent and is fairly well understood; far less is known about the inhibitory cue(s) associated with spindle poles. Here, we show that an inhibitory cue associated with centrosomes triggers a local, Aurora A kinase-dependent inhibition of cortical ECT-2 recruitment (Figure 7A). Our studies suggest that the posteriorly shifted spindle triggers preferential displacement of ECT-2 from the posterior cortex by Aurora A-dependent phosphorylation of the ECT-2 PH domain, though the evidence for this phosphorylation event is indirect. This local inhibition of cortical ECT-2 recruitment triggers anterior-directed, myosin-dependent cortical flows that amplify ECT-2 asymmetric accumulation, likely generating positive feedback (Figure 7B). Inhibition of cortical accumulation of ECT-2 by Aurora A influences both centralspindlin-independent and -dependent furrowing. These results provide insights into the mechanism of generation of cortical asymmetries during polarization and cytokinesis and extends the commonalities between the two processes.

Figure 7

Download asset Open asset

Model of local symmetry breaking of cortical ECT-2 and amplification by cortical flows.

(A) Summary graph of cortical ECT-2 accumulation at the time of maximal ECT-2 asymmetry as a function of distance between the centrosome and the proximal cortex. Centrosome-cortex distance was measured using either a mCherry:Tubulin or Nomarski image. Three embryos from each treatment listed in the legend were chosen at random. Source data is available (Figure 7—source data 1). (B) During both polarization and cytokinesis, centrosome-derived AIR-1 locally inhibits cortical ECT-2 accumulation, triggering symmetry breaking and generation of cortical flows directed away from the site of symmetry breaking. ECT-2 dynamically associates with components that undergo cortical flow, leading to its further depletion from the posterior cortex and its anterior accumulation. During polarity establishment, the centrosomes are small and closely apposed to the cortex, whereas during anaphase the centrosomes are large and accumulate significant AIR-1 that modulate cortical association of ECT-2. During polarization, the pseudocleavage furrow forms at the boundary between low and high ECT-2 and myosin levels; in cytokinesis, this site correlates with the centralspindlin-dependent furrow that correlates with the spindle midzone. (C) Proposed genetic pathway (D).

Figure 7—source data 1 This data lists the distance of each anterior or posterior centrosome to its respective membrane (anterior or posterior) and the amount of ECT-2 accumulation on the respective membrane. Individual centrosomes were plotted.: https://cdn.elifesciences.org/articles/83992/elife-83992-fig7-data1-v2.zip
Download elife-83992-fig7-data1-v2.zip

Centrosomal AIR-1 regulates cortical association of ECT-2

In the early C. elegans embryo, shortly after fertilization, the RHO-1 GEF ECT-2 is enriched on the cortex relative to the cytoplasm. Following an initial period of uniform cortical association, as the paternally contributed centrosome approaches the cortex, ECT-2 accumulation at the nascent posterior cortex is reduced, coupled with an accompanying increase of ECT-2 in the nascent anterior cortex leading to the asymmetric distribution of ECT-2 and establishment of the primary body axis (Motegi and Sugimoto, 2006). The centrosome-induced reduction of posterior ECT-2 involves AIR-1 (Zhao et al., 2019; Kapoor and Kotak, 2019; Klinkert et al., 2019).

A C-terminal fragment of ECT-2, lacking the regulatory BRCT domains and the catalytic GEF domain, largely recapitulates the asymmetric distribution of endogenous ECT-2 throughout the early divisions. The PH domain contained in this fragment contains a sequence, predicted to form a loop, that contains several putative sites for phosphorylation by AIR-1. Mutation of these sites to non-phosphorylatable alanine residues enhances ECT-2 cortical association and renders it resistant to the inhibitory action of AIR-1; conversely, phosphomimetic mutations reduce its cortical association. We hypothesize that AIR-1 phosphorylates these sites thereby inhibiting ECT-2 cortical association. The flexible loop containing these putative sites is predicted to lie adjacent to the surface where phosphoinositides bind in structurally characterized PH domains (Reckel et al., 2017). The strong effect of phosphomimetic and non-phosphorylatable mutations on ECT-2 accumulation is striking, given the predicted flexibility of this loop, lending additional support to their relevance.

Our results implicating AIR-1 in polarity establishment are largely in concordance with those of Zhao et al., 2019, with the exception that these authors did not observe an effect of AIR-1 depletion on ECT-2 localization during polarity establishment. In that regard, our results align with those of Kapoor and Kotak, 2019. Multiple lines of evidence implicate AIR-1 in inhibiting ECT-2 cortical localization. These include (i) changes in cortical accumulation of ECT-2 upon depletion of SAPS-1 and AIR-1 (ECT-2^C is also enhanced upon AIR-1 depletion); (ii) increased cortical accumulation of GFP:ECT-2^C_3A as compared to GFP:ECT-2^C; (iii) susceptibility of GFP:ECT-2^C, but not GFP:ECT-2^C_3A to GBP:mCherry:AIR-1; (iv) a phosphomimetic substitution of a predicted AIR-1 site in ECT-2, T634, attenuates its cortical accumulation and function.

During both polarization and cytokinesis, centrosomes, by virtue of their role in concentrating and activating AIR-1 (Hamill et al., 2002; Joukov et al., 2010), appear to act as cues for ECT-2 displacement. This assertion is based on several earlier studies (Kapoor and Kotak, 2019; Klinkert et al., 2019) and the observations shown here. Specifically, during both processes, ECT-2 displacement requires the core centrosomal component SPD-5, which is required to recruit AIR-1 to centrosomes (Hamill et al., 2002), but ECT-2 displacement is not inhibited by depolymerization of microtubules and it does not require the AIR-1 activator TPXL-1 (see below). During cytokinesis, centrosomes preferentially inhibit ECT-2 cortical association in the posterior of the embryo. This difference results from the asymmetric position of the spindle that arises as a consequence of embryo polarization.

To summarize our results related to cytokinesis, we plotted cortical ECT-2 accumulation as a function of centrosome-cortical distance in representative embryos from a variety of experimental conditions (Figure 7A). These data demonstrate that, during cytokinesis, ECT-2 accumulation and centrosome proximity inversely correlate over a range of distances up to ~20 µm (40% of embryo length). The inhibitory effect of centrosomes on ECT-2 accumulation decays rapidly in the range of 10–18 µm from the cortex, which correspond to the positions of the posterior and anterior centrosomes in wild-type embryos (Kemphues et al., 1988), respectively. These findings are consistent with our earlier work demonstrating that spindle elongation regulates furrowing during cytokinesis, although we previously inferred that the inhibition was due to astral microtubules, which are proximal to the cortex (Dechant and Glotzer, 2003), rather than centrosomes which are not.

Cortical flows amplify ECT-2 asymmetry

Centrosomes and AIR-1 kinase promote ECT-2 asymmetry during both polarity establishment and cytokinesis. During polarity establishment in myosin-depleted embryos which lack cortical flows, the zone of inhibition of ECT-2 accumulation in the posterior shrinks considerably; similar observations have reported previously (Motegi and Sugimoto, 2006). During cytokinesis, perturbations, such as microtubule disassembly and ZYG-9 depletion, enhance cortical flows creating a larger posterior zone of inhibition and enhance the accumulation of ECT-2 in the anterior. The enhancement of these flows is likely the result of a combination of a reduction in the distance between the posterior centrosome to the cortex and the removal of a centrosome in the anterior domain which would otherwise suppress contractility in that region. The ability of ECT-2^C_3A to polarize in the presence, but not the absence of myosin, further establishes the role of flows in amplifying ECT-2 asymmetry. These results suggest that in wild-type embryos, Aurora A-dependent symmetry breaking is amplified by anterior-directed cortical flows that enhance ECT-2 asymmetry. Further, we speculate that centrosomal AIR-1 not only breaks symmetry, but that this regulation of ECT-2 by centrosomal AIR-1 continues throughout anaphase. These findings extend our understanding of the mechanistic basis for the requirement for myosin in polarity establishment (Guo and Kemphues, 1996). Furthermore, asymmetric ECT-2 is predicted to enhance asymmetric assembly of cortical myosin, which is predicted to accentuate cortical flows, further amplifying ECT-2 asymmetry.

Long-range flows result from the viscosity of the cortical actomyosin network (Mayer et al., 2010). It is unlikely that ECT-2 is concentrated by these flows itself, as it exchanges rapidly, on the order of seconds. Rather, we favor a model in which the association of ECT-2 with the cortex involves interactions with cortical component(s) that are concentrated by cortical flows (Mayer et al., 2010). This would explain the anterior accumulation of ECT-2 as polarization proceeds and the enhanced ECT-2 asymmetry during cytokinesis in embryos with exaggerated flows. The PH domain of ECT-2 has been shown to bind the actomyosin-binding protein anillin (Frenette et al., 2012), making it a candidate contributor to ECT-2 cortical anchoring. Anterior PAR proteins may also contribute to ECT-2 accumulation. Interestingly, anterior PAR proteins are known to promote cortical flow (Munro et al., 2004). Further analysis of the interactions required for ECT-2 association will be a focus of future studies.

Although RhoGEF activators accelerate cortical flows, basal levels of myosin appear sufficient to drive the cortical flows that polarize ECT-2. For example, inactivation of NOP-1 slows cortical flows and delays the polarization of anterior PAR proteins, but it does not preclude asymmetric accumulation of ECT-2 (Rose et al., 1995; Tse et al., 2012) and this study. Cortical flow-mediated concentration of ECT-2 may be central to polarization of the embryo, analogous to flow-induced re-orientation of PAR protein polarity (Mittasch et al., 2018).

During anaphase, asymmetric ECT-2 accumulation is also myosin-dependent, presumably due to its role in generating cortical flows. During cytokinesis, basal myosin levels appear to be sufficient to promote asymmetric ECT-2 accumulation, as ECT-2 asymmetry increases during anaphase even when both NOP-1 and CYK-4 are attenuated. We infer that bulk, cortical ECT-2 has a low level of RhoGEF activity.

In summary, during both polarization and cytokinesis, the cortex undergoes large, furrow-inducing, flows of contractile cortical material due to the influence of the centrosomes on ECT-2 localization. During polarization the two centrosomes are tiny and they interact with the male pronucleus (Malone et al., 2003). These immature centrosomes are usually closely associated with the posterior cortex (Delattre et al., 2006; Pelletier et al., 2006; Cowan and Hyman, 2004). By contrast, during the first cytokinesis, the two centrosomes are large, they contain dramatically more AIR-1 kinase (Hannak et al., 2001), and they are separated from each other by the length of the spindle and they lie at least 10 µm from the nearest cortex (Figure 7A). Despite the wide differences in size, components, and positions of the immature and mature centrosomes, roughly similar patterns of ECT-2 accumulation and contractility are induced during polarization and cytokinesis.

Cell cycle regulation of ECT-2 cortical localization

The temporal pattern of ECT-2 cortical accumulation is dynamic. ECT-2 is initially uniformly distributed on the cortex and becomes progressively asymmetric through the first interphase. Cortical ECT-2 accumulation then declines throughout maintenance phase, though asymmetry remains. During cytokinesis, ECT-2 levels on the cortex decline, though it is preferentially displaced from the posterior cortex, coincident with spindle elongation.

While many of these changes in cortical localization reflect changes controlled by AIR-1 and centrosome position, other factors appear to influence cortical accumulation of ECT-2. For example, cell cycle regulated, global changes in ECT-2 accumulation are apparent in AIR-1-depleted embryos. Furthermore, the ability of GBP::mCherry::AIR-1 to induce GFP-ECT-2C displacement fluctuates with the cell cycle, suggesting the susceptibility of ECT-2 to AIR-1 may be cell cycle regulated. For example, Cdk1 phosphorylation of ECT-2 might inhibit its phosphorylation by AIR-1.

TPXL-1 is not required for ECT-2 asymmetry

In addition to examining the role of AIR-1 in regulating cortical association of ECT-2, we also examined the role of one of its activators, TPXL-1 (Ozlü et al., 2005). TPXL-1 targets AIR-1 to astral microtubules, which are well positioned to modulate cortical ECT-2. However, TPXL-1 makes a more modest contribution to ECT-2 asymmetry than does AIR-1. Additionally, the effect observed upon TPXL-1 depletion might be, at least in part, an indirect consequence of the role of TPXL-1 in regulating spindle assembly and positioning. Indeed, TPXL-1 is dispensable for ECT-2 asymmetry when spindle poles are close to the cortex. This finding is consistent with earlier work that revealed that several AIR-1 functions, particularly those at the centrosome, are TPXL-1 independent (Ozlü et al., 2005; Zhao et al., 2019).

A previous study demonstrated that TPXL-1 is involved in polar relaxation during cytokinesis (Mangal et al., 2018). There are several differences between that study and the present one. Here, we assayed ECT-2 accumulation, whereas the previous study assayed accumulation of the RHO-1 effector anillin in embryos that were depleted of both myosin II and the principle RhoGAPs that inhibit RHO-1, RGA-3/4. In that context, TPXL-1 appears to inhibit accumulation of anillin specifically at the anterior cortex; anillin does not strongly accumulate on the posterior cortex of either wild-type or TPXL-1-depleted embryos (Mangal et al., 2018). Thus, AIR-1 appears to regulate Rho-mediated contractility at multiple steps, which are differentially dependent on both TPXL-1 and NMY-2.

Centralspindlin can facilitate ECT-2 cortical association

Despite the inhibition of ECT-2 accumulation at the posterior cortex, which is further potentiated when the spindle is displaced to the posterior, a furrow can form at a site adjacent to the spindle midzone in the embryo posterior. Formation of this furrow requires centralspindlin (Werner et al., 2007). Despite the overall inhibition of ECT-2 cortical association in the posterior, our results suggest an initial centralspindlin accumulation recruits and activates ECT-2 to induce furrowing at this site.

These results indicate that cortical recruitment of ECT-2 can occur by two distinct modes. One mode is independent of centralspindlin, and is inhibited by AIR-1. The other mode is dependent on centralspindlin and is less sensitive to AIR-1 activity. The centralspindlin-dependent mode of ECT-2 accumulation correlates with a reduced mobile fraction of ECT-2 as compared to the ECT-2 that accumulates on the cortex more broadly. Centralspindlin-dependent accumulation of cortical ECT-2 is reminiscent of its regulation in cultured human cells. In HeLa cells, Ect2 does not constitutively associate with the cortex, due to Cdk1-mediated phosphorylation of a basic region C-terminal to the PH domain (Su et al., 2011). During anaphase in these cells, equatorial accumulation of ECT-2 requires centralspindlin (Su et al., 2011). Thus, in both systems, phosphorylation of the C-terminus of ECT-2 attenuates its cortical association. However, the kinases are distinct (Cdk1 in human cells vs. AIR-1 in C. elegans) whereas the phosphoregulation appears global in human cells and spatially regulated in C. elegans blastomeres.

Polar relaxation in C. elegans and mammalian cells

The results shown here indicate that inhibition of ECT-2 accumulation by centrosomal Aurora A at spindle poles can induce furrow formation. The conservation of centrosomal Aurora A and ECT-2 suggests that this phenomenon could be generalizable. In mammalian cells, Aurora A may function in parallel to Cdk1 (Su et al., 2011) to antagonize cortical association of Ect2 at certain times of the cell cycle. A recent report, relying primarily on chemical inhibitors of Aurora kinases, indicates that Aurora B promotes clearance of F-actin from the polar cortex (Ramkumar et al., 2021). However, these inhibitors are not highly selective for Aurora A vs. Aurora B at the concentrations used in cell-based assays (de Groot et al., 2015).

Remarkably, the highly similar kinases, Aurora A and Aurora B, regulate cytokinesis in quite different manners. Aurora A primarily acts to inhibit contractility at cell poles through ECT-2, while Aurora B primarily promotes contractility at the equator through centralspindlin (Basant et al., 2015).

Limitations of this study

While our data implicate AIR-1 in phosphorylation of ECT-2, we have not directly shown that AIR-1 phosphorylates ECT-2. Additionally, while mutations of these sites affect ECT-2 localization, we have not shown that these sites are phosphorylated in vivo.

Materials and methods

Plasmid and strain construction

MosSCI transgenic strains

Request a detailed protocol

C. elegans strains expressing GFP:ECT-2^C and GFP:ECT-2^C-3A, and GFP:ECT-2^C-6A using the promoter and 3’utr elements from pie-1. These transgenes were integrated into ttTi5605 on Chromosome II using MosSCI (Frøkjær-Jensen et al., 2012; Frøkjaer-Jensen et al., 2008). A complete list of strains used in this study is available (Supplementary file 1). Plasmids were built using Gibson Assembly (Gibson et al., 2009). A complete list of plasmids used in this study is available (Supplementary file 2). A complete list of oligonucleotides used in this study is available (Supplementary file 3).

CRISPR

Request a detailed protocol

mNG was integrated at the C-terminus of endogenous ECT-2 using CRISPR (Dickinson et al., 2015). ECT-2^T634E was mutated at the endogenous locus by CRISPR using Cas9 ribonucleoprotein [IDT] complexes (crRNA sequence: ACGCTGTCTCGAATGTAAAC) and a single-stranded oligodeoxynucleotide (CAGCGAGAAGAATgaAATGCAACGTCTAGCTCGTCAtGCGTCGTTTGCgAGTTTACATTC) (Paix et al., 2015; Dokshin et al., 2018).

RNAi vector construction

Request a detailed protocol

ECT-2(244-536aa)/L4440 and TPXL-1(50-395aa)/L4440 were amplified from N2 lysates and inserted into L4440 between SacI and SpeI by Gibson Assembly (Gibson et al., 2009). For TBA-2 + TBB-2/L4440, the TBB-2 fragment from TBB-2/L4440 (Ahringer library) was amplified and inserted into the KpnI site of TBA-2/L4440 (Ahringer library) by ligation.

RNAi

Request a detailed protocol

Ahringer library: par-3, air-1, spd-5, saps-1, dhc-1, perm-1, nop-1, and zyg-9 (Kamath et al., 2003).

ect-2 [this study], tpxl-1 [this study], tba-2+tbb-2 [this study], cyk-4 (Zhang and Glotzer, 2015), gα (from Pierre Gönczy), and nmy-2 (from Ed Munro).

For all experiments, RNAi plasmids were transformed into HT115 competent cells and cultured for 6–8 hr in LB + ampicillin at 37°C; cultures or mixtures thereof were seeded onto NGM plates containing ampicillin and 1 mM IPTG and incubated overnight at room temperature before storage at 4° C.

For ect-2: RNAi plates were seeded with a mixed ECT-2:L4440 culture (1:2 or 1:10). For experiments, L4 worms were transferred to plates about 20 hr prior to imaging.

For perm-1: RNAi plates were seeded with a mixed PERM-1:L4440 culture (1:20) (Carvalho et al., 2011). For experiments, L4 worms were transferred to plates about 16 hr prior to imaging.

