- Email: [email protected]
Gastrulation: PARtaking of the Bottle
Current Biology, Vol. 13, R223–R225, March 18, 2003, ©2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/S0960-9822(03)00155-6
Gastrulation: PARtaking of the Bottle Aaron P. Putzke and Joel H. Rothman
Gastrulation in C. elegans embryos involves ingression of individual cells, but is driven by apical constriction of the kind that promotes migration of epithelial cell sheets. Recent work shows that PAR proteins, known for their role in polarization and unequal cell division, are also associated with the polarization that establishes this apical constriction.
How poor are they who have not gastrulation! What wound did ever heal but by ingression? (apologies to William Shakespeare)
We all begin as a cluster of rather featureless cells; our final forms are sculpted into specialized tissue layers which achieve their final state by the concerted movements of cells. In triploblastic animals, multiple layers initially arise from a cluster of cells by ingression or gastrulation, a process that has attracted the attention of developmental biologists since His proposed in 1874 that invagination is propagated by forces arising from cell division [1]. Although gastrulation looks dramatically different in various beasts — such as echinoderms [2], insects [3] and vertebrates [4] (Figure 1A) — it generally involves the coordinated inward movement of epithelialized cell sheets. But some creatures, such as nematodes, present an extreme case of simplicity in which ingression of individual, rather than sheets, of cells sink into the embryonic interior apparently independently during gastrulation. Recent studies on embryos of the nematode Caenorhabditis elegans indicate that the cellular mechanisms promoting ingression may be common to animals with widely divergent modes of gastrulation. A common theme in gastrulation is that invaginating cells constrict at their apical ends, resulting in wedge-shape ‘bottle cells’. Cell autonomous apical constriction [5], triggered by cell-fate specification [6], has recently been found also to drive gastrulation in C. elegans. Unexpectedly, the ‘PAR’ proteins, known from their role in generating the anterior-posterior cell polarity that creates unequal daughters during asymmetric cleavage of the zygote, are also implicated in the apical-basal asymmetry of ingressing cells [6]. One model describing the mechanisms of cell ingression during gastrulation posits that cells adjacent to ingressing cells move toward the site of internalization, causing the cell sheet to buckle [7]. Another popular view proposes that a continuous actomyosin ‘purse-string’ within a ring of cells contracts, causing cells within the ring to ingress. In the most commonly held model, ingressing cells change their shape by Department of MCD Biology, University of California, Santa Barbara, California 93106, USA. E-mail: [email protected]
Dispatch
constriction of their apical ends, resulting in bending of the sheet and their inward movement toward the opposite (basal) surface. The mechanics of such apical constriction was modeled over 55 years ago [8] with brass bars and rubber bands representing a layer of epithelial cells: by changing the tension on one side of the layer, the sheet would buckle inward, thus mimicking ingression. Ingression by apical constriction may be energetically more favorable than other means of gap closure, such as the purse-string model. A study using computer algorithms to mimic gastrulating embryos showed that, when a cell constricts apically, it propagates a wave through the surrounding sheet of cells, causing an inward buckling [9]. Morphological support for apical constriction in a developing organism came from the observation of bottle cells, for example during primary invagination of epithelial cells in sea urchins [1], in which bottle cells have been shown to be essential for gastrulation [2]. Moreover, molecular evidence for apical constriction has been reported in Drosophila, in which it was found that actin and myosin localize to apical cortices, concomitant with cell shape changes during ventral furrow formation [10,11]. The simplicity of C. elegans gastrulation, which begins with the ingression of just two cells — the gut precursor cells or ‘endoderm pair’ (Figure 1A) — makes it a highly manipulable system with which to study the mechanisms that drive cell ingression (with the caveat that one must suffer an onslaught of jokes from epitheliocentric researchers, who would have us believe that what these worm embryos do