Current Biology
Dispatches A
Homeostatic response to hypoxia is regulated by the N-end rule pathway in plants. Nature 479, 415–418.
B Light
Asymmetric distribution of plant growth hormones
ERF VII transcription factors
Apical hook
Low oxygen
Proteolytic destruction of ERF VIIs Oxygen, light Current Biology
Figure 1. Early seedling development is guided by light and oxygen. (A) Skotomorphogenesis versus photomorphogenesis. In the absence of light, plants develop long hypocotyls and an apical hook, bending down their yellow cotyledons and protecting the stem-cell like apical meristem. The apical hook is maintained longer in a low oxygen environment. In the light, cotyledons develop functional chloroplasts and hypocotyls stay short, but roots grow longer to provide minerals for further growth. Photo by Annkatrin Rose. (B) Light acts as a master regulator of morphogenesis. Light prevents the asymmetric distribution of auxin and other growth hormones in the apical hook region. In absence of oxygen, ERF VIIs accumulate and stabilize the apical hook. Light can overrule the effect of low oxygen on ERF VII degradation, presumably by activating a different ERF VII degradation pathway.
access to fresh air will eventually allow for a low level of light, providing energy for a slow transition to photomorphogenic development. In a low oxygen atmosphere, hook opening does not happen. Surprisingly, this apparent lack of morphogenic reaction is actually yet another survival technique of plants. Without light, the seedlings die under normoxic conditions within a certain time window. In absence of oxygen, however, they survive much longer. Maybe seedlings wait for rain and wind to strip off the last layer of soil. The survival effect of the hypoxic response has also been noticed when plants lack oxygen due to flooding. In the latter case, plants appear to spend less of their energy reserves, which may enhance survival [10]. One may safely assume that without light, plants will eventually die in any air environment. However, light can also overrule the low oxygen response, by triggering degradation of the ERF VII transcription factors that are normally stabilized under hypoxic conditions. Clearly, with light, plants can produce their own oxygen, approaching the ideal state of sunshine and fresh air. REFERENCES 1. Mazzella, M.A., Casal, J.J., Muschietti, J.P., and Fox, A.R. (2014). Hormonal networks involved in apical hook development in
darkness and their response to light. Front. Plant Sci. 5, 00052. 2. Abbas, M., Berckhan, S., Rooney, D.J., Gibbs, D.J., Vicente Conde, J., Sousa Correia, C., Bassel, G.W., Marin-de la Rosa, N., Leo´n, J., Alabadı´, D., et al. (2015). Oxygen sensing coordinates photomorphogenesis to facilitate seedling survival. Curr. Biol. 25, 1483–1488. 3. Gibbs, D.J., Lee, S.C., Isa, N.M., Gramuglia, S., Fukao, T., Bassel, G.W., Correia, C.S., Corbineau, F., Theodoulou, F.L., Bailey-Serres, J., and Holdsworth, M.J. (2011).
4. Licausi, F., Kosmacz, M., Weits, D.A., Giuntoli, B., Giorgi, F.M., Voesenek, L.A., Perata, P., and van Dongen, J.T. (2011). Oxygen sensing in plants is mediated by an N-end rule pathway for protein destabilization. Nature 479, 419–422. 5. Gibbs, D.J., Md Isa, N., Movahedi, M., Lozano-Juste, J., Mendiondo, G.M., Berckhan, S., Marin-de la Rosa, N., Vicente Conde, J., Sousa Correia, C., Pearce, S.P., et al. (2014). Nitric oxide sensing in plants is mediated by proteolytic control of group VII ERF transcription factors. Mol. Cell 53, 369–379. 6. Weits, D.A., Giuntoli, B., Kosmacz, M., Parlanti, S., Hubberten, H.M., Riegler, H., Hoefgen, R., Perata, P., van Dongen, J.T., and Licausi, F. (2014). Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Nat. Commun. 5, 3425. 7. Varshavsky, A. (2011). The N-end rule pathway and regulation by proteolysis. Protein Sci. 20, 1298–1345. 8. Graciet, E., and Wellmer, F. (2010). The plant N-end rule pathway: structure and functions. Plant J. 15, 447–453. 9. Gibbs, D.J., Bacardit, J., Bachmair, A., and Holdsworth, M.J. (2014). The eukaryotic N-end rule pathway: conserved mechanisms and diverse functions. Trends Cell Biol. 24, 603–611. 10. Riber, W., Muller, J.T., Visser, E.J., Sasidharan, R., Voesenek, L.A., and Mustroph, A. (2015). The greening after extended darkness1 is an N-end rule pathway mutant with high tolerance to submergence and starvation. Plant Physiol. 167, 1616–1629.