For all experiments except as indicated, L4 hermaphrodites were transferred to RNAi plates for ~24 hr before imaging.

RNAi-treated embryos were phenotypically assessed for sufficient depletion. These phenotypes were assessed using Nomarski or mCherry:Tubulin. Embryos that did not show these phenotypes were excluded from the analysis:

air-1(RNAi): Mitotic spindle fails to assemble properly; hypercontractility during polarization (Hannak et al., 2001).
cyk-4(RNAi): Loss of bundling of anti-parallel microtubules upon anaphase onset, exaggerated anaphase B (Jantsch-Plunger et al., 2000).
dhc-1(RNAi): Lack of pronuclear migration and centrosomes remained in posterior during polarization and cytokinesis (Gönczy et al., 1999).
ect-2(RNAi) (partial): Lack of cortical ruffling during polarization (Motegi and Sugimoto, 2006).
gα(RNAi): Absence of spindle rocking during anaphase and shorter aster separation (Gotta and Ahringer, 2001).
nmy-2(RNAi) (partial): Lack of pseudocleavage and contractility during polarization (Munro et al., 2004).
nop-1(RNAi): Lack of pseudocleavage (Rose et al., 1995; Tse et al., 2012).
par-2(RNAi): Symmetric first division (Kemphues et al., 1988).
par-3(RNAi): Symmetric first division (Etemad-Moghadam et al., 1995).
saps-1(RNAi): Absence of spindle rocking during anaphase (Afshar et al., 2010).
spd-5(RNAi): Lack of pronuclear migration and spindle assembly failure (Hamill et al., 2002).
tba-2 + tbb-2(RNAi) + nocodazole: Centrosomes remained in posterior during polarization and cytokinesis and asters did not separate during cytokinesis (Tsai and Ahringer, 2007).
tpxl-1(RNAi): Small mitotic spindle (Ozlü et al., 2005).
zyg-9(RNAi): Spindle assembles in posterior of the embryo (Matthews et al., 1998).

Imaging

Mounting

Request a detailed protocol

Generally, gravid C. elegans were dissected in egg salts (5 mM HEPES pH 7.4, 118 mM NaCl, 40 mM KCl, 3.4 mM MgCl_2, 3.4 mM CaCl₂) and mounted on 2% agarose pads. For tubulin experiments, C. elegans were dissected in egg salts plus 50 μg/mL nocodazole or DMSO.

For MLN8237 inhibitor experiment: After perm-1 depletion, C. elegans were dissected in embryonic imaging medium (50% L-15 Leibovitz’s Medium [Gibco], 10% 5 mg/mL Inulin, 20% FBS [Gibco], 20% 250 mM HEPES pH 7.5) on a Poly-D-Lysine (R&D Systems)-coated coverslip mounted on a slide with double-sided tape (3M) used as spacer. Additional imaging media was flowed into fill the chamber. Fifty μM MLN8237 or DMSO was flowed into the embryonic chamber at 100 s post NEB while imaging.

For imaging early polarization: Worms were dissected into embryonic imaging medium (50% L-15 Leibovitz’s Medium [Gibco], 10% 5 mg/mL Inulin, 20% FBS, 20% 250 mM HEPES, pH 7.5) on a Poly-D-Lysine (R&D Systems)-coated coverslip mounted on a slide with double-sided tape (3M) used as spacer. Additional media was flowed into fill the chamber before imaging.

Microscopy

Request a detailed protocol

Time-lapse recordings of embryos were acquired with a Zeiss Axioimager M1 equipped with a Yokogawa CSU-X1 spinning disk unit (Solamere) using 50 mW 488 and 561 nm lasers (Coherent) and a Prime 95B camera (Photometrics). Images were taken at 63× using Micromanager software. ECT-2:mNG was imaged at 10 s intervals during cytokinesis and at 20 s intervals during polarization. NMY-2:GFP was imaged at 10 s intervals during polarization and cytokinesis. In all cases, a Nomarski image was acquired at each time point. Nuclear envelope breakdown was scored based on ECT-2 fluorescence and/or the appearance of the nucleus in the Nomarski images. For the NMY-2:GFP particle tracking, GFP was imaged only at the cortex by taking 4-1 µm step slices at a stream rate beginning at anaphase onset.

FRAP experiments were performed on a Zeiss LSM 880 equipped with AiryScan on confocal setting using a Plan-Apochromat 63×/1.40 Oil DIC M27 objective. Images were taken at 100 ms intervals.

Image analysis

All embryos were processed and analyzed using FIJI (Schindelin et al., 2012). Further analysis and graphs were generated using RStudio with the following packages: tidyverse, gcookbook, ggplot2, broom, ggpubr, rlang, mgcv, and zoo (Wickham et al., 2019).

ECT-2 accumulation

Request a detailed protocol

For each embryo, at every time point, the average intensities of the background and cytoplasm were measured. The embryonic cortex was identified by performing a Gaussian blur on a copy of the embryo, using the FIJI magic wand to select the embryo interior. This area was converted to a line, transferred to the original image and then the line was expanded to a width of 50 pixels and straightened from the middle of the anterior of the embryo clockwise around the perimeter (Figure 1—figure supplement 1). From the image containing the straightened cortical region, a 3 pixel-wide line was drawn perpendicular to the cortex and the maximum intensity was recorded at each position along the line. Accumulation was calculated for every position using (maximum intensity-background)/(cytoplasm-background). The measurements of anterior and posterior accumulation reflect the average accumulation in the anterior 20% or the posterior 20% of the embryo, respectively.

The perimeter was measured as the length of the ROI used to identify the embryo cortex at each time point. The perimeter was normalized to the maximum perimeter of each embryo. To assess boundary length, ECT-2:mNG accumulation across all positions along the perimeter of the embryo was fitted to a regression model by generalized additive model (GAM) (prediction accumulation) in R Studio using the mgcv:gam function. The average cortical accumulation over the anterior 60% of the embryo was calculated and a threshold was set at 85% of this average. The number of positions that fell below the threshold were counted and the boundary length was set as the fraction of positions below threshold.

The accumulation profile of ECT-2 was calculated from the average of the regression models at the time during cytokinesis at which each embryo exhibited maximally asymmetric ECT-2. This estimate is plotted as a function of relative position along the cortex of the embryo with 0 as the anterior and 0.5 as the posterior.

For the ect-2(RNAi) partial depletion experiments, embryo intensity was measured as the average pixel intensity of the embryo after subtracting the background at NEBD.

Furrow kinetics

Request a detailed protocol

The positions of the furrow tips in NMY-2:GFP, NMY-2:mKate, or ECT-2:mNG embryos were measured at 10 s intervals starting at NEBD or anaphase. The distance between the furrow tips were normalized to maximum embryo width [furrow width]. Furrow width was plotted as a rolling average (period = 3) over time for individual embryos.

For the partial ect-2(RNAi) experiments: To measure ingression initiation, the length of time between anaphase onset and 90% width was measured and plotted. For furrow failure, the fraction of embryos that did not reach a furrow width of 0% were counted and plotted.

FRAP

Request a detailed protocol

The FRAP images were analyzed using FIJI. The FRAP Analysis Script source is https://imagej.net/imagej-wiki-static/Analyze_FRAP_movies_with_a_Jython_script.

The coefficients from the fitted recovery curves of each individual embryo were averaged for each condition and the average curve was plotted.

Centrosome-cortical distance

Request a detailed protocol

Three embryos of each treatment indicated in the legend were chosen at random and, the time point of maximally asymmetric ECT-2 accumulation was selected for measurement. Centrosome distance to the nearest cortex was measured using the mCherry:Tubulin or Nomarski images. This distance and the average ECT-2 accumulation at the nearest cortex were plotted.

Particle tracking

Request a detailed protocol

Myosin particles were tracked using the Trackmate plugin in FIJI (Ershov et al., 2022).

Statistical tests

Request a detailed protocol

Ninety-five percent confidence intervals are shown as shaded ribbons in the ECT-2 cortical accumulation graphs. Additionally, t-tests were performed in R Studio to test the significance of ECT-2 accumulation between two treatments (i.e. RNAi depletion and wild-type) at every time point imaged as indicated in the figure legends. t-Tests were also performed to test the significance of the length of ECT-2 boundaries. Labels on the graphs are defined as follows: ‘ns’, p>0.05; *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Structural analysis

Request a detailed protocol

The MatchMaker function of UCSF Chimera (Pettersen et al., 2004) (https://www.cgl.ucsf.edu/chimera/) was used to align the predicted PH domain of C.e. ECT-2 (https://alphafold.com/entry/Q9U364) to 5OC7 (Reckel et al., 2017).

Materials availability

Request a detailed protocol

Strains (Supplementary file 1) and plasmids (Supplementary file 2) generated in this study are available from the author upon request. Images, analysis scripts, image quantification data, and plasmid maps will be deposited in Zenodo and associated with the DOI:10.7554/eLife.83992.

Data availability

The data generated or analyzed during this study are included in the manuscript and in the associated source data files. Image files, raw image analysis data, image analysis scripts, R data files and plasmid sequences are uploaded to Zenodo at https://doi.org/10.5281/zenodo.7415982.

The following data sets were generated

1. Longhini KM
2. Glotzer M
(2022) Zenodo
Aurora A and cortical flows promote polarization and cytokinesis by inducing asymmetric ECT-2 accumulation - Data Archive.
https://doi.org/10.5281/zenodo.7415982

