Heads or Tails

Developmental Cell, Vol. 3, 157–166, August, 2002, Copyright 2002 by Cell Press

Heads or Tails: Cell Polarity and Axis Formation in the Early Caenorhabditis elegans Embryo Rebecca Lyczak, Jose´-Eduardo Gomes, and Bruce Bowerman1 Institute of Molecular Biology University of Oregon 1370 Franklin Boulevard Eugene, Oregon 97403

In C. elegans, the first embryonic axis is established shortly after fertilization and requires both the microtubule and microfilament cytoskeleton. Cues from sperm-donated centrosomes result in a cascade of events that polarize the distribution of widely conserved PAR proteins at the cell cortex. The PAR proteins in turn polarize the cytoplasm and position mitotic spindles. Lessons learned from C. elegans should improve our understanding of how cells become polarized and divide asymmetrically during development. Five Asymmetric Cell Divisions Make Most of a Nematode Asymmetric cell divisions are of substantial importance during the development of most organisms (Horvitz and Herskowitz, 1992). For example, in spite of the ability of isolated early mouse embryo cells to regulate and produce an entire embryo, even the first mitotic division of a mouse zygote is now known to be asymmetric, segregating embryonic from extraembryonic cell fates (Piotrowska et al., 2001; Piotrowska and ZernickaGoetz, 2001). Stem cells also divide unequally, further motivating efforts to better understand this fundamental developmental process. Cell divisions in the early C. elegans embryo are notable for their obvious asymmetry, making the early embryo a useful model system for genetic studies of cell polarity during mitosis (Sulston et al., 1983). The first mitotic division of the worm zygote, called P0, is unequal, producing a larger anterior daughter called AB, and a smaller posterior daughter called P1. Asymmetries during this first cleavage can be observed using Nomarski optics, as shown in Figure 1. Prior to fertilization, a mature and activated oocyte enters the spermathecum, where sperm reside and diploid life begins (reviewed in McCarter et al., 1999). The oocyte pronucleus moves away from the pole that first enters the spermathecum, and the oocyte adopts an elongated, more oval shape during this process. Sperm entry usually occurs as an oocyte enters the spermathecum, opposite the end occupied by the oocyte pronucleus. This sequence of events often results in the two parental pronuclei residing at opposite poles. Subsequently, the maternal pronucleus migrates to meet its paternal partner near the posterior pole. Together, the pronuclei move to the center and rotate such that the first mitotic spindle forms in alignment with the long axis. During anaphase, the spindle becomes displaced to the posterior, resulting in an unequal division. The P1 and AB daughters differ 1

Correspondence: [email protected]

Review

in size, cell cycle timing, spindle orientation, and cell fate potential (reviewed in Rose and Kemphues, 1998; Doe and Bowerman, 2001). Subsequently, P1 but not AB, and three immediate descendants of P1, divide asymmetrically (Sulston et al., 1983; Schierenberg, 1987). Four of these divisions appear to be intrinsically polarized, although one is induced, in part by Wnt signaling (reviewed in Bowerman and Shelton, 1999; Thorpe et al., 2000). These early unequal divisions produce six “founder” cells (Figure 2), first noticed because each divides more equally to produce descendants that divide more synchronously (Sulston et al., 1983). Each founder is born committed to the production of a specific subset of the cells that make a nematode larva. Notch signals influence the development of the first born founder cell, AB, but the fates of the remaining founders appear to be determined for the most part upon birth, presumably due to the unequal partitioning of embryonic determinants (Priess and Thomson, 1987; Schierenberg, 1987; reviewed in Bowerman, 1998). Most of the recent progress we review has been made in understanding the asymmetric divisions of P0 and P1. After Sperm Entry: Establishing an AnteriorPosterior Body Axis in P0 In C. elegans, the first embryonic axis is the anteriorposterior (A-P) axis, established shortly after fertilization along the long axis of the one-cell zygote. Sperm entry triggers three events: completion of oocyte meiosis I and II, production of a protective eggshell, and specification of an A-P axis (Figure 3). Upon entry, the spermdonated pronucleus, with its associated centrosome, generates a cytoplasmic flux that moves this complex to one pole of the oval zygote (Hird et al., 1996; Hird and White, 1993; Goldstein and Hird, 1996). Regardless of where fertilization occurs, the pole of the zygote occupied by the paternal pronucleus, with its associated centrosomes, becomes posterior (Albertson, 1984; Goldstein and Hird, 1996). The opposite pole becomes anterior, apparently by default. The sperm pronucleus/centrosome complex (SPCC) appears to specify the A-P axis, orchestrating changes both in the cytoplasm and at the cortex of the one-cell zygote (see Figure 3). The SPCC directs cytoplasmic streaming, during which internal cytoplasm flows toward and cortical cytoplasm flows away from the SPCC (Hird and White, 1993). At the same time, polarized changes in the cortex occur during a process called pseudocleavage: the anterior cortex of the embryo ruffles extensively while the posterior becomes smooth. Although pseudocleavage appears to be dispensable for normal development (Rose et al., 1995), mutations that disrupt polarity often interfere with it (Rappleye et al., 2002; Golden et al., 2000; Kirby et al., 1990). Polarizing the PAR Proteins in Response to Sperm Cues Sperm-triggered specification of the A-P axis results in an asymmetric distribution of early developmental

Developmental Cell 158

Figure 2. Founder Cells Are Established Early in C. elegans Embryogenesis By the eight-cell stage, four of the six founder cells are present; the last two appear after the division of P3. The founder cells arise from a series of asymmetric cell divisions. As determinants are asymmetrically partitioned during these divisions, each founder cell has different fate potential.

Figure 1. The First Two Mitotic Divisions of the Caenorhabditis elegans Embryo Anterior is to the left, ventral at the bottom, in each image. The maternal and paternal pronuclei usually appear at opposite poles, and migrate to meet near the posterior. They rotate and move to the center, with the first mitotic spindle forming along the A-P axis and moving slightly toward the posterior during anaphase. The smaller posterior daughter divides after the larger anterior daughter; the AB mitotic spindle remains transverse while the P1 spindle rotates to align with the A-P axis. Due to eggshell constraints, the four-cell stage blastomeres adopt a stereotyped arrangement that defines the dorsal-ventral axis (ABp, dorsal; EMS, ventral).

regulators, perhaps the most important being the cortical PAR proteins, themselves required for most A-P asymmetry in the one-cell zygote (reviewed in Golden, 2000). PAR-1 and PAR-2 localize to the posterior cortex in P0, while PAR-3 and PAR-6 localize to the anterior (Figure 3). These two cortical domains abut at a common boundary, roughly midway along the A-P axis (Boyd et al., 1996; Etemad-Moghadam and Kemphues, 1995; Guo and Kemphues, 1995; Hung and Kemphues, 1999).