Cell Migration: Recoiling from an Embrace Miriam A. Genuth and Orion D. Weiner* Cardiovascular Research Institute and Department of Biochemistry and Biophysics, University of California San Francisco, San Francisco, CA 94158, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2015.05.015
For proper spacing or rapid dispersion, some migratory cells are guided by repulsive collisions with their neighbors. A new study reveals that a surprising intercellular coupling of leading edge actin networks forms the basis of mutual repulsion in Drosophila hemocytes. Proper control of migration is necessary for development. In addition to using soluble cues for guidance, some cells
R566 Current Biology 25, R549–R568, June 29, 2015 ª2015 Elsevier Ltd All rights reserved
take advantage of direct interactions with their neighbors. One of the most well-characterized interactions is contact
Current Biology
Dispatches inhibition of locomotion (CIL), in which a cell ceases its forward motion upon contact with another cell [1]. This is frequently followed by repolarization and migration in the opposite direction. Interest in CIL has recently blossomed following the discovery of its necessity for multiple developmental migration events in vivo, such as neural crest and Cajal-Retzius cell migration, where it is used to ensure proper cell dispersal [2–4]. Furthermore, metastatic cell lines have been observed to exhibit homotypic CIL interactions with themselves, but not stromal cells, suggesting a possible mechanism for tumor invasiveness [5,6]. The dominant model of CIL has been focal signal-induced activation of Rho and actomyosin-based contraction at the site of cell–cell contact (Figure 1A). Here, local activation of signaling cascades, such as the planar cell polarity pathway, provides the cue and local actomyosin contractility provides the force to push each cell away from the site of contact. It is not known whether this is the mechanism that underlies CIL in all contexts. In a recent Cell paper, Davis et al. [7] report the careful observation of actin dynamics in Drosophila hemocytes and demonstrate a different mechanism at play for CIL in this setting. Drosophila hemocytes (macrophagelike cells) develop from the head mesoderm and then distribute evenly throughout the embryo under the ventral surface in a matter that is thought to depend on CIL [4]. Whereas the efficiency of CIL in other contexts does not depend on the orientation of cell collisions (front to front, front to back, front to side) [2,6], Davis et al. [7] found that hemocytes only undergo a CIL response when two active lamellipodia come into contact. Furthermore, careful tracking of hemocyte CIL in vivo showed that colliding hemocytes initially accelerate towards each other before slowing and withdrawing at two to three times the speed of retraction in freely moving cells. Live-cell microscopy of actin and adhesion reporters revealed the cytoskeletal dynamics that underlie this retraction. When lamellipodia first came into contact, an adhesion marker was rapidly recruited to the site of contact. Then there was a pronounced reduction in the rate of retrograde actin flow in a corridor immediately behind the putative
A
B
Neural crest CIL model
Hemocyte CIL model Approach
Approach
Signalling
Spring loading PCP
PCP
Force
Force
RhoA
RhoA
Simultaneous recoil
Reversal
Microfilament
Cytoskeletal tension
Myosin II
Adhesion Current Biology
Figure 1. Different modes of CIL. (A) In Xenopus neural crest, N-cadherin engagement activates planar cell polarity (PCP) signalling, leading to focal activation of RhoA and myosin II at the site of cell–cell contact and local retraction [2]. This mode of CIL can occur with any orientation of collision. (B) Drosophila hemocytes utilize the contractile forces that generate retrograde flow for retraction in CIL [7]. When cells collide, adhesions couple the cytoskeleton to the plasma membrane, leading to an increase in lamellar tension that causes the cells to recoil when the adhesions are released. This mode of CIL only occurs for lamellar-to-lamellar (i.e. front-to-front) collisions.
