Current Biology
Dispatches 8. Nelson, X.J., and Jackson, R.R. (2009). Aggressive use of Batesian mimicry by an antlike jumping spider. Biol. Lett. 5, 755–757.
12. Nelson, X.J., and Jackson, R.R. (2006). Visionbased innate aversion to ants and ant mimics. Behav. Ecol. 17, 676–681.
9. Peka´r, S., and Jiros, P. (2011). Do ant mimics imitate cuticular hydrocarbons of their models? Anim. Behav. 82, 1193– 1199.
13. Kitamura, T., and Imafuku, M. (2015). Behavioural mimicry in flight path of Batesian intraspecific polymorphic butterfly Papilio polytes. Proc. R. Soc. B 282, 20150483.
10. Durkee, C.A., Weiss, M.R., and Uma, D.B. (2011). Ant mimicry lessens predation on a north american jumping spider by larger salticid spiders. Environ. Entomol. 40, 1223– 1231.
14. Peka´r, S., Jarab, M., Fromhage, L., and Herberstein, M.E. (2011). Is the evolution of inaccurate mimicry a result of selection by a suite of predators? A case study using myrmecomorphic spiders. Am. Nat. 178, 124–134.
11. Ramesh, A., Vijayan, S., Sreedharan, S., Somanathan, H., and Uma, D. (2016). Similar yet different: differential response of a praying mantis to ant-mimicking spiders. Biol. J. Linn. Soc. 119, 158–165.
15. Wang, M.-Y., Vasas, V., Chittka, L., and Yen, S.-H. (2017). Sheep in wolf’s clothing: multicomponent traits enhance the success of mimicry in spider-mimicking moths. Anim. Behav. 127, 219–224.
16. Uma, D., Durkee, C., Herzner, G., and Weiss, M. (2013). Double deception: ant-mimicking spiders elude both visually- and chemicallyoriented predators. PLoS ONE 8, e79660. 17. Nelson, X.J., Li, D., and Jackson, R.R. (2006). Out of the frying pan and into the fire: a novel trade-off for Batesian mimics. Ethology 112, 270–277. 18. Valkonen, J.K., and Mappes, J. (2014). Resembling a viper: implications of mimicry for conservation of the endangered smooth snake. Conserv. Biol. 28, 1568–1574. 19. Nelson, X.J., and Jackson, R.R. (2006). Compound mimicry and trading predators by the males of sexually dimorphic Batesian mimics. Proc. R. Soc. B Biol. Sci. 273, 367–372.
Insect Vision: A Neuron that Anticipates an Object’s Path Mark A. Frye Department of Integrative Biology and Physiology, University of California Los Angeles, Los Angeles, CA 90095, USA Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2017.08.049
Dragonflies are superb aerial predators, plucking tiny insect prey from the sky. This ability depends on a visual system that has fascinated scientists for decades, and now one of its visual-target-detecting neurons has been shown to anticipate the image path of prey. I am an avid baseball fan, in part because the game is very hard. Consider that a good pitcher can hurl the ball to home plate in about 420 milliseconds. Minimally processing the visual image, sending a motor command, and swinging the bat takes 300 milliseconds. That’s the blink of an eye, and does not even take into account the time required to evaluate the pitch and to decide whether and how to swing. It is no wonder that failing 70% of the time nevertheless puts a player in the Hall of Fame. The batter must integrate a mere 120 milliseconds worth of visual information about the ball’s trajectory to predict its future path and plan an action. We do not have a clear understanding of how such a prediction is made in any animal visual system, but dragonflies provide a tractable system for studying the comparable feat of visually directed prey capture. A new study by Weiderman et al. [1] has revealed a remarkable neuron
in the dragonfly optic lobe that changes its response properties to account for the future path of a prey-like target. Dragonflies make their living preying upon airborne insect fastballs. Circling an alpine pond on a sunny day, a dragonfly careening into a cloud of swarming insects requires only about 45 milliseconds to perceive and react to a change in the path of a single prey — which sounds, and indeed is, pretty fast! But the animal can cut down that delay to as little as 4 milliseconds by considering the most likely path of its ensuing meal, using what is called an ‘internal model’ [2]. Internal models of our own body dynamics or the dynamics of an arcing baseball make commonplace activities such as catching a deep outfield fly ball possible. But how are internal predictions about movement in the world encoded in the brain? Does a predatory dragonfly rely on neuronal circuits that are finely tuned
R1076 Current Biology 27, R1060–R1080, October 9, 2017 ª 2017 Elsevier Ltd.
