- Email: [email protected]
Chromosome Segregation: Centromeres Get Bent
Dispatch R159
Dispatches
Chromosome Segregation: Centromeres Get Bent Work over the last several decades has shown that kinetochores play an active part in chromosome segregation, while the chromatin and, more to the point, the DNA have gathered little attention. In two intriguing papers, the Bloom and Khodjakov groups show that intercentromeric chromatin plays a much more active part in chromosome segregation than previously suspected. Jonas F. Dorn and Paul S. Maddox ‘‘Indeed, the role in mitosis of the chromosome arms, which carry most of the genetic material, may be compared with that of a corpse at a funeral: they provide the reason for the proceedings but do not take an active part in them.’’ (Mazia, 1961 [1])
This famous quote from Dan Mazia summarizes our current understanding of chromosome segregation. During mitosis, interactions between the mitotic spindle and chromosomes regulate and facilitate chromosome segregation. Mitotic chromosomes attach to microtubules via a complex protein structure termed the kinetochore, which assembles on the centromeric DNA of each sister chromosome. Microsurgical experiments in which the chromosome arms were removed suggested that the ‘chromosome’ itself is dispensable for microtubule-based kinetochore motility [2]. Furthermore, purified kinetochore proteins devoid of DNA are competent to interact with and, in fact, move on microtubules [3]. Together, these results seem to confirm Mazia’s famous quote. Aside from passive oscillatory movement coupled to microtubule growth and shortening, the only activity of the chromosome considered noteworthy was the fact that it is stretched toward opposite poles. Since chromosome stretching decreases when microtubules are removed, it seemed natural to conclude that the intercentromeric chromatin behaves like an elastic spring under the forces of microtubules attached to kinetochores. Yeh et al. [4], in a recent issue of Current Biology, have now shed light on the geometry of the spring by investigating the curious ‘cohesin paradox’ in budding yeast (see below). In complementary work in vertebrate
arek et al. [5] have cells, Lonc investigated the physical properties of the intercentromeric region. The Cohesin Paradox The cohesin complex consists of four proteins (Smc1, Smc3, Scc1 and Scc3) that are hypothesized to form 50 nm rings around strands of DNA from two sister chromatids. These rings are
A
cleaved at anaphase onset, allowing sister chromatids to segregate. Prior to anaphase, cohesin holds the sisters together along their entire length, preventing premature separation and, thereby, aneuploidy. Interestingly, in budding yeast, chromatin immunoprecipitation studies demonstrated that there is a higher density of cohesin in the region of the centromere compared with the chromosome arms [6,7]. In the light of these data, it is perhaps surprising that budding yeast sister centromeres are separated by as much as 700 nm during metaphase [8,9]. These
B Chromosome arm Microtubule
Kinetochore
Centromeric chromatin Cohesin
C
D
Current Biology
Figure 1. The chromatin loop. Fluorescent labeling of cohesin (green) in the canonical chromosome arrangement (A) should give rise to a disk-shaped fluorescence distribution (B). If the chromosomes adopt a cruciform shape with intramolecular chromatin loops (C), the fluorescence intensity would follow a tube-like distribution (D).
Current Biology Vol 18 No 4 R160
is spring-like and can be stretched by the forces of microtubule attachment to sister kinetochores. Thus, the molecular structure of the budding yeast centromeric region described by Yeh et al. [4] provides an elegant explanation of the cohesin paradox and is consistent with the canonical view of the mechanical properties of the kinetochore spring.