adhesion. A stress fiber grew from the base of the lamellipod through the corridor to the adhesion linking the two cells together, and then the two cells simultaneously recoiled following a loss of adhesion. On the basis of these observations, Davis et al. [7] propose an ‘intercellular actin clutch’ model for CIL that is very similar to the ‘molecular clutch’ thought to underlie traction stresses at focal adhesions [8]. In this model, the adhesion physically couples the actin cytoskeleton to the plasma membrane (Figure 1B). This binding inhibits the movement of microfilaments and thereby slows retrograde flow. The cytoskeletal contractile forces that were previously spent generating retrograde flow can now pull through the intercellular adhesions, producing the acceleration that is seen upon initial cell contact. The intercellular adhesions enable contractile tension to build across the cell–cell junction (i.e. loading the ‘spring’) until the adhesions fail and the stored energy in
the spring is released, resulting in recoil of the cells away from one other. This model leads to a number of testable predictions. One is that lamellar tension should be higher in cells undergoing CIL than in freely migrating cells. Another is that formation of the stress fiber connecting the cells should be crucial for CIL. To test the first prediction, the authors performed laser abscission experiments where they ablated either the leading edge of a freely migrating cell or an adhesive puncta in colliding cells and then measured the rate of recoil of the plasma membrane. The membrane recoiled at double the rate in the colliding cells, equivalent to a threefold increase in lamellar tension. Interestingly, ablations in the colliding cells, but not the freely moving cells, led not only to a local membrane retraction but also to a rearward movement of the cell body. These data suggest that there is a specific regulatory step that controls the timing of the release of adhesions.
Current Biology 25, R549–R568, June 29, 2015 ª2015 Elsevier Ltd All rights reserved R567
Current Biology
Dispatches To test the role of the stress fiber in CIL, the authors analyzed the velocity and actin dynamics of colliding cells lacking myosin II (the motor that generates the tension in the stress fiber) or a formin (Diaphanous, an actin nucleator that builds the actin network of the stress fiber). While the myosin II mutant hemocytes lacked a stress fiber and exhibited defects in CIL, they also had chronically reduced rates of retrograde flow and lamellar retraction defects when freely migrating, making it difficult to determine whether myosin II plays a specific role in CIL. The Diaphanous mutants proved more informative. These cells had normal rates of retrograde flow while freely migrating. However, when Diaphanous mutant hemocytes collided, they exhibited unpredictable changes in retrograde flow rate that were not well correlated between the cells. They often failed to form a stress fiber and did not exhibit CIL. These data suggest that proper coordination of actin dynamics and formation of a stress fiber is necessary for CIL in hemocytes. Finally, the authors analyzed the spatial distribution of the hemocytes in Diaphanous mutants and found that these cells were not as evenly dispersed as in wild-type embryos, providing a nice confirmation that CIL is necessary for the proper migration of hemocytes. While Davis et al. [7] provide a compelling overarching framework with their molecular clutch model, many interesting features remain to be explored. Almost none of the molecular players driving these complicated actin dynamics has been identified. Also, why only contact between active lamellipodia is sufficient for CIL in hemocytes has yet to be determined. One possible explanation is that the unique actin dynamics of lamellipodia are necessary for engagement of the clutch and loading of the spring. If this is the case, then the CIL responses observed in other contexts with less restrictive geometries [2,5,6] would use a different mechanism for generating repulsive forces, such as