by the experience of catching prey? Or can prediction be manifest within a single brain cell hardwired at birth? Dragonfly target vision is served by remarkably sophisticated visual processes. The optics of the dragonfly’s large compound eye give a visual acuity that is high for an insect, rivaling that of a mammal [3]. The high acuity eye supplies information to neurons residing deep in the dragonfly optic lobes that are only excited by the movement of contrasting objects considerably smaller than the solid angle that would fill one optical pixel, and this occurs even when the contrasting object is viewed against a cluttered moving background [4]. One of these neurons, called the ‘contrifugal small target movement detector number one’ (CSTMD1), has been shown to selectively track a single moving target while ignoring others [5], similar to the way its flying host snatches only one from a
Current Biology
Dispatches cloud of gnats swarming around the edge of a pond. One peculiar property of CSTMD1 is that its spiking frequency increases nonlinearly over the length of long trajectories of object motion across the eye [6,7]. The further the object moves, the higher the firing rate climbs, but the process is rather slow, operating over many hundreds of milliseconds — one or two orders of magnitude longer than the reactionary turns that these animals produce when chasing prey. The function of this slow spatial facilitation phenomenon was not known, but cannot reasonably be coupled directly to the pre-motor control of rapid reactionary turns during real-time pursuit. Rather, Weiderman et al. [1] were intrigued by the possibility that spatial facilitation was encoding the future path of the object based on its current trajectory — thereby producing a sort of internal model of the prey’s path. Weiderman et al. [1] recorded the activity of CSTMD1 using a sharp micropipette electrode in order to make fine-scale measurements of membrane dynamics and action potentials. The dragonfly subject viewed a high performance computer monitor that presented stimuli consisting of a tiny black dot moving along a long path, with the path broken into two variable segments called the ‘primer’ and the ‘probe’ (Figure 1). The primer segment was meant to induce spatial facilitation, in essence to prime CSTMD1 with object movement outside of the cell’s receptive field boundary. The authors systematically varied the path and speed of the primer. The second stimulus segment, the probe, was presented directly within the cell’s receptive field. Weiderman et al. [1] found that the primer induced a spiking response to the probe that was much stronger than that for the probe alone. More generally, by offsetting the primer and probe on different linear paths, or by spacing them on the same path interrupted by different time delays, the authors showed that CSTMD1 develops a sort of ‘hot spot’ of increased sensitivity directly in the path of the object (Figure 1, top). The hot spot is focused in front of the object’s immediate linear path, so it can only be activated if the object continues along its
current trajectory into the future. In essence, CSTMD1 receives information about an object’s path while the object is moving outside of the cell’s own receptive field, and somehow assembles an internal model of object dynamics so that the cell’s spiking response is heightened if the object continues on the ‘predicted’ path. The authors next explored additional properties of the predictive hot spot, and demonstrated that the cell shows increased sensitivity to the contrast of the object (Figure 1, middle). They found that not only does the cell’s increased sensitivity to firing action potentials predict the path of the object, but the same process also increases the object’s visual salience or ‘detectability’ against the visual surroundings. One notable property of CSTMD’s receptive field is that it is not selective for any particular direction of object motion — it encodes orthogonal directions equally well. By systematically varying the angular offset between the primer path and the probe path, the authors found that the direction of the primer bestows greater sensitivity to the probe in the same direction (Figure 1, bottom): the primer makes the receptive field strongly directionally tuned. The range of possible directional ‘learning’ is not meager, but rather can operate over a span of at least 90 degrees: CSTMD1 can thus be primed to become maximally sensitive to either upward or perpendicular rightward object paths. The response properties of the CSTMD1 neuron of the dragonfly optic lobe are highly sophisticated by comparison to visual feature detectors found in other animals. Like CSTMD1, neurons of the feline cortex and the fly lobula show strong selectivity for small objects, and are inhibited by motion of the visual panorama, but neither show spatial facilitation outside the receptive field [8,9]. Neurons of the mouse retina and fly lobula plate are strongly directionally selective (for review see [10]), but neither show directional plasticity like CSTMD1. To my knowledge, no other cell class or single identified neuron has been shown to have rapidly adapting properties so well suited to predict the future path of an object. It will be interesting to see how this research impacts our understanding
Predictive ‘hot spot’
Primer
Probe
Receptive field Computer display Contrast enhancement
Directional tuning
Current Biology
Figure 1. Dynamic tuning properties of the CSTMD1 neuron. Small contrasting objects were presented on a computer display in front of the dragonfly. Top: target motion outside the receptive field (dashed line) is called a ‘primer’ and target motion inside the receptive field is called the ‘probe’. After the primer, spiking responses of the probe show a strong ‘hot spot’ (red) of increased object sensitivity directly in the path of the probe, flanked by inhibitory surround regions (blue). Middle: a primer causes the probe response to be more sensitive to object contrast, increasing the apparent saliency of a low contrast object. Bottom: a primer moving along one linear path increases the directional tuning of the probe response along the same direction.