PAR-4 and PAR-5 are not polarized in distribution but are enriched throughout the cortex (Morton et al., 2002; Watts et al., 2000). Thus far, the identities of the PAR proteins only hint at their mechanism of action. PAR-1 and PAR-4 are predicted ser/thr kinases, while PAR-6 and PAR-3 have one and three PDZ domains, respectively (reviewed in Rose and Kemphues, 1998). PAR-5 is a 14-3-3 protein (Morton et al., 2002), while PAR-2 is largely novel but includes a RING finger domain (Levitan et al., 1994). The anterior complex of PAR-3 and PAR-6, which also includes an atypical protein kinase C called PKC-3, is widely conserved and regulates cell polarity in many different settings (reviewed in Doe and Bowerman, 2001). The PAR-1 kinase is conserved and regulates cell polarity in mammals and insects (see below). Whether relatives of PAR-2, -4, and -5 regulate cell polarity elsewhere is not known, but PAR-4 and PAR-5 are conserved in other organisms. As the PAR proteins have been reviewed extensively, we do not provide an exhaustive account of them, but instead focus on recent advances. Microtubules, Microfilaments, and A-P Axis Specification How sperm cues polarize the anterior and posterior cortical PAR domains remains unclear, but an intact actomyosin cytoskeleton is required (Shelton et al., 1999; Guo and Kemphues, 1996; Hird and White, 1993; Hill and Strome, 1988, 1990; Strome and Wood, 1983). After chemical disruption of microfilaments with cytochalasin D (and after dsRNA-mediated interference of a nonmuscle myosin II heavy chain gene, nmy-2, or a myosin light chain gene, mlc-4), cytoplasmic flows and pseudocleavage are eliminated, pronuclei meet near the center of the zygote, and the first cleavage is symmetric. Reducing nmy-2 and mlc-4 function also results in PAR-3 spreading throughout the cortex, in PAR-2 accumulating in a reduced cortical patch near the sperm pronucleus, and in delayed localization of P granules to the posterior

Review 159

a pronucleus, a centrosome, and possibly other factors, but the sperm pronucleus is dispensable for establishing the A-P axis (Sadler and Shakes, 2000). This leaves the sperm-donated centrosomes and associated astral microtubules (MTs) as the most obvious candidates for providing cue(s) that break the initial symmetry of the embryo, and mutations that delay sperm aster maturation severely disrupt axis specification (O’Connell et al., 2000). Further support for MTs having an important role in A-P axis specification comes from the discovery that arrested meiotic spindle MTs can reverse the A-P axis in mutant embryos lacking components of the anaphase promoting complex (APC). These mutants arrest at metaphase of meiosis I, shortly after fertilization (Golden et al., 2000). Sperm asters never form, and MTs associated with the arrested and frayed meiotic spindle, usually present near the cortex at the pole opposite that occupied by the paternal pronucleus, provide a partial polarity cue (Wallenfang and Seydoux, 2000). PAR-2 and PAR-3 become reversed in these mutants, although cytoplasmic flows and P granule localization are absent. It seems likely that both sperm astral MTs and other centrosome-dependent processes are involved in the initial specification of an A-P axis in the one-cell C. elegans zygote.

Figure 3. Establishment of the Anterior-Posterior Axis The sperm pronuclear/centrosome complex(es) triggers cortical rearrangements and cytoplasmic flows to polarize the axis. During this phase, the cortical PAR proteins become asymmetrically distributed, the oocyte pronucleus (o) migrates to the posterior, and the germline P granules move to the posterior. Polarization culminates in posterior displacement of the spindle and an asymmetric first cleavage in which determinants are segregated to one or the other daughter. This cleavage results in a larger anterior daughter AB, and a smaller posterior daughter P1.

pole (Guo and Kemphues, 1996; Shelton et al., 1999). Although disruption of the actin cytoskeleton clearly perturbs axis formation, we still do not understand how microfilament and PAR protein function are coupled. While the C terminus of PAR-1 binds NMY-2, the significance of this interaction remains unknown (Guo and Kemphues, 1996). The Sperm-Donated Centrosome Specifies an A-P Axis in P0 How the SPCC influences cortical microfilaments is not known, but the sperm-donated centrosome appears to mediate this process. Upon fertilization, sperm donate

Getting Intimate with the Cortex: The Anaphase Promoting Complex and a Connection to the Cell Cycle? Recent work suggests that the APC and its target separin are required for the sperm-donated centrosomes to interact with the microfilament-rich cortex and specify a posterior pole. Weak mutations in components of the APC permit progression through meiosis but prevent formation of an A-P axis (Rappleye et al., 2002). In weak APC mutants, the SPCC does not stay close to the cortex, and PAR-2 fails to accumulate at the cortex. Thus, close contact between centrosomes and astral MTs with the cortex may be important for specifying a posterior pole. Aroian and colleagues suggest that such contact somehow excludes PAR-3 from the posterior cortex, allowing PAR-2 to accumulate. In their model, PAR localization occurs downstream of events that require a close association of the pronucleus with the cortex. Consistent with this view, the close association of the paternal pronucleus with the cell cortex is not affected in par-3 mutants (Rappleye et al., 2002). Nevertheless, par-2 and par-3 mutants have partial defects in cytoplasmic streaming (Shelton et al., 1999; Kirby et al., 1990), suggesting that PAR proteins may be involved in the initial establishment of an A-P axis, in addition to mediating subsequent asymmetries. Does Dad Contribute More Than a Positional Cue? The initial cue for polarity requires both maternally and paternally contributed factors. While the sperm-donated centrosome plays a central role, maturation of functional centrosomes requires the strictly maternal gene spd-2 (O’Connell et al., 2000). Furthermore, mutations in the APC that disrupt polarity show both maternal and paternal contributions (Rappleye et al., 2002). Thus, both sperm and egg factors are required for the earliest

Developmental Cell 160

events in axis specification. Sperm may provide only a centriolar scaffold, upon which maternal factors assemble a cue for polarity. Alternatively, other paternal factors might participate more directly in axis specification. Regardless of the extent to which sperm-donated factors influence axis formation, it should be possible to identify paternal effect mutations that disrupt polarity. Only one strictly paternal effect mutation has been identified, in the spe-11 gene, which encodes a novel protein required for the completion of meiosis by the maternal pronucleus, for the secretion of an eggshell, and for P0 mitotic spindle positioning (Browning and Strome, 1996; Hill et al., 1989). Other genes required for embryogenesis are provided at least in part through paternal contributions, but a role for these factors in establishing an A-P axis has not been reported (Lee et al., 2001; Go¨nczy et al., 1999b; O’Connell et al., 1998). In Drosophila, maternal and paternal gene products also appear to undergo extensive interactions during early embryogenesis (reviewed in Fitch et al., 1998). In both nematodes and insects, sperm apparently contribute much more than a haploid genome to the earliest moments in diploid life. However, it is not known whether the same or different processes require paternal gene products during early embryogenesis in Drosophila and C. elegans. Further study of both maternal and paternal effect mutants should shed light on this intriguing and still largely unexplored issue. In Drosophila, both the A-P and the dorsal-ventral body axes are specified well before fertilization, during oogenesis (reviewed in St Johnston and Nusslein-Volhard, 1992). Despite this substantial difference in the timing of body axis specification, the Drosophila ortholog of C. elegans PAR-1 is localized to, and required for specification of, the posterior pole of the embryo (Riechmann et al., 2002; Shulman et al., 2000; Tomancak et al., 2000). D-Par-1 also is required for earlier steps in oocyte development, including a reorganization of MT organizing centers, and a later reorganization of MTs that participate in moving determinants to the posterior pole (Cox et al., 2001; Huynh et al., 2001; Shulman et al., 2000). Finally, mammalian homologs of PAR-1 have been shown to destabilize MTs (Drewes et al., 1997), although it is not clear whether this activity relates to the role of D-Par-1 in organizing oocyte MTs. In C. elegans, PAR-1 is present at the cortex only in the posterior (Guo and Kemphues, 1995), and posterior astral MTs in P0 often appear shorter than anterior astral MTs late in mitosis (see Severson et al., 2000). However, it is not known whether destabilization of posterior astral MTs is important for posterior displacement of the P0 spindle. While D-Par-1 and C. elegans PAR-1 both are required for A-P axis formation and both localize to the posterior, it is not yet clear whether these two kinases participate in similar or distinct processes during early insect and worm development (reviewed in Kemphues, 2000). Polarity and Mitotic Spindle Positioning in P0 The specification of an A-P axis in P0 results in posterior displacement of the first mitotic spindle, and an unequal cleavage that produces differently sized daughters. To position the P0 spindle, a dynein- and dynactin-dependent rotation aligns the forming spindle with the A-P