planar cell polarity signaling to locally recruit myosin, independent of intercellular adhesions. This paper observes that the actin networks of the colliding cells become physically coupled and undergo synchronous changes but doesn’t demonstrate that this coupling is necessary for CIL. Based on the data presented a bead coated with an adhesion molecule, i.e. a substrate that can be pulled on but has no actin dynamics, might be sufficient to induce CIL. In fact, a bead coated in the Eph receptor ligand ephrin-A5 is sufficient to induce CIL in metastatic prostate cancer cells [5]. Clonal analyses of collisions between wild-type cells and cells lacking cytoskeletal regulators such as myosin II will yield valuable insight into the role of coupled actin dynamics in this process. Just as there are multiple mechanisms by which a cell can be guided by a soluble gradient, there are likely to be multiple mechanisms of CIL. The ‘intercellular actin clutch’ model provides a compelling mechanism for ensuring that both cells undergo an equivalent repulsive response. The prior models of CIL, where receptor engagement leads to activation of RhoA and local retraction, do not impose a requirement for coordinated mutual repulsion. In fact, a recent paper from Lin et al. [6] analyzed collisions of pairs of metastatic breast cancer cells and found that CIL would occur in either one cell or both cells at rates that appeared purely probabilistic, suggesting that independent decision-making does occur. A point of particular interest will be the analysis of cells such as Cajal-Retzius and neural crest cells, which consistently exhibit CIL in both partners in a collision and yet lack a requirement for leading edge contact [2,3]. These cells could either make robust, independent decisions or utilize an asymmetric actin clutch. Neural crest has an interesting additional wrinkle in that these cells are quite adhesive in vivo, so while cell contact does lead to the repolarization of protrusions away from the site of contact, it does not always lead to cell separation
R568 Current Biology 25, R549–R568, June 29, 2015 ª2015 Elsevier Ltd All rights reserved
[9]. These data suggest that there may be an additional mechanism to stabilize adhesions in these cells. Davis et al.’s [7] work has provided a provocative new model for CIL, and the tools they utilize could lead to a deeper understanding of CIL in other contexts.
REFERENCES 1. Abercrombie, M., and Heaysman, J.E.M. (1953). Observations on the social behaviour of cells in tissue culture: I. Speed of movement of chick heart fibroblasts in relation to their mutual contacts. Exp. Cell Res. 5, 111–131. 2. Carmona-Fontaine, C., Matthews, H.K., Kuriyama, S., Moreno, M., Dunn, G.A., Parsons, M., Stern, C.D., and Mayor, R. (2008). Contact inhibition of locomotion in vivo controls neural crest directional migration. Nature 456, 957–961. 3. Villar-Cervin˜o, V., Molano-Mazo´n, M., Catchpole, T., Valdeolmillos, M., Henkemeyer, M., Martı´nez, L.M., Borrell, V., and Marı´n, O. (2013). Contact repulsion controls the dispersion and final distribution of Cajal-Retzius cells. Neuron 77, 457–471. 4. Davis, J.R., Huang, C.-Y., Zanet, J., Harrison, S., Rosten, E., Cox, S., Soong, D.Y., Dunn, G.A., and Stramer, B.M. (2012). Emergence of embryonic pattern through contact inhibition of locomotion. Development 139, 4555–4560. 5. Astin, J.W., Batson, J., Kadir, S., Charlet, J., Persad, R.A., Gillatt, D., Oxley, J.D., and Nobes, C.D. (2010). Competition amongst Eph receptors regulates contact inhibition of locomotion and invasiveness in prostate cancer cells. Nat. Cell Biol. 12, 1194–1204. 6. Lin, B., Yin, T., Wu, Y.I., Inoue, T., and Levchenko, A. (2015). Interplay between chemotaxis and contact inhibition of locomotion determines exploratory cell migration. Nat. Commun. 6, 6619. 7. Davis, J.R., Luchici, A., Mosis, F., Thackery, J., Salazar, J.A., Mao, Y., Dunn, G.A., Betz, T., Miodownik, M., and Stramer, B.M. (2015). Inter-cellular forces orchestrate contact inhibition of locomotion. Cell 161, 361–373. 8. Gardel, M.L., Sabass, B., Ji, L., Danuser, G., Schwarz, U.S., and Waterman, C.M. (2008). Traction stress in focal adhesions correlates biphasically with actin retrograde flow speed. J. Cell Biol. 183, 999–1005. 9. Theveneau, E., Marchant, L., Kuriyama, S., Gull, M., Moepps, B., Parsons, M., and Mayor, R. (2010). Collective chemotaxis requires contact-dependent cell polarity. Dev. Cell 19, 39–53.