References

1. Afshar K
2. Werner ME
3. Tse YC
4. Glotzer M
5. Gönczy P
(2010) Regulation of cortical contractility and spindle positioning by the protein phosphatase 6 PPH-6 in one-cell stage C. elegans embryos
Development 137:237–247.
https://doi.org/10.1242/dev.042754
- PubMed
- Google Scholar
1. Basant A
2. Lekomtsev S
3. Tse YC
4. Zhang D
5. Longhini KM
6. Petronczki M
7. Glotzer M
(2015) Aurora B kinase promotes cytokinesis by inducing centralspindlin oligomers that associate with the plasma membrane
Developmental Cell 33:204–215.
https://doi.org/10.1016/j.devcel.2015.03.015
- PubMed
- Google Scholar
1. Basant A
2. Glotzer M
(2018) Spatiotemporal regulation of rhoa during cytokinesis
Current Biology 28:R570–R580.
https://doi.org/10.1016/j.cub.2018.03.045
- PubMed
- Google Scholar
1. Bringmann H
2. Hyman AA
(2005) A cytokinesis furrow is positioned by two consecutive signals
Nature 436:731–734.
https://doi.org/10.1038/nature03823
- PubMed
- Google Scholar
1. Burkard ME
2. Maciejowski J
3. Rodriguez-Bravo V
4. Repka M
5. Lowery DM
6. Clauser KR
7. Zhang C
8. Shokat KM
9. Carr SA
10. Yaffe MB
11. Jallepalli PV
(2009) Plk1 self-organization and priming phosphorylation of hscyk-4 at the spindle midzone regulate the onset of division in human cells
PLOS Biology 7:e1000111.
https://doi.org/10.1371/journal.pbio.1000111
- PubMed
- Google Scholar
1. Canman JC
2. Lewellyn L
3. Laband K
4. Smerdon SJ
5. Desai A
6. Bowerman B
7. Oegema K
(2008) Inhibition of rac by the gap activity of centralspindlin is essential for cytokinesis
Science 322:1543–1546.
https://doi.org/10.1126/science.1163086
- PubMed
- Google Scholar
1. Carvalho A
2. Olson SK
3. Gutierrez E
4. Zhang K
5. Noble LB
6. Zanin E
7. Desai A
8. Groisman A
9. Oegema K
(2011) Acute drug treatment in the early C. elegans embryo
PLOS ONE 6:e24656.
https://doi.org/10.1371/journal.pone.0024656
- PubMed
- Google Scholar
1. Chan E
2. Nance J
(2013) Mechanisms of CDC-42 activation during contact-induced cell polarization
Journal of Cell Science 126:1692–1702.
https://doi.org/10.1242/jcs.124594
- PubMed
- Google Scholar
(2020) Polar relaxation by dynein-mediated removal of cortical myosin II
The Journal of Cell Biology 219:1250.
https://doi.org/10.1083/jcb.201903080
- PubMed
- Google Scholar
1. Chen M
2. Pan H
3. Sun L
4. Shi P
5. Zhang Y
6. Li L
7. Huang Y
8. Chen J
9. Jiang P
10. Fang X
11. Wu C
12. Chen Z
(2019) Structure and regulation of human epithelial cell transforming 2 protein
PNAS 117:1027–1035.
https://doi.org/10.1073/pnas.1913054117
- PubMed
- Google Scholar
1. Chen A
2. Ulloa Severino L
3. Panagiotou TC
4. Moraes TF
5. Yuen DA
6. Lavoie BD
7. Wilde A
(2021) Inhibition of polar actin assembly by astral microtubules is required for cytokinesis
Nature Communications 12:2409.
https://doi.org/10.1038/s41467-021-22677-0
- PubMed
- Google Scholar
1. Cowan CR
2. Hyman AA
(2004) Centrosomes direct cell polarity independently of microtubule assembly in C. elegans embryos
Nature 431:92–96.
https://doi.org/10.1038/nature02825
- PubMed
- Google Scholar
1. Dechant R
2. Glotzer M
(2003) Centrosome separation and central spindle assembly act in redundant pathways that regulate microtubule density and trigger cleavage furrow formation
Developmental Cell 4:333–344.
https://doi.org/10.1016/s1534-5807(03)00057-1
- PubMed
- Google Scholar
1. de Groot CO
2. Hsia JE
3. Anzola JV
4. Motamedi A
5. Yoon M
6. Wong YL
7. Jenkins D
8. Lee HJ
9. Martinez MB
10. Davis RL
11. Gahman TC
12. Desai A
13. Shiau AK
(2015) A cell biologist’s field guide to aurora kinase inhibitors
Frontiers in Oncology 5:285.
https://doi.org/10.3389/fonc.2015.00285
- PubMed
- Google Scholar
(2006) Sequential protein recruitment in C. elegans centriole formation
Current Biology 16:1844–1849.
https://doi.org/10.1016/j.cub.2006.07.059
- PubMed
- Google Scholar
(2013) Engineering the Caenorhabditis elegans genome using cas9-triggered homologous recombination
Nature Methods 10:1028–1034.
https://doi.org/10.1038/nmeth.2641
- PubMed
- Google Scholar
(2015) Streamlined genome engineering with a self-excising drug selection cassette
Genetics 200:1035–1049.
https://doi.org/10.1534/genetics.115.178335
- PubMed
- Google Scholar
(2018) Robust genome editing with short single-stranded and long, partially single-stranded DNA donors in Caenorhabditis elegans
Genetics 210:781–787.
https://doi.org/10.1534/genetics.118.301532
- PubMed
- Google Scholar
1. Ershov D
2. Phan MS
3. Pylvänäinen JW
4. Rigaud SU
5. Le Blanc L
6. Charles-Orszag A
7. Conway JRW
8. Laine RF
9. Roy NH
10. Bonazzi D
11. Duménil G
12. Jacquemet G
13. Tinevez JY
(2022) TrackMate 7: integrating state-of-the-art segmentation algorithms into tracking pipelines
Nature Methods 19:829–832.
https://doi.org/10.1038/s41592-022-01507-1
- PubMed
- Google Scholar
(1995) Asymmetrically distributed par-3 protein contributes to cell polarity and spindle alignment in early C. elegans embryos
Cell 83:743–752.
https://doi.org/10.1016/0092-8674(95)90187-6
- PubMed
- Google Scholar
1. Fievet BT
2. Rodriguez J
3. Naganathan S
4. Lee C
5. Zeiser E
6. Ishidate T
7. Shirayama M
8. Grill S
9. Ahringer J
(2012) Systematic genetic interaction screens uncover cell polarity regulators and functional redundancy
Nature Cell Biology 15:103–112.
https://doi.org/10.1038/ncb2639
- PubMed
- Google Scholar
1. Frenette P
2. Haines E
3. Loloyan M
4. Kinal M
5. Pakarian P
6. Piekny AJ
(2012) An anillin-ect2 complex stabilizes central spindle microtubules at the cortex during cytokinesis
PLOS ONE 7:e34888.
https://doi.org/10.1371/journal.pone.0034888
- PubMed
- Google Scholar
(2008) Single-copy insertion of transgenes in Caenorhabditis elegans
Nature Genetics 40:1375–1383.
https://doi.org/10.1038/ng.248
- PubMed
- Google Scholar
(2012) Improved mos1-mediated transgenesis in C. elegans
Nature Methods 9:117–118.
https://doi.org/10.1038/nmeth.1865
- PubMed
- Google Scholar
1. Gibson DG
2. Young L
3. Chuang RY
4. Venter JC
5. Hutchison CA
6. Smith HO
(2009) Enzymatic assembly of DNA molecules up to several hundred kilobases
Nature Methods 6:343–345.
https://doi.org/10.1038/nmeth.1318
- PubMed
- Google Scholar
1. Goldstein B
2. Hird SN
(1996) Specification of the anteroposterior axis in Caenorhabditis elegans
Development 122:1467–1474.
https://doi.org/10.1242/dev.122.5.1467
- PubMed
- Google Scholar
(1999) Cytoplasmic dynein is required for distinct aspects of MTOC positioning, including centrosome separation, in the one cell stage Caenorhabditis elegans embryo
The Journal of Cell Biology 147:135–150.
https://doi.org/10.1083/jcb.147.1.135
- PubMed
- Google Scholar
1. Gotta M
2. Ahringer J
(2001) Distinct roles for galpha and gbetagamma in regulating spindle position and orientation in Caenorhabditis elegans embryos
Nature Cell Biology 3:297–300.
https://doi.org/10.1038/35060092
- PubMed
- Google Scholar
1. Guo S
2. Kemphues KJ
(1996) A non-muscle myosin required for embryonic polarity in Caenorhabditis elegans
Nature 382:455–458.
https://doi.org/10.1038/382455a0
- PubMed
- Google Scholar
(2007) Centrosomes promote timely mitotic entry in C. elegans embryos
Developmental Cell 12:531–541.
https://doi.org/10.1016/j.devcel.2007.02.015
- PubMed
- Google Scholar
(2002) Centrosome maturation and mitotic spindle assembly in C. elegans require SPD-5, a protein with multiple coiled-coil domains
Developmental Cell 3:673–684.
https://doi.org/10.1016/s1534-5807(02)00327-1
- PubMed
- Google Scholar
1. Hannak E
2. Kirkham M
3. Hyman AA
4. Oegema K
(2001) Aurora-A kinase is required for centrosome maturation in Caenorhabditis elegans
The Journal of Cell Biology 155:1109–1116.
https://doi.org/10.1083/jcb.200108051
- PubMed
- Google Scholar
(2016) A conserved motif provides binding specificity to the PP2A-B56 phosphatase
Molecular Cell 63:686–695.
https://doi.org/10.1016/j.molcel.2016.06.024
- PubMed
- Google Scholar
(2019) Anterior-enriched filopodia create the appearance of asymmetric membrane microdomains in polarizing C. elegans zygotes
Journal of Cell Science 132:jcs230714.
https://doi.org/10.1242/jcs.230714
- PubMed
- Google Scholar
1. Hird SN
2. White JG
(1993) Cortical and cytoplasmic flow polarity in early embryonic cells of Caenorhabditis elegans
The Journal of Cell Biology 121:1343–1355.
https://doi.org/10.1083/jcb.121.6.1343
- PubMed
- Google Scholar
1. Jantsch-Plunger V
2. Gönczy P
3. Romano A
4. Schnabel H
5. Hamill D
6. Schnabel R
7. Hyman AA
8. Glotzer M
(2000) Cyk-4: a rho family gtpase activating protein (GAP) required for central spindle formation and cytokinesis
The Journal of Cell Biology 149:1391–1404.
https://doi.org/10.1083/jcb.149.7.1391
- PubMed
- Google Scholar
(2010) Centrosomal protein of 192 kda (CEP192) promotes centrosome-driven spindle assembly by engaging in organelle-specific aurora A activation
PNAS 107:21022–21027.
https://doi.org/10.1073/pnas.1014664107
- PubMed
- Google Scholar
1. Jumper J
2. Evans R
3. Pritzel A
4. Green T
5. Figurnov M
6. Ronneberger O
7. Tunyasuvunakool K
8. Bates R
9. Žídek A
10. Potapenko A
11. Bridgland A
12. Meyer C
13. Kohl SAA
14. Ballard AJ
15. Cowie A
16. Romera-Paredes B
17. Nikolov S
18. Jain R
19. Adler J
20. Back T
21. Petersen S
22. Reiman D
23. Clancy E
24. Zielinski M
25. Steinegger M
26. Pacholska M
27. Berghammer T
28. Bodenstein S
29. Silver D
30. Vinyals O
31. Senior AW
32. Kavukcuoglu K
33. Kohli P
34. Hassabis D
(2021) Highly accurate protein structure prediction with alphafold
Nature 596:583–589.
https://doi.org/10.1038/s41586-021-03819-2
- PubMed
- Google Scholar
1. Kamath RS
2. Fraser AG
3. Dong Y
4. Poulin G
5. Durbin R
6. Gotta M
7. Kanapin A
8. Le Bot N
9. Moreno S
10. Sohrmann M
11. Welchman DP
12. Zipperlen P
13. Ahringer J
(2003) Systematic functional analysis of the Caenorhabditis elegans genome using rnai
Nature 421:231–237.
https://doi.org/10.1038/nature01278
- PubMed
- Google Scholar
1. Kapoor S
2. Kotak S
(2019) Centrosome aurora A regulates rhogef ECT-2 localisation and ensures A single PAR-2 polarity axis in C. elegans embryos
Development 146:dev174565.
https://doi.org/10.1242/dev.174565
- PubMed
- Google Scholar
(1988) Identification of genes required for cytoplasmic localization in early C. elegans embryos
Cell 52:311–320.
https://doi.org/10.1016/s0092-8674(88)80024-2
- PubMed
- Google Scholar
1. Kishi K
2. Sasaki T
3. Kuroda S
4. Itoh T
5. Takai Y
(1993) Regulation of cytoplasmic division of Xenopus embryo by rho p21 and its inhibitory GDP/GTP exchange protein (rho GDI)
Journal of Cell Biology 120:1187–1195.
https://doi.org/10.1083/jcb.120.5.1187
- PubMed
- Google Scholar
1. Klinkert K
2. Levernier N
3. Gross P
4. Gentili C
5. von Tobel L
6. Pierron M
7. Busso C
8. Herrman S
9. Grill SW
10. Kruse K
11. Gönczy P
(2019) Aurora a depletion reveals centrosome-independent polarization mechanism in Caenorhabditis elegans
eLife 8:e44552.
https://doi.org/10.7554/eLife.44552
- PubMed
- Google Scholar
1. Kotak S
2. Afshar K
3. Busso C
4. Gönczy P
(2016) Aurora a kinase regulates proper spindle positioning in C. elegans and in human cells
Journal of Cell Science 129:3015–3025.
https://doi.org/10.1242/jcs.184416
- PubMed
- Google Scholar
1. Lee KY
2. Green RA
3. Gutierrez E
4. Gomez-Cavazos JS
5. Kolotuev I
6. Wang S
7. Desai A
8. Groisman A
9. Oegema K
(2018) Cyk-4 functions independently of its centralspindlin partner ZEN-4 to cellularize oocytes in germline syncytia
eLife 7:e36919.
https://doi.org/10.7554/eLife.36919
- PubMed
- Google Scholar
1. Lewellyn L
2. Dumont J
3. Desai A
4. Oegema K
(2010) Analyzing the effects of delaying aster separation on furrow formation during cytokinesis in the Caenorhabditis elegans embryo
Molecular Biology of the Cell 21:50–62.
https://doi.org/10.1091/mbc.e09-01-0089
- PubMed
- Google Scholar
(2012) The rhogap domain of CYK-4 has an essential role in rhoa activation
Current Biology 22:213–219.
https://doi.org/10.1016/j.cub.2011.12.019
- PubMed
- Google Scholar
1. Malone CJ
2. Misner L
3. Le Bot N
4. Tsai MC
5. Campbell JM
6. Ahringer J
7. White JG
(2003) The C. elegans hook protein, ZYG-12, mediates the essential attachment between the centrosome and nucleus
Cell 115:825–836.
https://doi.org/10.1016/s0092-8674(03)00985-1
- PubMed
- Google Scholar
1. Mangal S
2. Sacher J
3. Kim T
4. Osório DS
5. Motegi F
6. Carvalho AX
7. Oegema K
8. Zanin E
(2018) TPXL-1 activates aurora A to clear contractile ring components from the polar cortex during cytokinesis
The Journal of Cell Biology 217:837–848.
https://doi.org/10.1083/jcb.201706021
- PubMed
- Google Scholar
(1998) ZYG-9, a Caenorhabditis elegans protein required for microtubule organization and function, is a component of meiotic and mitotic spindle poles
The Journal of Cell Biology 141:1159–1168.
https://doi.org/10.1083/jcb.141.5.1159
- PubMed
- Google Scholar
1. Mayer M
2. Depken M
3. Bois JS
4. Jülicher F
5. Grill SW
(2010) Anisotropies in cortical tension reveal the physical basis of polarizing cortical flows
Nature 467:617–621.
https://doi.org/10.1038/nature09376
- PubMed
- Google Scholar
(2004) Aurora kinases link chromosome segregation and cell division to cancer susceptibility
Current Opinion in Genetics & Development 14:29–36.
https://doi.org/10.1016/j.gde.2003.11.006
- PubMed
- Google Scholar
1. Mittasch M
2. Gross P
3. Nestler M
4. Fritsch AW
5. Iserman C
6. Kar M
7. Munder M
8. Voigt A
9. Alberti S
10. Grill SW
11. Kreysing M
(2018) Non-invasive perturbations of intracellular flow reveal physical principles of cell organization
Nature Cell Biology 20:344–351.
https://doi.org/10.1038/s41556-017-0032-9
- PubMed
- Google Scholar
(2012) A genome-wide rnai screen for enhancers of PAR mutants reveals new contributors to early embryonic polarity in Caenorhabditis elegans
Genetics 192:929–942.
https://doi.org/10.1534/genetics.112.143727
- PubMed
- Google Scholar
1. Motegi F
2. Sugimoto A
(2006) Sequential functioning of the ECT-2 rhogef, RHO-1 and CDC-42 establishes cell polarity in Caenorhabditis elegans embryos
Nature Cell Biology 8:978–985.
https://doi.org/10.1038/ncb1459
- PubMed
- Google Scholar
1. Motegi F
2. Zonies S
3. Hao Y
4. Cuenca AA
5. Griffin E
6. Seydoux G
(2011) Microtubules induce self-organization of polarized PAR domains in Caenorhabditis elegans zygotes
Nature Cell Biology 13:1361–1367.
https://doi.org/10.1038/ncb2354
- PubMed
- Google Scholar
(2004) Cortical flows powered by asymmetrical contraction transport PAR proteins to establish and maintain anterior-posterior polarity in the early C. elegans embryo
Developmental Cell 7:413–424.
https://doi.org/10.1016/j.devcel.2004.08.001
- PubMed
- Google Scholar
1. Murthy K
2. Wadsworth P
(2008) Dual role for microtubules in regulating cortical contractility during cytokinesis
Journal of Cell Science 121:2350–2359.
https://doi.org/10.1242/jcs.027052
- PubMed
- Google Scholar
1. Niiya F
2. Xie X
3. Lee KS
4. Inoue H
5. Miki T
(2005) Inhibition of cyclin-dependent kinase 1 induces cytokinesis without chromosome segregation in an ECT2 and mgcracgap-dependent manner
The Journal of Biological Chemistry 280:36502–36509.
https://doi.org/10.1074/jbc.M508007200
- PubMed
- Google Scholar
1. Nishimura Y
2. Yonemura S
(2006) Centralspindlin regulates ECT2 and rhoa accumulation at the equatorial cortex during cytokinesis
Journal of Cell Science 119:104–114.
https://doi.org/10.1242/jcs.02737
- PubMed
- Google Scholar
(2005) An essential function of the C. elegans ortholog of TPX2 is to localize activated aurora a kinase to mitotic spindles
Developmental Cell 9:237–248.
https://doi.org/10.1016/j.devcel.2005.07.002
- PubMed
- Google Scholar
(2015) High efficiency, homology-directed genome editing in Caenorhabditis elegans using CRISPR-cas9 ribonucleoprotein complexes
Genetics 201:47–54.
https://doi.org/10.1534/genetics.115.179382
- PubMed
- Google Scholar
1. Pelletier L
2. Ozlü N
3. Hannak E
4. Cowan C
5. Habermann B
6. Ruer M
7. Müller-Reichert T
8. Hyman AA
(2004) The Caenorhabditis elegans centrosomal protein SPD-2 is required for both pericentriolar material recruitment and centriole duplication
Current Biology 14:863–873.
https://doi.org/10.1016/j.cub.2004.04.012
- PubMed
- Google Scholar
(2006) Centriole assembly in Caenorhabditis elegans
Nature 444:619–623.
https://doi.org/10.1038/nature05318
- PubMed
- Google Scholar
1. Pettersen EF
2. Goddard TD
3. Huang CC
4. Couch GS
5. Greenblatt DM
6. Meng EC
7. Ferrin TE
(2004) UCSF chimera -- a visualization system for exploratory research and analysis
Journal of Computational Chemistry 25:1605–1612.
https://doi.org/10.1002/jcc.20084
- PubMed
- Google Scholar
1. Portier N
2. Audhya A
3. Maddox PS
4. Green RA
5. Dammermann A
6. Desai A
7. Oegema K
(2007) A microtubule-independent role for centrosomes and aurora A in nuclear envelope breakdown
Developmental Cell 12:515–529.
https://doi.org/10.1016/j.devcel.2007.01.019
- PubMed
- Google Scholar
1. Ramkumar N
2. Patel JV
3. Anstatt J
4. Baum B
(2021) Aurora B-dependent polarization of the cortical actomyosin network during mitotic exit
EMBO Reports 22:e52387.
https://doi.org/10.15252/embr.202152387
- PubMed
- Google Scholar
1. Rappaport R
(1985) Repeated furrow formation from a single mitotic apparatus in cylindrical sand dollar eggs
The Journal of Experimental Zoology 234:167–171.
https://doi.org/10.1002/jez.1402340120
- PubMed
- Google Scholar
1. Reckel S
2. Gehin C
3. Tardivon D
4. Georgeon S
5. Kükenshöner T
6. Löhr F
7. Koide A
8. Buchner L
9. Panjkovich A
10. Reynaud A
11. Pinho S
12. Gerig B
13. Svergun D
14. Pojer F
15. Güntert P
16. Dötsch V
17. Koide S
18. Gavin AC
19. Hantschel O
(2017) Structural and functional dissection of the DH and ph domains of oncogenic BCR-ABL tyrosine kinase
Nature Communications 8:2101.
https://doi.org/10.1038/s41467-017-02313-6
- PubMed
- Google Scholar
1. Reich JD
2. Hubatsch L
3. Illukkumbura R
4. Peglion F
5. Bland T
6. Hirani N
7. Goehring NW
(2019) Regulated activation of the PAR polarity network ensures a timely and specific response to spatial cues
Current Biology 29:1911–1923.
https://doi.org/10.1016/j.cub.2019.04.058
- PubMed
- Google Scholar
(2015) Kinetochore-localized PP1-sds22 couples chromosome segregation to polar relaxation
Nature 524:489–492.
https://doi.org/10.1038/nature14496
- PubMed
- Google Scholar
1. Rose LS
2. Lamb ML
3. Hird SN
4. Kemphues KJ
(1995) Pseudocleavage is dispensable for polarity and development in C. elegans embryos
Developmental Biology 168:479–489.
https://doi.org/10.1006/dbio.1995.1096
- PubMed
- Google Scholar
1. Schindelin J
2. Arganda-Carreras I
3. Frise E
4. Kaynig V
5. Longair M
6. Pietzsch T
7. Preibisch S
8. Rueden C
9. Saalfeld S
10. Schmid B
11. Tinevez JY
12. White DJ
13. Hartenstein V
14. Eliceiri K
15. Tomancak P
16. Cardona A
(2012) Fiji: an open-source platform for biological-image analysis
Nature Methods 9:676–682.
https://doi.org/10.1038/nmeth.2019
- PubMed
- Google Scholar
(2011) Targeting of the rhogef ECT2 to the equatorial membrane controls cleavage furrow formation during cytokinesis
Developmental Cell 21:1104–1115.
https://doi.org/10.1016/j.devcel.2011.11.003
- PubMed
- Google Scholar
(2015) Caenorhabditis elegans aurora a kinase is required for the formation of spindle microtubules in female meiosis
Molecular Biology of the Cell 26:4187–4196.
https://doi.org/10.1091/mbc.E15-05-0258
- PubMed
- Google Scholar
1. Tatsumoto T
2. Xie X
3. Blumenthal R
4. Okamoto I
5. Miki T
(1999) Human ect2 is an exchange factor for rho gtpases, phosphorylated in G2/M phases, and involved in cytokinesis
The Journal of Cell Biology 147:921–928.
https://doi.org/10.1083/jcb.147.5.921
- PubMed
- Google Scholar
1. Tsai MC
2. Ahringer J
(2007) Microtubules are involved in anterior-posterior axis formation in C. elegans embryos
The Journal of Cell Biology 179:397–402.
https://doi.org/10.1083/jcb.200708101
- PubMed
- Google Scholar
1. Tse YC
2. Werner M
3. Longhini KM
4. Labbe JC
5. Goldstein B
6. Glotzer M
(2012) Rhoa activation during polarization and cytokinesis of the early Caenorhabditis elegans embryo is differentially dependent on NOP-1 and CYK-4
Molecular Biology of the Cell 23:4020–4031.
https://doi.org/10.1091/mbc.E12-04-0268
- PubMed
- Google Scholar
1. Wagner E
2. Glotzer M
(2016) Local rhoa activation induces cytokinetic furrows independent of spindle position and cell cycle stage
The Journal of Cell Biology 213:641–649.
https://doi.org/10.1083/jcb.201603025
- PubMed
- Google Scholar
(2007) Astral signals spatially bias cortical myosin recruitment to break symmetry and promote cytokinesis
Current Biology 17:1286–1297.
https://doi.org/10.1016/j.cub.2007.06.070
- PubMed
- Google Scholar
1. White JG
2. Borisy GG
(1983) On the mechanisms of cytokinesis in animal cells
Journal of Theoretical Biology 101:289–316.
https://doi.org/10.1016/0022-5193(83)90342-9
- PubMed
- Google Scholar
1. Wickham H
2. Averick M
3. Bryan J
4. Chang W
5. McGowan L
6. François R
7. Grolemund G
8. Hayes A
9. Henry L
10. Hester J
11. Kuhn M
12. Pedersen T
13. Miller E
14. Bache S
15. Müller K
16. Ooms J
17. Robinson D
18. Seidel D
19. Spinu V
20. Takahashi K
21. Vaughan D
22. Wilke C
23. Woo K
24. Yutani H
(2019) Welcome to the tidyverse
Journal of Open Source Software 4:1686.
https://doi.org/10.21105/joss.01686
- Google Scholar
(2009) Polo-like kinase 1 directs assembly of the hscyk-4 rhogap/ect2 rhogef complex to initiate cleavage furrow formation
PLOS Biology 7:e1000110.
https://doi.org/10.1371/journal.pbio.1000110
- PubMed
- Google Scholar
Book
1. Wolpert L
(1960)
The mechanics and mechanism of cleavage

In: Wolpert L, editors. International Review of Cytology. Elsevier Science. pp. 163–216.
- Google Scholar
(2007) Asymmetric inheritance of mother versus daughter centrosome in stem cell division
Science 315:518–521.
https://doi.org/10.1126/science.1134910
- PubMed
- Google Scholar
(2005) An ECT2-centralspindlin complex regulates the localization and function of rhoa
The Journal of Cell Biology 170:571–582.
https://doi.org/10.1083/jcb.200501097
- PubMed
- Google Scholar
1. Zhang D
2. Glotzer M
(2015) The rhogap activity of CYK-4/mgcracgap functions non-canonically by promoting rhoa activation during cytokinesis
eLife 4:e08898.
https://doi.org/10.7554/eLife.08898
- PubMed
- Google Scholar
1. Zhao W
2. Fang G
(2005) Mgcracgap controls the assembly of the contractile ring and the initiation of cytokinesis
PNAS 102:13158–13163.
https://doi.org/10.1073/pnas.0504145102
- PubMed
- Google Scholar
1. Zhao P
2. Teng X
3. Tantirimudalige SN
4. Nishikawa M
5. Wohland T
6. Toyama Y
7. Motegi F
(2019) Aurora-A breaks symmetry in contractile actomyosin networks independently of its role in centrosome maturation
Developmental Cell 48:631–645.
https://doi.org/10.1016/j.devcel.2019.02.012
- PubMed
- Google Scholar
1. Zhou X
2. Zheng Y
(2013) Cell type-specific signaling function of rhoa gtpase: lessons from mouse gene targeting
The Journal of Biological Chemistry 288:36179–36188.
https://doi.org/10.1074/jbc.R113.515486
- PubMed
- Google Scholar
1. Zorba A
2. Buosi V
3. Kutter S
4. Kern N
5. Pontiggia F
6. Cho YJ
7. Kern D
(2014) Molecular mechanism of aurora A kinase autophosphorylation and its allosteric activation by TPX2
eLife 3:e02667.
https://doi.org/10.7554/eLife.02667
- PubMed
- Google Scholar

Decision letter

Mohan K Balasubramanian

Reviewing Editor; University of Warwick, United Kingdom
Anna Akhmanova

Senior Editor; Utrecht University, Netherlands

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

[Editors' note: this paper was reviewed by Review Commons.]