Figure 4. Model for Spindle Positioning in P0 The first mitotic spindle is displaced posteriorly during anaphase, due to unequal forces applied to the anterior and posterior centrosomes (arrows). This unequal force distribution requires the PAR proteins, which position the DEP domain protein LET-99 within a lateral cortical stripe, closer to the posterior pole. Microtubules appear to interact with cortical dynein to exert forces on the centrosomes that position the spindle. Microtubule/cortex interactions are weakest in regions of LET-99 localization (thin lines), resulting in a stronger net force on the spindle from the posterior. Also required for spindle positioning is the LIN-5 protein, which localizes to centrosomes, and ␣ subunits of the heterotrimeric G proteins (localized around the entire cortex and at spindle poles; not shown).

axis (Go¨nczy et al., 1999a; Skop and White, 1998). Subsequently, the cortical poles of the zygote exert unequal pulling forces on each end of the spindle (Grill et al., 2001; see Figure 4). Posterior displacement of the P0 spindle, but not the initial rotation that aligns the spindle along the A-P axis, requires the PAR proteins. For example, the spindle remains centrally located in both par-2 and par-3 mutants due to equal forces being applied to each pole (Grill et al., 2001). In par-2, PAR-3 expands to the posterior (Etemad-Moghadam and Kemphues, 1995), and a smaller net force is applied to both poles (Grill et al., 2001). Conversely, in par-3, PAR-2 extends anteriorly (Boyd et al., 1996), and both poles are pulled by a stronger force, similar in magnitude to that applied only to the posterior centrosome in a wild-type embryo (Grill et al., 2001). PAR-3 has been proposed to interact with astral MTs that contact the anterior cortex, preventing spindle rotation in AB at the two-cell stage (Rose and Kemphues, 1997; Cheng et al., 1995; Etemad-Moghadam and Kemphues, 1995). Perhaps the same interactions in P0 result in a smaller net force on the anterior centrosome. Alternatively, the PARs may act indirectly to position the P0 spindle, localizing or regulating factors that interact with MTs. Downstream of the PARs: Mediators of Spindle Positioning in P0 Recent work has led to the identification of proteins that appear to act downstream of the PARs to influence posterior displacement of the P0 spindle (Figure 4). For example, the novel coiled-coil protein LIN-5 localizes to centrosomes in an MT-dependent manner, and is required for posterior displacement of the P0 spindle but not for the polarized distribution of the PAR proteins (Lorson et al., 2000). Also required for spindle positioning are heterotrimeric G proteins subunits. Two redundant G␣ subunits, GOA-1 and GPA-16, localize to the centrosomal asters and to the cell cortex, and are required for posterior displacement of the P0 spindle (Gotta and Ahringer, 2001). A protein called RIC-8 acts positively with GOA-1 during this process (Miller and Rand, 2000). Despite defects in spindle alignment along

Review 161

the A-P axis when these genes are inactivated, other asymmetries in P0 appear normal, suggesting that GOA-1, GPA-16, and LIN-5 mediate one subset of the asymmetries specified by the PARs (Gotta and Ahringer, 2001; Lorson et al., 2000; Miller and Rand, 2000). How the PAR proteins regulate their activity, and how LIN-5 and G␣ subunits influence spindle position, is not known. It is intriguing that heterotrimeric G protein subunits, which frequently require upstream activation by serpentine transmembrane receptors, regulate these apparently intrinsic asymmetries. The Worm Embryo Gets a Stripe, for Posterior Displacement of the P0 Spindle Another downstream effector of PAR polarity is the DEP domain protein LET-99, enriched at the cortex within a narrow band positioned slightly to the posterior in P0 and P1 (Rose and Kemphues, 1997; Tsou et al., 2002; see Figure 4). Proper localization of this posterior band requires the par genes, and LET-99 is required for P0 and P1 spindle positioning. In let-99 mutants, the ends of the P0 spindle swing back and forth, and posterior displacement is defective though not absent. The abnormal swinging movements require dynein, suggesting that LET-99 regulates dynein-mediated forces on astral MTs (Tsou et al., 2002). Rose and colleagues have proposed that cortical forces on astral microtubules are lowest in regions where LET-99 levels are highest (Figure 4). Thus, in P0, anterior and lateral forces pull on the anterior centrosome, while the LET-99 band in the posterior results in a weakening of lateral forces on the posterior centrosome. The sum of these forces results in a greater pull from the posterior cortex on the posterior centrosome, compared to counteracting forces applied on the anterior centrosome, contributing to posterior displacement of the P0 spindle. Thus, LET-99 links the PARs to the generation of asymmetric forces. Nevertheless, the P0 spindle does move to the posterior in let99 mutants, with more severe spindle orientation defects appearing at the two-cell stage. Thus, LET-99 makes only a partial contribution to P0 spindle positioning. Its role(s) at the two-cell stage is more difficult to interpret, as the defects could be indirect consequences of earlier problems in P0. In adult nematodes, the DEP domain protein EGL10 regulates the frequency of egg laying, opposing the activity of GOA-1 (Koelle and Horvitz, 1996). Therefore, the DEP domain in LET-99 might spatially regulate the uniformly cortical and centrosomal G␣ proteins GOA-1 and GPA-16 to influence asymmetric spindle movement. However, DEP domains also are found in other proteins, such as Disheveled, which are not necessarily involved in G protein-mediated processes (Wong et al., 2000; Tree et al., 2002), and EGL-10 also has an RGS domain implicated in regulation of G protein function (Koelle and Horvitz, 1996). Regulating Mitotic Spindle Orientation in P1 In addition to their influence on posterior displacement of the P0 spindle, the PAR proteins are required for orienting mitotic spindles in AB and P1 (Cheng et al., 1995; Kemphues et al., 1988). In a wild-type embryo, AB divides slightly before P1, and the AB spindle remains

