Bacterial Cytokinesis: FzlA Frizzes FtsZ Filaments for Fission Force

Current Biology Vol 20 No 23 R1024

in females could trigger apoptosis at different times during prophase depending on which particular gene is silenced. However, MSUC at variable unsynapsed chromatin regions may be insufficient to explain the elimination of oocytes in many of the DSB and SC mutants. It is possible that both unrepaired DSBs and unsynapsed chromatin can trigger sustained ATR activation. High ATR activity might be incompatible with oocyte survival beyond the pachytene or diplotene stages, independently from its possible effect on gene expression. The involvement of ATR in the detection of SC and DSB defects seems to be similar in males and females. This may pose the danger that a female-type meiotic prophase checkpoint response would trigger apoptosis in all spermatocytes due to lack of SC formation and delayed DSB repair on the X/Y pair in males. Thus, during the course of the evolution of sex chromosomes in mammals an alternative prophase surveillance mechanism had to emerge in males. It appears that the solution came in the form of a mechanism that acts at an earlier stage than the female checkpoint controls. This surveillance mechanism is exceptionally robust because elimination of defective spermatocytes does not depend on the function of proteins that are involved in the monitoring of essential meiotic processes, i.e. synapsis and DSB repair. Instead, the sensory mechanisms, in particular ATR activation, are required for progression beyond stage IV/mid-pachytene. It is likely that this male stage-IV death mechanism can be inactivated only by multiple mutations affecting redundant killer genes on the Y and X chromosomes, two of which have been identified by Royo et al. [9]. How do ZFY1/2 proteins kill? Gene duplication during evolution probably generated Zfy1 and Zfy2, which encode very similar zinc finger-type transcription factors [10]. Zfy1 is the mouse homolog of human ZFY, and there are other ZFY family members on the X chromosome (Zfx) and on autosomes (Zfa). The normal biological role of Zfy1/2 proteins remains elusive, although they are expressed during embryogenesis in somatic cells and primordial germ cells of the genital ridge, in meiosis before pachytene, and later in spermatids during spermiogenesis [11]. Thus, to

understand how Zfy1 and Zfy2 kill, one must know the target genes regulated by them. Another question, perhaps even therapeutically relevant, concerns options to down-regulate ZFY1/2 expression when that is considered pathological. Infertility in men affects about 5–7% of couples [10,12]. Among those patients are XYY males, which show very frequent Y–Y synapsis. It is not far-fetched to speculate that MSCI failure and ZFY expression causes or at least significantly contributes to azoospermia seen in these men. Because ZFY1 and ZFY2 are expressed at various stages during germ cell development, they may have essential functions during gametogenesis. Thus, it is possible that the mid-pachytene surveillance mechanism cannot be inactivated without deleterious affects on gametogenesis, which would make this quality control mechanism inescapable during male meiosis. References 1. Handel, M.A., and Schimenti, J.C. (2010). Genetics of mammalian meiosis: regulation, dynamics and impact on fertility. Nat. Rev. Genet. 11, 124–136. 2. Neale, M.J., and Keeney, S. (2006). Clarifying the mechanics of DNA strand exchange in meiotic recombination. Nature 442, 153–158. 3. Cromie, G.A., and Smith, G.R. (2007). Branching out: meiotic recombination and its regulation. Trends Cell Biol. 17, 448–455. 4. Costa, Y., and Cooke, H. (2007). Dissecting the mammalian synaptonemal complex using targeted mutations. Chromosome Res. 15, 579. 5. Burgoyne, P.S., Mahadevaiah, S.K., and Turner, J.M. (2009). The consequences of asynapsis for mammalian meiosis. Nat. Rev. Genet. 10, 207–216.