is not truly ‘gastrulation’). It has been suggested that proper cell contacts are essential for gastrulation in C. elegans [12], implying that interactions between ingressing cells and their neighbors are essential for this movement. It was recently shown, however, that gastrulation movements of the endoderm pair can occur in culture with a small number of cells, and that cell-autonomous apical constriction of these two cells drives their inward migration [5]. Such a cell-autonomous mechanism may also occur during Drosophila gastrulation where, prior to ingression, the apical constriction of ventral cells occurs stochastically, perhaps suggesting that cell contacts may not be necessary for apical constriction to take place [3]. As in other animals, microfilaments — but not microtubules — are essential for gastrulation in C. elegans. A non-muscle myosin localizes to the apical edges of the ingressing endoderm cells, and pharmacological studies suggest that myosin activity is required for gastrulation [5], further demonstrating the importance of the actomyosin system in nematode gastrulation. But active migration of cells surrounding the ingressing cells is not essential for their movement into the interior. At the time of their ingression, the pair of endoderm cells is flanked on the anterior and posterior sides, respectively, by a mesodermal and a germline progenitor (Figure 1A); these flanking cells congress to fill the gap left on the surface following migration of the
Dispatch R224
A
Ea Ep P4 MS
B
AB
P1
Current Biology
ingressing cells. The flanking cells do not move together by chemotaxis or by actively undergoing protrusion-based motility during ingression of the endoderm pair [5]. Rather, apical constriction of the ingressing endoderm cells — which has been detected by observing the convergence of fluorescent microspheres adhering to their surfaces — apparently draws the flanking cells towards each other, as the endoderm cells shrink away from the embryonic surface [5]. The polarity that results in apical constriction of the endoderm pair is an intrinsic property of these cells, as the position of the flanking cells does not alter this polarity. A number of studies have shown that, in C. elegans, cells are specified before they are sent into the interior of the embryo, and that gastrulation requires proper cell-fate specification. For example, inhibition of zygotic transcription blocks ingression of the endoderm pair [13], as do mutations in a number of maternal and zygotic genes that are essential for specifying the identity of the endoderm progenitor cell [14]. Specification of cells as mesodermal is sufficient to activate their ingression: a mutation that transforms ectodermal progenitors, which do not normally ingress, into mesodermal precursors, confers the potential to undergo ectopic ingression [6]. In contrast, mutations in two genes, gad-1 and emb-5, result in premature division of the endoderm pair and their failure to ingress, without preventing specification of the endoderm [15,16]. These phenotypes suggested that GAD-1 and EMB-5 might function specifically in gastrulation . Recent (unpublished) findings by our group, however, suggest that both proteins may instead be required to set the transcriptional activity of the zygotic endoderm-specifying end genes [14]; the threshold for activation of gastrulation is apparently higher than for specification of the endoderm progenitor fate. Thus, factors specifically involved in C. elegans gastrulation have yet to be identified. The relationship between cell-fate specification and gastrulation is similarly seen in Drosophila (reviewed in [17]). For
Figure 1. Gastrulation movements and changes in PAR protein localization. (A) Gastrulation events in three model organisms: lateral view of a sea urchin embryo with ingression of bottle cells (wedge-shaped cells); ventral furrow formation in Drosophila (ingressing cells shaded gray); and lateral view of C. elegans embryo. The intestinal precursor cells — the ‘endoderm pair’ Ea and Ep — ingress, while mesoderm (MS) and germline (P4) descendants move to overlie Ea and Ep. (Black arrows indicate movement of ingressing cells.) (B) Localization of PAR-2 (red) and PAR-3 (green) in C. elegans two-cell and 26-cell embryos, respectively. In two-cell embryos, PAR-3 excludes PAR-2 from the anterior of P1. In the 26-cell embryos, PAR-3 and PAR-2 have switched from anterior-posterior polarity to apical-basal polarity (shown here only for the endoderm pair). PAR-3 excludes PAR-2 from the apical surface of each cell.