transversely oriented, perpendicular to the longer A-P axis (Figure 1). In P1, the centrosome/nuclear complex rotates during prophase, aligning the spindle with the A-P axis (Hyman, 1989; Hyman and White, 1987). PAR-2 and PAR-3 oppose each other in regulating spindle orientation at the two-cell stage (Cheng et al., 1995; Kemphues et al., 1988). In par-2 mutants, both spindles are transverse, while in par-3 both rotate. Both spindles also rotate in par-2; par-3 double mutants, indicating that PAR-2 and PAR-3 are not required for rotation, but are required for restricting rotation to P1 (Cheng et al., 1995). PAR-2 and PAR-3 presumably regulate downstream factors that more directly influence spindle orientation, although when this regulation occurs is not clear. The polarized distributions of PAR-2 and PAR-3 in P0 are reiterated in the germline precursors P1, P2, and P3 at the two-cell, four-cell, and eight-cell stages, respectively. However, temperature upshift experiments using conditional par alleles have failed to reveal requirements for the PAR proteins beyond the one-cell stage (Morton et al., 1992). Perhaps the PARs influence proteins at the one-cell stage, which then act later in development. Alternatively, the PARs may act repeatedly to influence polarity in each germline precursor. Studies of two genes called ooc-3 and ooc-5 provide some evidence in support of PAR proteins functioning beyond the one-cell stage. OOC-3, a novel transmembrane protein, and OOC-5, a torsin-related protein, both localize to the ER (Basham and Rose, 1999, 2001; Pichler et al., 2000). In ooc-3 and ooc-5 mutants, PAR-3 spreads throughout the cortex of P1, and the nuclear/centrosome complex in P1 fails to rotate, as in par-2 mutants. Like par-2; par-3 double mutants, ooc-3; par-3 double mutants show partial restoration of P1 rotation, suggesting that mislocalized PAR-3 prevents rotation in these mutants. Because OOC-3 and OOC-5 are ER proteins, they could perhaps regulate the trafficking of membrane proteins that influence cell polarity. In wild-type embryos, PAR-3 is enriched throughout the cortex in AB, where it may also prevent rotation (Cheng et al., 1995). Presumably, other factors act in P1 to promote rotation, even though PAR-3 is present throughout the anterior portion of the P1 cortex (Etemad-Moghadam and Kemphues, 1995). Does the P1 Mitotic Spindle Become Polarized and Maintain the Axis of Cell Polarity during Mitosis? Rotation of the mitotic spindle in P1 is thought to be important for the unequal segregation of cell fate determinants to the daughters of P1. Consistent with this view, the failure of P1 spindle rotation in let-99 and G␣ mutants results in a failure to segregate P granules to only one P1 daughter (Gotta and Ahringer, 2001; Miller and Rand, 2000; Rose and Kemphues, 1997). An intriguing exception to the idea that rotation is required for the proper segregation of embryonic determinants has recently been described (Gomes et al., 2001). In mutant embryos lacking the function of a maternal gene called spn-4, the P1 spindle remains transversely oriented, but the axis of polarity in P1 becomes respecified to align with the transverse spindle axis. As a result, P granules and PAR-2 are properly segregated

Developmental Cell 162

Figure 5. Polarity and Spindle Positioning in P1, a Two-Step Model (A) Polarity phenotypes of mutations that prevent rotation in P1. In wild-type, PAR-2 and P granules localize to the posterior pole and the spindle aligns along the A-P axis. In ooc-3, ooc-5, let-99, goa-1, gpa-16, and spn-4 mutants, the P1 spindle fails to rotate. Disruption of these genes, except for spn-4, results in missegregation of determinants when P1 divides. OOC-3 and OOC-5 are required to polarize the PARs in P1. In spn-4, the axis of polarity reorients to align with the spindle. (B) A two-step model for the polarization of P1. Cortical and cytoplasmic polarity are established first. This requires the function of the ooc genes and may also involve the par genes and the microfilament cytoskeleton. During this first polarization step, both spindle poles are captured by a remnant site left from cytokinesis, and the nucleus moves anteriorly (Hyman, 1989; Hyman and White, 1987). Subsequently, we propose that polarization of the spindle results in the release of one pole and rotation. The heterotrimeric G proteins GOA-1 and GPA-16, along with RIC-8 and LET-99, are required for both polarization and rotation of the spindle, while spn-4 is required only for rotation. Polarization of the spindle, the second step in this model, may be important for adjusting the axis of polarity as the AB and P1 spindles elongate and flop over due to constraints imposed by the eggshell. This ensures the proper segregation of determinants along the final axis of division, which can vary considerably from embryo to embryo.

to one daughter, as in wild-type. These findings suggest a role for the mitotic spindle in maintaining a proper axis of cell polarity in P1 late in mitosis. Based on the different mutant phenotypes of P1 spindle rotation-defective mutants (Figure 5A), we propose a two-step model for initiating and maintaining polarity in P1 (Figure 5B). In this model, the first step is a cortical and cytoplasmic polarization, which at least indirectly requires the PAR proteins and also requires the ER proteins OOC-3 and OOC-5. This first polarization reestablishes PAR protein asymmetries and moves P granules to the posterior cortex. In a second step, we suggest that the mitotic spindle itself becomes polarized, with the “polarized” end subsequently influencing polarity. LET-99 and GOA-1/GPA-16 can be viewed as being required for both polarization and rotation of the spindle,

and SPN-4 specifically for rotation (Gomes et al., 2001). In our model, either the “polarized” spindle end is released, or only the nonpolarized end is captured, by a cortical complex that exerts force to rotate the forming P1 spindle (Hyman, 1989; Hyman and White, 1987). Our model also predicts that spindle polarization is required for rotation, but that rotation is not required for spindle polarization. To date, there is no evidence for polarization of the P0 spindle, and the A-P axis in P0 appears to be established independent of spindle position. However, in P0 the spindle remains aligned with the A-P axis throughout mitosis. By contrast, in P1 the mitotic spindle is forced to slant along a diagonal axis late in mitosis (Figure 5B). This occurs in response to the AB spindle becoming longer than the eggshell is wide, forcing it to slant over along an

Review 163

Figure 6. Polarity Mediators Act Downstream of the PARs in P0 Downstream of the PARs, polarity mediator proteins localize asymmetrically in the cytoplasm of P0 and restrict determinants to specific blastomeres. The mediator MEX-5 is present at high levels in anterior cytoplasm. PAR-1 prevents MEX-1 from accumulating in posterior cytoplasm, while MEX-5/6 prevent PIE-1 from accumulating in anterior cytoplasm. MEX-5/6 may promote the degradation of PIE-1 in the anterior half of the embryo (see text).