C
A
B
D
Current Biology
Figure 2. Is the intercentromeric chromatin elastic or plastic? A bipolar metaphase spindle (A) was treated with monastrol to create a monopolar spindle with attached chromosomes (B). When microtubules are depolymerized in this arrangement, elastic chromatin would result in sister kinetochores returning to a back-to-back orientation (C), while plastic chromatin will not recover the previous orientation on its own (D).
findings give rise to the ‘cohesin paradox’: the chromosomal region of the highest cohesin concentration is also the region with the largest sister chromatid separation. To investigate this paradox, Yeh et al. [4] first determined the cytological localization of cohesin. Canonical views of chromosome arrangement in metaphase budding yeast predict a disk-shaped distribution of cohesin perpendicular to the spindle axis (Figure 1A,B). By using a GFP–cohesin fusion protein, Yeh et al. [4] found that, in metaphase, cohesin localizes in a tube along the spindle axis (Figure 1D). This result could be due to either an unexpected arrangement of metaphase chromosomes or an unexpected, additional localization of the cohesin complex. Models for ‘normal’ cohesin function predict a stable association with chromatin. Using fluorescence redistribution after photobleaching (FRAP), Yeh et al. [4] confirmed that the tube-localized cohesin was in fact stable. Therefore, a new model of chromosome arrangement was needed; one that incorporates intra- rather than inter-sister pairing of chromatin fibers (Figure 1C) [10]. In one possible model, DNA from a single sister forms a loop at the centromere and cohesin molecules clamp this loop, thus creating a cruciform conformation. This model, if confirmed, could account for both
the biochemical data showing increased cohesin at centromeres and the microscopy data. To test this model further the authors investigated the geometry of budding yeast metaphase chromosomes in molecular detail. Chromosome conformation capture (3C) is a technique that allows the higher order structure of DNA in vivo to be inferred [11]. By cross-linking DNA and proteins in vivo, followed by restriction digestion and PCR amplification of specific fragments, DNA loops formed by protein–DNA interactions can be detected. Yeh et al. [4] used 3C to determine the arrangement of centromeric DNA and found that centromeric chromatin indeed adopts a cross-shaped conformation in metaphase. The single centromeric nucleosome of each sister chromatid [12] is at the apex of a loop that extends ~12.5 kb from the true metaphase plate toward the pole. Disruption of cohesin binding at the centromere leads to a decrease in chromosome looping and interkinetochore distance, indicating that cohesin holds the two strands of the same chromatid together, thus forcing the centromeres apart. Furthermore, chromosome looping is also decreased by loss of kinetochore–microtubule interactions. These results imply that the cohesin-stabilized centromere loop
Elastic or Plastic? arek et al. The second study, from Lonc [5], examined the mechanical properties of the intercentromeric region. Most models of the mitotic spindle [13–15] depict the intercentromeric region of the chromosome as an elastic spring. arek et al. [5] predicted that, if the Lonc intercentromeric region were bent by external forces, it should spring back in the absence of forces exerted by microtubules attached to kinetochores (Figure 2). In a clever series of arek et al. [5] experiments, Lonc forced spindle poles to collapse, thus bending bipolar attached sister kinetochores so that both face a single pole. They then removed load from these kinetochores by depolymerizing microtubules. Analysis by light and electron microscopy showed that the inter-kinetochore region remained bent and did not ‘spring’ back into a ‘back-to-back’ configuration. arek et al. [5] went on to show Lonc that microtubule-based force is required not only to bend but also to straighten the sister centromeres. In turn, back-to-back orientation of sister centromeres is required for robust spindle formation. These observations indicate the intercentromeric region is plastic (undergoes permanent structural rearrangement during deformation) rather than elastic (returns to the original state after deformation). We conclude that, in mitosis, the chromosome is more like a corpse at a crime scene than one at a funeral. It may only be a passive participant, but it is clearly interesting to investigate, and its properties do shape the proceedings. The next step in the investigation is to solve the elasto–plastic paradox implied by the arek studies. We suspect Yeh and Lonc that both properties are required for accurate chromosome segregation: an elastic kinetochore–DNA junction and a plastic intercentromeric region.