Dispatches A
Homeostatic response to hypoxia is regulated by the N-end rule pathway in plants. Nature 479, 415–418.
B Light
Asymmetric distribution of plant growth hormones
ERF VII transcription factors
Apical hook
Low oxygen
Proteolytic destruction of ERF VIIs Oxygen, light Current Biology
Figure 1. Early seedling development is guided by light and oxygen. (A) Skotomorphogenesis versus photomorphogenesis. In the absence of light, plants develop long hypocotyls and an apical hook, bending down their yellow cotyledons and protecting the stem-cell like apical meristem. The apical hook is maintained longer in a low oxygen environment. In the light, cotyledons develop functional chloroplasts and hypocotyls stay short, but roots grow longer to provide minerals for further growth. Photo by Annkatrin Rose. (B) Light acts as a master regulator of morphogenesis. Light prevents the asymmetric distribution of auxin and other growth hormones in the apical hook region. In absence of oxygen, ERF VIIs accumulate and stabilize the apical hook. Light can overrule the effect of low oxygen on ERF VII degradation, presumably by activating a different ERF VII degradation pathway.
access to fresh air will eventually allow for a low level of light, providing energy for a slow transition to photomorphogenic development. In a low oxygen atmosphere, hook opening does not happen. Surprisingly, this apparent lack of morphogenic reaction is actually yet another survival technique of plants. Without light, the seedlings die under normoxic conditions within a certain time window. In absence of oxygen, however, they survive much longer. Maybe seedlings wait for rain and wind to strip off the last layer of soil. The survival effect of the hypoxic response has also been noticed when plants lack oxygen due to flooding. In the latter case, plants appear to spend less of their energy reserves, which may enhance survival [10]. One may safely assume that without light, plants will eventually die in any air environment. However, light can also overrule the low oxygen response, by triggering degradation of the ERF VII transcription factors that are normally stabilized under hypoxic conditions. Clearly, with light, plants can produce their own oxygen, approaching the ideal state of sunshine and fresh air. REFERENCES 1. Mazzella, M.A., Casal, J.J., Muschietti, J.P., and Fox, A.R. (2014). Hormonal networks involved in apical hook development in
darkness and their response to light. Front. Plant Sci. 5, 00052. 2. Abbas, M., Berckhan, S., Rooney, D.J., Gibbs, D.J., Vicente Conde, J., Sousa Correia, C., Bassel, G.W., Marin-de la Rosa, N., Leo´n, J., Alabadı´, D., et al. (2015). Oxygen sensing coordinates photomorphogenesis to facilitate seedling survival. Curr. Biol. 25, 1483–1488. 3. Gibbs, D.J., Lee, S.C., Isa, N.M., Gramuglia, S., Fukao, T., Bassel, G.W., Correia, C.S., Corbineau, F., Theodoulou, F.L., Bailey-Serres, J., and Holdsworth, M.J. (2011).
4. Licausi, F., Kosmacz, M., Weits, D.A., Giuntoli, B., Giorgi, F.M., Voesenek, L.A., Perata, P., and van Dongen, J.T. (2011). Oxygen sensing in plants is mediated by an N-end rule pathway for protein destabilization. Nature 479, 419–422. 5. Gibbs, D.J., Md Isa, N., Movahedi, M., Lozano-Juste, J., Mendiondo, G.M., Berckhan, S., Marin-de la Rosa, N., Vicente Conde, J., Sousa Correia, C., Pearce, S.P., et al. (2014). Nitric oxide sensing in plants is mediated by proteolytic control of group VII ERF transcription factors. Mol. Cell 53, 369–379. 6. Weits, D.A., Giuntoli, B., Kosmacz, M., Parlanti, S., Hubberten, H.M., Riegler, H., Hoefgen, R., Perata, P., van Dongen, J.T., and Licausi, F. (2014). Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Nat. Commun. 5, 3425. 7. Varshavsky, A. (2011). The N-end rule pathway and regulation by proteolysis. Protein Sci. 20, 1298–1345. 8. Graciet, E., and Wellmer, F. (2010). The plant N-end rule pathway: structure and functions. Plant J. 15, 447–453. 9. Gibbs, D.J., Bacardit, J., Bachmair, A., and Holdsworth, M.J. (2014). The eukaryotic N-end rule pathway: conserved mechanisms and diverse functions. Trends Cell Biol. 24, 603–611. 10. Riber, W., Muller, J.T., Visser, E.J., Sasidharan, R., Voesenek, L.A., and Mustroph, A. (2015). The greening after extended darkness1 is an N-end rule pathway mutant with high tolerance to submergence and starvation. Plant Physiol. 167, 1616–1629.