example, the twist gene, which is expressed in the ventralmost cells of the fly embryo prior to ventral furrow formation, is required to specify mesodermal cell fate and gastrulation. In the absence of Twist protein, ingression and the subsequent differentiation of ventral cells do not occur. While the machinery that drives gastrulation in C. elegans is not well understood, a possible link has been made between proteins that direct asymmetric cell division and the asymmetric behavior of ingressing cells in C. elegans [6]. The PAR proteins were originally identified by their requirement in establishing cell polarity and directing asymmetric cell divisions in C. elegans, and orthologs have since been shown to function in cell polarity in Drosophila, Xenopus and mammals. PAR-3 and PAR-6 localize at the anterior cortex of the dividing zygote, and PAR-2 is reciprocally localized at the posterior cortex [18]. This reciprocal localization shifts from anterior-posterior to apical-basal polarity after the two-cell stage (Figure 1B) [6]. Asymmetric PAR protein localization at the cortex depends on cell contact: PAR-2 localizes to regions where cells are in contact, whereas PAR-3 and PAR–6 are present in regions that are not in contact (Figure 1B). Just as posterior-specific localization of PAR-2 in the zygote requires anterior PAR-3, the exclusion of PAR-2 from the cell-contact-free apical regions in later embryos depends on apical PAR-3. It has been suggested that PAR-3 localization is used to distinguish the basal from lateral surfaces of cells surrounding the blastocoel; in the absence of PAR-3, the lateral surfaces of neighboring cells lose contact and extra blastocoel-like voids are created. Establishment of differential adhesion through cell polarity was previously suggested for blastocyst formation and differentiation of trophectoderm in mammals [19]. Could this be applied to cell polarity changes during gastrulation as well? It is possible that PAR-3 helps to localize actin and myosin apically, resulting in the apical constriction that drives gastrulation. It will be interesting to
Current Biology R225
learn how localization of PAR-3 in internalizing cells might regulate apical constriction and ingression. The formation of bottle cells by apical constriction is widely used for diverse processes, including gastrulation, neurulation and embryonic wound healing, and the mechanisms that direct formation of bottle cells during ingression is likely to be conserved in many organisms, regardless of whether sheets of cells, or individual cells are involved. In the case of embryonic wound healing in Xenopus, cells are brought together to seal the wound by apical constriction and ingression, as opposed to protrusion-based motility [20]. When a rectangular piece of ectodermal tissue is removed from a Xenopus embryo, the wound does not close via an actomyosin purse-string mechanism; rather, cells exposed by the missing ectoderm constrict apically and ingress, pulling the overlying cells together to seal the wound. Intriguingly, the apical constriction used in embryo wound healing differs from the protrusion-based motility system in adult wound healing [20]. One conspicuous difference between the two mechanisms is the extent of cellular differentiation. Is it possible that movement of undifferentiated cells favors ingression, while movement of fully differentiated cells is favored by a protrusion-based motility mechanism? As much as we have learned about bottle-cell formation during gastrulation and wound healing, and the mechanisms driving the ingression of cells, we still lack detailed knowledge of the mechanisms that regulate these events. In this regard, it will be of interest to unveil the molecular switches that initiate apical constriction and the signals that may override those switches in fully differentiated cells. References 1. Davidson, L.A., Koehl, M.A.R., Keller, R. and Oster, G.F. (1995). How do sea-urchins invaginate - using biomechanics to distinguish between mechanisms of primary invagination. Development 121, 2005–2018. 2. Kimberly, E.L. and Hardin, J. (1998). Bottle cells are required for the initiation of primary invagination in the sea urchin embryo. Dev. Biol. 204, 235–250. 3. Sweeton, D., Parks, S., Costa, M. and Wieschaus, E. (1991). Gastrulation in Drosophila - the formation of the ventral furrow and posterior midgut invaginations. Development 112, 775–789. 4. Purcell, S.M. and Keller, R. (1993). A different type of amphibian mesoderm morphogenesis in Ceratophrys-Ornata. Development 117, 307–317. 5. Lee, J.-Y. and Goldstein, B. (2003). Mechanisms of cell positioning during C. elegans gastrulation. Development 130, 307–320. 6. Nance, J. and Priess, J.R. (2002). Cell polarity and gastrulation in C. elegans. Development 129, 387–397. 7. Burke, R.D., Myers, R.L., Sexton, T.L. and Jackson, C. (1991). Cell movements during the initial phase of gastrulation in the sea-urchin embryo. Dev. Biol. 146, 542–557. 8. Lewis, W.H. (1947). Mechanics of invagination. Anat. Rec. 97, 139–156. 9. Odell, G.M., Oster, G., Alberch, P. and Burnside, B. (1981). The mechanical basis of morphogenesis. 1. Epithelial folding and invagination. Dev. Biol. 85, 446–462. 10. Kiehart, D.P., Ketchum, A., Young, P., Lutz, D., Alfenito, M.R., Chang, X.J., Awobuluyi, M., Pesacreta, T.C., Inoue, S., Stewart, C.T., et al. (1990). Contractile proteins in Drosophila development. Ann. N. Y. Acad. Sci. 582, 233–251. 11. Young, P.E., Pesacreta, T.C. and Kiehart, D.P. (1991). Dynamic changes in the distribution of cytoplasmic myosin during Drosophila embryogenesis. Development 111, 1–14. 12. Skiba, F. and Schierenberg, E. (1992). Cell lineages, developmental timing and spatial pattern-formation in embryos of free-living soil nematodes. Dev. Biol. 151, 597–610.
