Concentrating on the mitotic spindle

In eukaryotes, the microtubule-based spindle drives chromosome segregation. In this issue, Schweizer et al. (2015; J. Cell Biol. http://dx.doi.org/10.1083/jcb.201506107) find that the spindle area is demarcated by a semipermeable organelle barrier. Molecular crowding, which is microtubule independent, causes the enrichment and/or retention of crucial factors in the spindle region. Their results add an important new feature to the models of how this structure assembles and is regulated.Mitosis is marked by the assembly of the mitotic spindle, a microtubule-based structure that facilitates accurate chromosome segregation. Many biochemical reactions are coupled to spindle assembly, from tubulin polymerization itself to the mitotic checkpoint, which inhibits chromosome disjunction until all the chromosomes are properly attached and aligned (Cleveland et al., 2003). Interestingly, these reactions are virtually unaltered over a broad morphological range; large spindles and small spindles follow roughly the same biochemical rules despite quite distinct geometries (Brown et al., 2007; Wühr et al., 2008). In this issue, Schweizer et al. report that the mitotic spindle area is delineated by membrane-bound organelles, generating a “spindle envelope” with unique molecular constituents compared with the surrounding cytoplasm. Spindle envelope–based molecular crowding provides an enticing hypothetical solution to the broad problem of confining mitotic biochemistry to a specific cellular space irrespective of cell size.Schweizer et al. (2015) used FRAP and fluorescence correlation spectroscopy (FCS) to measure the mobility of specific proteins. The authors found that tubulin and the Mad2 spindle assembly checkpoint protein were enriched in the spindle area in a microtubule polymer-independent manner. Given that the mobility of these proteins outside and within the spindle area was the same, changes in local concentration were likely a result of a barrier effect. In support of this hypothesis, modeling their FCS data with a fenestrated barrier separating the spindle area from the cytoplasm reproduced the measured FCS results. To test if a barrier surrounding the spindle area was important for spindle function, the authors disrupted the envelope area by laser microsurgery and found chromosome segregation errors consistent with defects in spindle assembly and kinetochore attachment monitoring. Thus, spindle envelope–based concentration of basal components in two critical spindle reactions, spindle assembly and mitotic checkpoint signaling, could mechanistically catalyze cell division.Molecular crowding can catalyze reactions and stabilize proteins by altering the local concentration of one or more rate-limiting components and is best characterized by membrane-bound organelles. Crowding by aggregation can greatly increase biochemical reactions without the need for a contiguous membrane (Weber and Brangwynne, 2012; Brangwynne, 2013) and this phenomenon can regulate cell cycle states (Lee et al., 2013). Here, Schweizer et al. (2015) propose that by creating a membrane-bound organelle exclusion zone, a spindle envelope could cause the molecular crowding of important spindle proteins and thereby their enrichment in the spindle area.The mitotic spindle is known to scale with cell size: smaller cells have smaller spindles (Levy and Heald, 2012). Spindle size scaling is prominent during development when repeated cell division without embryonic growth results in cells that can be several orders of magnitude smaller than that of the zygote. Recently, cytoplasm volume and tubulin concentration was shown to be an important factor in spindle size scaling; however, a curious exception to the size scaling rule is that there seems to be an upper limit to spindle size, resulting in stable spindle size when a threshold cell size is reached (Wühr et al., 2008; Good et al., 2013; Hazel et al., 2013). A spindle envelope would provide mechanisms to maintain increased local tubulin concentration independent of the absolute amount available in the cell. The net effect would be that spindle size scales in very large cells to the spindle envelope size rather than cell size in a manner analogous to chromosome size scaling to nuclear size independently of cell size (Fig. 1 A). Clearly this is a more complex problem and factors such as tubulin protein production and polymerization cofactors (such as the Tog family of proteins) clearly play an important role (Slep, 2009). However, spindle envelope–based molecular crowding could provide an elegant solution to a biochemical problem.Open in a separate windowFigure 1.Molecular crowding in the mitotic spindle. Schematic view of how a spindle envelope could mitigate spindle size scaling during development (A) or cell cycle control (B) in a common cytoplasm. (A) A spindle envelope (black dotted line) that excludes large membrane-bound organelles (yellow) could locally increase the concentration of spindle proteins (depicted as a red background) such as tubulin to control spindle size independent of cell size during early development. (B) Neighboring nuclei in a common cytoplasm (e.g., the syncytial mitotic gonad in C. elegans) could have differing mitotic states by restricting the diffusion of important regulatory proteins such as Mad2 (similar coloring as in A).A spindle envelope could also provide a cell biological solution to another developmental problem—independent cell cycle control of separate nuclei within a single cytoplasm (syncytia). For example, the mitotic region of the Caenorhabditis elegans germline contains germ cell precursors that divide independently of one another in a common cytoplasm. In some cases, two neighboring dividing nuclei can have different biochemistry, one arrested in metaphase because of a kinetochore microtubule attachment defect while the other is progressing into anaphase (Gerhold et al., 2015). By restricting the diffusive radius of signaling molecules like Mad2, a steep threshold of checkpoint activity can be maintained, allowing independent cell cycle control even in a common cytoplasm (Fig. 1 B).The mitotic spindle has long been known to exclude large membrane-bound organelles, even in the absence of microtubule polymer, leading to a hypothesized nontubulin-based “spindle matrix.” A spindle matrix would be an excellent candidate to underlie the spindle envelope. The molecular nature of a spindle matrix, however, has never been agreed upon with candidate mechanisms ranging from nonprotein macromolecules to actin (Pickett-Heaps et al., 1984; Chang et al., 2004). A convincing argument can be made that the Skeletor/Megator/Chromator proteins first identified in Drosophila melanogaster constitute a spindle matrix (Walker et al., 2000; Qi et al., 2004; Rath et al., 2004; Schweizer et al., 2014). These proteins are large and are found in the nucleus in interphase and as microtubule-independent fibrous structures in and around the spindle in mitosis. Depletion of these proteins results in mitotic errors; however, these may or may not be caused by a role as the spindle matrix.Schweizer et al. (2015) evaluated Megator (as a representative member of the complex) as a possible basis for generating the spindle envelope. FRAP and FCS showed that, like tubulin and Mad2, Megator is concentrated in the spindle envelope region independent of microtubules. However, Megator in the spindle region had slower diffusive properties compared with that around the cell periphery. Thus, unlike tubulin and Mad2, the mobility of Megator within the spindle was altered, indicating that Megator likely forms a high molecular weight complex with its binding partners Skeletor and Chromator in the spindle area, which may help form the spindle envelope.The sum of these results lead to a possible model whereby the Skeletor/Megator/Chromator proteins complex together and subsequently support a spindle envelope independent of microtubules. The spindle envelope excludes large membrane-bound organelles, leading to increased concentration of mitotic reaction constituents and thus ultimately catalyzing cell division. It will be exciting in the future to determine if the spindle area is indeed subject to molecular crowding in the purest of forms (solvent exclusion) and how this effect drives cell division.     