Cell Migration: Recoiling from an Embrace Miriam A. Genuth and Orion D. Weiner* Cardiovascular Research Institute and Department of Biochemistry and Biophysics, University of California San Francisco, San Francisco, CA 94158, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2015.05.015
For proper spacing or rapid dispersion, some migratory cells are guided by repulsive collisions with their neighbors. A new study reveals that a surprising intercellular coupling of leading edge actin networks forms the basis of mutual repulsion in Drosophila hemocytes. Proper control of migration is necessary for development. In addition to using soluble cues for guidance, some cells
R566 Current Biology 25, R549–R568, June 29, 2015 ª2015 Elsevier Ltd All rights reserved
take advantage of direct interactions with their neighbors. One of the most well-characterized interactions is contact
Current Biology
Dispatches inhibition of locomotion (CIL), in which a cell ceases its forward motion upon contact with another cell [1]. This is frequently followed by repolarization and migration in the opposite direction. Interest in CIL has recently blossomed following the discovery of its necessity for multiple developmental migration events in vivo, such as neural crest and Cajal-Retzius cell migration, where it is used to ensure proper cell dispersal [2–4]. Furthermore, metastatic cell lines have been observed to exhibit homotypic CIL interactions with themselves, but not stromal cells, suggesting a possible mechanism for tumor invasiveness [5,6]. The dominant model of CIL has been focal signal-induced activation of Rho and actomyosin-based contraction at the site of cell–cell contact (Figure 1A). Here, local activation of signaling cascades, such as the planar cell polarity pathway, provides the cue and local actomyosin contractility provides the force to push each cell away from the site of contact. It is not known whether this is the mechanism that underlies CIL in all contexts. In a recent Cell paper, Davis et al. [7] report the careful observation of actin dynamics in Drosophila hemocytes and demonstrate a different mechanism at play for CIL in this setting. Drosophila hemocytes (macrophagelike cells) develop from the head mesoderm and then distribute evenly throughout the embryo under the ventral surface in a matter that is thought to depend on CIL [4]. Whereas the efficiency of CIL in other contexts does not depend on the orientation of cell collisions (front to front, front to back, front to side) [2,6], Davis et al. [7] found that hemocytes only undergo a CIL response when two active lamellipodia come into contact. Furthermore, careful tracking of hemocyte CIL in vivo showed that colliding hemocytes initially accelerate towards each other before slowing and withdrawing at two to three times the speed of retraction in freely moving cells. Live-cell microscopy of actin and adhesion reporters revealed the cytoskeletal dynamics that underlie this retraction. When lamellipodia first came into contact, an adhesion marker was rapidly recruited to the site of contact. Then there was a pronounced reduction in the rate of retrograde actin flow in a corridor immediately behind the putative
A
B
Neural crest CIL model
Hemocyte CIL model Approach
Approach
Signalling
Spring loading PCP
PCP
Force
Force
RhoA
RhoA
Simultaneous recoil
Reversal
Microfilament
Cytoskeletal tension
Myosin II
Adhesion Current Biology
Figure 1. Different modes of CIL. (A) In Xenopus neural crest, N-cadherin engagement activates planar cell polarity (PCP) signalling, leading to focal activation of RhoA and myosin II at the site of cell–cell contact and local retraction [2]. This mode of CIL can occur with any orientation of collision. (B) Drosophila hemocytes utilize the contractile forces that generate retrograde flow for retraction in CIL [7]. When cells collide, adhesions couple the cytoskeleton to the plasma membrane, leading to an increase in lamellar tension that causes the cells to recoil when the adhesions are released. This mode of CIL only occurs for lamellar-to-lamellar (i.e. front-to-front) collisions.