Thank you for submitting your article "Aurora A and cortical flows promote polarization and cytokinesis by inducing asymmetric ECT-2 accumulation" for consideration by eLife. Your article has been reviewed by 3 peer reviewers at Review Commons and an eLife referee, and the evaluation at eLife has been overseen by a Reviewing Editor (Mohan Balasubramanian) and Anna Akhmanova as the Senior Editor.

Based on the previous reviews and the revisions, the manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

The eLife expert raised some points which are transmitted verbatim below. I have read them carefully. I think the vast majority of points can be addressed by rewriting and providing explanations or toning down some of the conclusions. Also, the referee has asked that you provide some more experimental details, statistical methods, and additional citations.

Points arising from your response letter and revision:

1. In a revised manuscript, I am not convinced with their interpretation of the phenotype of air-1(RNAi);par-2(RNAi) zygotes. Given that either single par-2(RNAi) or single air-1(RNAi) abolished the anterior-enriched distribution of ECT-2::mNG (Figure 4 supplement 2), this data simply indicates the indispensable roles of both PAR-2 and AIR-1 in ECT-2 asymmetry, but they cannot conclude that ECT-2 asymmetry in air-1(RNAi) condition is due to PAR-2-dependent weaker cortical flows. Indeed, the anterior/the posterior ratio of ECT-2::mNG in air-1(RNAi) zygotes shown in Figure 4 supplement 2 is very close to 1.0 throughout mitosis, which is quite different from that in Figure 4A. This discrepancy should be addressed in the final manuscript.

2. The authors' response "While cytokinesis generally involves an equatorial contractile ring, furrow formation can be driven by an asymmetric – i.e. non-equatorial – accumulation of actomyosin. This behavior is exemplified during pseudocleavage during which the entire anterior cortex is enriched for actomyosin and the posterior is depleted of myosin (Figure 1 Supplement 2). Several published studies provide evidence that the asymmetric pattern of myosin accumulation contributes to cytokinesis (PMID 22918944, 17669650)."

The role of switching off the cortical flow from the P-to-A alone mode to the bidirectional mode in cytokinetic furrow formation has been reported in many papers (PMID: 27719759, 29963981, 32497213, etc.) as mentioned above. A simple unidirectional asymmetry is not sufficient in discussing the spatial regulation of cytokinesis.

3. The authors' response "However, as seen in e.g. ZYG-9 depleted embryos, ECT-2 is recruited to the posterior cortex in a centralspindlin-dependent manner". I don't understand the logic here. Has this been directly tested by, for example, depletion of ZEN-4 or CYK-4 in zyg-9(b244) mutant embryos? In Figure 3C (zyg-9(b244)), no particular enrichment of ECT-2 was observed at the posterior furrow, which is formed by centralspindlin.

4. "In contrast, the Gomez-Cavasos paper (PMID 32619481) shows in figure S2 that the PH domain is required for cortical localization of ECT-2; this paper does not focus extensively on the cortical accumulation of ECT-2". I think Gomez-Cavasos should also be cited as these provide complementary information as to the role of the PH domain.

5. In addition to the above, please answer either by rewriting or with experiments the points raised by the eLife referee below. As I read it, some experiments on Ect2 localization are required to firmly test your models. All other points raised might involve a balanced discussion of various observations and raising the limitations of the work.

eLife referee comments verbatim:Asymmetric actomyosin contractility plays key roles in various cellular activities. In C. elegans embryos, both the post-fertilization polarity establishment and cytokinesis depend on this process, which is known to be under the control of Rho GTPase. In this manuscript, the authors studied the molecular mechanism of symmetry breaking, focusing on ECT-2 RhoGEF, the crucial upstream activator of Rho, and its regulation by AIR-1, Aurora A kinase. Although the roles of these molecules both in polarity establishment and cytokinesis have already been reported, it remains unclear whether and how AIR-1 might regulate the activity of ECT-2.

By a combination of genetic manipulation and high-quality quantitive microscopy, the authors compared various perturbations and concluded that centrosomes and the cortical flow driven by actomyosin network play roles in the asymmetric cortical localization of ECT-2 while astral microtubules and TPXL-1, a conserved Aurora A regulator that recruits AIR-1 to the astral microtubules, are not essential for the ECT-2 asymmetry during cytokinesis in contrary to the previous reports. Then, the authors tested the hypothesis that AIR-1 induces cortical asymmetry by directly phosphorylating ECT-2 and presented the in vivo phenotypes of the ECT-2 constructs with mutations at the putative phosphorylation sites, which were consistent with their hypothesis.

(Influence of the cortical asymmetry at the mitotic entry)

There is a flaw in their interpretation of the results of various perturbations. Their model in Figure 7B cytokinesis depicts that NMY-2 is absent at the cell cortex during earlier stages of mitosis (pro~meta). This is not precise and misleading. In normal embryos, even after the pseudo-furrowing settles, NMY-2 doesn't disappear from the cell cortex and, importantly, is kept anteriorly enriched though at slightly lower intensity in smaller patches than in the earlier phase (eg. Tse et al. 2012 PMID:22918944, Figure 3). Actin filaments also remain more enriched in the anterior cortex than in the posterior cortex. This is a clear difference from the post-fertilization polarity establishment, in which uniform distribution of the actomyosin network is maintained until the entry of sperm breaks the symmetry.

During the early stages of mitosis, the cortical flow is suspended due to the inhibitory activity of CDK1. The onset of anaphase cancels this inhibition and triggers the contraction of the actomyosin network. If the symmetry of cortical actomyosin is already broken as in the normal embryos, even uniform activation of actomyosin throughout the cell will result in a cortical flow towards the region with a denser actomyosin network (= the anterior cap). If the cortical symmetry is not broken for some reason, it needs to be broken to cause the cortical flow. In theory, the target of symmetry breakage can be any component of the contractile actomyosin network.

The authors observed attenuated ECT-2 asymmetry during cytokinesis in the embryos depleted of SPD-5 (Figure 2C), PAR-3 (Figure 2D), PAR-2 (Figure 2 Supplement 1), NMY-2 (Figure 2 Supplement 1), and AIR-1 (Figure 4A, Figure 4 Supplement 2). In all these cases, the cortical ECT-2 at NEB was found more symmetric (A:P ratio at NEB =1.3, 1.0, 1.2, 1.3, and 1.1, respectively) than the normal embryos (1.4~1.6). In almost all the cases where ECT-2 was asymmetric at NEB (or metaphase), with Galpha(RNAi) as an exception, ECT-2 asymmetry during cytokinesis was normal or enhanced (Tubulin(RNAi)+nocodazole, nop-1(it142), zyg-9(b244), dhc-1(RNAi), tpxl-1(RNAi), tpxl-1(RNAi);zyg-9(b244), and saps^-1(RNAi)). There is a simple and strong correlation between the ECT-2 asymmetry upon mitotic entry and the ECT-2 asymmetry during cytokinesis.

The authors challenge the roles of astral microtubules and dynein in the polar relaxation during cytokinesis based on the observations of the cortical flow in the embryos in which microtubules and spindles are drastically messed up (Tubulin(RNAi)+nocodazole, zyg-9(b244), dhc-1(RNAi)). However, starting with the asymmetric actomyosin network that was successfully established after the fertilization, the anterior-directed cortical flow should occur spontaneously upon reactivation of the contractility following the anaphase onset even without any additional cue. The ECT-2 asymmetry during cytokinesis in these embryos can just be reflecting the fact that these treatments didn't completely disrupt the post-fertilization cortical polarity. Indeed, the ECT-2 asymmetry at the mitotic entry in embryos in Tub(RNAi)+noc and dhc-1(RNAi) embryos was more intense than in the normal embryos (A:P ratio at NEB = 1.9 and 1.8, respectively).

The role of switching of the cortical flow from the P-to-A alone mode to the bidirectional mode in cytokinetic furrow formation has been reported in many papers (PMID: 27719759, 29963981, 32497213, etc.). In this sense, the influence of the anterior centrosome/aster on the anterior cortex is crucial for the spatial control of cytokinesis. This must be a rationale for Mangal 2018 (PMID:29311228) to focus on the role of TPXL-1 in the clearing of anillin from the anterior cortex. Acceleration of furrow formation in correlation with the anterior shift of the anterior centrosome/aster has also been reported (PMID: 32497213). No such test has been performed for ECT-2 localization in this manuscript.

(Phosphorylation of ECT-2 by AIR-1)

Although the authors claim the role of centrosomes based on the spatial correlation, no direct evidence has been provided for the positive role of centrosomes in these embryos. For example, if the ECT-2 asymmetry during anaphase is regulated by centrosomes, disruption of the centrosomes after anaphase onset should disrupt the cortical ECT-2 asymmetry, which is turning over in seconds. In this sense, treatment with MLN8237 (Figure 4-Supplement 1) at metaphase is highly interesting. A caveat here is that MLN8237 is not really specific to Aurora A. It also inhibits Aurora B-INCENP at Ki = 27 nM (just 5~27-fold larger than the Ki for Aurora A) (de Groot et al. 2015). The concentration of MLN8237 used (20 uM) is 700x higher than the Ki for Aurora B-INCENP. The phenotype might be due to the inhibition of Aurora B. Indeed, the ECT-2 signal at the midzone, which is likely to depend on centralspindlin and Aurora B, was lost by the MLN8237 treatment. A mild delay in the removal of ECT-2 from the posterior cortex might have been caused by the inhibition of Aurora B/AIR-2 in addition to or instead of AIR-1.

Point mutations at the phosphorylation sites on ECT-2 are expected to compensate for the above issue of specificity and strengthen the author's theory of direct regulation of ECT-2 by AIR-1. However, as the authors admit in the "Limitations of this study" section, evidence for the in vivo phosphorylation of the putative sites is missing. In addition, it has not been tested whether AIR-1 can phosphorylate these sites. The Glotzer group has revealed key phospho-regulatory mechanisms in cytokinesis (PMID: 15282614, 15854913, 17488623, 19468300). In these works, they showed both the in vivo phosphorylation of the putative phospho-acceptor sites and the in vitro phosphorylation by a protein kinase as well as the in vivo phenotypes of the point mutants of the phosphorylation sites. Currently, what we can conclude from the data in Figures 5 and 6B is that some mutations in a loop in the ECT-2 PH domain mildly affect the cortical association of ECT-2. The gap between this and the phosphorylation of these sites by AIR-1 is huge.

Figure 6 tests AIR-1 depletion in nop-1(it142) mutant embryos (A) and the phenotype of the endogenous T634E mutation of ect-2 (B). Is 'Figure 6. AIR-1 is involved in central-spindle-dependent furrowing' (page 42) or 'AIR-1 affects centralspindlin-dependent furrowing' (page 56) an appropriate title for this figure? The dependency on centralspindlin or the central spindle has not been directly tested. Although synthetic defects in NOP-1 and centralspindlin suppress furrow formation, this doesn't necessarily mean that all the residual cortical activity in nop-1(it142) embryos relies on centralspindlin or the central spindle. In nop-1(it142) embryos, the NMY-2 cortical accumulation is globally weakened in comparison with the wild-type embryos. Additional depletion of AIR-1 might promote furrowing by facilitating the cortical recruitment of ECT-2 or prevent cytokinesis by suppressing the aster/centrosome-dependent pathway, which may or may not be dependent on centralspindlin. By the way, how does ECT-2 behave in these embryos?

Recommendation for authors:

The feedback loop from myosin to the biochemically upstream regulator RhoGEF is highly intriguing. Focusing on the mechanisms for this phenomenon might be more fruitful than spending time trying to obtain evidence for the phosphorylation on the sites that are not conserved during evolution and only weakly match with the consensus for the Aurora phosphorylation.

(page 17) "As during polarization, basal myosin levels appear to suffice, as ECT-2 asymmetry still increases during anaphase when both NOP-1 and CYK-4 are inactivated, indicating that bulk, cortical ECT-2 has a low level of RhoGEF activity." Difficult to understand the logical structure due to repeated "as".

Other points

(page 3) "Despite progress in our understanding of cytokinesis, gaps remain in our understanding of the mechanism by which asters spatially regulate RhoA activation." This is correct. However, considering the feedback loop shown by this work (dependence of ECT-2 asymmetry on NMY-2), this might be misleading. The point of regulation of the cortical contraction could be anywhere in the loop (RhoGEF, RhoGAP, Rho, formin, Rho-kinase, myosin phosphatase, myosin-II, actin, actin-bundling proteins, …).

(page 25 method of measuring Boundary Length) "ECT-2:mNG accumulation across all positions along the perimeter of the embryo was fitted to a regression model by GAM (Prediction Accumulation) in R Studio." It is not clear what the 'regression model' was and what the "GAM (Prediction Accumulation) in R Studio". Provide the precise (mathematical) description of the model and the proper reference to the method as well as the R source code.

(page 26 statistical test) It reads that t-test was performed between treatments at every time point. I am not sure whether this is an appropriate approach. Data from different time points from a time series are correlated. It is not clear how this should be reflected in handling the issue of multiple comparisons. Can't we get advice from an expert on statistics?

(Figure Supplement 1 and Figure 5F) Average Accumulation (same as 'Anterior:Posterior Ratio'?) is only shown. Isn't it better to include 'Accumulation on Anterior Cortex' and 'Accumulation on Posterior Cortex' as well for consistency with the other figures?

https://doi.org/10.7554/eLife.83992.sa1

Author response

1. General Statements

We thank the reviewers for their thoughtful and helpful comments. In general, the reviews were highly positive, although their reviews indicated parts of the manuscript that needed further clarification. We have made extensive changes that improve the clarity and rigor of this submission. We have performed several additional experiments which have extended our analysis in several ways detailed below. None of the conclusions have changed.

The following is a list of eight major changes implemented during the revisions. Point-by-point responses to the reviewers comments follow on subsequent pages.

1. The reviews made clear that we needed to more explicitly discuss the AIR-1 depletion phenotype. This phenotype is complex, it does not result in a complete loss of asymmetry, unlike, for example, depletion of the centrosome component SPD-5. This is because, in AIR-1 depleted embryos, a PAR-2 and cortical flow-dependent pathway induces PAR-2 accumulation at both anterior and posterior poles that induces flows from each pole to the lateral region (Reich 2019, Kapoor 2019, Zhao 2019, Klinkert 2019; PMIDs 31155349, 31636075, 30861375, 30801250). These flows also modulate ECT-2 localization. To clarify this point which came up in multiple reviews, we now include an explanation of the complexity of the AIR-1 phenotype and we present an analysis of ECT-2 localization in embryos depleted of both AIR-1 and PAR-2.

2. In addition to the 95% confidence intervals that were present on our graphs, we now include indications of the results of statistical tests of significance to the results of different treatments.

3. We have revised the analysis ECT-2 accumulation in two ways. First, in the previous draft, we assessed the anterior accumulation over the anterior 40% and the posterior 15% of the embryo. We have revised this analysis comparing the anterior and posterior 20% of the cortex, respectively. This is simpler and more logical in contexts where embryos are symmetric. In addition, we altered the measurements of the length of the posterior boundary. Previously we used a common threshold value, below which we counted pixels to assess boundary length. During the revisions, we noticed that this value was not appropriate for our mutant transgenes which accumulated to higher levels. Therefore, we revised our analysis pipeline such that, for each embryo, we measure the average intensity of the cortex in the anterior 60% of the embryo. We set a threshold of 0.85* this average anterior intensity value. As before, cortical positions below this threshold contribute to the boundary length. This is a more robust and simpler means of evaluating the size of the posterior domain. Neither of these changes affect any of our conclusions, but they are simpler and more rigorous.

4. Most of our figures include quantification of the degree of ECT-2 asymmetry as well as the average anterior and posterior accumulation of ECT-2 as a function of time. While the images show the intensity profiles across the embryo, previously, we did not explicitly show a quantification of the average intensity of ECT-2 as a function of position along the embryo. A new graph, Figure 2Bv, shows this for control embryos and embryos in which tubulin is depleted and depolymerized. This shows that the MT depolymerization results in lower accumulation at the posterior of the embryo and higher accumulation at the anterior.

5. We provide documentary and quantitative evidence that ZYG-9 depletion induces potent cortical flows (Figure 3c and Figure 3, supplement 3), further bolstering the central role of cortical flows in inducing ECT-2 asymmetry.

6. As requested by reviewer 2 (R2b), we have included the analysis of ECT-2 distribution in Gα depleted embryos. As expected due to the lack of spindle elongation, the displacement of ECT-2 from the posterior cortex is greatly attenuated.

7. As requested by reviewer 2 (R2d), we now show that ECT-2C fragments accumulate on the cortex in embryos depleted of ECT-2.

8. One other important point raised by several reviewers concerns the behavior of the ECT-2 T634E allele. This allele, due to the substitution of a phosphomimetic residue, accumulates on the cortex at about 50% the level of the wild-type version. To investigate the possibility that this quantitative difference was the cause of the phenotype, we depleted both the wildtype and mutant ECT-2 constructs by RNAi (these are the sole sources of ECT-2 in the animals). First, we find that wild-type ECT-2 can be depleted to 20% of wild type levels with only a 13% rate of cytokinesis failure (when T634E is depleted to 20%, embryos fail more than 50% of the time). Thus the two-fold reduction in cortical ECT-2 seen in T634E not likely highly significant (ECT-2 is not haploinsufficient). In addition, embryos with ECT-2 T634E initiate ingression in a timely manner, but the furrows ingress more slowly than wildtype. In contrast, depletion of ECT-2 to 20% results in a delay in furrow initiation, but once these furrows form, they ingress at rates similar rates to wild-type. Thus, the T634E variant exhibits a behavior that is quite distinct from that resulting from a (strong) reduction in the levels of wild-type ECT-2.

2. Point-by-point description of the revisions

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

Summary

R1a In this study the authors addressed how Ect2 localization is controlled during polarization and cytokinesis in the one-cell C. elegans embryo. Ect2 is a central regulator of cortical contractility and its spatial and temporal regulation is of uttermost importance. After fertilization, the centrosome induces removal of Ect2 from the posterior plasma membrane. During cytokinesis Ect2 activity is expected to be high at the cell equator and low at the cell poles. Similarly to polarization, the centrosome provides an inhibitory signal during cytokinesis that clears contractile ring components from the cell poles. Whether and how the centrosomes regulate Ect2 localization is not know and investigated in the study.

This is an accurate summary of the goals of this study.