apparently randomly chosen diagonal axis. This forces a complementary slanting of the P1 spindle. Thus, polarization of the P1 spindle may be needed to segregate determinants, given the variability observed in P1 spindle orientations late in mitosis. How Much Alike Are P1 and a Drosophila Neuroblast? It is interesting to compare the asymmetric division of P1 to neuroblast (NB) stem cell divisions in the developing Drosophila central nervous system. An apical complex of Bazooka/D-Par-6/aPKC in NBs, related to the anterior complex of PAR-3/PAR-6/PKC-3 in P0 and P1, appears to regulate cell polarity and mitotic spindle orientation (reviewed in Doe and Bowerman, 2001). In wild-type NBs, the axes of cortical polarity and spindle alignment are in register by metaphase. Reducing the function of the apical complex results in the axes of cortical polarity and spindle polarity failing to align by metaphase. However, later in mitosis, these axes do align and asymmetric division is restored, a process dubbed “telophase rescue” (Cai et al., 2001; Peng et al., 2000). Thus, NB polarity mutants at least superficially resemble P1 in spn-4 mutant embryos, restoring asymmetric cell division late in mitosis. Furthermore, loss-of-function mutations in a Drosophila heterotrimeric G protein ␣ subunit called G␣I disrupt NB polarity (Schaefer et al., 2001), much as GOA-1/GPA-16 are required for P1 spindle rotation and polarization (see above). Finally, a protein called PINS is localized apically and required for proper NB spindle orientation (see Schaefer et al., 2001). PINS has GoLoCo domains, which in other proteins are known to bind heterotrimeric G protein ␣ subunits, again implicating the regulation of G protein ␣ subunits during spindle positioning. The similarities of Drosophila NB divisions to the asymmetric division of P1 tempt one to speculate that polarization of NB mitotic spindles may be involved in telophase rescue, with G␣I required for spindle polarization. However, NBs and P1 undergo what appear to be very different divisions. In both P0 and P1, the mitotic spindle appears roughly symmetrical, other than some differences in astral MT length. In NBs, the mitotic spindle becomes distinctly polarized in structure late in mitosis. One centrosome becomes larger than the other (Spana and Doe, 1995), and the distances from each centrosome to the cleavage furrow within a single NB

differ substantially (Kaltschmidt et al., 2000). Thus, an asymmetric spindle, rather than spindle displacement, appears largely responsible for generating unequal NB divisions. It will be interesting to learn more about the different strategies used by other cell types to achieve unequal cell division during development. Cortical Polarity, Cytoplasmic Polarity, and Cell Fate Determinants In addition to influencing spindle position, the PAR proteins are required for the asymmetric distribution of cell fate determinants to P1 and AB descendants. Several maternally expressed regulators accumulate to high levels only in P1 descendants, or only in AB descendants (posterior and anterior group proteins; see Schubert et al., 2000). For example, high levels of the transcription factors SKN-1 (Bowerman et al., 1993) and PAL-1 (Hunter and Kenyon, 1996) are detected only in P1 descendants, while the Notch homolog GLP-1 is present at high levels only in AB descendants (Evans et al., 1994). A third group of regulators is restricted within the P1 lineage to germline precursors (Schubert et al., 2000). These “germline proteins” include a key player in germline development called PIE-1 (Mello et al., 1996). The asymmetric distributions of anterior, posterior, and germline regulatory proteins require the par genes (Huang et al., 2002; Schubert et al., 2000; Bowerman et al., 1993, 1997; Crittenden et al., 1996). Several proteins appear to act downstream of the PARs to mediate the asymmetric distributions of anterior, posterior, and germline proteins. These “polarity mediators” include several related CCCH finger proteins: MEX-1 (Guedes and Priess, 1996), the closely related MEX-5 and MEX-6 proteins (Schubert et al., 2000), and POS-1 (Tabara et al., 1999). Two other polarity mediators are unrelated putative RNA binding proteins: the KH domain protein MEX-3 (Draper et al., 1996) and the RNA recognition motif (RRM) protein SPN-4 (Gomes et al., 2001). Mutations in any one of these essential polarity mediators result in the improper localization of cell fate determinants. In mex-1, mex-5, and spn-4 mutants, SKN-1 is mislocalized to anterior cells (Gomes et al., 2001; Guedes and Priess, 1996; Schubert et al., 2000), while in spn-4 and mex-3 mutants, PAL-1 is mislocalized to anterior cells (Huang et al., 2002). All of these polarity mediators are cytoplasmic proteins that become polarized in distribution along the A-P axis during, or shortly

Developmental Cell 164

after, asymmetric cell divisions (Draper et al., 1996; Guedes and Priess, 1996; Schubert et al., 2000; Tabara et al., 1999). Intriguingly, the cytoplasmic distributions of MEX-5 and PIE-1 precisely respect the boundary defined by the anterior and posterior cortical PAR proteins (Figure 6). High levels of MEX-5 are present only to the anterior of this boundary, throughout the cytoplasm, while high levels of PIE-1 are found only to the posterior (Schubert et al., 2000). The ser/thr kinase PAR-1 restricts high levels of MEX-5 to anterior cytoplasm, and MEX-5 and MEX-6 restrict PIE-1 to the posterior cytoplasm (Schubert et al., 2000). The polarized cytoplasmic distributions of other polarity mediators also require the par genes, and the cortical polarization of PAR proteins is unaffected in mutants lacking polarity mediators. Thus, the one-cell zygote becomes polarized at the cortex, in response to sperm entry, with cortical polarity converted into cytoplasmic polarity. Cell Polarity and Cell Fate: Translational Regulation The identification of cytoplasmic mediators of cell polarity has begun to reveal pathways that connect PAR functions to the regulation of cell fate determinants, and hence to the establishment of founder cell identities (Schubert et al., 2000; Huang et al., 2002). As most maternally expressed mRNAs are distributed throughout early embryonic cells in C. elegans (Seydoux and Fire, 1994), polarity mediator proteins presumably influence either translation or protein stability to generate asymmetric distributions of cell fate determinants. The KH domains in MEX-3 and the RNA recognition motif (RRM) in SPN-4 suggest that these polarity mediators influence the expression of cell fate determinants posttranscriptionally (Draper et al., 1996; Gomes et al., 2001). Moreover, the 3⬘ untranslated regions of the glp-1 and pal-1 mRNAs have been shown to be necessary and sufficient for restricting protein expression to anterior and posterior cells, respectively (Evans et al., 1994; Hunter and Kenyon, 1996). However, it is not yet known whether MEX-3 or SPN-4 play direct roles in regulating the translation of specific mRNAs. Cell Polarity and Cell Fate: Protein Stability In addition to translational regulation, the regulation of protein stability appears likely to play a prominent role in the no doubt complex transition from cortical and cytoplasmic polarity to different founder cell identities. This conclusion follows from studies of the CCCH finger proteins PIE-1 and MEX-5/6. PIE-1 appears to repress all transcription in germline precursors as it is sequentially segregated to the germline during early embryonic cell divisions (Batchelder et al., 1999; Tenenhaus et al., 1998). MEX-5 and MEX-6 are required redundantly to restrict high levels of PIE-1 to the posterior during the first embryonic division (Figure 6), and this regulation appears to be reiterated in subsequent asymmetric divisions of germline precursors (Schubert et al., 2000). The first of the two CCCH fingers in PIE-1, called ZF1, is required for the asymmetric cytoplasmic distribution of PIE-1 after cell division (Reese et al., 2000). In PIE-1 mutants lacking ZF1, PIE-1 protein left in anterior cells after division fails to disappear, suggesting that ZF1