adhesion. A stress fiber grew from the base of the lamellipod through the corridor to the adhesion linking the two cells together, and then the two cells simultaneously recoiled following a loss of adhesion. On the basis of these observations, Davis et al. [7] propose an ‘intercellular actin clutch’ model for CIL that is very similar to the ‘molecular clutch’ thought to underlie traction stresses at focal adhesions [8]. In this model, the adhesion physically couples the actin cytoskeleton to the plasma membrane (Figure 1B). This binding inhibits the movement of microfilaments and thereby slows retrograde flow. The cytoskeletal contractile forces that were previously spent generating retrograde flow can now pull through the intercellular adhesions, producing the acceleration that is seen upon initial cell contact. The intercellular adhesions enable contractile tension to build across the cell–cell junction (i.e. loading the ‘spring’) until the adhesions fail and the stored energy in
the spring is released, resulting in recoil of the cells away from one other. This model leads to a number of testable predictions. One is that lamellar tension should be higher in cells undergoing CIL than in freely migrating cells. Another is that formation of the stress fiber connecting the cells should be crucial for CIL. To test the first prediction, the authors performed laser abscission experiments where they ablated either the leading edge of a freely migrating cell or an adhesive puncta in colliding cells and then measured the rate of recoil of the plasma membrane. The membrane recoiled at double the rate in the colliding cells, equivalent to a threefold increase in lamellar tension. Interestingly, ablations in the colliding cells, but not the freely moving cells, led not only to a local membrane retraction but also to a rearward movement of the cell body. These data suggest that there is a specific regulatory step that controls the timing of the release of adhesions.
Current Biology 25, R549–R568, June 29, 2015 ª2015 Elsevier Ltd All rights reserved R567
Current Biology
Dispatches To test the role of the stress fiber in CIL, the authors analyzed the velocity and actin dynamics of colliding cells lacking myosin II (the motor that generates the tension in the stress fiber) or a formin (Diaphanous, an actin nucleator that builds the actin network of the stress fiber). While the myosin II mutant hemocytes lacked a stress fiber and exhibited defects in CIL, they also had chronically reduced rates of retrograde flow and lamellar retraction defects when freely migrating, making it difficult to determine whether myosin II plays a specific role in CIL. The Diaphanous mutants proved more informative. These cells had normal rates of retrograde flow while freely migrating. However, when Diaphanous mutant hemocytes collided, they exhibited unpredictable changes in retrograde flow rate that were not well correlated between the cells. They often failed to form a stress fiber and did not exhibit CIL. These data suggest that proper coordination of actin dynamics and formation of a stress fiber is necessary for CIL in hemocytes. Finally, the authors analyzed the spatial distribution of the hemocytes in Diaphanous mutants and found that these cells were not as evenly dispersed as in wild-type embryos, providing a nice confirmation that CIL is necessary for the proper migration of hemocytes. While Davis et al. [7] provide a compelling overarching framework with their molecular clutch model, many interesting features remain to be explored. Almost none of the molecular players driving these complicated actin dynamics has been identified. Also, why only contact between active lamellipodia is sufficient for CIL in hemocytes has yet to be determined. One possible explanation is that the unique actin dynamics of lamellipodia are necessary for engagement of the clutch and loading of the spring. If this is the case, then the CIL responses observed in other contexts with less restrictive geometries [2,5,6] would use a different mechanism for generating repulsive forces, such as
