View metadata, citation and similar papers at core.ac.uk brought to you by CORE

R754 Dispatch provided by Elsevier - Publisher Connector

Cytoskeleton: Anatomy of an organizing center Laura G. Marschall and Tim Stearns

One component of the yeast spindle pole body, Spc42p, SPB are duplicated each cycle, such that a new has been found to form a crystalline array within one of is created next to the existing one. Unlike DNA the central layers of the structure; the Spc42p crystal replication, where the structure of the molecule immedi- might provide a scaffold around which the spindle pole ately suggests a templated mechanism of replication, body is assembled, and could be involved in regulating there is no obvious template in either the or the size of the spindle pole body. SPB. Two recent papers suggest that the SPB might be assembled from the inside out, starting with a central Address: Department of Biological Sciences, Stanford University, Stanford, California 94305, USA. crystalline array of a single protein [1], to which other components are added [2]. Current Biology 1997, 7:R754–R756 http://biomednet.com/elecref/09609822007R0754 Early studies showed that, when viewed by thin-section © Current Biology Ltd ISSN 0960-9822 electron microscopy, the SBP appears to consist of three layers (reviewed in [3]). The central plaque is embedded Most cells in your body have an extensive in the , the inner plaque is on the nuclear network that is busily transporting and vesicles side of the central plaque, and the outer plaque is on the from one place to another, and separating chromosomes in cytoplasmic side of the central plaque. and meiosis. For these microtubule ‘tracks’ to be nucleated from the outer plaque of the SPB extend into effective, they must be present in the right number and the and are involved in orienting the nucleus orientation at the right time. The temporal and spatial during cell division and mating (reviewed in [4]). Micro- organization of microtubules is determined, in part, by the tubules nucleated from the inner plaque of the SPB form microtubule organizing center, a specialized organelle that the mitotic and meiotic spindles [4]. exists in one copy per cell. In fungi, the microtubule organizing center is the spindle pole body (SPB), a laminar Recent cryoelectron microscopy studies by Bullitt and structure embedded in the nuclear envelope. In animal coworkers [1] have now been able to resolve at least six cells, the main microtubule organizing center is the cen- layers within the SPB. Starting from the cytoplasmic face trosome, a nucleus-associated organelle that consists of of the SPB and moving toward the nuclear face, these and pericentriolar material. Although the SPB layers are as follows. The first layer is the outer plaque, and centrosome are structurally different, they are func- which interfaces with the cytoplasmic microtubules. Then tionally equivalent: both nucleate the polymerization of come two intermediate layers, IL1 and IL2. The fourth microtubules, anchor those microtubules, and duplicate layer is the central plaque, which spans the nuclear once per . envelope. Fifth is a beaded layer of the inner plaque (IP1). Sixth and last is the capped microtubule layer (IP2), Two aspects of the assembly of these organizing centers which anchors the nuclear microtubules (Figure 1). An are remarkable. First, both the centrosome and the SPB unexpected finding was that there is considerable have the capacity to change size and nucleation capacity variation in the lateral dimension — in the plane of the during the course of the cell cycle, and during develop- nuclear envelope — of the SPB, whereas the vertical ment. For example, the centrosome nucleates a radial dimension — through the layers — remained constant. array of microtubules in interphase, becomes larger and nucleates more microtubules at the onset of mitosis, then To get more information about SPB structure from their becomes smaller and nucleates few microtubules at samples, Bullitt et al. [1] used tomographic reconstruction telophase, finally returning to its interphase state. The to examine heparin-extracted SPBs. In this technique, a Saccharomyces cerevisiae SPB does not appear to go through series of electron microscope images is taken, each with such radical changes in size and nucleation capacity over the sample tilted a different amount with respect to the the course of a cell cycle, but it can change size in electron beam. These images are then reconstructed by response to cell ploidy. The SPB in diploid cells is computer, giving a three-dimensional representation of approximately twice as large as that in haploid cells, pre- the object. The reconstructed object can be ‘sectioned’ by sumably because the SPB must organize more micro- computer, revealing aspects of structure that may not be tubules for mitosis in the diploid cell. visible in standard thin-section electron microscopy. This technique has been used to great effect on the animal The second remarkable aspect of organizing center centrosome [5,6], resulting in the identification of putative assembly is the way that both the centrosome and the microtubule-nucleating structures in the pericentriolar Dispatch R755

Figure 1

Schematic representation of the spindle pole body (SPB). The representation is diagrammatic and is not meant to imply stoichiometric relationships amongst the Cytoplasmic (a) (b) microtubules proteins that comprise the SPB. The inset Tub4p, shows a hexagonal layer of Spc42p within the Spc97p, Spc98p SPB core: (a) a slice through the tomographic Outer plaque reconstruction of an SBP core; (b) computed IL1 Spc42p (c) Fourier transform of (a); (c) two-dimensional IL2 projection map of (a) with imposed symmetry. Central plaque 33 (Inset reproduced with permission from [1]). Nuclear envelope Calmodulin Spc110p Inner plaque IP1 Spc97p, Spc98p, IP2 Tub4p