of object motion detection in other model systems — the field is expanding rapidly, particularly in the fly and mouse model systems. But neither of these animals has evolved to pursue small targets. We may need to first look at other insect aerial predators (such as the killer fly [11]) for a neuron that is as highly specialized as CSTMD1 for selecting and predicting the path of airborne ‘fastballs’.
Current Biology 27, R1060–R1080, October 9, 2017 R1077
Current Biology
Dispatches REFERENCES
4. Nordstro¨m, K., Barnett, P.D., and O’Carroll, D.C. (2006). Insect detection of small targets in moving visual clutter. PLoS Biol. 4, e54.
8. Kato, H., Bishop, P.O., and Orban, G.A. (1978). Hypercomplex and simple/complex cell classifications in cat striate cortex. J. Neurophysiol. 41, 1071–1095.
1. Wiederman, S.D., Fabian, J.M., Dunbier, J.R., and O’Carroll, D.C. (2017). A predictive focus of gain modulation encodes target trajectories in insect vision. eLife 6, e26478.
5. Wiederman, S.D., and O’Carroll, D.C. (2013). Selective attention by an insect visual neuron. Curr. Biol. 23, 156–161.
2. Mischiati, M., Lin, H.-T., Herold, P., Imier, E., Olberg, R., and Leonardo, A. (2015). Internal models direct dragonfly interception steering. Nature 517, 333–338.
6. Nordstro¨m, K., Bolzon, D.M., and O’Carroll, D.C. (2011). Spatial facilitation by a highperformance dragonfly target-detecting neuron. Biol. Lett. 7, 588–591.
10. Mauss, A.S., Vlasits, A., Borst, A., and Feller, M. (2017). Visual circuits for direction selectivity. Annu. Rev. Neurosci. 40, 211–230.
3. Harland, D.P. (2000). Optical cues and visionbased discrimination mechanisms underlying predatory versatility in jumping spiders. Doctoral thesis, Zoology (University of Canterbury).
7. Dunbier, J.R., Wiederman, S.D., Shoemaker, P.A., and O’Carroll, D.C. (2012). Facilitation of dragonfly target-detecting neurons by slow moving features on continuous paths. Front. Neural Circuits 6, 79.
11. Wardill, T.J., Knowles, K., Barlow, L., Tapia, G., Nordstro¨m, K., Olberg, R.M., and GonzalezBellido, P.T. (2015). The killer fly hunger games: target size and speed predict decision to pursuit. Brain Behav. Evol. 86, 28–37.