13.
14.
15.
16.
17. 18.
19. 20.
Powell-Coffman, J.A., Knight, J. and Wood, W.B. (1996). Onset of C. elegans gastrulation is blocked by inhibition of embryonic transcription with an RNA polymerase antisense RNA. Dev. Biol. 178, 472–483. Zhu, J., Hill, R.J., Heid, P.J., Fukuyama, M., Sugimoto, A., Priess, J.R. and Rothman, J.H. (1997). end-1 encodes an apparent GATA factor that specifies the endoderm precursor in Caenorhabditis elegans embryos. Genes Dev. 11, 2883–2896. Knight, J.K. and Wood, W.B. (1998). Gastrulation initiation in Caenorhabditis elegans requires the function of gad-1, which encodes a protein with WD repeats. Dev. Biol. 198, 253–265. Nishiwaki, K., Sano, T. and Miwa, J. (1993). Emb-5, a gene required for the correct timing of gut precursor cell-division during gastrulation in Caenorhabditis elegans, encodes a protein similar to the yeast nuclear-protein Spt6. Mol. Gen. Genet. 239, 313–322. Castanon, I. and Baylies, M.K. (2002). A Twist in fate: evolutionary comparison of Twist structure and function. Gene 287, 11–22. Etemad-Moghadam, B., Guo, S. 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. Watson, A.J. and Barcroft, L.C. (2001). Regulation of blastocyst formation. Front. Biosci. 6, D708–D730. Davidson, L.A., Ezin, A.M. and Keller, R. (2002). Embryonic wound healing by apical contraction and ingression in Xenopus laevis. Cell Motil. Cytoskeleton 53, 163–176.
Gastrulation: PARtaking of the Bottle Aaron P. Putzke and Joel H. Rothman
Gastrulation in C. elegans embryos involves ingression of individual cells, but is driven by apical constriction of the kind that promotes migration of epithelial cell sheets. Recent work shows that PAR proteins, known for their role in polarization and unequal cell division, are also associated with the polarization that establishes this apical constriction.
How poor are they who have not gastrulation! What wound did ever heal but by ingression? (apologies to William Shakespeare)
We all begin as a cluster of rather featureless cells; our final forms are sculpted into specialized tissue layers which achieve their final state by the concerted movements of cells. In triploblastic animals, multiple layers initially arise from a cluster of cells by ingression or gastrulation, a process that has attracted the attention of developmental biologists since His proposed in 1874 that invagination is propagated by forces arising from cell division [1]. Although gastrulation looks dramatically different in various beasts — such as echinoderms [2], insects [3] and vertebrates [4] (Figure 1A) — it generally involves the coordinated inward movement of epithelialized cell sheets. But some creatures, such as nematodes, present an extreme case of simplicity in which ingression of individual, rather than sheets, of cells sink into the embryonic interior apparently independently during gastrulation. Recent studies on embryos of the nematode Caenorhabditis elegans indicate that the cellular mechanisms promoting ingression may be common to animals with widely divergent modes of gastrulation. A common theme in gastrulation is that invaginating cells constrict at their apical ends, resulting in wedge-shape ‘bottle cells’. Cell autonomous apical constriction [5], triggered by cell-fate specification [6], has recently been found also to drive gastrulation in C. elegans. Unexpectedly, the ‘PAR’ proteins, known from their role in generating the anterior-posterior cell polarity that creates unequal daughters during asymmetric cleavage of the zygote, are also implicated in the apical-basal asymmetry of ingressing cells [6]. One model describing the mechanisms of cell ingression during gastrulation posits that cells adjacent to ingressing cells move toward the site of internalization, causing the cell sheet to buckle [7]. Another popular view proposes that a continuous actomyosin ‘purse-string’ within a ring of cells contracts, causing cells within the ring to ingress. In the most commonly held model, ingressing cells change their shape by Department of MCD Biology, University of California, Santa Barbara, California 93106, USA. E-mail: [email protected]
Dispatch