Dispatch R161
References 1. Mazia, D. (1961). Mitosis and the physiology of cell division. In The Cell: Biochemistry, Physiology, Morphology, J. Brachet and A.E. Mirsky, eds. (New York: Academic Press), pp. 77–412. 2. Faruki, S., Cole, R.W., and Rieder, C.L. (2002). Separating centrosomes interact in the absence of associated chromosomes during mitosis in cultured vertebrate cells. Cell Motil. Cytoskeleton 52, 107–121. 3. Westermann, S., Wang, H.W., Avila-Sakar, A., Drubin, D.G., Nogales, E., and Barnes, G. (2006). The Dam1 kinetochore ring complex moves processively on depolymerizing microtubule ends. Nature 440, 565–569. 4. Yeh, E., Haase, J., Paliulis, L.V., Joglekar, A., Bond, L., Bouck, D., Salmon, E.D., and Bloom, K. (2008). Pericentric chromatin is organized into an intramolecular loop in mitosis. Curr. Biol. 18, 81–90. 5. Loncarek, J., Kisurina-Evgenieva, O., Vinogradova, T., Hergert, P., La Terra, S., Kapoor, T.M., and Khodjakov, A. (2007). The centromere geometry essential for keeping mitosis error free is controlled by spindle forces. Nature 450, 745–749. 6. Blat, Y., and Kleckner, N. (1999). Cohesins bind to preferential sites along yeast chromosome
7.
8.
9.
10.
11.
12.
13.
III, with differential regulation along arms versus the centric region. Cell 98, 249–259. Weber, S.A., Gerton, J.L., Polancic, J.E., DeRisi, J.L., Koshland, D., and Megee, P.C. (2004). The kinetochore is an enhancer of pericentric cohesin binding. PLoS Biol. 2, 1340–1353. He, X.W., Asthana, S., and Sorger, P.K. (2000). Transient sister chromatid separation and elastic deformation of chromosomes during mitosis in budding yeast. Cell 101, 763–775. Pearson, C.G., Maddox, P.S., Salmon, E.D., and Bloom, K. (2001). Budding yeast chromosome structure and dynamics during mitosis. J. Cell Biol. 152, 1255–1266. Bloom, K., Sharma, S., and Dokholyan, N.V. (2006). The path of DNA in the kinetochore. Curr. Biol. 16, R276–R278. Dekker, J., Rippe, K., Dekker, M., and Kleckner, N. (2002). Capturing chromosome conformation. Science 295, 1306–1311. Furuyama, S., and Biggins, S. (2007). Centromere identity is specified by a single centromeric nucleosome in budding yeast. Proc. Natl. Acad. Sci. USA 104, 14706–14711. Maddox, P., Straight, A., Coughlin, P., Mitchison, T.J., and Salmon, E.D. (2003). Direct
observation of microtubule dynamics at kinetochores in Xenopus extract spindles: implications for spindle mechanics. J. Cell Biol. 162, 377–382. 14. Gardner, M.K., Pearson, C.G., Sprague, B.L., Zarzar, T.R., Bloom, K., Salmon, E.D., and Odde, D.J. (2005). Tension-dependent regulation of microtubule dynamics at kinetochores can explain metaphase congression in yeast. Mol. Biol. Cell 16, 3764–3775. 15. Civelekoglu-Scholey, G., Sharp, D.J., Mogilner, A., and Scholey, J.M. (2006). Model of chromosome motility in Drosophila embryos: Adaptation of a general mechanism for rapid mitosis. Biophys. J. 90, 3966–3982.
