Edwin M Munro et al. (2005) International Worm Meeting "Realistic computer simuations that predict cellular dynamics from molecular interactions."

Garrett Odell, Edwin M Munro

[International Worm Meeting, 2005]

Cell fate determination in the early C. elegans embryo depends on a suite of complex cellular processes that underlie cell polarization, cell signaling, regulated expression and segregation of cell fate determinants, control of asymmetric cell division, etc. Over the last several decades, the research community has made enormous strides towards identifying the key players involved in these processes and characterizing their functional interactions. As this empirical database grows at an ever-increasing rate, a new challenge is emerging: to assemble mechanistic explanations of how complex cellular behaviors emerge from networks of known molecular interactions. The magnitude of this challenge is difficult to overstate: even the simplest cellular processes involve heterogeneous local biochemical and mechanical interactions among thousands of players, distributed over space- the emergent consequences of these interactions are simply too complicated for human intuition alone to predict. To help meet this challenge, we are developing a computational simulation arena a virtual embryo into which one can install known molecular players (e.g. cytoskeletal polymers, motors, regulatory molecules, etc) and organelles (e.g. the nucleus, centrosomes, etc), specify their biochemical and/or force-generating interactions, and predict the cellular-level emergent consequences of these interactions. Our long-term goal is to develop a user-friendly computational resource and make it freely available to the research community. Our shorter-term approach is to develop prototype simulation tools in the context of specific case studies drawn from our own experimental research and through collaborations with other labs pursuing related problems. In this presentation, we describe our general approach and illustrate it using two ongoing research projects: 1) How force generating interactions among dynamically unstable microtubules, centrosomes, motors, tip-tracking proteins and cortical factors position mitotic spindles; 2) How force-generating interactions among short-lived cortical actin filaments, cross-linkers and myosin motor complexes, together with biochemical interactions among PAR proteins and PAR-actomyosin cortex interactions, could bring about the establishment and maintenance of cortical PAR domains during polarization of the zygote.

Edwin M Munro et al. (2005) International Worm Meeting "Cortical mechanics meet regulatory biochemistry: modeling cell polarization in the C. elegans embryo"

Eliana Hechter, Edwin M Munro

[International Worm Meeting, 2005]

Polarization of the C. elegans zygote involves a complex network of interactions among PAR proteins and the cortical actomyosin cytoskeleton. Many of the key players are now known; their interactions are being rapidly characterized, and the outlines of a polarization mechanism have begun to emerge. But the complexity and distributed nature of these interactions makes it difficult to move beyond the outline to a deeper understanding of polarization. To address this issue, we have developed a computational model that integrates cytoskeletal mechanics and regulatory biochemistry to predict observed polarization dynamics from known interactions among PAR proteins and components of the actomyosin cytoskeleton. Here, we focus on local actomyosin dynamics and mechanisms that govern the initiation and extent of cortical flow and PAR protein movement during polarization, highlighting key insights, experimental predictions and confirmations: The model predicts that local actomyosin dynamics (the cyclic formation and disappearance of actomyosin foci) observed during polarization in living embryos emerge as a robust consequence of the local mechanochemistry of cross-linked actomyosin networks. Quantitative tunings of model parameters that govern F-actin assembly/disassembly, cross-link formation and myosin activity predict a range of contractility/cortical flow phenotypes that encompass those seen during genetic or pharmacological perturbations of polarization, including those caused by PAR mutants. We have confirmed some of these predictions experimentally. For quantitative parameter tunings that produce normally observed local actomyosin dynamics, the model predicts that interactions among anterior and posterior PAR proteins, and between PAR proteins and actomyosin, are required to convert a transient local weakening of the cortex (to mimic the sperm cue) into sustained cortical flow and polarization. Finally, analysis of the model reveals an unanticipated mechanism that could limit the extent of cortical flow as seen during normal polarization and thus determine the size of the anterior PAR domain. This mechanism works only for parameter tunings that support local contractile instabilities and actomyosin focus formation. As predicted, defeating this contractile instability experimentally (without abolishing contractility) causes an acute anterior shift of the anterior PAR domain. We are now using this model as a sophisticated working hypothesis to guide further experimental analysis of the polarization mechanism.