9. Keles, M.F., and Frye, M.A. (2017). Object detecting neurons in Drosophila. Curr. Biol. 27, 680–687.
Cell Biology: Capturing Formin’s Mechano-Inhibition Dimitrios Vavylonis* and Brandon G. Horan Department of Physics, Lehigh University, Bethlehem, PA 18015, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2017.08.020
Formins polymerize actin filaments for the cytokinetic contractile ring. Using in vitro reconstitution of fission yeast contractile ring precursor nodes containing formins and myosin, a new study shows that formin-mediated polymerization is strongly inhibited upon the capture and pulling of actin filaments by myosin, a result that has broad implications for cellular mechanosensing. Fundamental cellular processes such as cell motility and cell division rely on the ability of the actin cytoskeleton to generate and respond to mechanical forces. Polymerization of networks or bundles of actin filaments close to the cell membrane can result in a local cell protrusion or contraction when newly generated actin filaments are pulled by myosin motor proteins. Local activation of actin polymerization is thus tightly regulated by cells. Formin proteins are key actin filament regulators that generate actin filaments for the actomyosin contractile ring during cytokinesis and also have important roles in a host of cellular processes as well as being specific targets of pathogenic bacteria [1–3]. Research in fission yeast has shown that the formin Cdc12 and the type II myosin Myo2 localize in a broad band of membrane-associated nodes during early stages of cytokinesis [4]. Actin filaments nucleated by Cdc12 are pulled by Myo2, generating the force required
to pull the nodes together into a contractile ring. A ‘search, capture, pull and release’ (SCPR) mechanism [5] was proposed in a computational model of this process; however, this behavior has not been tested in an experiment with controlled conditions in vitro, leaving many mechanistic questions unanswered. The first minimal reconstitution of ‘search, capture and pull’ has now been achieved in a new study by Zimmermann et al. [6] (Figure 1A). In addition to establishing basic biophysical properties of the process, the authors discovered that Cdc12-mediated actin polymerization is drastically reduced when Myo2 pulls on the actin filament. This novel mechanoregulatory mechanism has important implications in broad areas of cell mechanosensing and contractility. Formins are multi-domain proteins that form dimers to nucleate actin filaments. They remain associated with the barbed end of the actin filament through their formin homology (FH) 2 domain that
R1078 Current Biology 27, R1060–R1080, October 9, 2017 ª 2017 Elsevier Ltd.
wraps around the filament and rotates around its helix as the filament elongates [1,7] (Figure 1). Predicted flexible FH1 domains extend from each FH2 domain and link at their amino terminus, which is frequently associated with proteins on the cell membrane. FH1 domains characteristically contain proline-rich regions that bind the protein profilin, which itself is bound to a large fraction of actin monomers in cells. Kinetic modeling and physical arguments suggest that the FH1 domain captures and directly transfers profilin–actin to the barbed end of the actin filament, speeding up polymerization by severalfold in a profilinconcentration-dependent manner [8]. Experiments with various formin constructs have supported key features of the transfer mechanism, such as a relationship between the proximity of an FH1 proline-rich sequence to the FH2 domain and its contribution to actin polymerization rate [9–11]. The proposed flexible nature of the FH1 domain that is required for direct transfer
Dispatches 8. Nelson, X.J., and Jackson, R.R. (2009). Aggressive use of Batesian mimicry by an antlike jumping spider. Biol. Lett. 5, 755–757.
12. Nelson, X.J., and Jackson, R.R. (2006). Visionbased innate aversion to ants and ant mimics. Behav. Ecol. 17, 676–681.
9. Peka´r, S., and Jiros, P. (2011). Do ant mimics imitate cuticular hydrocarbons of their models? Anim. Behav. 82, 1193– 1199.
13. Kitamura, T., and Imafuku, M. (2015). Behavioural mimicry in flight path of Batesian intraspecific polymorphic butterfly Papilio polytes. Proc. R. Soc. B 282, 20150483.
10. Durkee, C.A., Weiss, M.R., and Uma, D.B. (2011). Ant mimicry lessens predation on a north american jumping spider by larger salticid spiders. Environ. Entomol. 40, 1223– 1231.
14. Peka´r, S., Jarab, M., Fromhage, L., and Herberstein, M.E. (2011). Is the evolution of inaccurate mimicry a result of selection by a suite of predators? A case study using myrmecomorphic spiders. Am. Nat. 178, 124–134.
11. Ramesh, A., Vijayan, S., Sreedharan, S., Somanathan, H., and Uma, D. (2016). Similar yet different: differential response of a praying mantis to ant-mimicking spiders. Biol. J. Linn. Soc. 119, 158–165.
15. Wang, M.-Y., Vasas, V., Chittka, L., and Yen, S.-H. (2017). Sheep in wolf’s clothing: multicomponent traits enhance the success of mimicry in spider-mimicking moths. Anim. Behav. 127, 219–224.
16. Uma, D., Durkee, C., Herzner, G., and Weiss, M. (2013). Double deception: ant-mimicking spiders elude both visually- and chemicallyoriented predators. PLoS ONE 8, e79660. 17. Nelson, X.J., Li, D., and Jackson, R.R. (2006). Out of the frying pan and into the fire: a novel trade-off for Batesian mimics. Ethology 112, 270–277. 18. Valkonen, J.K., and Mappes, J. (2014). Resembling a viper: implications of mimicry for conservation of the endangered smooth snake. Conserv. Biol. 28, 1568–1574. 19. Nelson, X.J., and Jackson, R.R. (2006). Compound mimicry and trading predators by the males of sexually dimorphic Batesian mimics. Proc. R. Soc. B Biol. Sci. 273, 367–372.