Institute for Research in Immunology and Cancer, Department of Pathology and Cell Biology, Universite´ de Montreal, P.O. Box 6128, Station Centre-Ville, Montre´al, Quebec H3C 3J7, Canada. E-mail: [email protected] DOI: 10.1016/j.cub.2007.12.044
Fly Fighting: Octopamine Modulates Aggression The genetics and neurobiology of Drosophila aggression are still poorly understood. A new study using an automated method to analyze one component of male fly aggression has shown that the biogenic amine octopamine plays a role in the modulation of aggressive behavior. Herman A. Dierick Aggressive behavior is a complex social interaction influenced by numerous internal and external factors. It is widespread in the animal kingdom and, in the case of male Drosophila melanogaster, was described in remarkable detail nearly 50 years ago [1]. A recent flurry of papers [2–8] has used this model organism to begin to elucidate the genetics and the neurocircuitry of aggressive behavior. The study of fly aggression has been time-consuming and tedious, given the lack of an automated assay that would allow it to be easily quantified. As reported recently in Current Biology [9], this obstacle has now been elegantly overcome by the development of an automated way of analysing one component of male fly aggression, the lunge. Lunges occur in most aggressive encounters of male fruit flies [10] and are unambiguously aggressive as males only perform this behavior when they fight. Hoyer et al. [9] showed that the lunge can be decomposed into three
phases (Figure 1): the attacking male first changes its body posture to a roughly 50 angle by rising on its hind-legs (phase 1); then it slams down on its opponent with its forelegs, reaching a head velocity of more than 250 mm sec21 (phase 2; note that a fly is w2.5 mm long); and finally, it tries, not always successfully, to pull its opponent towards it with its forelegs (phase 3). The first two phases only take on average 46 milliseconds, a fact nicely illustrated with high-velocity movies. With a regular camera, a lunge takes just three frames, and Hoyer et al. [9] developed software, called ‘lunge count’, to count the number of lunges from this characteristic video sequence. They applied this analysis on the last 15 minutes of 30 minutes of videotape of one pair of males in a small arena with a food patch (the presence of food greatly enhances aggression [10]). The software also registers walking distance and size on the basis of the fly’s surface area. Because the software is somewhat sloppy at low fighting levels, they
Figure 1. The three phases of the lunge. In phase 1 (top), the male fly rises on his hind legs, changing his body posture to a w50 angle. In phase 2 (middle), the attacking male slams his front legs down on the opponent reaching a head velocity of w260 mm sec21. In phase 3 (bottom), the attacking male tries to pull its opponent towards it with its forelegs. (Photographs courtesy of Susanne Hoyer.)
Dispatches
Chromosome Segregation: Centromeres Get Bent Work over the last several decades has shown that kinetochores play an active part in chromosome segregation, while the chromatin and, more to the point, the DNA have gathered little attention. In two intriguing papers, the Bloom and Khodjakov groups show that intercentromeric chromatin plays a much more active part in chromosome segregation than previously suspected. Jonas F. Dorn and Paul S. Maddox ‘‘Indeed, the role in mitosis of the chromosome arms, which carry most of the genetic material, may be compared with that of a corpse at a funeral: they provide the reason for the proceedings but do not take an active part in them.’’ (Mazia, 1961 [1])
This famous quote from Dan Mazia summarizes our current understanding of chromosome segregation. During mitosis, interactions between the mitotic spindle and chromosomes regulate and facilitate chromosome segregation. Mitotic chromosomes attach to microtubules via a complex protein structure termed the kinetochore, which assembles on the centromeric DNA of each sister chromosome. Microsurgical experiments in which the chromosome arms were removed suggested that the ‘chromosome’ itself is dispensable for microtubule-based kinetochore motility [2]. Furthermore, purified kinetochore proteins devoid of DNA are competent to interact with and, in fact, move on microtubules [3]. Together, these results seem to confirm Mazia’s famous quote. Aside from passive oscillatory movement coupled to microtubule growth and shortening, the only activity of the chromosome considered noteworthy was the fact that it is stretched toward opposite poles. Since chromosome stretching decreases when microtubules are removed, it seemed natural to conclude that the intercentromeric chromatin behaves like an elastic spring under the forces of microtubules attached to kinetochores. Yeh et al. [4], in a recent issue of Current Biology, have now shed light on the geometry of the spring by investigating the curious ‘cohesin paradox’ in budding yeast (see below). In complementary work in vertebrate
arek et al. [5] have cells, Lonc investigated the physical properties of the intercentromeric region. The Cohesin Paradox The cohesin complex consists of four proteins (Smc1, Smc3, Scc1 and Scc3) that are hypothesized to form 50 nm rings around strands of DNA from two sister chromatids. These rings are
A
cleaved at anaphase onset, allowing sister chromatids to segregate. Prior to anaphase, cohesin holds the sisters together along their entire length, preventing premature separation and, thereby, aneuploidy. Interestingly, in budding yeast, chromatin immunoprecipitation studies demonstrated that there is a higher density of cohesin in the region of the centromere compared with the chromosome arms [6,7]. In the light of these data, it is perhaps surprising that budding yeast sister centromeres are separated by as much as 700 nm during metaphase [8,9]. These
B Chromosome arm Microtubule
Kinetochore
Centromeric chromatin Cohesin
C
D
Current Biology
Figure 1. The chromatin loop. Fluorescent labeling of cohesin (green) in the canonical chromosome arrangement (A) should give rise to a disk-shaped fluorescence distribution (B). If the chromosomes adopt a cruciform shape with intramolecular chromatin loops (C), the fluorescence intensity would follow a tube-like distribution (D).