Edwin M Munro et al. (2003) International Worm Meeting "A contractile actomyosin network drives cortical flows that establish and maintain Par domains and AP polarities in the early C. elegans embryo"

Edwin M Munro, Jeremy Nance, Jim Priess

[International Worm Meeting, 2003]

In C. elegans embryos the anterior/posterior axis sets up in response to a polarizing cue supplied by the sperm centrosome and its associated microtubules. During pseudocleavage, in response to this cue, the egg initiates transient cytoplasmic flows that carry cortical cytoplasm towards the anterior pole and interior cytoplasm towards the posterior pole. During the same period, the egg sets up and then maintains anterior and posterior cortical domains enriched respectively in the anterior Par proteins (Par-3, Par-6 and Pkc-3) and the posterior Par proteins (Par-2 and Par-1). However, the mechanisms underlying cortical flows and their relationship to the establishment and maintenance of par polarities remain unclear. Here we show that a contractile meshwork containing F-actin and the non-muscle myosin NMY-2 are responsible for driving cortical flows in P0 and that these flows transport anterior Par proteins within the cortex to establish an anterior domain enriched in both NMY-2 and the anterior Pars. Similar flows of both NMY-2 and anterior Par proteins establish an enrichment of NMY-2 and anterior Pars in P1. We show that the organization and activity of the acto-myosin cortex is regulated both by the cell cycle clock, by Par protein activity and locally by the sperm-donated MTOC. In embryos depleted of anterior Par proteins by RNAi, cortical NMY-2 flows locally away from the sperm MTOC but NMY-2 does not flow globally or cap to the anterior. In Par-2 RNAi embryos, NMY-2 caps initially, but then returns to the posterior cortex after the end of pseudocleavage. In support of recent results (Cuenca et al 2003, Development, 130(7), 1255-65), our data points to distinct mechanisms for the initial establishment of the anterior Par domain during pseudocleavage and its subsequent maintenance during prometaphase and metaphase prior to first cleavage and implicates Par-dependent myosin contractility and cortical flows in both mechanisms. These data suggest a modular mechanism for the establishment of Par domains in response to polarizing cues that could be readily redeployed in other cellular contexts and may help to explain the ubiquitous roles of Par proteins in establishing cellular polarities.

Yde, Sarah E. et al. (2021) International Worm Meeting "Using the C. elegans zygote to study principles of actin cytoskeleton self-organization"

Munro, Edwin, Zaidel-Bar, Ronen, Ray, Shinjini, Yde, Sarah E., Kovar, David, Kadzik, Rachel S.

[International Worm Meeting, 2021]

The dynamic actin cytoskeleton is continually remodeled during organism development, assembling and disassembling functionally diverse filamentous (F-) actin networks at precisely the right time and place. A major question is how a cell can organize and maintain multiple F-actin networks with diverse architectures and dynamics from a common pool of actin and actin-binding proteins (ABPs) within a single cytoplasm. We use the C. elegans zygote to study this question in vivo, and purified C. elegans proteins to reconstitute cytoskeletal dynamics in vitro. We hypothesize that a series of self-organization mechanisms facilitates the differential recruitment and activation of ABPs that determine actin filament architectures and dynamics of different F-actin networks. A key determinant of F-actin dynamics and architecture is the length of filaments within a network. A powerful regulator of actin filament length is the ABP capping protein (CP), which binds the fast-growing barbed ends of actin filaments to prevent polymerization and depolymerization. Using both 'bulk' and single-molecule/filament assays,we have biochemically characterized the dynamics of C. elegans ceCP on actin filament barbed ends in vitro. We find that ceCP has a high affinity for actin filament barbed ends and a slow off-rate relative to the lifetime of an actin filament in a C. elegans zygote. To characterize the biological activity of ceCP, we compared this in vitro data to the dynamics of ceCP in the C. elegans zygote using a powerful in vivo single-molecule approach. We found that the lifetime of ceCP single molecules on barbed ends in a zygote is much shorter than in vitro, suggesting the presence of regulatory mechanisms that act on ceCP in the zygote. We also characterize ceCP perturbation phenotypes at both the whole-network and single-filament level, and determine how ceCP regulates the balance of actin assembly and contributes to the self-sorting of other ABPs to different F-actin networks. Upon depletion of ceCP in a zygote, assembly of some F-actin networks is increased, while others are diminished, indicating a role for ceCP in the proper distribution of actin amongst networks. Additionally, we are characterizing other ABPs using similar combined in vitro/in vivo approaches, with the ultimate goal of using the C. elegans zygote to determine the minimal components for self-organization of multiple distinct F-actin networks within a common cytoplasm.