planar cell polarity signaling to locally recruit myosin, independent of intercellular adhesions. This paper observes that the actin networks of the colliding cells become physically coupled and undergo synchronous changes but doesn’t demonstrate that this coupling is necessary for CIL. Based on the data presented a bead coated with an adhesion molecule, i.e. a substrate that can be pulled on but has no actin dynamics, might be sufficient to induce CIL. In fact, a bead coated in the Eph receptor ligand ephrin-A5 is sufficient to induce CIL in metastatic prostate cancer cells [5]. Clonal analyses of collisions between wild-type cells and cells lacking cytoskeletal regulators such as myosin II will yield valuable insight into the role of coupled actin dynamics in this process. Just as there are multiple mechanisms by which a cell can be guided by a soluble gradient, there are likely to be multiple mechanisms of CIL. The ‘intercellular actin clutch’ model provides a compelling mechanism for ensuring that both cells undergo an equivalent repulsive response. The prior models of CIL, where receptor engagement leads to activation of RhoA and local retraction, do not impose a requirement for coordinated mutual repulsion. In fact, a recent paper from Lin et al. [6] analyzed collisions of pairs of metastatic breast cancer cells and found that CIL would occur in either one cell or both cells at rates that appeared purely probabilistic, suggesting that independent decision-making does occur. A point of particular interest will be the analysis of cells such as Cajal-Retzius and neural crest cells, which consistently exhibit CIL in both partners in a collision and yet lack a requirement for leading edge contact [2,3]. These cells could either make robust, independent decisions or utilize an asymmetric actin clutch. Neural crest has an interesting additional wrinkle in that these cells are quite adhesive in vivo, so while cell contact does lead to the repolarization of protrusions away from the site of contact, it does not always lead to cell separation
R568 Current Biology 25, R549–R568, June 29, 2015 ª2015 Elsevier Ltd All rights reserved
[9]. These data suggest that there may be an additional mechanism to stabilize adhesions in these cells. Davis et al.’s [7] work has provided a provocative new model for CIL, and the tools they utilize could lead to a deeper understanding of CIL in other contexts.
REFERENCES 1. Abercrombie, M., and Heaysman, J.E.M. (1953). Observations on the social behaviour of cells in tissue culture: I. Speed of movement of chick heart fibroblasts in relation to their mutual contacts. Exp. Cell Res. 5, 111–131. 2. Carmona-Fontaine, C., Matthews, H.K., Kuriyama, S., Moreno, M., Dunn, G.A., Parsons, M., Stern, C.D., and Mayor, R. (2008). Contact inhibition of locomotion in vivo controls neural crest directional migration. Nature 456, 957–961. 3. Villar-Cervin˜o, V., Molano-Mazo´n, M., Catchpole, T., Valdeolmillos, M., Henkemeyer, M., Martı´nez, L.M., Borrell, V., and Marı´n, O. (2013). Contact repulsion controls the dispersion and final distribution of Cajal-Retzius cells. Neuron 77, 457–471. 4. Davis, J.R., Huang, C.-Y., Zanet, J., Harrison, S., Rosten, E., Cox, S., Soong, D.Y., Dunn, G.A., and Stramer, B.M. (2012). Emergence of embryonic pattern through contact inhibition of locomotion. Development 139, 4555–4560. 5. Astin, J.W., Batson, J., Kadir, S., Charlet, J., Persad, R.A., Gillatt, D., Oxley, J.D., and Nobes, C.D. (2010). Competition amongst Eph receptors regulates contact inhibition of locomotion and invasiveness in prostate cancer cells. Nat. Cell Biol. 12, 1194–1204. 6. Lin, B., Yin, T., Wu, Y.I., Inoue, T., and Levchenko, A. (2015). Interplay between chemotaxis and contact inhibition of locomotion determines exploratory cell migration. Nat. Commun. 6, 6619. 7. Davis, J.R., Luchici, A., Mosis, F., Thackery, J., Salazar, J.A., Mao, Y., Dunn, G.A., Betz, T., Miodownik, M., and Stramer, B.M. (2015). Inter-cellular forces orchestrate contact inhibition of locomotion. Cell 161, 361–373. 8. Gardel, M.L., Sabass, B., Ji, L., Danuser, G., Schwarz, U.S., and Waterman, C.M. (2008). Traction stress in focal adhesions correlates biphasically with actin retrograde flow speed. J. Cell Biol. 183, 999–1005. 9. Theveneau, E., Marchant, L., Kuriyama, S., Gull, M., Moepps, B., Parsons, M., and Mayor, R. (2010). Collective chemotaxis requires contact-dependent cell polarity. Dev. Cell 19, 39–53.