Insect Vision: A Neuron that Anticipates an Object’s Path Mark A. Frye Department of Integrative Biology and Physiology, University of California Los Angeles, Los Angeles, CA 90095, USA Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2017.08.049
Dragonflies are superb aerial predators, plucking tiny insect prey from the sky. This ability depends on a visual system that has fascinated scientists for decades, and now one of its visual-target-detecting neurons has been shown to anticipate the image path of prey. I am an avid baseball fan, in part because the game is very hard. Consider that a good pitcher can hurl the ball to home plate in about 420 milliseconds. Minimally processing the visual image, sending a motor command, and swinging the bat takes 300 milliseconds. That’s the blink of an eye, and does not even take into account the time required to evaluate the pitch and to decide whether and how to swing. It is no wonder that failing 70% of the time nevertheless puts a player in the Hall of Fame. The batter must integrate a mere 120 milliseconds worth of visual information about the ball’s trajectory to predict its future path and plan an action. We do not have a clear understanding of how such a prediction is made in any animal visual system, but dragonflies provide a tractable system for studying the comparable feat of visually directed prey capture. A new study by Weiderman et al. [1] has revealed a remarkable neuron
in the dragonfly optic lobe that changes its response properties to account for the future path of a prey-like target. Dragonflies make their living preying upon airborne insect fastballs. Circling an alpine pond on a sunny day, a dragonfly careening into a cloud of swarming insects requires only about 45 milliseconds to perceive and react to a change in the path of a single prey — which sounds, and indeed is, pretty fast! But the animal can cut down that delay to as little as 4 milliseconds by considering the most likely path of its ensuing meal, using what is called an ‘internal model’ [2]. Internal models of our own body dynamics or the dynamics of an arcing baseball make commonplace activities such as catching a deep outfield fly ball possible. But how are internal predictions about movement in the world encoded in the brain? Does a predatory dragonfly rely on neuronal circuits that are finely tuned
R1076 Current Biology 27, R1060–R1080, October 9, 2017 ª 2017 Elsevier Ltd.
by the experience of catching prey? Or can prediction be manifest within a single brain cell hardwired at birth? Dragonfly target vision is served by remarkably sophisticated visual processes. The optics of the dragonfly’s large compound eye give a visual acuity that is high for an insect, rivaling that of a mammal [3]. The high acuity eye supplies information to neurons residing deep in the dragonfly optic lobes that are only excited by the movement of contrasting objects considerably smaller than the solid angle that would fill one optical pixel, and this occurs even when the contrasting object is viewed against a cluttered moving background [4]. One of these neurons, called the ‘contrifugal small target movement detector number one’ (CSTMD1), has been shown to selectively track a single moving target while ignoring others [5], similar to the way its flying host snatches only one from a
Current Biology
Dispatches cloud of gnats swarming around the edge of a pond. One peculiar property of CSTMD1 is that its spiking frequency increases nonlinearly over the length of long trajectories of object motion across the eye [6,7]. The further the object moves, the higher the firing rate climbs, but the process is rather slow, operating over many hundreds of milliseconds — one or two orders of magnitude longer than the reactionary turns that these animals produce when chasing prey. The function of this slow spatial facilitation phenomenon was not known, but cannot reasonably be coupled directly to the pre-motor control of rapid reactionary turns during real-time pursuit. Rather, Weiderman et al. [1] were intrigued by the possibility that spatial facilitation was encoding the future path of the object based on its current trajectory — thereby producing a sort of internal model of the prey’s path. Weiderman et al. [1] recorded the activity of CSTMD1 using a sharp micropipette electrode in order to make fine-scale measurements of membrane dynamics and action potentials. The dragonfly subject viewed a high performance computer monitor that presented stimuli consisting of a tiny black dot moving along a long path, with the path broken into two variable segments called the ‘primer’ and the ‘probe’ (Figure 1). The primer segment was meant to induce spatial facilitation, in essence to prime CSTMD1 with object movement outside of the cell’s receptive field boundary. The authors systematically varied the path and speed of the primer. The second stimulus segment, the probe, was presented directly within the cell’s receptive field. Weiderman et al. [1] found that the primer induced a spiking response to the probe that was much stronger than that for the probe alone. More generally, by offsetting the primer and probe on different linear paths, or by spacing them on the same path interrupted by different time delays, the authors showed that CSTMD1 develops a sort of ‘hot spot’ of increased sensitivity directly in the path of the object (Figure 1, top). The hot spot is focused in front of the object’s immediate linear path, so it can only be activated if the object continues along its