Zonies, Seth A. et al. (2009) International Worm Meeting "PAR-2-dependent mechanisms of polarity maintenance are sufficient to initiate polarity in C. elegans zygotes."

Hao, Yingsong, Seydoux, Geraldine, Zonies, Seth A., Motegi, Fumio

[International Worm Meeting, 2009]

Polarization of the C. elegans zygote has been proposed to occur in two successive phases. Polarity is initiated by a powerful flow of cortical nonmuscle myosin (NMY-2) and F-actin, which sweeps anterior PARs away from the MTOC during pronuclear migration. During mitosis, polarity is maintained by PAR-2, which localizes on the cortex next to the MTOC and prevents myosin and the anterior PARs from flowing back to the posterior (Munro et al 2004, Cuenca et al., 2003). We have obtained evidence that the PAR-2 "maintenance mechanism" is sufficient to initiate polarity when initiation fails. In a screen for temperature-sensitive polarity mutants, we isolated ax751, a partial loss-of-function mutant in ect-2. ect-2 is required for the actomyosin contractility that powers cortical flows during polarity initiation. As expected, ax751 zygotes fail to develop strong cortical flows during initiation, but unexpectedly still localize PAR-2 to the posterior, and form a myosin cap in the anterior during pronuclear centration. Formation of the myosin cap depends on par-2, and correlates with an overall increase in cortical myosin and a dramatic increase in phosphorylation of the myosin light chain in the cytoplasm. Analysis of ax751; par-2(RNAi) zygotes suggests that PAR-2 creates the anterior myosin cap by resisting myosin accumulation specifically in the posterior cortex. In ax751 zygotes, par-2 is also essential for PAR-3 asymmetry at the cortex and for PIE-1 asymmetry in the cytoplasm. Together with other experiments in our lab (See abstract by Motegi and Seydoux), our data suggest that polarity in the zygote depends on two redundant mechanisms that reinforce each other. A first mechanism (par-2-independent) uses cortical flows to clear myosin and the anterior PARs from the posterior (Munro et al., 2004). A second mechanism depends on loading of PAR-2 in the posterior cortex, which in turn excludes myosin and the anterior PARs. We speculate that this second mechanism may be the primary mechanism used to polarize the P1 blastomere, since P1 polarization occurs in ax751 embryos but not in par-2 embryos.

Ji Liu et al. (2006) Development & Evolution Meeting "ts alleles of NMY-2 shed light on its many functions in controlling early embryo polarity and asymmetry"

Lisa Maduzia, Craig Mello, Masaki Shirayama, Ji Liu

[Development & Evolution Meeting, 2006]

The nonmuscle myosin NMY-2 is essential for the establishment of anterior/posterior (A-P) polarity in the worm. In response to the sperm/centrosome cue, an NMY-2-dependent cortical flow appears to play a key role in establishing the asymmetric cortical localizations of the PAR proteins (Munro et al, 2004). However, it is not clear whether (or how) NMY-2 functions at later times to control cell cycle timing, movements of the mitotic spindle, and the maintenance of PAR protein localization. From our ts-embryonic lethal screens we isolated two ts alleles of nmy-2(ne3409) and nmy-2(ne1490), that may be useful in addressing some of these questions. Both alleles respond quickly to temperature shift and the two mutations fall into the same coiled-coil region of the protein. However, curiously, ne1490 exhibits a more severe polarity phenotype, and a less severe cytokinesis phenotype than ne3409. Temperature shift studies on these alleles indicate that NMY-2 function influences cell-cycle timing at the 2-cell stage, is required for maintenance of PAR protein localization, and contributes to P2/EMS signaling in the 4-cell stage embryo. We will report on our progress in analyzing these newly discovered roles for NMY-2.