R1b The authors start by filming endogenously-tagged Ect2 and find that Ect2 localizes asymmetrically, with high anterior and low posterior membrane levels during polarization and cytokinesis. They reveal that the centrosome together with myosin-dependent flows results in asymmetric Ect2 localization. Previous studies had suggested that Air1, clears Ect2 from the posterior during polarization and the authors expand those finding by showing that Air1 function is also required to displace Ect2 from the posterior membrane during cytokinesis.

To elucidate if Ect2 displacement is induced by phosphorylation of Ect2 by Air1, the authors investigate the localization of a C-terminal Ect2 fragment containing the membrane binding PH domain. When the predicted Air1 phosphorylation sites are mutated to alanine, the Ect2 fragment still localizes asymmetrically but exhibits increased membrane accumulation.

Finally, they investigate the functional role of Air-1 during furrow ingression. They demonstrate that embryos deficient of Air1 and NOP1 have impaired furrow ingression. Lastly, the authors sought to confirm that there is a direct effect of Air1 on Ect2 function by generating a phosphomimetic point mutation of Ect2 using Crispr. They find that the membrane localization of phosphomimetic Ect2 is reduced and consequently furrow ingression is impaired.

This is an accurate summary of our results.

Major comments

R1c It is not convincing that the six putative phosphorylation sites are targeted by the Air1. If Air1 phosphorylation displaces Ect2 from the membrane, a reduction in Ant/Post Ect2 ratio is expected in the phosphodeficient mutants, like after air1 RNAi. However this is not observed for cytokinesis or polarization (Figure 5D(i); E). This suggests that phosphorylation of those sites is not essential for the asymmetric Ect2 localization.

In otherwise wild-type embryos, phosphorylation of these sites is not required for asymmetric ECT-2 localization. Non-phosphorylatable ECT-2 variants exhibit asymmetric localization because these proteins relocalize due to myosin-directed flows. To test the role of phosphorylation, we examine the distribution of ECT-2 and ECT-2C fragments in myosin-depleted embryos in which the flows are blocked, under these conditions, transient local depletion is observed with the phosphorylatable variants, Figure 5E.

While AIR-1 promotes normal polarity establishment, as shown in several recent papers, cortical changes nevertheless occur in the absence of AIR-1. Specifically, a parallel PAR-2 dependent pathway induces weaker flows from both poles toward the equator. To further substantiate the effect of PAR-2 accumulation on ECT-2 accumulation in AIR-1 depleted embryos, we assayed ECT-2 accumulation in air-1(RNAi); par-2(RNAi) embryos (Figure 4, supplement 2). These results show that ECT-2 is nearly symmetric in these double depleted embryos. In addition we have edited the text to describe the unusual bi-polar PAR-2 accumulation that occurs in AIR-1 depleted embryos.

R1d The authors aim to demonstrate that phosphorylation of the identified sites is important for cytokinesis. For this they investigate contractile ring ingression in the phosphomimetic point mutation. Since ring ingression is slower and fails in nop1 mutant they authors conclude that this demonstrates a functional importance of this site. I am not surprised that embryos ingress slower in this mutant since Ect2 localization to the membrane is reduced. This however does not show that this phosphorylation site is the target of the centrosome signal. Importantly, authors would need to demonstrate that Rho signaling and thus Ect2 activity, is increased at the poles, when phosphodeficient Ect2 is the only Ect2 in the embryo.

The fact that a phosphomimetic residue at this site leads to reduced membrane localization is highly relevant, as we suggest that phosphorylation of this site contributes to the mechanism by which AIR-1 generates asymmetric ECT-2. Given the role of AIR-1 in regulating polarity, a version of ECT-2 that can not be phosphorylated would be predicted to be dominant lethal, necessitating a conditional expression strategy which does not currently exist in the early C. elegans embryo system (indeed we were unable to recover a T-> A allele at this site, despite extensive efforts). To avoid this issue, we used a viable, fertile, hypomorphic allele that is predicted to be less responsive to AIR-1 activity. The goal of this experiment was to evaluate whether the putative AIR-1 sites affect not only the NOP-1 pathway for furrow ingression, but also impact furrowing that is centralspindlin-dependent.

To complement this finding have performed experiments in which ECT-2 was partially depleted We used RNAi to partially deplete ECT-2 and ECT-2 T634E and measured the total embryo fluorescence of each ECT-2 variant and the kinetics of furrow ingression. Partial depletion of wt ECT-2, to ~ 20% of control levels leads to delay in furrow formation and all but 2/18 (11%) of embryos complete cell division. In contrast, a similar depletion of ECT-2T634E depletion results in a failure of furrow ingression in ~52 % of embryos. Furthermore, while ECT-2T634E embryos initiate furrowing with normal kinetics, they exhibit a slower rate of furrow ingression, in contrast, partial depletion of WT ECT-2 results in a delay in furrow initiation, but once initiated, the rate of furrow ingression is not significantly affected. These results demonstrate that ECT-2T634E behavior can not simply be explained by a modest reduction in membrane binding.

R1e The authors use the Aurora A inhibitor MLN8237: It was shown prior (De Groot et al., 2015) that this inhibitor is not highly specific for Aurora A, and that it also inhibits Aurora B. Thus experiments need to be repeated with MK5108 or MK8745. They should also be conducted during polarization. Why does Aurora A inhibition not abolish asymmetry? That would be expected?

The role of AIR-1 in symmetry breaking during polarization is previously published, including with chemical inhibitors (Reich 2019, Kapoor 2019, Zhao 2019, Klinkert 2019, PMID 31155349, 31636075, 30861375, 30801250). ECT-2 localization depends on both the spatial regulation of AIR-1 activity and the distribution of cortical factors that contribute to ECT-2 cortical association, as a result of cortical flows. During acute, chemical perturbation of AIR-1 it is likely that these factors, which were polarized prior to drug treatment, remain polarized, allowing the residual cortical ECT-2 to remain asymmetric. The reviewer is correct about the specificity of MLN8237 and we do not rely on it alone to demonstrate the role of AIR-1. Rather this experiment is a complement to our AIR-1 depletion studies, which are sufficient to establish specificity. We present this experiment merely to show that AIR-1 acutely regulates ECT-2 during cytokinesis in embryos that were entirely unperturbed during polarization.

R1f There is no statistical analysis of the results in the entire study. For all claims stating a change in Ant/Post Ect2 ratio or Ect2 membrane localization selected time points should be statistically compared: for example the main point of Figure 1 is that Ect2 becomes more asymmetric during anaphase. Thus a statistical analysis of the Ect2 ratio at anaphase onset (t=0s) and eg. t=90 s after anaphase onset should be performed; or Figure 3A nop-1 mutant Ant/Post Ect2 ratio during polarization: again statistical analysis of control and nop-1 mutant embryos is needed at a particular time point.

All of the graphs were presented with the mean of ~10 embryos per condition and included the 95% confidence intervals. In the revised manuscript, we have included tests of statistical significance, at each time point. While non-overlapping confidence intervals generally suggest statistical significance, we include these analyses on the graphs as it can be difficult to assess statistical significance when the confidence intervals overlap.

R1g The aim of Figure 2B is to demonstrate that Ect2 localization is independent of microtubules, however they still observe some microtubules with the Cherry-tubulin marker and those are even very close to the membrane and therefore could very well influence Ect2 on the membrane. Therefore I am not convinced that this experiment rules out that microtubules have no role in regulating Ect2 localization.

We do not exclude that microtubules play a contributing role in ECT-2 phosphoregulation, but rather we conclude that the primary cue is the centrosome. Indeed, microtubules can play an important role in controlling spindle positioning which affects the proximity of the centrosome to the cortex.

The manuscript states, “Despite significant depletion of tubulin and near complete depolymerization of microtubules (Figure 2B, insets), we observed strong displacement of ECT-2 from a broad region of the posterior cortex during anaphase (Figure 2B).” Thus, despite dramatic reductions in microtubules, not only does ECT-2 become polarized, it becomes hyperpolarized. In contrast, were microtubules directly involved in ECT-2 displacement, one would expect a reduction in polarization as a result microtubule depolymerization. Conversely, though SPD-5 depleted embryos contain far more microtubules than embryos in which microtubule assembly is suppressed, ECT-2 is not polarized in SPD-5 depleted embryos. Thus in the manuscript, we conclude, “Collectively, these studies suggest that ECT-2 asymmetry during anaphase is centrosome-directed.” This conclusion is well supported by the results shown.

R1h Throughout the paper the authors should tone down their statement that Air1 breaks symmetry by phosphorylating Ect2, since phosphorylation of Ect2 by Air2 is not shown.

We agree with this comment and will make the necessary edits to the text. Indeed, this is the reason why we had included the final section in our original draft, “Limitations of this study” which makes this point explicitly.

R1i I understand that the establishment of Ect2 asymmetry is important for polarization. However, how does asymmetric Ect2 localization result in more active Ect2 at the cell equator, which is required for the formation of the active RhoA zone? Would we not expect an accumulation of Ect2 at the cell equator, or if that is not the case more active Ect2 at the equator versus the poles?

The pseudocleavage furrow forms as a result of the anterior enrichment of active RHO-1 and its downstream effectors. There is no evidence for a local accumulation of active RHO-1 specifically at the site of the pseudocleavage furrow. Rather, this furrow forms at the boundary between the portion of the embryo where RHO-1 is active and the posterior of the embryo where RHO-1 is far less active (Figure 1 Supplement 2). We suggest that aster-directed furrowing during cytokinesis likewise results from asymmetric accumulation of the same components, without them necessarily being specifically enriched solely at the furrow.

While cytokinesis generally involves an equatorial contractile ring, furrow formation can be driven by an asymmetric – i.e. non-equatorial – accumulation of actomyosin. This behavior is exemplified during pseudocleavage during which the entire anterior cortex is enriched for actomyosin and the posterior is depleted of myosin (Figure 1 Supplement 2). Several published studies provide evidence that the asymmetric pattern of myosin accumulation contributes to cytokinesis (PMID 22918944, 17669650).

Minor comments

R1j Can the authors explain why the quantification of Ant/Post Ect2 ratio in control embryos differs in different figures? For example: in Figure 1D (i) a slight increase of Ect2 asymmetry ratio is seen at around 80 s after anaphase onset. In comparison, in Figure 2C (i) this increase is not obvious. Are those different genetic backgrounds?

In figure 1 D, time 0 begins at anaphase onset, whereas in 2C, time 0 is specified at the time of nuclear envelope breakdown (NEBD). The duration between NEBD and anaphase onset is ~130 sec and an increase in ECT-2 polarization is observed at 220 s post NEBD, ie 90 sec post anaphase onset comparable to that seen in Figure 1D.

R1k One key point of the paper is that myosin-dependent cortical flows amplify Ect2 asymmetry during polarization and cytokinesis. During polarization the data is convincing, however during cytokinesis Ect2 ratio is only slightly decreased after nmy-2 depletion, again is this decrease even significant?

Figure 3 supplement 1 shows a significant difference in ECT-2 asymmetry between control and myosin-depleted embryos.

R1l In the introduction: "Centralspindlin both induces relief of ECT-2 auto-inhibition and promotes Ect2 recruitment to the plasma membrane" it should be added 'Equatorial' membrane, since Ect2 membrane binding is, to my knowledge, not compromised in centralspindlin mutants or in Ect2 mutants that cannot bind centralspindlin.

Generally speaking, the reviewer is correct that cortical accumulation of ECT-2 globally is centralspindlin independent. However, as seen in e.g. ZYG-9 depleted embryos, ECT-2 is recruited to the posterior cortex in a centralspindlin-dependent manner. Thus centralspindlin can promote ECT-2 accumulation to the cortex and the site of that accumulation will be dictated by the position of the spindle midzone.

R1m Labels in the figures are often very small eg Figure 1 (ii-v) and difficult to read. In addition it is easier for the reader if the proteins shown in the fluorescent images is also labeled in the figure (eg Figure 2B add NG-Ect2).

These useful suggestions have been incorporated.

R1n Material and methods it should be mentioned which IPTG concentration was used.

The IPTG concentration (1 mM) has been added to the revised text.

R1o The authors speculate that the Air1 phosphorylation sites in Ect2 PH domain prevent binding to phospholipid due the negative charge. At the same time, the authors propose that the PH domain binds to a more stable protein on the membrane, which is swept along with the cortical flows and they propose anillin could be that additional binding partner. I might miss something, but do the authors suggest Ect2 has two binding partners: anillin and the phospholipids? It would be necessary to explain this better.

The authors should test if anillin represents the suggested myosin II dependent Ect2 anchor. For this they should check if Ect2 localization to the membrane is altered upon on anillin RNAi.

This summary of our model is largely correct, though we do not know the identity of the more stable cortical anchor(s). While we suspect the PH domain binds to a phospholipid, ECT-2 cortical localization also requires ~100 residues C-terminal to the PH domain. It is likely that this domain interacts with a cortical component.

In preliminary experiments, ECT-2 accumulation is not strictly anillin-dependent. However, functional redundancy may obscure a contribution of anillin. Anillin was mentioned simply because of the evidence for a physical interaction between ECT-2 and anillin (Frenete PMID 22514687). In the revised manuscript we also include the possibility that ECT-2 accumulations involves one or more anterior PAR proteins. The identity of the cortical anchor(s) is an interesting question for future studies. We consider this question beyond the scope of the current manuscript.

R1p The title of Figure 3 does not fit the statement the authors want to make, since the key point is how Ect2 polarization is affected and not membrane localization in general.

Thank you for this suggestion. The title has been changed to “Cortical flows contribute to asymmetric cortical accumulation of ECT-2”.

R1q In Figure 4A/C. After air1 depletion the authors observe a reduction in Ect2 asymmetry. Why are the centrosomes not marked in the figures? Because they cannot be detected? The authors would also need to show that the mitotic spindle and centrosomes are no altered by air1 RNAi in the zyg9 mutant. Otherwise the observed effect might be indirect.

Centrosomes are perturbed by depletion of AIR-1 (Hannak, PMID 11748251), but they are still detectable and their positions will be added to figure 4. As has been extensively demonstrated, AIR-1 depletion does lead to attenuated spindles and defects in spindle assembly, some of which are also seen TPXL-1 depleted embryos. These consequences of AIR-1 depletion do complicate the analysis, but this is typical of factors that regulate many processes. This is one of the key reasons why we used ZYG-9 depletion in combination with AIR-1 depletion to overcome these indirect effects.

R1r The authors state that tpxl-1 depletion attenuates Ect2 asymmetry, this is not seen in the quantification (Figure 4Bi). The main phenotype they observe is that Ect2 levels on the membrane increase (Figure 4 ii and iii). They go on testing the function of tpxl1 by depleting tpxl1 in the zyg9 mutant, where the centrosomes are close to the posterior cortex. Here they see no effect on Ect2 asymmetry. Based on that they conclude that tpxl1 has no role in this process. To me this finding is not surprising since the centrosome is close the cortex in zyg9 mutant embryos. Therefore sufficient amounts of active Air1 could reach the membrane and displace Ect2. Thus an amplification of the inhibitory signal by tpxl1 on astral microtubules might not be required. The authors need to mention this possibility and tone down their statment (also in the discussion) that tpxl1 is not required for this process.

In the text, we state, “Cortical ECT-2 accumulation is enhanced by TPXL-1 depletion, though the degree of ECT-2 asymmetry is unaffected (Figure 4B).… we observed robust depletion of ECT-2 at the posterior pole in zyg-9 embryos depleted of TPXL-1, but not AIR-1 (Figure 4C). We conclude that while AIR-1 is a major regulator of the asymmetric accumulation of ECT-2, the TPXL-1/AIR-1 complex does not play a central role in this process.” We consider this to be an accurate description of the results. In sum, we have found no evidence that TPXL-1 contributes to generating ECT-2 asymmetry, beyond its well established role in regulating spindle length and position. The are several other processes that are known to be AIR-1 dependent and TPXL-1 independent; these primarily involve the centrosome (Ozlu, PMID 16054030). Given that TPXL-1 associates with astral microtubules, the fact that microtubule depletion can enhance ECT-2 asymmetry also argues against a requirement for TPXL-1.

R1s It was shown that the C-terminus of Ect2 is sufficient and the PH domain is required for Ect2 membrane localization in C. elegans (Chan and Nance, 2013; Gomez-Cavazos et al., 2020). Papers should be cited.

Thank you for this helpful comment. Chan and Nance 2013 indeed shows that the ECT-2 C-term is sufficient to localize to the cell cortex. In contrast, the Gomez-Cavasos paper (PMID 32619481) shows in figure S2 that the PH domain is required for cortical localization of ECT-2; this paper does not focus extensively on cortical accumulation of ECT-2. We have cited Chan and Nance in the revised manuscript.

R1t The authors find that nmy-2 depletion results in loss of asymmetry for the Ect2 C-term and Ect2 3A fragment during polarization. Why is the same experiment not shown for cytokinesis?

Strong depletion of NMY-2 prevents polarity establishment, resulting in symmetric spindles, which in turn results in symmetric ECT-2 accumulation. Thus, the requested experiment would not provide significant additional information.

R1u Air1 is targeted to GFP-C-term Ect2 fragment via GFP-binding to determine the influence on GFP-C-term Ect2 localization (Figure 5F). They state that they see a reduction of Ect2 C-term but not of C-term 3A after targeting. The reader has to compare Figure 5D with F. Since the differences are not big, they need to compare the Ect2 C-term and Ect2 C-term 3A with and without Air1 targeting in the same graph (plus statistics). Otherwise this statement is not convincing.

It is not straightforward to directly compare ECT-2C in the presence and absence of GBP-mCherryAIR-1, because the GBP:AIR-1 fusion protein recruits a large fraction of ECT-2C to the centrosome. For this reason we think it is best to compare the behavior over time of ECT-2C and ECT-2C3A in the presence of GBP-mCherry-AIR-1. At the onset of anaphase, these two fragments localize similarly, but they then diverge over time.

R1v In Figure 6A the authors determine the contribution of air1 to furrowing. For this they deplete air1 in the nop1 mutant. According to previous studies, air1 mutants have a monopolar spindle. How can the authors analyze the function of air1 in cytokinesis when the spindle is monopolar? Did the authors do partial air1 depletion? They authors need to show that there is not major effect on the spindle and centrosome for their conditions. For comparison air1(RNAi) alone has to be included, otherwise the experiment is not conclusive.

AIR-1 depletion does not result in a monopolar spindle in C. elegans embryos, though the spindle is attenuated and disorganized (PMID 9778499). TPXL-1 depletion also results in short, well organized spindles (PMID 19889842). The concerns are the reason we performed the ZYG-9 depletion experiments in Figure 4C to ensure the centrosomes are proximal to the cortex.

R1w Upon air1(RNAi) in the nop1 mutant NMY2 intensity seems decreased and not increased. Can the authors comment on that, since that is opposite of what is expected.