constriction of their apical ends, resulting in bending of the sheet and their inward movement toward the opposite (basal) surface. The mechanics of such apical constriction was modeled over 55 years ago [8] with brass bars and rubber bands representing a layer of epithelial cells: by changing the tension on one side of the layer, the sheet would buckle inward, thus mimicking ingression. Ingression by apical constriction may be energetically more favorable than other means of gap closure, such as the purse-string model. A study using computer algorithms to mimic gastrulating embryos showed that, when a cell constricts apically, it propagates a wave through the surrounding sheet of cells, causing an inward buckling [9]. Morphological support for apical constriction in a developing organism came from the observation of bottle cells, for example during primary invagination of epithelial cells in sea urchins [1], in which bottle cells have been shown to be essential for gastrulation [2]. Moreover, molecular evidence for apical constriction has been reported in Drosophila, in which it was found that actin and myosin localize to apical cortices, concomitant with cell shape changes during ventral furrow formation [10,11]. The simplicity of C. elegans gastrulation, which begins with the ingression of just two cells — the gut precursor cells or ‘endoderm pair’ (Figure 1A) — makes it a highly manipulable system with which to study the mechanisms that drive cell ingression (with the caveat that one must suffer an onslaught of jokes from epitheliocentric researchers, who would have us believe that what these worm embryos do is not truly ‘gastrulation’). It has been suggested that proper cell contacts are essential for gastrulation in C. elegans [12], implying that interactions between ingressing cells and their neighbors are essential for this movement. It was recently shown, however, that gastrulation movements of the endoderm pair can occur in culture with a small number of cells, and that cell-autonomous apical constriction of these two cells drives their inward migration [5]. Such a cell-autonomous mechanism may also occur during Drosophila gastrulation where, prior to ingression, the apical constriction of ventral cells occurs stochastically, perhaps suggesting that cell contacts may not be necessary for apical constriction to take place [3]. As in other animals, microfilaments — but not microtubules — are essential for gastrulation in C. elegans. A non-muscle myosin localizes to the apical edges of the ingressing endoderm cells, and pharmacological studies suggest that myosin activity is required for gastrulation [5], further demonstrating the importance of the actomyosin system in nematode gastrulation. But active migration of cells surrounding the ingressing cells is not essential for their movement into the interior. At the time of their ingression, the pair of endoderm cells is flanked on the anterior and posterior sides, respectively, by a mesodermal and a germline progenitor (Figure 1A); these flanking cells congress to fill the gap left on the surface following migration of the
Dispatch R224
A
Ea Ep P4 MS
B
AB
P1
Current Biology
ingressing cells. The flanking cells do not move together by chemotaxis or by actively undergoing protrusion-based motility during ingression of the endoderm pair [5]. Rather, apical constriction of the ingressing endoderm cells — which has been detected by observing the convergence of fluorescent microspheres adhering to their surfaces — apparently draws the flanking cells towards each other, as the endoderm cells shrink away from the embryonic surface [5]. The polarity that results in apical constriction of the endoderm pair is an intrinsic property of these cells, as the position of the flanking cells does not alter this polarity. A number of studies have shown that, in C. elegans, cells are specified before they are sent into the interior of the embryo, and that gastrulation requires proper cell-fate specification. For example, inhibition of zygotic transcription blocks ingression of the endoderm pair [13], as do mutations in a number of maternal and zygotic genes that are essential for specifying the identity of the endoderm progenitor cell [14]. Specification of cells as mesodermal is sufficient to activate their ingression: a mutation that transforms ectodermal progenitors, which do not normally ingress, into mesodermal precursors, confers the potential to undergo ectopic ingression [6]. In contrast, mutations in two genes, gad-1 and emb-5, result in premature division of the endoderm pair and their failure to ingress, without preventing specification of the endoderm [15,16]. These phenotypes suggested that GAD-1 and EMB-5 might function specifically in gastrulation . Recent (unpublished) findings by our group, however, suggest that both proteins may instead be required to set the transcriptional activity of the zygotic endoderm-specifying end genes [14]; the threshold for activation of gastrulation is apparently higher than for specification of the endoderm progenitor fate. Thus, factors specifically involved in C. elegans gastrulation have yet to be identified. The relationship between cell-fate specification and gastrulation is similarly seen in Drosophila (reviewed in [17]). For