Y Hao et al. (2005) International Worm Meeting "Analysis of the RING finger protein PAR-2 implicates the ubiquitination machinery in polarity maintenance in the C. elegans zygote."

Y Hao, T Frazier, L Boyd, G Seydoux

[International Worm Meeting, 2005]

Polarization of the C. elegans zygote begins when a wave of actomyosin contractility sweeps the PAR-3/PAR-6/PKC-3 complex towards the anterior. During this time, the RING finger protein PAR-2 accumulates on the posterior cortex. PAR-2, although not essential to initiate polarity, is required to maintain polarity and prevent the return of PAR-3/PAR-6/PKC-3 (Munro et al., 2004; Cuenca et al., 2003).PAR-2 is an unusual protein with a RING finger domain and an ATP binding domain (Levitan et al., 1994; Boyd et al., 1996). We show that the RING finger, but not the ATP binding domain, is essential in vivo. We also identify a domain in PAR-2 necessary and sufficient for cortical localization. This domain contains several PKC phosphorylation sites. PKC phosphorylates PAR-2 in vitro, and mutations in the PKC sites cause GFP:PAR-2 to localize throughout the cortex in vivo. In contrast, mutations in the RING finger cause GFP:PAR-2 to become hypersensitive to PKC-3. PAR-2 can become ubiquitinated in vitro, suggesting that, like other RING finger proteins, PAR-2 may function as a ubiquitin ligase. We propose that phosphorylation of PAR-2 by PKC-3 prevents PAR-2 from localizing to the cortex and that the RING finger of PAR-2 allows it to overcome this inhibition in the posterior.

Zonies, Seth et al. (2010) C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany "Polarization of the zygote by microtubules and PAR-2"

Motegi, Fumio, Zonies, Seth, Seydoux, Geraldine

[C. elegans: Development and Gene Expression, EMBL, Heidelberg, Germany, 2010]

Polarization of the C. elegans zygote involves the segregation of PAR proteins to distinct anterior (PAR-3, PAR-6 and PKC-3) and posterior (PAR-1 and PAR-2) domains. Anterior and posterior PAR proteins compete with each other for access to the cortex; this competition is biased by the sperm-donated MTOC (microtubule-organizing center). The MTOC induces cortical flows that mobilize anterior PARs away from the MTOC (Munro et al., 2004). We recently found that the MTOC also directly recruits PAR-2 to the posterior cortex, and that PAR-2 in turn induces myosin flows to displace anterior PARs. Before polarization, PKC-3 phosphorylates PAR-2, keeping PAR-2 in the cytoplasm. We have found that PAR-2 binds microtubules directly. When bound to MTOC-nucleated microtubules, PAR-2 becomes protected from phosphorylation by PKC-3, and is allowed to access the cortex nearest the MTOC. Once on the cortex, PAR-2 triggers myosin flows, which transport the anterior PAR complex away from PAR-2 in a positive feedback loop. Our genetic evidence suggests that PAR-2 does this by interfering with PAR-3-dependent recruitment of myosin to the cortex during mitosis. We will report on our progress in understanding the mechanisms by which PAR proteins control their own localization using myosin flows.

Williams, Falina J et al. (2013) International Worm Meeting "Wnts and VANG-1/Van Gogh control cell fates in the Q lineage."