This is expected as previous studies have shown that NOP-1 contributes to RHO-1 activation during polarization and cytokinesis (Tse, PMID 22918944). (NOP stands for No Pseudocleavage).

R1x In Figure 6B they introduce a phosphomimetic point mutation in S634 [sic, T634] in the endogenous Ect2 locus. It not clear why the authors chose this site out of the six putative sites and why they only chose one and not 3 or 6 sites? This needs some explanation.

In our early work with ECT-2 transgenes, we found that a T634E mutation strongly affected cortical ECT-2C, so we decided to assess its affect on the function and localization of endogenous ECT-2. While we were able to recover a T634E variant, we were not able to recover a T634A variant, despite considerable effort. Based on these experiences, we anticipated that we would be unable to recover a mutant version of ECT-2 in which all sites were changed to phosphomimetic.

R1y In the model (Figure 7) no astral microtubules are shown during pronuclear meeting and metaphase. Astral microtubules are present at this stage and should be added to the schematic.

MTs will be added to the figure.

Reviewer #1 (Significance (Required)):

R1z. The centrosomes inhibit cortical contractility during polarization and cytokinesis in the one-cell C. elegans embryo. Centrosome localized Air1 was proposed to be part of this inhibitory signal, however the phosphorylation target of Air1 is not known. The identification of Ect2 as a phosphorylation target of Air1 would be a great advancement in the field. However, the presented manuscript lacks convincing data that Ect2 is the phosphorylation target of Air1 during polarization and cytokinesis.

We explicitly acknowledge that we have not directly shown that AIR-1 phosphorylates ECT-2. However, we have shown that (i) AIR-1 inhibits cortical ECT-2 localization, (ii) the negative regulator of AIR-1, SAPS^-1, promotes AIR-1 cortical accumulation, (iii) that the cortical localization domain of ECT-2 has putative AIR-1 sites, which, when mutated to non-phosphorylatable residues leads to increased cortical accumulation of ECT-2 (and (iv) phosphomimetic residues reduce its cortical accumulation), and (v) that these AIR-1 sites are required to render GFP-ECT-2C responsive to GBP-AIR-1. For these reasons we feel that our data makes a strong, albeit indirect, case that AIR-1 regulates ECT-2, even though we clearly acknowledge that we do not directly show that AIR-1 directly phosphorylates ECT-2. Direct proof would require the demonstration that AIR-1 phosphorylates ECT-2 in vivo. This would be difficult to show as ECT-2 phosphorylation is likely transient, it likely affects only a subset of the total ECT-2 pool, and it likely results in loss of membrane association of ECT-2. As it it not possible to synchronize C. elegans embryos, biochemical analysis would be very difficult. Even a phosphor-specific antibody for the putative ECT-2 phosphosites might not be particularly informative, as it would be predicted to give a diffuse cytoplasmic signal.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

R2a. In this work, Longhini and Glotzer investigate the localization of an essential regulator of polarity and cytokinesis, RhoGEF ECT-2, in the one-cell C. elegans embryo. The authors show that centrosome localized Aurora A kinase (AIR-1 in C. elegans) and myosin-dependent cortical flows are critical in asymmetric ECT-2 accumulation at the membrane. Since membrane interaction of ECT-2 is dependent on the Pleckstrin homology domain present at the Cterminus of ECT-2, they further analyzed the importance of putative AIR-1 consensus sites present in this domain.

The authors linked the relevance of these sites in controlling ECT-2 localization and its significance on cytokinesis. The manuscript is well written, the work is interesting, and the data quality is high.

We thank the reviewer for their critique.

Major comments:

R2b. In Figure 2, the authors claim that the centrosomes and the position of the mitotic spindle are critical in regulating the asymmetric enrichment of ECT-2 at the membrane. To test the relevance of spindle positioning on ECT-2 localization, the authors depleted PAR-3 and PAR-2. The authors observed that the ECT-2 asymmetry is affected in these settings. However, PAR-3 or PAR-2 depletion impacts polarity, which is critical for many cellular processes, including spindle positioning. Can the authors try to specifically misposition the spindle without affecting polarity? For instance, by depleting Galpha/GPR-1/2 and assessing the impact of such depletion on ECT-2 localization.

Thank reviewer for good suggestion. We have performed the suggested experiment (presented in Figure 2, supplement 2). As one might predict, ECT-2 starts out polarized as Gα is not required for polarity establishment. During anaphase, ECT-2 becomes more symmetric in Gα depleted embryos as compared to wild-type.

R2c. I wonder why the intensity of ECT-2 at the anterior and posterior membrane decreases in air-1(RNAi) post anaphase onset (Figure 4A)? Moreover, I fail to observe a significant asymmetric distribution of ECT-2 in embryos depleted for PERM-1. Therefore it appears that the difference between DMSO and MLN8237-treated embryos is not substantial (at least in the images)?

We do not have a complete or rigorous explanation for all the changes in cortical ECT-2, but they are highly reproducible. We speculate that there are cell cycle regulated changes in ECT-2 accumulation, in addition to its regulation by AIR-1. For example, in figure 1, a strong reduction in both anterior and posterior cortical ECT-2 is evident beginning at approximately -350 sec, which may reflect the initial stages of Cdk1 activation. This may result from cell cycle regulated modulation of ECT-2, as there is evidence that mammalian ECT-2 is subject to a very potent inhibition membrane association by Cdk1 (PMID 22172673). Alternatively, there could be cell cycle modulation of the cortical factor that serves as the “co-anchor” of ECT-2. The ability of GBP-AIR-1 to induce GFP-ECT-2C dissociation also appears cell cycle regulated.

Consistent with a cell cycle regulated component, note that NEBD is delayed in AIR-1 depleted embryos (PMID 17669650, 17419991, 30861375). This delay results in a shorter interval between NEBD and e.g. the peak in Cdk1 activation, explaining the earlier decrease in AIR-1(RNAi) embryos vs. control, relative to NEBD.

Our quantitative analysis indicates a significant increase of cortical ECT-2 upon treatment with MLN8237. In addition, the quantitation in the previous version did show a significant polarization of ECT-2 in PERM-1-depleted embryos prior to treatment. We have revised this figure to simply show an acute increase in cortical ECT-2 upon drug treatment, as the focus of this experiment was solely to show that ECT-2 cortical accumulation is acutely responsive to chemical inhibition during cytokinesis in otherwise normal embryos.

–The data in Figure 5 and 6 are exciting but raise a few concerns:

R2d. The authors show that ECT-2C localization mimics the localization of endogenous tagged ECT-2. However, all these analyses with ECT-2C and various mutants are performed in the presence of endogenous ECT-2. Can the author check the localization of these mutant strains in conditions where the endogenous proteins are depleted? I understand that the cortical flow would be perturbed in conditions where endogenous ECT-2 is depleted. However, I suspect that one can analyze the anaphase-specific distribution.

We have examined ECT-2C localization in embryos depleted of ECT-2. Cortical localization of ECT-2C is not dependent upon endogenous ECT-2. This result is now shown in figure 5 supplement 1. However, as the reviewer suggested, embryos depleted of ECT-2 do not show a high degree of ECT-2C asymmetry as ECT-2 is required for the cortical flows that amplify the symmetry breaking during polarization. During cytokinesis, ECT-2C does show a modest change in localization at the poles; the extent of the polar reduction is limited and the changes are symmetric as ECT-2 displacement causes spindles to be symmetrically positioned and limits their elongation during anaphase.

R2e. Can the author comment on why ECT-2C does not accumulate at a similar level as ECT-2C(3A or 6A) at the cell membrane when AIR-1 is depleted (compare Figure 5D with Supplemental Figure 5)?

When ECT-2C(3A or 6A) are expressed in otherwise wild-type embryos, embryo polarization occurs, resulting in anterior-directed flows that concentrate the factor(s) that enables the anterior enrichment of ECT-2 (and ECT-2C 3A/6A). By contrast, when AIR-1 is depleted, most embryos exhibit a “bipolar” phenotype in which PAR-2 is recruited to both anterior and posterior poles, and the actomyosin network becomes somewhat concentrated laterally (PMID 30801250, 30861375, 31636075). The differential positioning of the actomyosin network in AIR-1 depleted embryos is likely responsible for the interesting difference that the reviewer points out. This section of the results states. “Nevertheless, these variants accumulated in an asymmetric manner. ECT-2C asymmetry temporally correlated with anteriorly-directed cortical flows (Figure 5 D,E), raising the possibility that asymmetric accumulation of endogenous ECT-2 drives flows that cause asymmetry of the transgene, irrespective of its phosphorylation status.”

R2f (c). Does the cortical localization of the ECT-2C(6A) mutant become symmetric upon further depletion of AIR-1? Of course, if the asymmetric distribution of ECT-2C(6A) is dependent on the presence of endogenous protein in the cellular milieu, the point raised earlier will help address this concern.

We have not performed this exact experiment with ECT-2C-3A though we have performed it with a longer ECT-2 C-terminal fragment (aa 559-924). As expected, due to the considerations described above, the asymmetry of ECT-2C-3A is reduced when AIR-1 is depleted. Likewise, ECT-2C-6A is becomes symmetric when endogenous ECT-2 is depleted due to the dependence of its asymmetry on cortical flows, as discussed above.

In the revised manuscript, we provide additional explanation of the AIR-1 depletion phenotype which will explain the origin of the asymmetric distribution of ECT-2.

R2g.. The authors predict that the AIR-1 mediated phosphorylation delocalizes ECT-2 from the polar region of the cell cortex. Since the posterior spindle pole is much closer to the posterior cortical region, the delocalization is much more robust at the posterior cell membrane. I wonder why targetting AIR-1 at the membrane (GBP-mCherry-AIR-1) does not entirely abolish GFP-ECT-2C membrane localization? Can the author include the localization of GBPmCherry-AIR-1 in the data? Also, do we know for sure if GBP-mCherry-AIR-1 is kinase active?

The GBP-mCherry-AIR-1 transgene was obtained from the Gönczy lab which demonstrated that it has some activity (PMID 30801250). Given that centrosomal AIR-1 (as compared to astral AIR-1) is the primary pool of AIR-1 responsible for modulating cortical ECT-2 levels, it is a not clear that the GBPfused form of AIR-1 is as active as the centrosomal pool of AIR-1; indeed we suspect it is significantly less active, similar to the manner in which TPXL-1/AIR-1 appears less active towards ECT-2 than centrosomal AIR-1. Indeed as the reviewer suggests, were this pool of AIR-1 highly active, we would expect that its cortical recruitment would preclude embryo polarization, and this transgene would cause lethality when expressed with a GFP-tagged cortical protein. These concerns notwithstanding, we do observe a specific reduction in the anterior accumulation of ECT-2C as compared to ECT-2C3A, suggesting that this form of the kinase has some ability to modulate ECT-2C.

Co-expression of GFP-ECT-2C with GBP-mCherry-AIR-1 induces the centrosomal/astral accumulation of GFP-ECT-2C, which is highly visible in the figure and not seen in the absence of GBP-mCherry-AIR-1. Not surprisingly, the co-expression also induces a cortical pool of GBP-mCherry-AIR-1 that is not seen in the absence of GFP-ECT-2C. These redistributions indicate formation of the complex between GFPECT-2C and GBP-mCherry-AIR-1. The mCherry-AIR-1 images could be added as insets to the figure, but in our opinion, they would not make a substantive contribution, given the dramatic accumulation of centrosomal GFP-ECT-2C.

R2h (e). The authors show that centrosomal enriched AIR-1 [spd-5(RNAi)], but not the astral microtubules localized AIR-1 [tpxl-1(RNAi)], is vital for ECT-2 membrane localization. Interestingly, the authors showed that AIR-1 acts in the centralspindlin-directed furrowing pathway (Figure 6A). I wonder if the authors can combine NOP-1 depletion with TPXL-1 depletion? I guess this will further help to exclude the function of TPXL-1 in the centralspindlin-directed furrowing pathway.

We would like to clarify that our data indicates that AIR-1 acts on both the centralspindlin-independent furrowing (e.g. the anterior furrow in 4C), as well as centralspindlin-dependent furrowing (Figure 6). While the experiment the reviewer proposes appears simple in theory, the interpretation is potentially a bit more complex, due to the role of TPXL-1 in spindle elongation, which can affect centralspindl-indirected furrowing. That said, there are two published experiments and one experiment in the manuscript that indicate that centralspindlin dependent furrowing can occur in TPXL-1 depleted embryos. First, Lewellyn et al. showed that while tpxl-1(RNAi) embryos furrow, tpxl-1(RNAi); zen-4(RNAi) embryos do not, suggesting centralspindlin can function in the absence of TPXL-1. Second, the same paper shows that embryos doubly depleted of TPXL-1 and GPR-1/2 exhibit multiple furrows. Our previous work has shown that furrowing in Galpha-depleted embryos is centralspindlin dependent (Dechant and Glotzer). Furthermore, in the current manuscript we found that embryos depleted of both TPXL-1 and ZYG-9 form posterior furrows (8/8 embryos, 6/8 furrows were strong furrows) although the appearance of these furrows is delayed, presumably due to the reduction in spindle elongation due to TPXL-1-depletion. As described in the manuscript, these posterior furrows have been previously shown to be centralspindlin dependent and NOP-1 independent.

In accordance with these results, and in direct response to the reviewer’s specific suggestion, we do observe furrowing in nop-1(it142); TPXL-1(RNAi) embryos (10/10 embryos furrow, 9/10 complete cytokinesis). Thus, all of the available results indicate that TPXL-1 is largely dispensable for centralspindlin dependent furrowing. However, the role of TPXL-1 in centralspindlin-dependent furrowing is not a focus of the manuscript, thus we do not favor including this result, as it distracts from the primary focus of the study.

R2i (f). Why do NMY-2-GFP cortical levels appear lower in 30% of the embryos that show various degrees of cytokinesis defects (Figure 6A)?

There are a number of possible origins of the variability. As shown in (Reich 2019, Kapoor 2019, Zhao 2019, Klinkert 2019, PMID 31155349, 31636075, 30861375, 30801250), AIR-1 depletion results in variable polarization (unpolarized PAR-2, bipolarized PAR-2, anterior PAR-2, posterior PAR-2). Furthermore, spindles in AIR-1 depleted embryos exhibit somewhat variable positioning. While we were unable to correlate these sources of variability with furrow formation, these results demonstrate that AIR-1 depletion impairs furrowing directed by centralspindlin, which was not entirely expected, given that (i) AIR-1 depletion potently suppresses NOP-1 dependent flows of cortical myosin, as evidenced by the loss of an anterior furrow in AIR-1(RNAi); nop-1(it142) embryos and (ii) centralspindlin directed furrowing can occur in the posterior in ZYG-9 depleted embryos both in the presence or absence of AIR-1 (Figure 4C).

R2j (g). The authors report that phosphomimetic mutation at the phospho-acceptor residue in ECT-2 impacts its cortical accumulation. This strain, together with NOP-1 depletion, affects furrow ingression. One explanation for this phenotype is that phosphomimetic mutant weakly accumulates at the membrane. However, one interesting observation is that ECT-2T634E enriches at the central spindle (Figure 6B, panel 120 sec), which somehow I could not find in the text. Could this additional localization of ECT2 at the central spindle contribute to the cytokinesis defects that the authors have observed? The microscopy images the authors have included show that ECT-2T634E significantly localizes at the equator at the time of furrow initiation. Can the authors add the localization of ECT2 wildtype and ECT-2T634E in NOP-1 depleted conditions where they see an apparent impact on the cytokinesis? Similarly, if the authors include the localization of NMY-2 in these conditions-it will further add more weightage to the data.

We regularly detect trace amounts of ECT-2 on the central spindle and this is slightly enhanced at in the ECT-2T634E mutant. However, given the large cytoplasmic pool of ECT-2, it seems unlikely that the slight enrichment of ECT-2 on the central spindle significantly affects the cortical pool of ECT-2, though the reduction in cortical ECT-2 may facilitate its enrichment on the central spindle.

As shown in figure 3B, depletion of NOP-1 does not dramatically affect cortical ECT-2 levels in wild-type embryos. Likewise, we did not observe a significant effect of NOP-1 depletion in ECT-2 T634E, thus we decided not to include this negative result.

As discussed in general point 8, we suggest the modest reduction in the membrane pool of ECT-2 is unlikely to be the primary cause of the T634E, but rather the ability of AIR-1 to modulate induce its relocalization. Consistent with this interpretation, the embryos that failed ingression tended to have more symmetric spindles, which could limit the residual cortical flows that facilitate furrow ingression.

Minor comments:

R2k -An explanation of how the timing of NEBD was analyzed in multiple settings would be helpful.

Depending on the experiment, we used either ECT-2:mNG fluorescence (it is excluded from the nucleus until NEBD) and/or the Nomarski images to score NEBD.

R2l -The authors mentioned on p. 6-'Despite significant depletion of tubulin…..during anaphase'. These experiments are performed in the near complete depolymerization of microtubules; thus, regular anaphase will not establish. I understand that the authors are monitoring localization wrt the timing similar to anaphase in the non-perturbed condition, and thus a bit of change in the sentence is required.

Thank you for highlighting this point. We have substituted “following mitotic exit” for “anaphase”. In these images, mitotic exit can be scored by the emergence of contractility.

R2m-After testing the relevance of SPD-5 (that primarily acts on PCM and not on centrioles)-the authors write on p. 6 that 'two classes of explanation…early embryo'. I did not understand the importance of this sentence here.

To clarify, we deleted the words “classes of” from the sentence in question and following that sentence we added the word, “first” indicating that we were explaining the first of the two possible explanations.

R2n-The observed impact of spd-5 (RNAi) on ECT-2 localization could be because of the effects of SPD-5 depletion on centrosomal AIR-1? The authors can link the impact of SPD-5 depletion not only with the centrosome but also with AIR-1 in the discussion.

Indeed, it is well established that SPD-5 is required for centrosomal AIR-1 (Hamill DR, et. Al Dev Cell 2002). The revised discussion now states, “Specifically, during both processes, ECT-2 displacement requires the core centrosomal component SPD-5, which is required to recruit AIR-1 to centrosomes{Hamill et al., 2002, #1201}, but ECT-2 displacement is not inhibited by depolymerization of microtubules and it does not require the AIR-1 activator TPXL-1 (see below).”

R2o-In the various Figure legends, sometimes the authors mention time '0' as anaphase, and other time as anaphase onset.

In all cases, anaphase onset was intended and the legends will be corrected.

Reviewer #2 (Significance (Required)):

R2p The manuscript is well written, the work is interesting, and the data quality is of good quality.

We thank the reviewer for their encouragement as well as for their thoughtful critique!