Current Biology Vol 18 No 4 R160
is spring-like and can be stretched by the forces of microtubule attachment to sister kinetochores. Thus, the molecular structure of the budding yeast centromeric region described by Yeh et al. [4] provides an elegant explanation of the cohesin paradox and is consistent with the canonical view of the mechanical properties of the kinetochore spring.
C
A
B
D
Current Biology
Figure 2. Is the intercentromeric chromatin elastic or plastic? A bipolar metaphase spindle (A) was treated with monastrol to create a monopolar spindle with attached chromosomes (B). When microtubules are depolymerized in this arrangement, elastic chromatin would result in sister kinetochores returning to a back-to-back orientation (C), while plastic chromatin will not recover the previous orientation on its own (D).
findings give rise to the ‘cohesin paradox’: the chromosomal region of the highest cohesin concentration is also the region with the largest sister chromatid separation. To investigate this paradox, Yeh et al. [4] first determined the cytological localization of cohesin. Canonical views of chromosome arrangement in metaphase budding yeast predict a disk-shaped distribution of cohesin perpendicular to the spindle axis (Figure 1A,B). By using a GFP–cohesin fusion protein, Yeh et al. [4] found that, in metaphase, cohesin localizes in a tube along the spindle axis (Figure 1D). This result could be due to either an unexpected arrangement of metaphase chromosomes or an unexpected, additional localization of the cohesin complex. Models for ‘normal’ cohesin function predict a stable association with chromatin. Using fluorescence redistribution after photobleaching (FRAP), Yeh et al. [4] confirmed that the tube-localized cohesin was in fact stable. Therefore, a new model of chromosome arrangement was needed; one that incorporates intra- rather than inter-sister pairing of chromatin fibers (Figure 1C) [10]. In one possible model, DNA from a single sister forms a loop at the centromere and cohesin molecules clamp this loop, thus creating a cruciform conformation. This model, if confirmed, could account for both
the biochemical data showing increased cohesin at centromeres and the microscopy data. To test this model further the authors investigated the geometry of budding yeast metaphase chromosomes in molecular detail. Chromosome conformation capture (3C) is a technique that allows the higher order structure of DNA in vivo to be inferred [11]. By cross-linking DNA and proteins in vivo, followed by restriction digestion and PCR amplification of specific fragments, DNA loops formed by protein–DNA interactions can be detected. Yeh et al. [4] used 3C to determine the arrangement of centromeric DNA and found that centromeric chromatin indeed adopts a cross-shaped conformation in metaphase. The single centromeric nucleosome of each sister chromatid [12] is at the apex of a loop that extends ~12.5 kb from the true metaphase plate toward the pole. Disruption of cohesin binding at the centromere leads to a decrease in chromosome looping and interkinetochore distance, indicating that cohesin holds the two strands of the same chromatid together, thus forcing the centromeres apart. Furthermore, chromosome looping is also decreased by loss of kinetochore–microtubule interactions. These results imply that the cohesin-stabilized centromere loop