Teuliere, Jerome, Williams, Falina J, Garriga, Gian

[International Worm Meeting, 2013]

Asymmetric cell division (ACD) is a process that generates cell diversity. Both intrinsic and extrinsic mechanisms can distribute developmental potential asymmetrically to generate daughter cells of different fates and position the cleavage furrow asymmetrically to generate cells of different sizes. Wnts are evolutionarily conserved secreted glycoproteins that are utilized throughout development and play a role in ACD.1 During C. elegans development, Wnts regulate asymmetric divisions by controlling the distributions of the b-catenin SYS-1 and the LEF/TCF POP-1.2 They can also regulate the orientation of the spindle.3 In other organisms, Wnts can also regulate cytoskeletal dynamics via the Planar Cell Polarity (PCP) pathway, which allows polarization of neighboring cells along an axis orthogonal to the apical-basal axis within an epithelial sheet.4 The ACDs of the Q.a and Q.p neuroblasts to give rise to a larger daughter that lives and a smaller cell destined to die. We find that two Frizzled homologs LIN-17 and MOM-5 together are necessary for both apoptotic fates, for the asymmetric distribution of POP-1 in the Q.a and Q.p daughter cells, and for the asymmetric position of the Q.a and Q.p furrows that produce daughter cells of different sizes. Reduction of Wnt signaling, however, fails to generate the same robust disturbance of POP-1 distribution and does not affect the furrow localization. Instead, reduction of Wnt signaling can result in a reversal of POP-1 asymmetry, a phenotype that is enhanced by loss of the Van Gogh homolog VANG-1, a component of the PCP pathway. 1.Munro and Bowerman. CSH Perspectives. 2009. 2.Mizumoto and Sawa. Trends in Cell Biology. 2007. 3.Walston and Hardin. Seminars in Cell and Developmental Biology. 2006. 4.Segalen and Bellaiche. Seminars in Cell and Developmental Biology. 2009.

Michael Werner et al. (2006) Development & Evolution Meeting "Regulation Of Cytokinesis By The Mitotic Spindle In The C.elegans Embryo"

Michael Werner, Michael Glotzer, Ed Munro

[Development & Evolution Meeting, 2006]

The anaphase spindle and asters coordinate actin polymerization and myosin activity so that the cleavage furrow generates two genetically identical daughter cells. Our earlier studies established that two redundant pathways regulate cleavage furrow formation (Dechant and Glotzer 2001). To gain insight into the molecular mechanisms underlying these two pathways, we have followed the dynamics of non-muscle myosin (NMY-2) fused to GFP in wild-type and mutant embryos. Prior to first metaphase, NMY-2 is asymmetrically enriched at the anterior cortex (Munro et al, 2004). We observe that during metaphase, NMY-2 dissociates from the cortex and quickly returns, accumulating in a broad cortical region in the anterior half of the embryo and concentrating at the site of furrow ingression. Recruitment of NMY-2 to the cortex during anaphase requires active RhoA. To determine how the position of the mitotic spindle affects myosin recruitment and furrow formation we destabilized microtubules pharmacologically or by RNAi to generate a small mitotic spindle that is positioned in the posterior and is perpendicular to the A-P axis. During anaphase, NMY-2 accumulates predominantly in the anterior half of the cortex. This accumulation is accompanied by a flow of cortical myosin away from the spindle, towards the anterior where an anterior furrow forms. A second furrow bisects the small spindle at the posterior cortex. These furrows are genetically separable. ZEN-4, a factor required for central spindle assembly, is required for the posterior, but not the anterior furrow. In contrast, the anterior furrow is more sensitive to depletion of regulators of the actomyosin network. These data suggest that mechanistically distinct pathways act simultaneously to position the cleavage furrow. One pathway involves inhibition of cortical myosin recruitment near, and flow away from, regions rich in microtubules. Formation of the anterior furrow requires that the cortical myosin meshwork is organized and non-uniformly distributed, the anterior flow contributes to this non uniform distribution. These data provide direct evidence for astral relaxation. In contrast, the second mechanism involves a local enrichment/activation of NMY-2 close to and dependent on the central spindle, irrespective of the overall pattern of cortical contractility.