Reviewer #3 (Evidence, reproducibility and clarity (Required)):

R3a Symmetry breaking is the process by which uniformity of the system is broken. Many biological systems, such as the body axes establishment and cell divisions in embryos, undergo symmetry breaking to pattern cellular interior design. C. elegans zygote has been a classic model system to study the molecular mechanism of symmetry breaking. Previous studies demonstrated critical roles of centrosomes and microtubules in breaking symmetry in the actin cytoskeleton during anterior-posterior polarization and cytokinesis. It, however, remains elusive how centrosomes and/or microtubules regulate the assembly and contractility of the actin cytoskeleton. Recent reports identified Aurora-A AIR-1 as the key centrosomal kinase that suppresses the function of the actin cytoskeleton, but little is known about a substrate of the kinase during symmetry breaking events.

Longhini and Glotzer proposed in this manuscript that RhoGEF ECT-2 plays a critical role in symmetry breaking of the actin cytoskeleton under the control of AIR-1 kinase. Kapoor and Kotak (2019) previously proposed the same GEF as a downstream effector of centrosomes, but this work did not provide direct evidence for ECT-2 as the AIR-1 effector. This manuscript identified three putative phospho-acceptor sites in the PH domain of ECT-2 that render ECT-2 responsive to inhibition by AIR-1. Although this manuscript lacks direct in vivo and in vitro evidence for phosphorylation of ECT-2 by AIR-1 kinase, the above findings reasonably support a model where in AIR-1 promotes the local inhibition of ECT-2 on the cortex. Design of the experiments, the quality of images, and data analysis are reasonable, and the main text was written very well. The main conclusion of this work will attract many readers in cell and developmental biology fields. I basically support its publication in the journals supported by Review Commons with minor revisions (see below).

We thank the reviewer for their encouraging remarks and helpful comments.

Minor comments

R3b (1) In Figures 2A and 2B, the authors claimed apparent correlation between spindle rocking and ECT-2 displacement. However, because both MTs and ECT-2 in Fig2AB images are blur, I cannot convince myself whether ECT-2 intensities on the cortex showed negative correlation with the distance between the posterior centrosome and the cortex. The authors may want to provide quantitative data set and use a statistical test to support this conclusion.

Only figure 2A focuses on the rocking. The important structure to assess is the position of the centrosome, as the astral arrays of microtubules are largely radially symmetric (except towards the spindle midzone). As this point in the manuscript were were not discriminating between the astral microtubules and the centrosomes, rather focusing on the overall position of the aster as a whole.

Figures 2B, 2D, Figure 2 Supplements 1 and 2, Figure 3C, and Figure 4B, summarized in figure 7A provide quantitive evidence that the centrosome-cortex distance is an important determinant of ECT-2 cortical accumulation.

R3c (2) Figure 2D would [sic; presumably should] show a ratio between the anterior/posterior pole and the lateral cortex.

The reviewer is presumably noticing that the lateral cortex is brighter than the poles when PAR-3 is depleted. While we agree with this assessment, the point of this experiment was to evaluate whether both centrosomes are equally capable of regulating cortical ECT-2 at the respective poles. It appears to us that comparing the anterior and posterior poles is the appropriate measurement to make to address this point and comparison of the poles to the lateral cortex in par-3(RNAi) vs control would be confusing to readers.

R3d (3) In Figure 3D, the authors need to clarify why they measured ECT-2 dynamics only within the "anterior pole". It would be reasonable to measure ECT-2 dynamics by FRAP and cortical high-speed live imaging on the posterior and the lateral cortex during symmetry breaking.

We measured ECT-2 recovery at a variety of sites with similar recovery kinetics. The comparison of ECT-2 dynamics on anterior and posterior furrows were shown in order to compare ECT-2 dynamics on centralspindlin-dependent and -independent furrows.

We now provide additional supplemental data on ECT-2 dynamics during symmetry breaking. When ECT-2 is polarized, the residual signal is too low to obtain a measure of its recovery.

R3e (4) In Figure 4 supplement, a difference between with or without ML8237 seems marginal. The authors need to show a statistical test to claim "rapid enhancement of cortical ECT-2 after ML8237 treatment".

We will provide a statistical analysis. As the inhibitor affects ECT-2 globally, the anterior/posterior ratio doesn’t change significantly. To avoid confusion, we now present total cortical ECT-2 levels upon anaphase onset in this experiment as this is the most relevant parameter.

R3f (5) I would strongly suggest the authors to clearly state in the first paragraph of discussion that "this working hypothesis is not supported by direct evidence for phosphorylation of ECT-2 by AIR-1 kinase in vitro and in vivo." It should be reasonable to weaken the statement "by Aurora A-dependent phosphorylation of the ECT-2 PH domain" in p13.

We agree with the underlying sentiment (as indicated by the “limitations” section that was present in the original version) and we have revised these sentences accordingly: “Our studies suggest that asymmetric, posteriorly-shifted, spindle triggers an initial focal displacement of ECT-2 from the posterior cortex by Aurora A-dependent phosphorylation of the ECT-2 PH domain, though the evidence for this phosphorylation event is indirect.”

Reviewer #3 (Significance (Required)):

See the second paragraph of the Evidence, Reproducibility, and Clarity section.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Based on the previous reviews and the revisions, the manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

The eLife expert raised some points which are transmitted verbatim below. I have read them carefully. I think the vast majority of points can be addressed by rewriting and providing explanations or toning down some of the conclusions. Also, the referee has asked that you provide some more experimental details, statistical methods, and additional citations.

Points arising from your response letter and revision:

1. In a revised manuscript, I am not convinced with their interpretation of the phenotype of air-1(RNAi);par-2(RNAi) zygotes. Given that either single par-2(RNAi) or single air-1(RNAi) abolished the anterior-enriched distribution of ECT-2::mNG (Figure 4 supplement 2), this data simply indicates the indispensable roles of both PAR-2 and AIR-1 in ECT-2 asymmetry, but they cannot conclude that ECT-2 asymmetry in air-1(RNAi) condition is due to PAR-2-dependent weaker cortical flows. Indeed, the anterior/the posterior ratio of ECT-2::mNG in air-1(RNAi) zygotes shown in Figure 4 supplement 2 is very close to 1.0 throughout mitosis, which is quite different from that in Figure 4A. This discrepancy should be addressed in the final manuscript.

These PAR-2 related experiments were not included in the first version of the manuscript, but they were added in response to the Review Commons referees. As is well documented in several papers (PMIDs 30801250, 30861375, 31636075, 31155349), AIR-1 depletion does not eliminate cortical reorganization, and PAR-2 accumulates, aberrantly, to the cortex in when AIR-1 is depleted. A reproducible accumulation of lateral ECT-2 is observed during mitosis in AIR-1 depleted embryos; which is associated with ectopic furrowing (see point 11, below). While this increase in ECT-2 is visible in the images, it is not surfaced by our standard quantification method which focuses on the anterior and posterior domains. Given this and the established interplay between AIR-1 and PAR-2, we felt it appropriate to include an analysis of ECT-2 localization in embryos deficient in both AIR-1 and PAR-2.

The referee states, “they cannot conclude that ECT-2 asymmetry in air-1(RNAi) condition is due to PAR-2-dependent weaker cortical flows.” We do not discuss PAR-2 in the context of cortical flows in the manuscript. The only sentences in the manuscript that explicitly discuss PAR-2 state ”Likewise, ECT-2 asymmetry during cytokinesis is reduced in embryos depleted of PAR-2.” and “The residual asymmetry in ECT-2 accumulation in AIR-1 depleted embryos is further reduced by co-depletion of PAR-2.”

While there are differences in these datasets, we consider them minor; when the ratios are plotted together only 4/35 timepoints show any degree of statistically significant difference. The pattern of anterior accumulation is virtually identical in the two experiments, but the experiment in the supplemental figure shows a slightly stronger ECT-2 accumulation in the posterior. While the images were processed in the same manner, the two figures were generated with strains with distinct markers: while both figures contain the same ECT-2:mNG allele, in Figure 4A it is paired with with mCh:Tub while Figure 4 supp 2 is paired with NMY-2:mKate, as the two figures were addressing different aspects of the phenotype.

2. The authors' response "While cytokinesis generally involves an equatorial contractile ring, furrow formation can be driven by an asymmetric – i.e. non-equatorial – accumulation of actomyosin. This behavior is exemplified during pseudocleavage during which the entire anterior cortex is enriched for actomyosin and the posterior is depleted of myosin (Figure 1 Supplement 2). Several published studies provide evidence that the asymmetric pattern of myosin accumulation contributes to cytokinesis (PMID 22918944, 17669650)."

The role of switching off the cortical flow from the P-to-A alone mode to the bidirectional mode in cytokinetic furrow formation has been reported in many papers (PMID: 27719759, 29963981, 32497213, etc.) as mentioned above. A simple unidirectional asymmetry is not sufficient in discussing the spatial regulation of cytokinesis.

We regret that the context of our answer was apparently not sufficiently clear. The original referee asked “However, how does asymmetric Ect2 localization result in more active Ect2 at the cell equator, which is required for the formation of the active RhoA zone? Would we not expect an accumulation of Ect2 at the cell equator, or if that is not the case more active Ect2 at the equator versus the poles?”

Our initial response was narrowly targeted to this question: “while cytokinesis generally involves an equatorial contractile ring, furrow formation can be driven by an asymmetric – i.e. non-equatorial – accumulation of actomyosin.” Our response implies that while bidirectional mode is the norm, there are cases in which a unilateral mode is sufficient (see bolded words in the response). The pseudocleavage furrow is a prime example of this behavior.

3. The authors' response "However, as seen in e.g. ZYG-9 depleted embryos, ECT-2 is recruited to the posterior cortex in a centralspindlin-dependent manner". I don't understand the logic here. Has this been directly tested by, for example, depletion of ZEN-4 or CYK-4 in zyg-9(b244) mutant embryos? In Figure 3C (zyg-9(b244)), no particular enrichment of ECT-2 was observed at the posterior furrow, which is formed by centralspindlin.

The underlying logic is that the posteriorly positioned, bipolar spindle induces the cortical accumulation of centralspindlin on the adjacent cortex. This pool of centralspindlin apparently recruits ECT-2 from the cytoplasm, despite the strong inhibitory activity of AIR-1 in this region. We infer that this binding may reflect a distinct mode of ECT-2 accumulation, as it appears impervious to AIR-1 activity. Indeed, FRAP experiments indicate this pool of ECT-2 has a larger immobile fraction than the cortical pool of ECT-2 elsewhere in the embryo.

In figure 3D of Werner et al., 2007, we demonstrated that the posterior furrow that forms in embryos with posterior furrows requires centralspindlin. In that particular experiment, MEL-26 depletion was used to induce the spindle to assemble in the posterior. In subsequent unpublished work, we have found that the posterior furrow similarly requires centralspindlin when ZYG-9 depletion is used to reposition the spindle (both MEL-26 and ZYG-9 enhance spindle assembly, albeit via distinct molecular mechanisms).

Additionally, in Tse et al., 2012, we showed that the posterior furrow is largely independent of NOP-1 and that embryos depleted of both NOP-1 and CYK-4 fail to furrow altogether.

ECT-2 is readily detected on the posterior furrow once they ingress further than the time point shown in Figure 3C, as shown in Figure 3 Supplement 4.

4. "In contrast, the Gomez-Cavasos paper (PMID 32619481) shows in figure S2 that the PH domain is required for cortical localization of ECT-2; this paper does not focus extensively on the cortical accumulation of ECT-2". I think Gomez-Cavasos should also be cited as these provide complementary information as to the role of the PH domain.

In this case, the focus of this section is which parts of ECT-2 are sufficient for membrane recruitment of ECT-2. That PH domains contribute to membrane accumulation of GEFs has been extensively documented (eg PMID 17007612 from 2006). For example, the PH and the C terminal regions of HsEct2 are required for its membrane accumulation (PMID 27926870). It would not be appropriate to cite Gomez-Cavasos, without citing these papers. The requirement for a PH domain in cortical association does not seem significant enough to warrant a summary of the literature, particularly since membrane binding is the canonical function of PH domains. The somewhat surprising result was that the PH domain was not sufficient, as previously shown by Chan and Nance, which is cited.

5. In addition to the above, please answer either by rewriting or with experiments the points raised by the eLife referee below. As I read it, some experiments on Ect2 localization are required to firmly test your models. All other points raised might involve a balanced discussion of various observations and raising the limitations of the work.

eLife referee comments verbatim:

Asymmetric actomyosin contractility plays key roles in various cellular activities. In C. elegans embryos, both the post-fertilization polarity establishment and cytokinesis depend on this process, which is known to be under the control of Rho GTPase. In this manuscript, the authors studied the molecular mechanism of symmetry breaking, focusing on ECT-2 RhoGEF, the crucial upstream activator of Rho, and its regulation by AIR-1, Aurora A kinase. Although the roles of these molecules both in polarity establishment and cytokinesis have already been reported, it remains unclear whether and how AIR-1 might regulate the activity of ECT-2.

By a combination of genetic manipulation and high-quality quantitive microscopy, the authors compared various perturbations and concluded that centrosomes and the cortical flow driven by actomyosin network play roles in the asymmetric cortical localization of ECT-2 while astral microtubules and TPXL-1, a conserved Aurora A regulator that recruits AIR-1 to the astral microtubules, are not essential for the ECT-2 asymmetry during cytokinesis in contrary to the previous reports. Then, the authors tested the hypothesis that AIR-1 induces cortical asymmetry by directly phosphorylating ECT-2 and presented the in vivo phenotypes of the ECT-2 constructs with mutations at the putative phosphorylation sites, which were consistent with their hypothesis.

This referee does not mention that we unambiguously demonstrate that AIR-1 plays an important role in generating cortical asymmetry during cytokinesis. In particular, we show that the ability of centrosomes to inhibit cortical contractility during anaphase depends on AIR-1. Furthermore, we show that AIR-1 is specifically required to induce the displacement of ECT-2 from the cortex during both polarization and cytokinesis. We identify sites on ECT-2 that when mutated to non-phosphorylatable residues increase ECT-2 membrane accumulation one of which, which when mutated to a phosphomimetic residue is sufficient to reduce the cortical accumulation of ECT-2. Further we show that a non-phosphorylatable substitution of these sites impacts the response of ECT-2 C-terminal fragments to an AIR-1:GBP fusion protein. Finally, although AIR-1 clearly regulates the bulk flow of cortical myosin during anaphase, this pool of myosin is largely dependent on a non-essential protein called NOP-1. Given the evolutionary novelty of NOP-1 we further showed that a putative AIR-1 phosphosite also regulates the ability of the conserved cytokinetic regulator centralspindlin to induce cytokinesis.

Author response image 1

Download asset Open asset

Depletion of MRCK-1 does not affect the distribution of cortical ECT-2 during anaphase.

MRCK-1 depletion was confirmed by observing the loss of the cortical cap of co-expressed NMY-2:mKate during metaphase.

(Influence of the cortical asymmetry at the mitotic entry)

There is a flaw in their interpretation of the results of various perturbations. Their model in Figure 7B cytokinesis depicts that NMY-2 is absent at the cell cortex during earlier stages of mitosis (pro~meta). This is not precise and misleading. In normal embryos, even after the pseudo-furrowing settles, NMY-2 doesn't disappear from the cell cortex and, importantly, is kept anteriorly enriched though at slightly lower intensity in smaller patches than in the earlier phase (eg. Tse et al. 2012 PMID:22918944, Figure 3). Actin filaments also remain more enriched in the anterior cortex than in the posterior cortex. This is a clear difference from the post-fertilization polarity establishment, in which uniform distribution of the actomyosin network is maintained until the entry of sperm breaks the symmetry.

During the early stages of mitosis, the cortical flow is suspended due to the inhibitory activity of CDK1. The onset of anaphase cancels this inhibition and triggers the contraction of the actomyosin network. If the symmetry of cortical actomyosin is already broken as in the normal embryos, even uniform activation of actomyosin throughout the cell will result in a cortical flow towards the region with a denser actomyosin network (= the anterior cap). If the cortical symmetry is not broken for some reason, it needs to be broken to cause the cortical flow. In theory, the target of symmetry breakage can be any component of the contractile actomyosin network.

While we dispute that there is a flaw in our interpretations, we do not dispute that NMY-2 does not disappear from the cortex during polarity maintenance, and have revised the model schematic in Figure 7B accordingly. However, our results indicate that this pool of myosin does not play a role in the organization of myosin upon anaphase onset and we have seen no evidence that this anterior cap is sufficient to direct anterior-directed cortical flows. As previously shown (PMID 21737681), polarized myosin clusters assemble normally during anaphase in MRCK-1 depleted embryos. MRCK-1 is known to be required for cortical myosin accumulation during the establishment phase (Figure 6, PMID 19923324).

Notably, the anterior cap of myosin during polarity maintenance (i.e. during mitosis) is qualitatively different from that during interphase. Myosin clusters during maintenance phase are far smaller and less clustered than those present during establishment phase or anaphase. Likewise the organization of actin filaments are quite different during these stages. During maintenance phase Arp-2/3 branched filaments predominate whereas a significant pool of unbranched Formin nucleated filaments assemble during establishment phase and anaphase. These distinct pools of myosin and actin appear to exhibit distinctly different abilities to induce contractions. To substantiate this assertion, we examined the asymmetry of ECT-2 distribution during cytokinesis in control and MRCK-1 depleted embryos. As indicated in the accompanying figure, MRCK-1 depletion has no significant impact on the distribution of ECT-2 during cytokinesis.

This result was predictable. As shown in Figure 2 Supplement 2, although Gα-depleted embryos enter anaphase with asymmetric ECT-2 (and there is no evidence to suggest that Gα-depletion perturbs the myosin cap during mitosis). Despite this preexisting asymmetry, as anaphase initiates, these embryos do not exhibit potent anterior-directed flows that lead to increasingly asymmetric ECT-2, rather ECT-2 becomes progressively more symmetric as anaphase proceeds. The progressive symmetrization of these embryos likely results from the symmetric position of the spindle and its reduced elongation during anaphase. These changes to spindle length and position result in the anterior and posterior centrosomes lying equidistant – and rather distal from the anterior and posterior cortices.

The referees comment “In theory, the target of symmetry breakage can be any component of the contractile actomyosin network” seems to suggest that symmetry breaking is a one time event and that the strength and duration of this symmetry breaking is immaterial. However, as the results of our study shows, dramatically exemplified in Figure 3C and summarized in Figure 7A, the nature of the symmetry breaking events during anaphase have dramatically different consequences on cortical ECT-2 and myosin depending upon the position of centrosomes relative to the cortex. Indeed, figure 2A provides an example of how cortical ECT-2 changes acutely in response to the position of the posterior aster as the spindle rocks. To emphasize this point, we added this sentence to the section of the discussion that focuses on cortical flows, “Further, we speculate that centrosomal AIR-1 not only breaks symmetry, but that this regulation of ECT-2 by centrosomal AIR-1 continues throughout anaphase.”

Author response image 2

Download asset Open asset

Summary figure comparing the average degree of ECT-2 asymmetry at NEBD/Anaphase Onset vs the maximal asymmetry of ECT-2 during anaphase.