Figure 1. Gastrulation movements and changes in PAR protein localization. (A) Gastrulation events in three model organisms: lateral view of a sea urchin embryo with ingression of bottle cells (wedge-shaped cells); ventral furrow formation in Drosophila (ingressing cells shaded gray); and lateral view of C. elegans embryo. The intestinal precursor cells — the ‘endoderm pair’ Ea and Ep — ingress, while mesoderm (MS) and germline (P4) descendants move to overlie Ea and Ep. (Black arrows indicate movement of ingressing cells.) (B) Localization of PAR-2 (red) and PAR-3 (green) in C. elegans two-cell and 26-cell embryos, respectively. In two-cell embryos, PAR-3 excludes PAR-2 from the anterior of P1. In the 26-cell embryos, PAR-3 and PAR-2 have switched from anterior-posterior polarity to apical-basal polarity (shown here only for the endoderm pair). PAR-3 excludes PAR-2 from the apical surface of each cell.
example, the twist gene, which is expressed in the ventralmost cells of the fly embryo prior to ventral furrow formation, is required to specify mesodermal cell fate and gastrulation. In the absence of Twist protein, ingression and the subsequent differentiation of ventral cells do not occur. While the machinery that drives gastrulation in C. elegans is not well understood, a possible link has been made between proteins that direct asymmetric cell division and the asymmetric behavior of ingressing cells in C. elegans [6]. The PAR proteins were originally identified by their requirement in establishing cell polarity and directing asymmetric cell divisions in C. elegans, and orthologs have since been shown to function in cell polarity in Drosophila, Xenopus and mammals. PAR-3 and PAR-6 localize at the anterior cortex of the dividing zygote, and PAR-2 is reciprocally localized at the posterior cortex [18]. This reciprocal localization shifts from anterior-posterior to apical-basal polarity after the two-cell stage (Figure 1B) [6]. Asymmetric PAR protein localization at the cortex depends on cell contact: PAR-2 localizes to regions where cells are in contact, whereas PAR-3 and PAR–6 are present in regions that are not in contact (Figure 1B). Just as posterior-specific localization of PAR-2 in the zygote requires anterior PAR-3, the exclusion of PAR-2 from the cell-contact-free apical regions in later embryos depends on apical PAR-3. It has been suggested that PAR-3 localization is used to distinguish the basal from lateral surfaces of cells surrounding the blastocoel; in the absence of PAR-3, the lateral surfaces of neighboring cells lose contact and extra blastocoel-like voids are created. Establishment of differential adhesion through cell polarity was previously suggested for blastocyst formation and differentiation of trophectoderm in mammals [19]. Could this be applied to cell polarity changes during gastrulation as well? It is possible that PAR-3 helps to localize actin and myosin apically, resulting in the apical constriction that drives gastrulation. It will be interesting to
Current Biology R225
learn how localization of PAR-3 in internalizing cells might regulate apical constriction and ingression. The formation of bottle cells by apical constriction is widely used for diverse processes, including gastrulation, neurulation and embryonic wound healing, and the mechanisms that direct formation of bottle cells during ingression is likely to be conserved in many organisms, regardless of whether sheets of cells, or individual cells are involved. In the case of embryonic wound healing in Xenopus, cells are brought together to seal the wound by apical constriction and ingression, as opposed to protrusion-based motility [20]. When a rectangular piece of ectodermal tissue is removed