current trajectory into the future. In essence, CSTMD1 receives information about an object’s path while the object is moving outside of the cell’s own receptive field, and somehow assembles an internal model of object dynamics so that the cell’s spiking response is heightened if the object continues on the ‘predicted’ path. The authors next explored additional properties of the predictive hot spot, and demonstrated that the cell shows increased sensitivity to the contrast of the object (Figure 1, middle). They found that not only does the cell’s increased sensitivity to firing action potentials predict the path of the object, but the same process also increases the object’s visual salience or ‘detectability’ against the visual surroundings. One notable property of CSTMD’s receptive field is that it is not selective for any particular direction of object motion — it encodes orthogonal directions equally well. By systematically varying the angular offset between the primer path and the probe path, the authors found that the direction of the primer bestows greater sensitivity to the probe in the same direction (Figure 1, bottom): the primer makes the receptive field strongly directionally tuned. The range of possible directional ‘learning’ is not meager, but rather can operate over a span of at least 90 degrees: CSTMD1 can thus be primed to become maximally sensitive to either upward or perpendicular rightward object paths. The response properties of the CSTMD1 neuron of the dragonfly optic lobe are highly sophisticated by comparison to visual feature detectors found in other animals. Like CSTMD1, neurons of the feline cortex and the fly lobula show strong selectivity for small objects, and are inhibited by motion of the visual panorama, but neither show spatial facilitation outside the receptive field [8,9]. Neurons of the mouse retina and fly lobula plate are strongly directionally selective (for review see [10]), but neither show directional plasticity like CSTMD1. To my knowledge, no other cell class or single identified neuron has been shown to have rapidly adapting properties so well suited to predict the future path of an object. It will be interesting to see how this research impacts our understanding
Predictive ‘hot spot’
Primer
Probe
Receptive field Computer display Contrast enhancement
Directional tuning
Current Biology
Figure 1. Dynamic tuning properties of the CSTMD1 neuron. Small contrasting objects were presented on a computer display in front of the dragonfly. Top: target motion outside the receptive field (dashed line) is called a ‘primer’ and target motion inside the receptive field is called the ‘probe’. After the primer, spiking responses of the probe show a strong ‘hot spot’ (red) of increased object sensitivity directly in the path of the probe, flanked by inhibitory surround regions (blue). Middle: a primer causes the probe response to be more sensitive to object contrast, increasing the apparent saliency of a low contrast object. Bottom: a primer moving along one linear path increases the directional tuning of the probe response along the same direction.
of object motion detection in other model systems — the field is expanding rapidly, particularly in the fly and mouse model systems. But neither of these animals has evolved to pursue small targets. We may need to first look at other insect aerial predators (such as the killer fly [11]) for a neuron that is as highly specialized as CSTMD1 for selecting and predicting the path of airborne ‘fastballs’.
Current Biology 27, R1060–R1080, October 9, 2017 R1077
Current Biology
Dispatches REFERENCES
4. Nordstro¨m, K., Barnett, P.D., and O’Carroll, D.C. (2006). Insect detection of small targets in moving visual clutter. PLoS Biol. 4, e54.
8. Kato, H., Bishop, P.O., and Orban, G.A. (1978). Hypercomplex and simple/complex cell classifications in cat striate cortex. J. Neurophysiol. 41, 1071–1095.
1. Wiederman, S.D., Fabian, J.M., Dunbier, J.R., and O’Carroll, D.C. (2017). A predictive focus of gain modulation encodes target trajectories in insect vision. eLife 6, e26478.
5. Wiederman, S.D., and O’Carroll, D.C. (2013). Selective attention by an insect visual neuron. Curr. Biol. 23, 156–161.
2. Mischiati, M., Lin, H.-T., Herold, P., Imier, E., Olberg, R., and Leonardo, A. (2015). Internal models direct dragonfly interception steering. Nature 517, 333–338.