6. Hunt, P.A., and Hassold, T.J. (2002). Sex matters in meiosis. Science 296, 2181–2183. 7. Kolas, N.K., Marcon, E., Crackower, M.A., Hoog, C., Penninger, J.M., Spyropoulos, B., and Moens, P.B. (2005). Mutant meiotic chromosome core components in mice can cause apparent sexual dimorphic endpoints at prophase or X-Y defective male-specific sterility. Chromosoma 114, 92–102. 8. Mahadevaiah, S.K., Bourc’his, D., de Rooij, D.G., Bestor, T.H., Turner, J.M., and Burgoyne, P.S. (2008). Extensive meiotic asynapsis in mice antagonises meiotic silencing of unsynapsed chromatin and consequently disrupts meiotic sex chromosome inactivation. J. Cell Biol. 182, 263–276. 9. Royo, H., Polikiewicz, G., Mahadevaiah, S.K., Prosser, H., Mitchell, M., Bradley, A., de Rooij, D.G., Burgoyne, P.S., and Turner, J.M. (2010). Evidence that meiotic sex chromosome inactivation is essential for male fertility. Curr. Biol. 20, 2117–2123. 10. Zambrowicz, B.P., Findley, S.D., Simpson, E.M., Page, D.C., and Palmiter, R.D. (1994). Characterization of the murine Zfy1 and Zfy2 promoters. Genomics 24, 406–408. 11. Zambrowicz, B.P., Zimmermann, J.W., Harendza, C.J., Simpson, E.M., Page, D.C., Brinster, R.L., and Palmiter, R.D. (1994). Expression of a mouse Zfy-1/lacZ transgene in the somatic cells of the embryonic gonad and germ cells of the adult testis. Development 120, 1549–1559. 12. de Boer, P., Giele, M., Lock, M.T., de Rooij, D.G., Giltay, J., Hochstenbach, R., and te Velde, E.R. (2004). Kinetics of meiosis in azoospermic males: a joint histological and cytological approach. Cytogenet. Genome Res. 105, 36–46. 13. Perry, J., Palmer, S., Gabriel, A., and Ashworth, A. (2001). A short pseudoautosomal region in laboratory mice. Genome Res. 11, 1826–1832.

Institute of Physiological Chemistry, Faculty of Medicine Carl Gustav Carus, Dresden University of Technology, Fiedlerstr. 42, MTZ, D-01307 Dresden, Germany. DOI: 10.1016/j.cub.2010.11.002

Bacterial Cytokinesis: FzlA Frizzes FtsZ Filaments for Fission Force Most bacteria divide by assembling filaments of the tubulin-like protein FtsZ into a cytokinetic ring, which then constricts. A recent study suggests that Caulobacter crescentus uses a novel regulator, FzlA, to activate ring constriction by inducing helical bundles of FtsZ filaments.

Tushar K. Beuria and William Margolin* The protein FtsZ is conserved in most bacteria, plant plastids, and many archaea, and is a structural homolog of tubulin that plays an important role in cell or organelle division [1]. Like tubulin, FtsZ assembles into polymers

in the presence of GTP, which is hydrolyzed upon assembly. FtsZ does not form microtubules, but FtsZ protofilaments tend to interact laterally and form straight bundles when incubated with ionic or protein cofactors [2]. High-resolution imaging of Escherichia coli cells suggests that the dividing ring, called the Z ring, is

Dispatch R1025

A

B

C

Current Biology

Figure 1. Speculative model for membrane constriction during bacterial cytokinesis. (A) Large circles denote a cross section of the cell at the cell division site, with arrows denoting progression through the cell cycle. FtsZ (small circles) and FzlA (triangles) coassemble into the Z ring (red polygon), anchored to the membrane via other proteins. Repeated cycles of GTP hydrolysis, filament curving (blue), filament breakage, and reassembly of new filaments at the membrane (green) ultimately exert inward constrictive force on the membrane [12]. FzlA enhances bending of FtsZ protofilaments, enhancing the constrictive force. (B,C) Septum-driven cytokinesis, as in, for example, Bacillus subtilis (B), may not require as many constrictive forces on the membrane as septum-independent constriction, as, for example, in C. crescentus (C). Thus, C. crescentus and its relatives may require additional factors to aid in powering constriction. The unconstricted Z ring (red) is triggered to constrict (blue) by an unknown mechanism and is pushed inwards by the growing division septum (green).