Elastic or Plastic? arek et al. The second study, from Lonc [5], examined the mechanical properties of the intercentromeric region. Most models of the mitotic spindle [13–15] depict the intercentromeric region of the chromosome as an elastic spring. arek et al. [5] predicted that, if the Lonc intercentromeric region were bent by external forces, it should spring back in the absence of forces exerted by microtubules attached to kinetochores (Figure 2). In a clever series of arek et al. [5] experiments, Lonc forced spindle poles to collapse, thus bending bipolar attached sister kinetochores so that both face a single pole. They then removed load from these kinetochores by depolymerizing microtubules. Analysis by light and electron microscopy showed that the inter-kinetochore region remained bent and did not ‘spring’ back into a ‘back-to-back’ configuration. arek et al. [5] went on to show Lonc that microtubule-based force is required not only to bend but also to straighten the sister centromeres. In turn, back-to-back orientation of sister centromeres is required for robust spindle formation. These observations indicate the intercentromeric region is plastic (undergoes permanent structural rearrangement during deformation) rather than elastic (returns to the original state after deformation). We conclude that, in mitosis, the chromosome is more like a corpse at a crime scene than one at a funeral. It may only be a passive participant, but it is clearly interesting to investigate, and its properties do shape the proceedings. The next step in the investigation is to solve the elasto–plastic paradox implied by the arek studies. We suspect Yeh and Lonc that both properties are required for accurate chromosome segregation: an elastic kinetochore–DNA junction and a plastic intercentromeric region.
Dispatch R161
References 1. Mazia, D. (1961). Mitosis and the physiology of cell division. In The Cell: Biochemistry, Physiology, Morphology, J. Brachet and A.E. Mirsky, eds. (New York: Academic Press), pp. 77–412. 2. Faruki, S., Cole, R.W., and Rieder, C.L. (2002). Separating centrosomes interact in the absence of associated chromosomes during mitosis in cultured vertebrate cells. Cell Motil. Cytoskeleton 52, 107–121. 3. Westermann, S., Wang, H.W., Avila-Sakar, A., Drubin, D.G., Nogales, E., and Barnes, G. (2006). The Dam1 kinetochore ring complex moves processively on depolymerizing microtubule ends. Nature 440, 565–569. 4. Yeh, E., Haase, J., Paliulis, L.V., Joglekar, A., Bond, L., Bouck, D., Salmon, E.D., and Bloom, K. (2008). Pericentric chromatin is organized into an intramolecular loop in mitosis. Curr. Biol. 18, 81–90. 5. Loncarek, J., Kisurina-Evgenieva, O., Vinogradova, T., Hergert, P., La Terra, S., Kapoor, T.M., and Khodjakov, A. (2007). The centromere geometry essential for keeping mitosis error free is controlled by spindle forces. Nature 450, 745–749. 6. Blat, Y., and Kleckner, N. (1999). Cohesins bind to preferential sites along yeast chromosome
7.
8.
9.
10.
11.
12.
13.