Nuclear/spindle microtubules Spc97p Spc98p

Nucleus Tub4p

Current Biology

material. Heparin-extraction of SPBs yields a core changes the distance between the inner and central structure that electron tomographic sectioning showed plaques [8], again indicating a vertical orientation of the contains a hexagonal lattice in the lateral dimension, corre- protein within the layer. This type of organization would sponding to layer IL2 [1] (Figure 1, inset). neatly explain the invariant vertical architecture of the SPB, as the layers and the connections between layers The presence of this crystalline hexagonal lattice immedi- would be defined by the intrinsic length of the major ately suggested to Bullitt et al. [1] that this layer might be protein(s) making them up. Also, the lateral dimension of made up of Spc42p, a SPB protein that had been shown the SPB could easily be varied in response to ploidy by previously to form a two-dimensional sheet when over- adding or subtracting subunits from the layers. The use of expressed in yeast [7]. This Spc42p sheet formed around proteins as biological ‘rulers’ to determine the length of the SPB, as if it were an extension of an existing layer of structures was first indicated in studies of the genetic the SPB. When Bullitt et al. [1] used electron microscopy control of the length of the tail of phage lambda [9], to examine these Spc42p sheets, they found a crystalline although in that case the ruler protein, the gene H protein, lattice with a hexagonal structure very similar to that in is probably not a main structural component of the tail. the IL2 layer, suggesting that IL2 may comprise, at least in part, hexagonally packed Spc42p. The Spc42p crystal There are some unanswered questions, of course. If a resulting from overexpression of the protein could be crystal of Spc42p is the main determinant of the lateral imaged at a much higher resolution than the SPB core, dimension of the SPB, then how is its size controlled? allowing predictions about the orientation of Spc42p in Similarly, new Spc42p subunits would be likely to add the layer. Spc42p is predicted to have a coiled-coil central preferentially to an existing crystalline array of other domain [7], and edge-on views of the SPB showed vertical Spc42p molecules rather than start a new crystal, so how striations, suggesting that the rod-like coiled-coil domain can a new SPB ever be initiated? Bullitt et al. [1] note that is oriented vertically in the crystal. there appears to be a ring of protein in the central plaque outside of the central core of Spc42p, and suggest that What role does this crystalline layer of Spc42p play in the perhaps this regulates crystal growth. This, of course, only assembly of the complex SPB? Bullitt et al. [1] propose pushes the issue back a step — what is this ring made of, that the Spc42p layer is the base structure of the SPB, and and how is its size determined? The duplication question that other layers of the SPB are added to that. Further, is more perplexing. The duplication cycle of the SPB has they propose that each of the visible layers of the SPB is been described in detail [10], and there are specific struc- made up of coiled-coil proteins interacting laterally. This tures associated with duplication, but there is nothing to is supported by work on another SPB component, indicate how it is that a new layer is initiated. It is possible Spc110p, which is also a coiled-coil protein, and which that one of the SPB structures associated with duplication, links the inner plaque to the central plaque. Remarkably, the half-bridge, nucleates a new crystal of Spc42p adjacent changing the length of the coiled-coil domain of Spc110p to the existing SPB. R756 Current Biology, Vol 7 No 12