composed of a loose association of bundled protofilaments with a width of about 110 nm [3]. In order to understand how FtsZ functions in bacterial cytokinesis, it is thus important know how the protein is bundled in cells, and how the bundling is regulated. New work by Goley et al. [4] has identified a protein, FzlA, which is important for cell division in Caulobacter crescentus and stimulates FtsZ protofilament bundling. Instead of being straight, these bundles are tightly wound helices [4]. The Z ring is highly dynamic. Although the ring appears to be static prior to constriction, cellular FtsZ is always rapidly shuttling between the ring and an extended helical form of FtsZ [5–7]. When purified FtsZ molecules linked to an artificial membrane tether are added to liposomes in the presence of GTP, they gradually coalesce into rings that partially constrict the membrane [8,9]. These forces may originate from mechanical curving of FtsZ protofilaments, because GTP hydrolysis induces a switch from a straight FtsZ filament to a curved filament [10]. Curved FtsZ filaments

are commonly observed under some conditions, such as on mica by atomic force microscopy [11]. The high turnover rate and curving of FtsZ protofilaments is the basis for one recently proposed model for bacterial cytokinesis. In this model, FtsZ protofilaments assemble as small straight bundles in a restricted area of the membrane. After GTP hydrolysis, the filaments curve in the same direction as the membrane. Because they are tethered to the membrane via interactions with membrane proteins, the resulting filament bending causes the membrane to constrict slightly. Multiple cycles of disassembly of these protofilaments and then rapid reassembly of straight filaments on the membrane could drive membrane constriction (Figure 1A) [12]. A switch to increased bundling of protofilaments or cooperative interactions of FtsZ at the membrane may also be important triggers of membrane constriction [13]. Although FtsZ filaments can exert forces on membranes, these forces need to be cell-cycle regulated. For example, constriction should only initiate when the septum synthesis machinery is completely assembled

and switched on. Moreover, the rate of ring constriction needs to be coordinated properly with the rate of septum synthesis and, in the case of Gram-negative bacteria, invagination of the outer membrane. A large number of transmembrane proteins, including penicillin-binding proteins, are recruited to the Z ring and are required for anchoring the ring to the membrane, synthesis of the division septum behind it, and eventual separation of the daughter cells [14]. It is likely, though unproven, that septum synthesis itself is an important contributor to the constriction force. If FtsZ protofilament bundling and/or curvature are necessary for Z ring function, then regulatory proteins should exist that induce this behavior. Indeed, proteins such as ZipA, ZapA, and SepF bind directly to FtsZ and stimulate FtsZ bundling [15–17]. These activities are antagonized by negatively acting FtsZ-binding proteins such as MinC, MipZ, SulA, ClpX, and EzrA, which either sequester FtsZ monomers or inhibit protofilament assembly and/ or bundling [14]. FtsA, a widely conserved actin homolog whose gene is usually cotranscribed with that of FtsZ, probably regulates protofilament

Current Biology Vol 20 No 23 R1026

bundling. A gain of function mutant of FtsA can break up FtsZ bundles in vitro upon ATP binding, and the resulting FtsZ protofilaments are curved [18], which potentially helps to power constriction. FzlA, another regulator of FtsZ, was discovered by Goley et al. [4] during a screen for fluorescent proteins in C. crescentus that bound to an elongated region of overproduced FtsZ. Several proteins, including FzlA, that localized to the resulting elongated cell constriction turned out to bind FtsZ directly. Cells depleted of FzlA grew but stopped dividing, indicating that FzlA is an important cell division protein. Some Z rings were present in the nondividing cells, and these contained other components of the ‘divisome’ cytokinetic machinery. This suggested that FzlA is not essential for Z-ring assembly or recruitment of later proteins, but instead enhances its integrity and ability to constrict. This idea is supported by the ability of FzlA to antagonize the inhibitory activity of MipZ. However, overproduction of FzlA induced FtsZ to form tight foci instead of rings, and cell division was inhibited; this suggests that the FzlA:FtsZ ratio is critical for function, and that too much FzlA can hyperstabilize FtsZ polymers. Consistent with this regulatory role, the cellular FzlA:FtsZ ratio was found to be highest before and during Z ring assembly and lower at later stages of cytokinesis. Given the strong effect of FzlA on Z rings, Goley et al. [4] then tested how it might affect FtsZ structures in vitro. Both sedimentation and light scattering assays demonstrated that FzlA binds to FtsZ directly and induces formation of higher order structures of FtsZ. Like other FtsZ bundling proteins, FzlA decreased FtsZ’s GTPase activity, which suggested that FtsZ is incorporated in relatively stable bundles. Most strikingly, negative stain electron microscopy revealed that FzlA:FtsZ ratios of 1:1 or 2:1 produced thick telephone-cord-like helical bundles with short helical pitches, along with structures that resembled half-pipes. The structures were dependent on both proteins. The cords and halfpipes were w50 nm in diameter, and thus were much more highly curved than previously characterized w300 nm diameter curved FtsZ filaments induced by FtsA. Their formation was inhibited at higher FzlA