III, with differential regulation along arms versus the centric region. Cell 98, 249–259. Weber, S.A., Gerton, J.L., Polancic, J.E., DeRisi, J.L., Koshland, D., and Megee, P.C. (2004). The kinetochore is an enhancer of pericentric cohesin binding. PLoS Biol. 2, 1340–1353. He, X.W., Asthana, S., and Sorger, P.K. (2000). Transient sister chromatid separation and elastic deformation of chromosomes during mitosis in budding yeast. Cell 101, 763–775. Pearson, C.G., Maddox, P.S., Salmon, E.D., and Bloom, K. (2001). Budding yeast chromosome structure and dynamics during mitosis. J. Cell Biol. 152, 1255–1266. Bloom, K., Sharma, S., and Dokholyan, N.V. (2006). The path of DNA in the kinetochore. Curr. Biol. 16, R276–R278. Dekker, J., Rippe, K., Dekker, M., and Kleckner, N. (2002). Capturing chromosome conformation. Science 295, 1306–1311. Furuyama, S., and Biggins, S. (2007). Centromere identity is specified by a single centromeric nucleosome in budding yeast. Proc. Natl. Acad. Sci. USA 104, 14706–14711. Maddox, P., Straight, A., Coughlin, P., Mitchison, T.J., and Salmon, E.D. (2003). Direct
observation of microtubule dynamics at kinetochores in Xenopus extract spindles: implications for spindle mechanics. J. Cell Biol. 162, 377–382. 14. Gardner, M.K., Pearson, C.G., Sprague, B.L., Zarzar, T.R., Bloom, K., Salmon, E.D., and Odde, D.J. (2005). Tension-dependent regulation of microtubule dynamics at kinetochores can explain metaphase congression in yeast. Mol. Biol. Cell 16, 3764–3775. 15. Civelekoglu-Scholey, G., Sharp, D.J., Mogilner, A., and Scholey, J.M. (2006). Model of chromosome motility in Drosophila embryos: Adaptation of a general mechanism for rapid mitosis. Biophys. J. 90, 3966–3982.
Institute for Research in Immunology and Cancer, Department of Pathology and Cell Biology, Universite´ de Montreal, P.O. Box 6128, Station Centre-Ville, Montre´al, Quebec H3C 3J7, Canada. E-mail: [email protected] DOI: 10.1016/j.cub.2007.12.044
Fly Fighting: Octopamine Modulates Aggression The genetics and neurobiology of Drosophila aggression are still poorly understood. A new study using an automated method to analyze one component of male fly aggression has shown that the biogenic amine octopamine plays a role in the modulation of aggressive behavior. Herman A. Dierick Aggressive behavior is a complex social interaction influenced by numerous internal and external factors. It is widespread in the animal kingdom and, in the case of male Drosophila melanogaster, was described in remarkable detail nearly 50 years ago [1]. A recent flurry of papers [2–8] has used this model organism to begin to elucidate the genetics and the neurocircuitry of aggressive behavior. The study of fly aggression has been time-consuming and tedious, given the lack of an automated assay that would allow it to be easily quantified. As reported recently in Current Biology [9], this obstacle has now been elegantly overcome by the development of an automated way of analysing one component of male fly aggression, the lunge. Lunges occur in most aggressive encounters of male fruit flies [10] and are unambiguously aggressive as males only perform this behavior when they fight. Hoyer et al. [9] showed that the lunge can be decomposed into three
phases (Figure 1): the attacking male first changes its body posture to a roughly 50 angle by rising on its hind-legs (phase 1); then it slams down on its opponent with its forelegs, reaching a head velocity of more than 250 mm sec21 (phase 2; note that a fly is w2.5 mm long); and finally, it tries, not always successfully, to pull its opponent towards it with its forelegs (phase 3). The first two phases only take on average 46 milliseconds, a fact nicely illustrated with high-velocity movies. With a regular camera, a lunge takes just three frames, and Hoyer et al. [9] developed software, called ‘lunge count’, to count the number of lunges from this characteristic video sequence. They applied this analysis on the last 15 minutes of 30 minutes of videotape of one pair of males in a small arena with a food patch (the presence of food greatly enhances aggression [10]). The software also registers walking distance and size on the basis of the fly’s surface area. Because the software is somewhat sloppy at low fighting levels, they
Figure 1. The three phases of the lunge. In phase 1 (top), the male fly rises on his hind legs, changing his body posture to a w50 angle. In phase 2 (middle), the attacking male slams his front legs down on the opponent reaching a head velocity of w260 mm sec21. In phase 3 (bottom), the attacking male tries to pull its opponent towards it with its forelegs. (Photographs courtesy of Susanne Hoyer.)