If we assume that a central crystal of Spc42p is the core of the nuclear side of the central plaque. It seems likely that the SPB, then how is the rest of the SPB assembled onto the outer layers will be organized using the same princi- that core? Recent work by Knop and Schiebel [2] provides ples revealed in the studies reviewed here, but there may a possible molecular basis for SPB assembly and organiza- still be some surprises in this remarkable structure. tion. They extracted yeast cells with a high salt buffer and identified a complex containing four proteins: Spc42p, References Spc110p, calmodulin and a 35 kDa protein whose identity 1. Bullitt E, Rout MP, Kilmartin JV, Akey CW: The yeast spindle pole body is assembled around a central crystal of Spc42p. Cell 1997, has not been determined. As described above, Spc110p is 89:1077-1086. the coiled-coil protein that links the central and inner 2. Knop M, Schiebel E: Spc98p and Spc97p of the yeast g - complex mediate binding to the spindle pole body via their plaques. Calmodulin binds to the carboxyl terminus of interaction with Spc110p. EMBO J 1997, in press. Spc110p and is found at the central plaque [11–13], thus it 3. Byers B: Cytology of the yeast life cycle. In The Molecular Biology is the carboxy-terminus of the rod-like Spc110p that is of the Yeast Saccharomyces: Life Cycle and Inheritance. Edited by Strathern JN, Jones EW, Broach JR. Cold Spring Harbor, New York: oriented towards the central plaque. Although it was not Cold Spring Harbor Laboratory Press; 1981:59-96. possible to demonstrate a direct interaction between 4. Botstein D, Amberg D, Huffaker T, Mulholland J, Adams A, Drubin D, Spc110p and Spc42p, the finding that the two proteins can Stearns T: The yeast . In The Molecular and Cellular Biology of the Yeast Saccharomyces cerevisiae. Cell Cycle and Cell be isolated in a complex suggests that the carboxyl termi- Biology. Edited by Pringle J, Broach J, Jones E. Cold Spring Harbor, nus of Spc110p interacts with the Spc42p layer, possibly New York: Cold Spring Harbor Press; 1997:1-90. 5. Moritz M, Braunfeld MB, Fung JC, Sedat JW, Alberts BM, Agard DA: through the unidentified 35 kDa protein [2]. Three-dimensional structural characterization of from early Drosophila embryos. J Cell Biol 1995, 130:1149-1159. 6. Moritz M, Braunfeld MB, Sedat JW, Alberts B, Agard DA: Microtubule All microtubule organizing centers nucleate the polymer- γ α β nucleation by -tubulin-containing rings in the centrosome. Nature ization of - and -tubulin into microtubules. Nucleation 1995, 378:638-640. is thought to be brought about by γ -tubulin, a specialized 7. Donaldson AD, Kilmartin JV: Spc42p: a phosphorylated component tubulin that is localized to the parts of microtubule orga- of the S. cerevisiae spindle pole body (SPB) with an essential function during SPB duplication. J Cell Biol 1996, 132:887-901. nizing centers that are in contact with microtubules, the 8. Kilmartin JV, Dyos SL, Kershaw D, Finch JT: A spacer protein in the inner and outer plaques in the SPB. Thus, any model of Saccharomyces cerevisiae spindle pole body whose transcript is γ cell cycle-regulated. J Cell Biol 1993, 123:1175-1184. SPB assembly must ultimately explain how -tubulin is 9. Katsura I, Hendrix RW: Length determination in bacteriophage connected to the other parts. Knop and Schiebel [2] lambda tails. Cell 1984, 39:691-698. present evidence that it is the amino terminus of Spc110p 10. Byers B, Goetsch L: Behavior of spindles and spindle plaques in the cell cycle and conjugation of Saccharomyces cerevisiae. J that links γ -tubulin to the SPB. In yeast, γ -tubulin is Bacteriol 1975, 124:511-523. encoded by the TUB4 gene [14–16], and Tub4p is found 11. Geiser JR, Sundberg HA, Chang BH, Muller EGD, Davis TN: The essential mitotic target of calmodulin is the 110-kilodalton as a complex with two other SPB proteins, Spc97p and component of the spindle pole body in Saccharomyces cerevisiae. Spc98p [17,18]. Both genetic and biochemical data indi- Mol Cell Biol 1993, 13:7913-7924. cated that the amino-terminal 200 amino acids of Spc110p 12. Stirling DA, Welch KA, Stark MJ: Interaction with calmodulin is required for the function of Spc110p, an essential component of are able to interact with Spc97p and Spc98p, but not the yeast spindle pole body. EMBO J 1994, 13:4329-4342. Tub4p [2]. 13. Spang A, Grein K, Schiebel E: The spacer protein Spc110p targets calmodulin to the central plaque of the yeast spindle pole body. J Cell Sci 1996, 109:2229-2237. A simple model for the basic organization of the SPB, 14. Sobel SG, Snyder M: A highly divergent gamma-tubulin gene is then, is that Spc42p forms a crystalline layer at the center essential for cell growth and proper microtubule organization in Saccharomyces cerevisiae. J Cell Biol 1995, 131:1775-1788. of the organelle, the carboxyl terminus of Spc110p inter- 15. Marschall L, Jeng R, Mulholland J, Stearns T: Analysis of Tub4p, a acts with Spc42p, and the amino terminus of Spc110p yeast gamma-tubulin-like protein: implications for microtubule interacts with the Tub4p (γ -tubulin) complex, which can organizing center function. J Cell Biol 1996, 134:443-454. 16. Spang A, Geissler S, Grein K, Schiebel E: γ -Tubulin-like Tub4p of interact with tubulin to nucleate microtubule polymeriza- Saccharomyces cerevisiae is associated with the spindle pole tion. An interesting corollary of this model is that it neatly body substructures that organize microtubules and is required for mitotic spindle formation. J Cell Biol 1996, 134:429-441. explains the invariant vertical structure of the SPB as 17. Geissler S, Pereira G, Spang A, Knop M, Soues S, Kilmartin J, being determined by the interactions of the component Schiebel E: The spindle pole body component Spc98p interacts proteins, and allows for changes in the lateral size of the with the gamma-tubulin-like Tub4p of Saccharomyces cerevisiae at the sites of microtubule attachment. EMBO J 1996, 15:3899- SPB by simply adding more protein assemblies to the 3911. existing structure. 18. Knop M, Pereira G, Geissler S, Grein K, Schiebel E: The spindle pole body component Spc97p interacts with the gamma-tubulin of Saccharomyces cerevisiae and functions in microtubule The SPB nucleates microtubules both from its nuclear organization and spindle pole body duplication. EMBO J 1997, and cytoplasmic faces. If Spc110p acts as the docking site 16:1550-1564. for the Tub4p-containing complex on the nuclear face of the SPB, then there must be a second protein that per- forms this function on the cytoplasmic face. Tub4p, Spc97p and Spc98p localize to the outer plaque as well as to the inner plaque; however, Spc110p is only found on