mediates the degradation of PIE-1 in anterior cytoplasm and cells. How MEX-5/6 interact with the PIE-1 ZF1 to degrade PIE-1 anteriorly is not known. MEX-5/6 also are required to restrict other determinants to the posterior of P0 (Schubert et al., 2000). While direct links and exact mechanisms have yet to be defined, a molecular understanding of the transition from cell polarity to cell fate in the early embryo appears close at hand. Cell Polarity and Cell Fate: The Bigger Picture The early worm embryo stands out in its use of highly asymmetric cell divisions to pattern cell fates, but such divisions occur in both prokaryotes and eukaryotes, and are likely to be of substantial importance during the development of most organisms (Horvitz and Herskowitz, 1992). Indeed, early embryos from several animal phyla undergo unequal cleavages that in at least some cases differentially segregate development potential to specific daughter cells (Sternberg and Fe´lix, 1997). Later, in both insects and mammals, asymmetric stem cell divisions generate a layered central nervous system (Chenn and McConnell, 1995; Doe and Bowerman, 2001). Lessons learned from C. elegans should continue to make important contributions to our understanding of this important developmental process. Acknowledgments We would like to thank P. Go¨nczy and G. Seydoux for helpful comments on the manuscript. R.L. was supported by a grant from the American Cancer Society (PF-00-097-010-MBC), J.-E.G. by the University of Lisbon in Portugal, and B.B. by grants from the NIH. References Albertson, D.G. (1984). Formation of the first cleavage spindle in nematode embryos. Dev. Biol. 101, 61–72. Basham, S.E., and Rose, L.S. (1999). Mutations in ooc-5 and ooc-3 disrupt oocyte formation and the reestablishment of asymmetric PAR protein localization in two-cell Caenorhabditis elegans embryos. Dev. Biol. 215, 253–263. Basham, S.E., and Rose, L.S. (2001). The Caenorhabditis elegans polarity gene ooc-5 encodes a Torsin-related protein of the AAA ATPase superfamily. Development 128, 4645–4656. Batchelder, C., Dunn, M.A., Choy, B., Suh, Y., Cassie, C., Shim, E.Y., Shin, T.H., Mello, C., Seydoux, G., and Blackwell, T.K. (1999). Transcriptional repression by the Caenorhabditis elegans germ-line protein PIE-1. Genes Dev. 13, 202–212. Bowerman, B. (1998). Maternal control of pattern formation in early Caenorhabditis elegans embryos. Curr. Top. Dev. Biol. 39, 73–117. Bowerman, B., and Shelton, C.A. (1999). Cell polarity in the early C. elegans embryo. Curr. Opin. Genet. Dev. 9, 390–395. Bowerman, B., Draper, B.W., Mello, C.C., and Priess, J.R. (1993). The maternal gene skn-1 encodes a protein that is distributed unequally in early C. elegans embryos. Cell 74, 443–452. Bowerman, B., Ingram, M.K., and Hunter, C.P. (1997). The maternal par genes and the segregation of cell fate specification activities in early Caenorhabditis elegans embryos. Development 124, 3815– 3826. Boyd, L., Guo, S., Levitan, D., Stinchcomb, D.T., and Kemphues, K.J. (1996). PAR-2 is asymmetrically distributed and promotes association of P granules and PAR-1 with the cortex in C. elegans embryos. Development 122, 3075–3084. Browning, H., and Strome, S. (1996). A sperm-supplied factor required for embryogenesis in C. elegans. Development 122, 391–404. Cai, Y., Chia, W., and Yang, X. (2001). A family of snail-related zinc finger proteins regulates two distinct and parallel mechanisms that

Review 165

mediate Drosophila neuroblast asymmetric divisions. EMBO J. 20, 1704–1714. Cheng, N.N., Kirby, C.M., and Kemphues, K.J. (1995). Control of cleavage spindle orientation in Caenorhabditis elegans: the role of the genes par-2 and par-3. Genetics 139, 549–559. Chenn, A., and McConnell, S.K. (1995). Cleavage orientation and the asymmetric inheritance of Notch1 immunoreactivity in mammalian neurogenesis. Cell 82, 631–641.

Hill, D.P., and Strome, S. (1988). An analysis of the role of microfilaments in the establishment and maintenance of asymmetry in Caenorhabditis elegans zygotes. Dev. Biol. 125, 15–84. Hill, D.P., and Strome, S. (1990). Brief cytochalasin-induced disruption of microfilaments during a critical interval in 1-cell C. elegans embryos alters the partitioning of developmental instructions to the 2-cell embryo. Development 108, 159–172.

Cox, D.N., Bingwei, L., Sun, T.-Q., Williams, L.T., and Jan, Y.N. (2001). Drosophila par-1 is required for oocyte differentiation and microtubule organization. Curr. Biol. 11, 75–87.

Hill, D.P., Shakes, D.C., Ward, S., and Strome, S. (1989). A spermsupplied product essential for initiation of normal embryogenesis in Caenorhabditis elegans is encoded by the paternal-effect embryonic-lethal gene, spe-11. Dev. Biol. 136, 154–166.

Crittenden, S.L., Rudel, D., Binder, J., Evans, T.C., and Kimble, J. (1996). Genes required for GLP-1 asymmetry in the early C. elegans embryo. Dev. Biol. 181, 36–46.

Hird, S.N., and White, J.G. (1993). Cortical and cytoplasmic flow polarity in early embryonic cells of Caenorhabditis elegans. J. Cell Biol. 121, 1343–1355.

Doe, C.Q., and Bowerman, B. (2001). Asymmetric cell division: fly neuroblast meets worm zygote. Curr. Opin. Cell Biol. 13, 68–75.

Hird, S.N., Paulsen, J.E., and Strome, S. (1996). Segregation of germ granules in living Caenorhabditis elegans embryos: cell-type-specific mechanisms for cytoplasmic localization. Development 122, 1303–1312.