6. Nordstro¨m, K., Bolzon, D.M., and O’Carroll, D.C. (2011). Spatial facilitation by a highperformance dragonfly target-detecting neuron. Biol. Lett. 7, 588–591.
10. Mauss, A.S., Vlasits, A., Borst, A., and Feller, M. (2017). Visual circuits for direction selectivity. Annu. Rev. Neurosci. 40, 211–230.
3. Harland, D.P. (2000). Optical cues and visionbased discrimination mechanisms underlying predatory versatility in jumping spiders. Doctoral thesis, Zoology (University of Canterbury).
7. Dunbier, J.R., Wiederman, S.D., Shoemaker, P.A., and O’Carroll, D.C. (2012). Facilitation of dragonfly target-detecting neurons by slow moving features on continuous paths. Front. Neural Circuits 6, 79.
11. Wardill, T.J., Knowles, K., Barlow, L., Tapia, G., Nordstro¨m, K., Olberg, R.M., and GonzalezBellido, P.T. (2015). The killer fly hunger games: target size and speed predict decision to pursuit. Brain Behav. Evol. 86, 28–37.
9. Keles, M.F., and Frye, M.A. (2017). Object detecting neurons in Drosophila. Curr. Biol. 27, 680–687.
Cell Biology: Capturing Formin’s Mechano-Inhibition Dimitrios Vavylonis* and Brandon G. Horan Department of Physics, Lehigh University, Bethlehem, PA 18015, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2017.08.020
Formins polymerize actin filaments for the cytokinetic contractile ring. Using in vitro reconstitution of fission yeast contractile ring precursor nodes containing formins and myosin, a new study shows that formin-mediated polymerization is strongly inhibited upon the capture and pulling of actin filaments by myosin, a result that has broad implications for cellular mechanosensing. Fundamental cellular processes such as cell motility and cell division rely on the ability of the actin cytoskeleton to generate and respond to mechanical forces. Polymerization of networks or bundles of actin filaments close to the cell membrane can result in a local cell protrusion or contraction when newly generated actin filaments are pulled by myosin motor proteins. Local activation of actin polymerization is thus tightly regulated by cells. Formin proteins are key actin filament regulators that generate actin filaments for the actomyosin contractile ring during cytokinesis and also have important roles in a host of cellular processes as well as being specific targets of pathogenic bacteria [1–3]. Research in fission yeast has shown that the formin Cdc12 and the type II myosin Myo2 localize in a broad band of membrane-associated nodes during early stages of cytokinesis [4]. Actin filaments nucleated by Cdc12 are pulled by Myo2, generating the force required
to pull the nodes together into a contractile ring. A ‘search, capture, pull and release’ (SCPR) mechanism [5] was proposed in a computational model of this process; however, this behavior has not been tested in an experiment with controlled conditions in vitro, leaving many mechanistic questions unanswered. The first minimal reconstitution of ‘search, capture and pull’ has now been achieved in a new study by Zimmermann et al. [6] (Figure 1A). In addition to establishing basic biophysical properties of the process, the authors discovered that Cdc12-mediated actin polymerization is drastically reduced when Myo2 pulls on the actin filament. This novel mechanoregulatory mechanism has important implications in broad areas of cell mechanosensing and contractility. Formins are multi-domain proteins that form dimers to nucleate actin filaments. They remain associated with the barbed end of the actin filament through their formin homology (FH) 2 domain that
R1078 Current Biology 27, R1060–R1080, October 9, 2017 ª 2017 Elsevier Ltd.
wraps around the filament and rotates around its helix as the filament elongates [1,7] (Figure 1). Predicted flexible FH1 domains extend from each FH2 domain and link at their amino terminus, which is frequently associated with proteins on the cell membrane. FH1 domains characteristically contain proline-rich regions that bind the protein profilin, which itself is bound to a large fraction of actin monomers in cells. Kinetic modeling and physical arguments suggest that the FH1 domain captures and directly transfers profilin–actin to the barbed end of the actin filament, speeding up polymerization by severalfold in a profilinconcentration-dependent manner [8]. Experiments with various formin constructs have supported key features of the transfer mechanism, such as a relationship between the proximity of an FH1 proline-rich sequence to the FH2 domain and its contribution to actin polymerization rate [9–11]. The proposed flexible nature of the FH1 domain that is required for direct transfer