concentrations and when nucleotides other than GTP were present, emphasizing the importance of a narrow stoichiometric range for optimal assembly of these structures. Higher magnification revealed that the helices are composed of three distinct parallel filaments. Given the ratios needed for helix formation, the authors suggest that FzlA dimers are sandwiched between two FtsZ protofilaments. FzlA by itself was shown by size exclusion chromatography to be monomeric, but it will be important to determine if it can dimerize. Recent structural work with SlmA, a DNA-binding protein that inhibits Z ring assembly by nucleoid occlusion, suggests that a DNA-bound SlmA dimer forms a similar sandwich between two FtsZ protofilaments, although the FtsZ filaments are antiparallel [19]. Although these FtsZ protofilaments are probably not functional and may be a way to sequester FtsZ from the Z ring, it is interesting that they also often assume a roughly helical shape. The work by Goley et al. [4] demonstrates that FzlA enhances Z ring integrity and promotes the formation of highly curved FtsZ filaments that are structurally distinct from the filaments induced by other known FtsZ regulators. It remains to be seen, of course, whether the striking FtsZ structures formed in vitro are physiologically relevant, particularly as the FzlA:FtsZ ratio in predivisional C. crescentus cells was measured at w1:6, lower than the ratios used for observing the helices. Nevertheless, the local concentrations may be different in the ring itself, and FzlA activity, such as dimerization, may itself be modified by a cell cycle cue. It is thus reasonable to propose that FzlA-mediated curvature of membrane-anchored FtsZ filaments may help to generate a force for membrane constriction during cytokinesis. Although FzlA is conserved in some other a-proteobacteria in addition to C. crescentus, the way C. crescentus divides suggests it may have a specific need for sharply curved FtsZ protofilaments. In contrast to bacteria such as E. coli or B. subtilis that synthesize a division septum, C. crescentus divides solely by coupled constriction of outer and inner membranes [20]. Therefore,