Draper, B.W., Mello, C.C., Bowerman, B., Hardin, J., and Priess, J.R. (1996). The maternal gene mex-3 encodes a KH domain protein and regulates blastomere identity in early C. elegans embryos. Cell 87, 205–216. Drewes, G., Ebneth, A., Preuss, U., Mandelkow, E.M., and Mandelkow, E. (1997). MARK, a novel family of protein kinases that phosphorylate microtubule-associated proteins and trigger microtubule disruption. Cell 89, 297–308. Etemad-Moghadam, S.G., and Kemphues, K.J. (1995). Asymmetrically distributed PAR-3 protein contributes to cell polarity and spindle alignment in early C. elegans embryos. Cell 83, 743–752. Evans, T.C., Crittenden, S.L., Kodoyianni, V., and Kimble, J. (1994). Translational control of maternal glp-1 mRNA establishes an asymmetry in the C. elegans embryo. Cell 77, 183–194. Fitch, K.R., Yasuda, G.K., Owens, K.N., and Wakimoto, B.T. (1998). Paternal effects in Drosophila: implications for mechanisms of early development. Curr. Top. Dev. Biol. 38, 1–34. Golden, A. (2000). Cytoplasmic flow and the establishment of polarity in C. elegans 1-cell embryos. Curr. Opin. Genet. Dev. 10, 414–420. Golden, A., Sadler, P.L., Wallenfang, M.R., Schumacher, J.M., Hamill, D.R., Bates, G., Bowerman, B., Seydoux, G., and Shakes, D.C. (2000). Metaphase to anaphase (mat) transition-defective mutants in Caenorhabditis elegans. J. Cell Biol. 151, 1469–1482. Goldstein, B., and Hird, S.N. (1996). Specification of the anteroposterior axis in Caenorhabditis elegans. Development 122, 1467–1474. Gomes, J.E., Encalada, S.E., Swan, K.A., Shelton, C.A., Carter, J.C., and Bowerman, B. (2001). The maternal gene spn-4 encodes a predicted RRM protein required for mitotic spindle orientation and cell fate patterning in early C. elegans embryos. Development 128, 4301– 4314. Go¨nczy, P., Pichler, S., Kirkham, M., and Hyman, A.A. (1999a). Cytoplasmic dynein is required for distinct aspects of MTOC positioning, including centrosome separation, in the one cell stage Caenorhabditis elegans embryo. J. Cell Biol. 147, 135–150. Go¨nczy, P., Schnabel, H., Kaletta, T., Amores, A.D., Hyman, T., and Schnabel, R. (1999b). Dissection of cell division processes in the one cell stage Caenorhabditis elegans embryo by mutational analysis. J. Cell Biol. 144, 927–946. Gotta, M., and Ahringer, J. (2001). Distinct roles for G␣ and G␤␥ in regulating spindle position and orientation in Caenorhabditis elegans embryos. Nat. Cell Biol. 3, 297–300. Grill, S.W., Go¨nczy, P., Stelzer, E.H., and Hyman, A.A. (2001). Polarity controls forces governing asymmetric spindle positioning in the Caenorhabditis elegans embryo. Nature 409, 630–633. Guedes, S., and Priess, J.R. (1996). The C. elegans MEX-1 protein is present in germline blastomeres and is a P granule component. Development 124, 731–739. Guo, S., and Kemphues, K.J. (1995). par-1, a gene required for establishing polarity in C. elegans embryos, encodes a putative ser/ thr kinase that is asymmetrically distributed. Cell 81, 611–620. Guo, S., and Kemphues, K.J. (1996). A non-muscle myosin required for embryonic polarity in Caenorhabditis elegans. Nature 382, 455–458.

Horvitz, H.R., and Herskowitz, I. (1992). Mechanisms of asymmetric cell division: two Bs or not two Bs, that is the question. Cell 68, 237–255. Huang, N.N., Mootz, D.E., Walhout, A.J.M., Vidal, M., and Hunter, C.P. (2002). MEX-3 interacting proteins link cell polarity to asymmetric gene expression in Caenorhabditis elegans. Development 129, 747–759. Hung, T.-J., and Kemphues, K.J. (1999). PAR-6 is a conserved PDZ domain-containing protein that colocalizes with PAR-3 in Caenorhabditis elegans embryos. Development 126, 127–135. Hunter, C.P., and Kenyon, C. (1996). Spatial and temporal controls target pal-1 blastomere-specification activity to a single blastomere lineage in C. elegans embryos. Cell 87, 217–226. Huynh, J.-R., Shulman, J.M., Benton, R., and St Johnston, D. (2001). PAR-1 is required for the maintenance of oocyte fate in Drosophila. Development 128, 1201–1209. Hyman, A.A. (1989). Centrosome movement in the early divisions of Caenorhabditis elegans: a cortical site determining centrosome position. J. Cell Biol. 109, 1185–1194. Hyman, A.A., and White, J.G. (1987). Determination of cell division axes in the early embryogenesis of Caenorhabditis elegans. J. Cell Biol. 105, 2123–2135. Kaltschmidt, J.A., Davidson, C.M., Brown, N.H., and Brand, A.H. (2000). Rotation and asymmetry of the mitotic spindle directs asymmetric cell division in the developing central nervous system. Nat. Cell Biol. 2, 7–12. Kemphues, K. (2000). PARsing embryonic polarity. Cell 101, 345– 348. Kemphues, K.J., Priess, J.R., Morton, D.G., and Cheng, N.S. (1988). Identification of genes required for cytoplasmic localization in early C. elegans embryos. Cell 52, 311–320. Kirby, C., Kusch, M., and Kemphues, K.J. (1990). Mutations in the par genes of C. elegans affect cytoplasmic reorganization during the first cell cycle. Dev. Biol. 142, 203–215. Koelle, M.R., and Horvitz, H.R. (1996). EGL-10 regulates G protein signaling in the C. elegans nervous system and shares a conserved domain with many mammalian proteins. Cell 84, 115–125. Lee, J., Jee, C., Lee, J.I., Lee, M.H., Koo, H.S., Chung, C.H., and Ahnn, J. (2001). A deubiquitinating enzyme, UCH/CeUBP130, has an essential role in the formation of a functional microtubule-organizing centre (MTOC) during early cleavage in C. elegans. Genes Cells 6, 899–911. Levitan, D.J., Boyd, L., Mello, C.C., Kemphues, K.K., and Stinchcomb, D.J. (1994). par-2, a gene required for blastomere asymmetry in Caenorhabditis elegans, encodes zinc-finger and ATP-binding motifs. Proc. Natl. Acad. Sci. USA 91, 6108–6112. Lorson, M.A., Horvitz, H.R., and van den Heuvel, S. (2000). LIN-5 is a novel component of the spindle apparatus required for chromosome segregation and cleavage plane specification in Caenorhabditis elegans. J. Cell Biol. 148, 73–86. McCarter, J., Bartlett, B., Dang, T., and Schedl, T. (1999). On the

Developmental Cell 166

control of oocyte meiotic maturation and ovulation in Caenorhabditis elegans. Dev. Biol. 205, 111–128.

J.R. (2000). MEX-5 and MEX-6 function to establish soma/germline asymmetry in early C. elegans embryos. Mol. Cell 5, 671–682.

Mello, C.C., Schubert, C., Draper, B., Zhang, W., Lobel, R., and Priess, J.R. (1996). The PIE-1 protein and germline specification in C. elegans embryos. Nature 382, 710–712.

Severson, A.F., Hamill, D.R., and Bowerman, B. (2000). The Aurora/ Ipl1p-related kinase AIR-2 is required for cytokinesis and functions at metaphase to recruit the kinesin-like protein ZEN-4/MKLP1 to the mitotic spindle midzone. Curr. Biol. 10, 1162–1171.

Miller, K.G., and Rand, J.B. (2000). A role for RIC-8 (Synembryn) and GOA-1 (G(o)␣) in regulating a subset of centrosome movements during early embryogenesis in Caenorhabditis elegans. Genetics 156, 1649–1660. Morton, D.G., Roos, J.M., and Kemphues, K.J. (1992). par-4, a gene required for cytoplasmic localization and determination of specific cell types in Caenorhabditis elegans embryogenesis. Genetics 130, 771–790. Morton, D.G., Shakes, D.C., Nugent, S., Dichoso, D., Wang, W., Golden, A., and Kemphues, K.J. (2002). The Caenorhabditis elegans par-5 gene encodes a 14-3-3 protein required for cellular asymmetry in the early embryo. Dev. Biol. 241, 47–58.