These values are not correlated.

The authors observed attenuated ECT-2 asymmetry during cytokinesis in the embryos depleted of SPD-5 (Figure 2C), PAR-3 (Figure 2D), PAR-2 (Figure 2 Supplement 1), NMY-2 (Figure 2 Supplement 1), and AIR-1 (Figure 4A, Figure 4 Supplement 2). In all these cases, the cortical ECT-2 at NEB was found more symmetric (A:P ratio at NEB =1.3, 1.0, 1.2, 1.3, and 1.1, respectively) than the normal embryos (1.4~1.6). In almost all the cases where ECT-2 was asymmetric at NEB (or metaphase), with Galpha(RNAi) as an exception, ECT-2 asymmetry during cytokinesis was normal or enhanced (Tubulin(RNAi)+nocodazole, nop-1(it142), zyg-9(b244), dhc-1(RNAi), tpxl-1(RNAi), tpxl-1(RNAi);zyg-9(b244), and saps^-1(RNAi)). There is a simple and strong correlation between the ECT-2 asymmetry upon mitotic entry and the ECT-2 asymmetry during cytokinesis.

Partial overlap in the requirements for polarization and asymmetric cytokinesis is expected, as these processes rely on a shared machinery. Additionally the degree of asymmetric accumulation of ECT-2 during anaphase depends on the asymmetric positioning of centrosomes, which depend on embryo polarization. In addition, there are exceptions to the correlation the referee cites, and they are highly informative. Gα – depleted embryos polarize normally and enter anaphase with control levels of ECT-2 asymmetry but, upon anaphase onset the degree of the asymmetry of ECT-2 declines. Conversely, the extent of ECT-2 asymmetry prior to anaphase onset in ZYG-9 depleted embryos is similar to wild-type yet during anaphase, ZYG-9-depleted embryos exhibit radically more highly asymmetric ECT-2 during anaphase than control embryos. Contrary to the assertion of the referee, ECT-2 asymmetry during mitosis and its asymmetry during cytokinesis are not well correlated.

The authors challenge the roles of astral microtubules and dynein in the polar relaxation during cytokinesis based on the observations of the cortical flow in the embryos in which microtubules and spindles are drastically messed up (Tubulin(RNAi)+nocodazole, zyg-9(b244), dhc-1(RNAi)). However, starting with the asymmetric actomyosin network that was successfully established after the fertilization, the anterior-directed cortical flow should occur spontaneously upon reactivation of the contractility following the anaphase onset even without any additional cue. The ECT-2 asymmetry during cytokinesis in these embryos can just be reflecting the fact that these treatments didn't completely disrupt the post-fertilization cortical polarity. Indeed, the ECT-2 asymmetry at the mitotic entry in embryos in Tub(RNAi)+noc and dhc-1(RNAi) embryos was more intense than in the normal embryos (A:P ratio at NEB = 1.9 and 1.8, respectively).

See the responses to points 6 and 8 above. Furthermore, “drastically messed up” it is not an accurate description of the spindles in ZYG-9 depleted embryos. These spindles exhibit normal bipolarity and morphology and they segregate chromosomes normally. These spindles are small and, due to a combination of the position of sperm entry and the bias in cortical forces, they assemble close to the posterior cortex.

Were dynein (or microtubules) required for polar relaxation during anaphase, it is reasonable to expect that dynein depletion (or MT disassembly) would result in a reduction in polar relaxation during anaphase, whereas the results show that dynein depletion results in a dramatic increase in polar relaxation.

The role of switching of the cortical flow from the P-to-A alone mode to the bidirectional mode in cytokinetic furrow formation has been reported in many papers (PMID: 27719759, 29963981, 32497213, etc.). In this sense, the influence of the anterior centrosome/aster on the anterior cortex is crucial for the spatial control of cytokinesis. This must be a rationale for Mangal 2018 (PMID:29311228) to focus on the role of TPXL-1 in the clearing of anillin from the anterior cortex. Acceleration of furrow formation in correlation with the anterior shift of the anterior centrosome/aster has also been reported (PMID: 32497213). No such test has been performed for ECT-2 localization in this manuscript.

We concur that the anterior centrosome regulates cortical behavior, including controlling ECT-2 localization. However, because the posterior centrosome is closer to the posterior cortex than their anterior counterparts, the equatorial directed myosin-dependent cortical flows are more pronounced in the posterior domain than in the anterior domain (e.g. Figure 3C). Indeed, the dramatic differences in ECT-2 and NMY-2 localization in ZYG-9 depleted embryos (Figure 3C) is likely to result from the combination of the close proximity of both centrosomes to the posterior cortex and the absence of the anterior centrosome from the anterior domain.

It is important, furthermore, to not only consider the flows of myosin foci that appear, but also the rate at which such foci assemble in the anterior and posterior domains. Two domains can exhibit similar rate of flows, but if the foci are smaller and less numerous in one domain than the other, the net accumulation of myosin in the two domains can differ dramatically.

Importantly, this manuscript focuses on the mechanisms that regulate both the assembly of myosin foci downstream of RHO-1 activation and the subsequent flows (centrosomal AIR-1 inhibiting ECT-2 accumulation), the finding that ECT-2 localization depends upon these flows, and the proposal that asymmetric ECT-2 could function to sustain these flows. These findings extend our current understanding of the mechanism and is consistent with all the data in the literature to our knowledge.

(Phosphorylation of ECT-2 by AIR-1)

Although the authors claim the role of centrosomes based on the spatial correlation, no direct evidence has been provided for the positive role of centrosomes in these embryos. For example, if the ECT-2 asymmetry during anaphase is regulated by centrosomes, disruption of the centrosomes after anaphase onset should disrupt the cortical ECT-2 asymmetry, which is turning over in seconds. In this sense, treatment with MLN8237 (Figure 4-Supplement 1) at metaphase is highly interesting. A caveat here is that MLN8237 is not really specific to Aurora A. It also inhibits Aurora B-INCENP at Ki = 27 nM (just 5~27-fold larger than the Ki for Aurora A) (de Groot et al. 2015). The concentration of MLN8237 used (20 uM) is 700x higher than the Ki for Aurora B-INCENP. The phenotype might be due to the inhibition of Aurora B. Indeed, the ECT-2 signal at the midzone, which is likely to depend on centralspindlin and Aurora B, was lost by the MLN8237 treatment. A mild delay in the removal of ECT-2 from the posterior cortex might have been caused by the inhibition of Aurora B/AIR-2 in addition to or instead of AIR-1.

We demonstrate that ECT-2 polarization during cytokinesis depends upon the core centrosomal component SPD-5, though it does not require the astral MTs that emanate from centrosomes.

We have shown the involvement of AIR-1 in ECT-2 regulation by both depletion of AIR-1 and its regulator SAPS^-1 and a gain of function approach using GBP-AIR-1.

The dramatic difference in ECT-2 localization in embryos deficient in ZYG-9 and AIR-1 as compared to embryos deficient in ZYG-9 and TPXL-1 further supports models in which the centrosomal, I.e. not the astral, pool of AIR-1 is relevant (TPXL-1 is required for AIR-1 to associate with astral microtubules).

The experiments with MLN8237 were merely included to complement the AIR-1 depletion studies and to show that acute inhibition can impact ECT-2 accumulation in otherwise unperturbed embryos. We have no basis to we assert, nor do we assert that these treatments fully inhibit AIR-1 activity. We also do not rule out some affect on AIR-2.

Point mutations at the phosphorylation sites on ECT-2 are expected to compensate for the above issue of specificity and strengthen the author's theory of direct regulation of ECT-2 by AIR-1. However, as the authors admit in the "Limitations of this study" section, evidence for the in vivo phosphorylation of the putative sites is missing. In addition, it has not been tested whether AIR-1 can phosphorylate these sites. The Glotzer group has revealed key phospho-regulatory mechanisms in cytokinesis (PMID: 15282614, 15854913, 17488623, 19468300). In these works, they showed both the in vivo phosphorylation of the putative phospho-acceptor sites and the in vitro phosphorylation by a protein kinase as well as the in vivo phenotypes of the point mutants of the phosphorylation sites. Currently, what we can conclude from the data in Figures 5 and 6B is that some mutations in a loop in the ECT-2 PH domain mildly affect the cortical association of ECT-2. The gap between this and the phosphorylation of these sites by AIR-1 is huge.

The mutational studies on ECT-2 must be considered in context of the loss of function studies of AIR-1 (reduction in AIR-1 activity resulting in increased ECT-2 accumulation; reduction in SAPS-1 activity resulting in increase in AIR-1 activity and an decrease in ECT-2 accumulation. Furthermore, these residues affect the response of an GFP-ECT-2 fusion protein to a GBP-AIR-1 fusion protein). While we would very much like to study the dynamic phosphorylation of ECT-2 in vivo (it is likely highly regulated in space and time) at this moment there are no tools available suitable to do so. The majority of the in vivo phosphorylation studies the referee mentions were performed in bulk lysates from human cells which can be synchronized; this is not possible with C. elegans embryos. Further, in the singular case where in vivo phosphorylation was shown in nematode embryos, a phospho-specific antibody was used to label a localized pool of ZEN-4; in this case, ECT-2 phosphorylation by AIR-1 triggers its delocalization. Demonstration of ECT-2 phosphorylation by AIR-1 in vitro would be nice, but neither a positive result nor a negative result would be particularly informative as kinases can be promiscuous in vitro and we can not rule out a requirement for other factors.

Figure 6 tests AIR-1 depletion in nop-1(it142) mutant embryos (A) and the phenotype of the endogenous T634E mutation of ect-2 (B). Is 'Figure 6. AIR-1 is involved in central-spindle-dependent furrowing' (page 42) or 'AIR-1 affects centralspindlin-dependent furrowing' (page 56) an appropriate title for this figure? The dependency on centralspindlin or the central spindle has not been directly tested. Although synthetic defects in NOP-1 and centralspindlin suppress furrow formation, this doesn't necessarily mean that all the residual cortical activity in nop-1(it142) embryos relies on centralspindlin or the central spindle. In nop-1(it142) embryos, the NMY-2 cortical accumulation is globally weakened in comparison with the wild-type embryos. Additional depletion of AIR-1 might promote furrowing by facilitating the cortical recruitment of ECT-2 or prevent cytokinesis by suppressing the aster/centrosome-dependent pathway, which may or may not be dependent on centralspindlin. By the way, how does ECT-2 behave in these embryos?

Furrow formation in wild-type embryos involves both NOP-1 and centralspindlin. Whereas depletion of NOP-1 has a dramatic effect on global cortical myosin, it has a limited effect on formation of the contractile ring, which ingresses to completion with near normal kinetics. Conversely, embryos deficient in centralspindlin subunits (CYK-4 or ZEN-4) form slowly ingressing furrows that only partially ingress. Embryos deficient in both NOP-1 and CYK-4 fail to form furrows. Thus, these two genes function in parallel pathways upstream of ECT-2. These findings are well established in the literature and also shown in figure 3B (PMID 22918944, 26252513, 32619481).

That said, as the referee indicates, depletion of AIR-1 increases cortical ECT-2 and induces a modest degree of hyper-contractility. For example, as shown in Figure 4A, AIR-1 depleted one-cell embryos form multiple furrows during anaphase. Formally, AIR-1 depletion could result in bypass suppression in embryos deficient in either CYK-4 and/or NOP-1. However, this hypercontractility depends upon NOP-1 during pseudocleavage (Figure 5E PMID 31636075) and during cytokinesis as embryos deficient in both AIR-1 and NOP-1 form a single furrow which is positioned at approximate midplane of the spindle (Figure 6A), suggesting it is centralspindlin directed. Indeed, it was precisely because AIR-1 depletion primarily enhances NOP-1 dependent contractility, we thought it would be relevant to examine whether AIR-1 has some impact of centralspindlin-directed furrowing. This was the underlying logic for the experiments shown in figure 7.

Nevertheless, to formally rule out the possibility of bypass suppression, we examined furrowing in embryos deficient in AIR-1, NOP-1, and CYK-4. Embryos deficient in all three factors (nop-1(it142); air-1(RNAi); cyk-4(RNAi)) fail to form furrows during anaphase (and pseudocleavage) (100% of embryos, n=8). The pattern of ECT-2 accumulation…Such embryos exhibit the membrane invaginations characteristic of embryos with weakened cortex (PMID 20808841). There were no significant differences in ECT-2 accumulation in (nop-1(it142); air-1(RNAi); cyk-4(RNAi)) embryos as compared to air-1(RNAi) embryos. This sentence was added to the results to reflect these findings, “To test whether AIR-1 depletion does not bypass the requirement for CYK-4 and NOP-1 in ECT-2 activation, we depleted both AIR-1 and CYK-4 in nop-1(it142) embryos. These triply deficient embryos fail to furrow during anaphase (100%, n=8).”

Given the facts above, the finding that AIR-1 depletion and a phosphomimetic substitution of T634 in ECT-2 affects furrowing behavior in NOP-1 deficient embryos is consistent with AIR-1 involvement in centralspindlin-dependent furrowing and ‘AIR-1 affecting central-spindlindependent furrowing’.

We thank the referee for pointing out the title on page 42, it has been revised to “AIR-1 is involved in centralspindlin-dependent furrowing.”

Regarding the inactivation of NOP-1 having only a very modest effect on cortical ECT-2 (figure 3B); depletion of AIR-1 results in a significant increase in the accumulation of ECT-2 on the posterior cortex (Figure 4B), and the phosphomimetic substitution of T634 in ECT-2 reduces its cortical accumulation.

Recommendation for authors:

The feedback loop from myosin to the biochemically upstream regulator RhoGEF is highly intriguing. Focusing on the mechanisms for this phenomenon might be more fruitful than spending time trying to obtain evidence for the phosphorylation on the sites that are not conserved during evolution and only weakly match with the consensus for the Aurora phosphorylation.

This suggestion is beyond the scope of the current manuscript.

(page 17) "As during polarization, basal myosin levels appear to suffice, as ECT-2 asymmetry still increases during anaphase when both NOP-1 and CYK-4 are inactivated, indicating that bulk, cortical ECT-2 has a low level of RhoGEF activity." Difficult to understand the logical structure due to repeated "as".

Thank you for pointing out this confusing sentence. We have revised it as follows:

“During cytokinesis, basal myosin levels appear to be sufficient to promote asymmetric ECT-2 accumulation, as ECT-2 asymmetry increases during anaphase even when both NOP-1 and CYK-4 are attenuated. We infer that bulk, cortical ECT-2 has a low level of RhoGEF activity."

Other points

(page 3) "Despite progress in our understanding of cytokinesis, gaps remain in our understanding of the mechanism by which asters spatially regulate RhoA activation." This is correct. However, considering the feedback loop shown by this work (dependence of ECT-2 asymmetry on NMY-2), this might be misleading. The point of regulation of the cortical contraction could be anywhere in the loop (RhoGEF, RhoGAP, Rho, formin, Rho-kinase, myosin phosphatase, myosin-II, actin, actin-bundling proteins, …).

We have changed the statement to "Despite progress in our understanding of cytokinesis, gaps remain in our understanding of the mechanism by which asters spatially regulate actomyosin contractility."

(page 25 method of measuring Boundary Length) "ECT-2:mNG accumulation across all positions along the perimeter of the embryo was fitted to a regression model by GAM (Prediction Accumulation) in R Studio." It is not clear what the 'regression model' was and what the "GAM (Prediction Accumulation) in R Studio". Provide the precise (mathematical) description of the model and the proper reference to the method as well as the R source code.

We have added the following text to the methods section “ECT-2:mNG accumulation across all positions along the perimeter of the embryo was fitted to a regression model by Generalized Additive Model (GAM) (Prediction Accumulation) in R Studio using the mgcv:gam function.” For additional information the reviewer could refer to https://www.rdocumentation.org/ packages/mgcv/versions/1.8-41/topics/gam

(page 26 statistical test) It reads that t-test was performed between treatments at every time point. I am not sure whether this is an appropriate approach. Data from different time points from a time series are correlated. It is not clear how this should be reflected in handling the issue of multiple comparisons. Can't we get advice from an expert on statistics?

While we are not experts on statistics, we believe that t-test is appropriate. Though data from different time points are correlated, the tests do not involve comparisons of data from neighboring time points. Rather we compare the data at a given time point between conditions. Regarding multiple comparisons, in each case, experimental results are compared to controls, no multiple comparisons are performed (ie we do not compare two different experimental treatments to each other). Finally, in each case, we provide 95% confidence intervals which provide a direct indication of the experimental noise.

(Figure Supplement 1 and Figure 5F) Average Accumulation (same as 'Anterior:Posterior Ratio'?) is only shown. Isn't it better to include 'Accumulation on Anterior Cortex' and 'Accumulation on Posterior Cortex' as well for consistency with the other figures?

Average accumulation is shown in cases where we focus on temporal regulation, as opposed to the spatial regulation. For example, in Figure 1Bi we show average accumulation to demonstrate the overall changes in ECT-2 accumulation during the cell cycle. This is also the case in Figure 4 Supplement 1, where we measure the overall changes in ECT-2:mNG in response to global AIR-1 inhibition with a chemical inhibitor. This is also the case in Figure 5F, where we track the overall accumulation of GFP:ECT-2C and GFP:ECT-2C-3A ,in the presence of GBP:AIR-1, which globally colocalizes with these C-terminal fragments of ECT-2.

https://doi.org/10.7554/eLife.83992.sa2

Article and author information

Author details

Katrina M Longhini

Department of Molecular Genetics and Cell Biology, University of Chicago, Chicago, United States

Contribution
Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review and editing

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-6600-7083
Michael Glotzer

Department of Molecular Genetics and Cell Biology, University of Chicago, Chicago, United States

Contribution
Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Methodology, Writing – original draft, Project administration, Writing – review and editing

For correspondence
mglotzer@uchicago.edu

Competing interests
No competing interests declared

"This ORCID iD identifies the author of this article:" 0000-0002-8723-7232

Funding

National Institute of General Medical Sciences (R35GM127091)

Katrina M Longhini

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Dan Dickinson (UT, Austin) for sharing the LP229 strain and Pierre Gönczy (EPFL) for sharing the strain expressing GBP::AIR-1. We also thank Pierre Gönczy, Ed Munro (U of Chicago), and Ashley Rich (Duke University) for helpful comments on the manuscript. Some strains were provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440). This work was supported by NIH grant R35GM127091.

Senior Editor

Anna Akhmanova, Utrecht University, Netherlands

Reviewing Editor

Mohan K Balasubramanian, University of Warwick, United Kingdom

Version history

Preprint posted: May 25, 2022 (view preprint)
Received: October 6, 2022
Accepted: December 12, 2022
Accepted Manuscript published: December 19, 2022 (version 1)
Version of Record published: December 29, 2022 (version 2)

Copyright

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.