whereas septum synthesis probably helps to push Z ring constriction in many bacteria, additional factors like FzlA may be needed to provide an extra mechanical push for C. crescentus constriction (Figure 1B,C). This extra power may not be as essential during slow growth because FzlA was not required for cell division when C. crescentus was grown in minimal medium [4]. This suggests that the demands of cytokinesis are higher during rapid growth and cells have evolved options to maintain Z ring integrity under these conditions. It is certain that the rapid growth in our knowledge of new FtsZ regulators will stimulate further evolution of bacterial cytokinesis models. References 1. Margolin, W. (2005). FtsZ and the division of prokaryotic cells and organelles. Nat. Rev. Mol. Cell. Biol. 6, 862–871. 2. Romberg, L., and Levin, P.A. (2003). Assembly dynamics of the bacterial cell division protein FtsZ: poised at the edge of stability. Annu. Rev. Microbiol. 57, 125–154. 3. Fu, G., Huang, T., Buss, J., Coltharp, C., Hensel, Z., and Xiao, J. (2010). In vivo structure of the E. coli FtsZ-ring revealed by photoactivated localization microscopy (PALM). PLoS One 5, e12682. 4. Goley, E.D., Dye, N.A., Werner, J.N., Gitai, Z., and Shapiro, L. (2010). Imaging-based identification of a critical regulator of FtsZ protofilament curvature in Caulobacter. Mol. Cell 39, 975–987. 5. Thanedar, S., and Margolin, W. (2004). FtsZ exhibits rapid movement and oscillation waves in helix-like patterns in Escherichia coli. Curr. Biol. 14, 1167–1173. 6. Anderson, D.E., Gueiros-Filho, F.J., and Erickson, H.P. (2004). Assembly dynamics of FtsZ rings in Bacillus subtilis and Escherichia coli and effects of FtsZ-regulating proteins. J. Bacteriol. 186, 5775–5581. 7. Peters, P.C., Migocki, M.D., Thoni, C., and Harry, E.J. (2007). A new assembly pathway for the cytokinetic Z ring from a dynamic helical structure in vegetatively growing cells of Bacillus subtilis. Mol. Microbiol. 64, 487–499. 8. Osawa, M., Anderson, D.E., and Erickson, H.P. (2008). Reconstitution of contractile FtsZ rings in liposomes. Science 320, 792–794. 9. Osawa, M., Anderson, D.E., and Erickson, H.P. (2009). Curved FtsZ protofilaments generate bending forces on liposome membranes. EMBO J. 28, 3476–3484. 10. Lu, C., Reedy, M., and Erickson, H.P. (2000). Straight and curved conformations of FtsZ are regulated by GTP hydrolysis. J. Bacteriol. 182, 164–170. 11. Mingorance, J., Tadros, M., Vicente, M., Gonzalez, J.M., Rivas, G., and Velez, M. (2005). Visualization of single Escherichia coli FtsZ filament dynamics with atomic force microscopy. J. Biol. Chem. 280, 20909–20914. 12. Li, Z., Trimble, M.J., Brun, Y.V., and Jensen, G.J. (2007). The structure of FtsZ filaments in vivo suggests a force-generating role in cell division. EMBO J. 26, 4694–4708. 13. Erickson, H.P. (2009). Modeling the physics of FtsZ assembly and force generation. Proc. Natl. Acad. Sci. USA 106, 9238–9243. 14. Adams, D.W., and Errington, J. (2009). Bacterial cell division: assembly, maintenance and disassembly of the Z ring. Nat. Rev. Microbiol. 7, 642–653.

Dispatch R1027

15. Gueiros-Filho, F.J., and Losick, R. (2002). A widely conserved bacterial cell division protein that promotes assembly of the tubulin-like protein FtsZ. Genes Dev. 16, 2544–2556. 16. Hale, C.A., Rhee, A.C., and de Boer, P.A. (2000). ZipA-induced bundling of FtsZ polymers mediated by an interaction between C-terminal domains. J. Bacteriol. 182, 5153–5166. 17. Singh, J.K., Makde, R.D., Kumar, V., and Panda, D. (2008). SepF increases the assembly and bundling of FtsZ polymers and stabilizes FtsZ protofilaments by binding along its length. J. Biol. Chem. 283, 31116–31124.

18. Beuria, T.K., Mullapudi, S., Mileykovskaya, E., Sadasivam, M., Dowhan, W., and Margolin, W. (2009). Adenine nucleotidedependent regulation of assembly of bacterial tubulin-like FtsZ by a hypermorph of bacterial actin-like FtsA. J. Biol. Chem. 284, 14079–14086. 19. Tonthat, N.K., Arold, S.T., Pickering, B., Van Dyke, M.W., Liang, S., Lu, Y., Beuria, T.K., Margolin, W., and Schumacher, M.A. (2010). Molecular mechanism by which the SlmA nucleoid occlusion factor keeps cytokinesis in check. EMBO J., in press. 20. Judd, E.M., Comolli, L.R., Chen, J.C., Downing, K.H., Moerner, W.E., and

Nuclear Migration: Rock and Roll Facilitated by Dynein and Kinesin The nucleus encounters other organelles as well as high cytoplasmic pressures during its migration within the cell. A new study describes how the action of kinesin and dynein motors is coordinated at the nuclear envelope to rock and roll the nucleus in Caenorhabditis elegans.