from a Xenopus embryo, the wound does not close via an actomyosin purse-string mechanism; rather, cells exposed by the missing ectoderm constrict apically and ingress, pulling the overlying cells together to seal the wound. Intriguingly, the apical constriction used in embryo wound healing differs from the protrusion-based motility system in adult wound healing [20]. One conspicuous difference between the two mechanisms is the extent of cellular differentiation. Is it possible that movement of undifferentiated cells favors ingression, while movement of fully differentiated cells is favored by a protrusion-based motility mechanism? As much as we have learned about bottle-cell formation during gastrulation and wound healing, and the mechanisms driving the ingression of cells, we still lack detailed knowledge of the mechanisms that regulate these events. In this regard, it will be of interest to unveil the molecular switches that initiate apical constriction and the signals that may override those switches in fully differentiated cells. References 1. Davidson, L.A., Koehl, M.A.R., Keller, R. and Oster, G.F. (1995). How do sea-urchins invaginate - using biomechanics to distinguish between mechanisms of primary invagination. Development 121, 2005–2018. 2. Kimberly, E.L. and Hardin, J. (1998). Bottle cells are required for the initiation of primary invagination in the sea urchin embryo. Dev. Biol. 204, 235–250. 3. Sweeton, D., Parks, S., Costa, M. and Wieschaus, E. (1991). Gastrulation in Drosophila - the formation of the ventral furrow and posterior midgut invaginations. Development 112, 775–789. 4. Purcell, S.M. and Keller, R. (1993). A different type of amphibian mesoderm morphogenesis in Ceratophrys-Ornata. Development 117, 307–317. 5. Lee, J.-Y. and Goldstein, B. (2003). Mechanisms of cell positioning during C. elegans gastrulation. Development 130, 307–320. 6. Nance, J. and Priess, J.R. (2002). Cell polarity and gastrulation in C. elegans. Development 129, 387–397. 7. Burke, R.D., Myers, R.L., Sexton, T.L. and Jackson, C. (1991). Cell movements during the initial phase of gastrulation in the sea-urchin embryo. Dev. Biol. 146, 542–557. 8. Lewis, W.H. (1947). Mechanics of invagination. Anat. Rec. 97, 139–156. 9. Odell, G.M., Oster, G., Alberch, P. and Burnside, B. (1981). The mechanical basis of morphogenesis. 1. Epithelial folding and invagination. Dev. Biol. 85, 446–462. 10. Kiehart, D.P., Ketchum, A., Young, P., Lutz, D., Alfenito, M.R., Chang, X.J., Awobuluyi, M., Pesacreta, T.C., Inoue, S., Stewart, C.T., et al. (1990). Contractile proteins in Drosophila development. Ann. N. Y. Acad. Sci. 582, 233–251. 11. Young, P.E., Pesacreta, T.C. and Kiehart, D.P. (1991). Dynamic changes in the distribution of cytoplasmic myosin during Drosophila embryogenesis. Development 111, 1–14. 12. Skiba, F. and Schierenberg, E. (1992). Cell lineages, developmental timing and spatial pattern-formation in embryos of free-living soil nematodes. Dev. Biol. 151, 597–610.
13.
14.
15.
16.
17. 18.
19. 20.
Powell-Coffman, J.A., Knight, J. and Wood, W.B. (1996). Onset of C. elegans gastrulation is blocked by inhibition of embryonic transcription with an RNA polymerase antisense RNA. Dev. Biol. 178, 472–483. Zhu, J., Hill, R.J., Heid, P.J., Fukuyama, M., Sugimoto, A., Priess, J.R. and Rothman, J.H. (1997). end-1 encodes an apparent GATA factor that specifies the endoderm precursor in Caenorhabditis elegans embryos. Genes Dev. 11, 2883–2896. Knight, J.K. and Wood, W.B. (1998). Gastrulation initiation in Caenorhabditis elegans requires the function of gad-1, which encodes a protein with WD repeats. Dev. Biol. 198, 253–265. Nishiwaki, K., Sano, T. and Miwa, J. (1993). Emb-5, a gene required for the correct timing of gut precursor cell-division during gastrulation in Caenorhabditis elegans, encodes a protein similar to the yeast nuclear-protein Spt6. Mol. Gen. Genet. 239, 313–322. Castanon, I. and Baylies, M.K. (2002). A Twist in fate: evolutionary comparison of Twist structure and function. Gene 287, 11–22. Etemad-Moghadam, B., Guo, S. 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. Watson, A.J. and Barcroft, L.C. (2001). Regulation of blastocyst formation. Front. Biosci. 6, D708–D730. Davidson, L.A., Ezin, A.M. and Keller, R. (2002). Embryonic wound healing by apical contraction and ingression in Xenopus laevis. Cell Motil. Cytoskeleton 53, 163–176.