Seydoux, G., and Fire, A. (1994). Soma-germline asymmetry in the distribution of embryonic RNAs in C. elegans development. Development 120, 2823–2834. Shelton, C.A., Carter, J.C., Ellis, G.C., and Bowerman, B. (1999). The nonmuscle myosin regulatory light chain gene mlc-4 is required for cytokinesis, anterior-posterior polarity, and body morphology during Caenorhabditis elegans embryogenesis. J. Cell Biol. 146, 439–451. Shulman, J.M., Benton, R., and St Johnston, D. (2000). The Drosophila homolog of PAR-1 organizes the oocyte cytoskeleton and directs Oskar mRNA localization to the posterior pole. Cell 101, 377–388.

O’Connell, K.F., Leys, C.M., and White, J.G. (1998). A genetic screen for temperature-sensitive cell-division mutants of Caenorhabditis elegans. Genetics 149, 1303–1321.

Skop, A.R., and White, J.G. (1998). The dynactin complex is required for cleavage plane specification in early Caenorhabditis elegans embryos. Curr. Biol. 8, 1110–1116.

O’Connell, K.F., Maxwell, K.N., and White, J.G. (2000). The spd-2 gene is required for polarization of the anteroposterior axis and formation of the sperm asters in the Caenorhabditis elegans zygote. Dev. Biol. 222, 55–70.

Spana, E.P., and Doe, C.Q. (1995). The prospero transcription factor is asymmetrically localized to the cell cortex during neuroblast mitosis in Drosophila. Development 121, 3187–3195.

Peng, C.Y., Manning, L., Alberson, R., and Doe, C.Q. (2000). The tumour-suppressor genes lgl and dlg regulate basal protein targeting in Drosophila neuroblasts. Nature 408, 596–600. Pichler, S., Go¨nczy, P., Schnabel, H., Pozniakowski, A., Ashford, A., Schnabel, R., and Hyman, A.A. (2000). OOC-3, a novel putative transmembrane protein required for establishment of cortical domains and spindle orientation in the P(1) blastomere of C. elegans embryos. Development 127, 2063–2073. Piotrowska, K., and Zernicka-Goetz, M. (2001). Role for sperm in spatial patterning of the early mouse embryo. Nature 409, 517–521. Piotrowska, K., Wianny, F., Pedersen, R.A., and Zernicka-Goetz, M. (2001). Blastomeres arising from the first cleavage division have distinguishable fates in normal mouse development. Development 128, 3739–3748. Priess, J.R., and Thomson, J.N. (1987). Cellular interactions in early C. elegans embryos. Cell 34, 85–100. Rappleye, C.A., Tagawa, A., Lyczak, R., Bowerman, B., and Aroian, R.V. (2002). The anaphase-promoting complex and separin are required for embryonic anterior-posterior axis formation. Dev. Cell 2, 195–206. Reese, K.J., Dunn, M.A., Waddle, J.A., and Seydoux, G. (2000). Asymmetric segregation of PIE-1 in C. elegans is mediated by two complementary mechanisms that act through separate PIE-1 protein domains. Mol. Cell 6, 445–455.

St Johnston, D., and Nusslein-Volhard, C. (1992). The origin of pattern and polarity in the Drosophila embryo. Cell 68, 201–219. Sternberg, P., and Fe´lix, M.-A. (1997). Evolution of cell lineage. Curr. Opin. Genet. Dev. 7, 543–550. Strome, S., and Wood, W.B. (1983). Generation of asymmetry and segregation of germ-line granules in early C. elegans embryos. Cell 35, 15–25. Sulston, J.E., Schierenberg, E., White, J.G., and Thomson, J.N. (1983). The embryonic cell lineage of the nematode Caenorhabditis elegans. Dev. Biol. 100, 64–119. Tabara, H., Hill, R.J., Mello, C.C., Priess, J.R., and Kohara, Y. (1999). pos-1 encodes a cytoplasmic zinc-finger protein essential for germline specification in C. elegans. Development 126, 1–11. Tenenhaus, C., Schubert, C., and Seydoux, G. (1998). Genetic requirements for PIE-1 localization and inhibition of gene expression in the embryonic germ lineage of Caenorhabditis elegans. Dev. Biol. 200, 212–224. Thorpe, C.J., Schlesinger, A., and Bowerman, B. (2000). Wnt signalling in Caenorhabditis elegans: regulating repressors and polarizing the cytoskeleton. Trends Cell Biol. 10, 10–17. Tomancak, P., Piano, F., Riechmann, V., Gunsalus, K.C., Kemphues, K.J., and Ephrussi, A. (2000). A Drosophila melanogaster homologue of Caenorhabditis elegans par-1 acts at an early step in embryonic axis formation. Nat. Cell Biol. 2, 458–460.

Riechmann, V., Gutierrez, G.J., Filardo, P., Nebreda, A.R., and Ephrussi, A. (2002). Par-1 regulates stability of the posterior determinant Oskar by phosphorylation. Nat. Cell Biol. 4, 337–342.

Tree, D.R.P., Shulman, J.M., Rousset, R., Scott, M.P., Gubb, D., and Axelrod, J.D. (2002). Prickle mediates feedback amplification to generate asymmetric planar cell polarity signaling. Cell 109, 371–381.

Rose, L.S., and Kemphues, K.J. (1997). The let-99 gene is required for proper spindle orientation during cleavage of the C. elegans embryo. Development 125, 1337–1346.

Tsou, M.B., Hayashi, A., DeBella, L.R., McGrath, G., and Rose, L.S. (2002). LET-99 determines spindle position and is asymmetrically enriched in response to PAR polarity cues. Development, in press.

Rose, L.S., and Kemphues, K.J. (1998). Early patterning of the C. elegans embryo. Annu. Rev. Genet. 32, 521–545.

Wallenfang, M.R., and Seydoux, G. (2000). Polarization of the anterior-posterior axis of C. elegans is a microtubule-directed process. Nature 408, 89–92.

Rose, L.S., Lamb, M.L., Hird, S.N., and Kemphues, K.J. (1995). Pseudocleavage is dispensable for polarity and development in C. elegans embryos. Dev. Biol. 168, 479–489. Sadler, P.L., and Shakes, D.C. (2000). A nucleate Caenorhabditis elegans sperm can crawl, fertilize oocytes and direct anterior-posterior polarization of the 1-cell embryo. Development 127, 355–366. Schaefer, M., Petronczki, M., Dorner, D., Forte, M., and Knoblich, J.A. (2001). Heterotrimeric G proteins direct two modes of asymmetric cell division in the Drosophila nervous system. Cell 107, 183–194. Schierenberg, E. (1987). Reversal of cellular polarity and early cellcell interaction in the embryo of Caenorhabditis elegans. Dev. Biol. 122, 452–463. Schubert, C.M., Lin, R., de Vries, C.J., Plasterk, R.H., and Priess,

Watts, J.L., Morton, D.G., Bestman, J., and Kemphues, K.J. (2000). The C. elegans par-4 gene encodes a putative serine-threonine kinase required for establishing embryonic asymmetry. Development 127, 1476–1475. Wong, H.C., Mao, J., Nguyen, J.T., Srinivas, S., Zhang, W., Liu, B., Li, L., Wu, D., and Zheng, J. (2000). Structural basis of the recognition of the Dishevelled DEP domain in the Wnt signaling pathway. Nat. Struct. Biol. 7, 1178–1184.