Xiaochang Zhang1,2 and Min Han1,2,* Migration of the nucleus relative to a cell body is a prominent cellular process that occurs during development in almost all eukaryotes [1]. A failure in nuclear migration often causes severe developmental defects and human diseases, such as lissencephaly, which is characterized by the lack of sulci and gyri in the brain [2]. Taking vertebrate neurogenesis and neuronal migration as examples, microtubules and their associated proteins, such as Lis1, Ndel, DCX and centrosomal proteins, are known to function during nuclear movement [3]. Although both dynein and kinesin have been implicated in nuclear migration, it has long been unclear how these microtubule motors, which move along the microtubule in opposite directions, are coordinated to apply their forces to the migrating nucleus [4]. A new study published in the Journal of Cell Biology by Fridolfsson and Starr [5] investigates the specific roles of kinesin and dynein complexes at the nuclear envelope during the bidirectional movements and the rolling of nuclei during development of the Caenorhabditis elegans hypodermis. Genetic studies in C. elegans identified functional interactions between the KASH-domain-containing proteins at the outer nuclear envelope and the SUN-domain-containing

proteins at the inner nuclear envelope and established their roles in connecting the nuclei to different components of the cytoskeleton [6,7]. The essential functions of SUN–KASH protein complexes in connecting the nuclear envelope with microtubule and actin networks have also been studied in other model organisms as well as in mammalian tissue culture cells [4]. In the Drosophila eye and zebrafish retina, the KASH proteins have been proposed to link the dynein/dynactin protein complex to the nuclear envelope [8,9]. In the developing mouse brain, the KASH protein Syne-2/nesprin-2 was found to interact with both dynein/ dynactin and kinesin motor complexes [10]. The most compelling evidence of interactions between KASH–SUN proteins and motor complexes came from studies in C. elegans: although a complex containing the KASH protein Zyg-12 and the SUN protein SUN1 interacts with dynein light chain and recruits the dynein–Lis1 protein complex to the nuclear envelope for pronuclear migration, germline nuclear anchorage/movement and other functions [11,12], a complex of the KASH protein UNC-83 and the SUN protein UNC-84 interacts with both kinesin and dynein complexes during nuclear migration in the hypodermal hyp7 cells [13,14] (Figure 1A). The interaction of a nuclear envelope protein with both kinesin and dynein

McAdams, H.H. (2005). Distinct constrictive processes, separated in time and space, divide Caulobacter inner and outer membranes. J. Bacteriol. 187, 6874–6882.

Department of Microbiology and Molecular Genetics, University of Texas Medical School at Houston, 6431 Fannin Street, Houston, TX 77030, USA. *E-mail: [email protected] DOI: 10.1016/j.cub.2010.10.052

motors is intriguing and led the field to investigate how the action of two motors that move in opposite directions can be coordinated to drive the presumably unidirectional nuclear movement. Fridolfsson and Starr [5] have now addressed this question by time-lapse imaging of the nuclear migration process in worm hyp7 cells. They found that loss of UNC-83 or UNC-84 completely blocked the initiation of nuclear migration, which is consistent with the prediction that the adaptor for motor proteins at the nuclear envelope is essential for motor function. The high-resolution imaging allowed the authors not only to observe the fast and slow phases of nuclear movement, but also to detect the surprising backward movement of nuclei (Figure 1A). Also strikingly, the authors observed a fast rolling behavior in a certain percentage of migrating nuclei. The observed bidirectional and rolling movements are proposed to be the mechanism by which a migrating nucleus releases the cytoplasmic pressure created by a build-up of organelles in front of the nucleus in the narrow cell body. Loss of kinesin function leads to a severe nuclear migration defect, including inhibition of the initiation of the movement. In comparison, the effect of losing dynein functions appears to be weaker and specific to certain aspects of the process. In worms with partially disrupted dynein function, the rolling of the nucleus was found to be compromised and the bidirectional nuclear movement was disrupted. By generating a transgene that produces a kinesin–UNC-83 fusion protein, the authors cleverly created a system that presumably retains the connection between kinesin and the nuclear envelope, but not that between