Ryanodine and contractile phenomena in the mitotic spindle

Spontaneous contraction of mouse heart muscle in tissue culture is stopped by ryanodine at 1–2 mg/ml. These concentrations also block mitosis. In the dividing Echinarachnius egg, anaphase movement of chromosomes is retarded and incomplete in ryanodine at 6–8 mg/ml, although some of the eggs complete 2 or 3 cleavages. Increasing the length of exposure prior to cleavage does not alter the character of response and only slightly increases the severity. These phenomena are consistent with an inhibition of traction fiber contraction by ryanodine, but are insufficient to demonstrate the existence of such a contractile phenomenon.     

Mechanisms and molecules of the mitotic spindle

In all eukaryotes, morphogenesis of the microtubule cytoskeleton into a bipolar spindle is required for the faithful transmission of the genome to the two daughter cells during division. This process is facilitated by the intrinsic polarity and dynamic properties of microtubules and involves many proteins that modulate microtubule organization and stability. Recent work has begun to uncover the molecular mechanisms behind these dynamic events. Here we describe current models and discuss some of the complex repertoire of factors required for spindle assembly and chromosome segregation.     

Bi-orienting chromosomes on the mitotic spindle

For the proper segregation of sister chromatids before cell division, each sister kinetochore must attach to microtubules that extend to opposite spindle poles. This process is called bipolar microtubule attachment or chromosome bi-orientation. The mechanism for chromosome bi-orientation lies at the heart of chromosome segregation, but is still poorly understood. Recent studies suggest that cells can promote bi-orientation by re-orienting kinetochore-spindle pole connections.     

Chromosome bi-orientation on the mitotic spindle

For proper chromosome segregation, sister kinetochores must attach to microtubules extending from opposite spindle poles prior to anaphase onset. This state is called sister kinetochore bi-orientation or chromosome bi-orientation. The mechanism ensuring chromosome bi-orientation lies at the heart of chromosome segregation, but is still poorly understood. Recent evidence suggests that mal-oriented kinetochore-to-pole connections are corrected in a tension-dependent mechanism. The cohesin complex and the Ipl1/Aurora B protein kinase seem to be key regulators for this correction. In this article, I discuss how cells ensure sister kinetochore bi-orientation for all chromosomes, mainly focusing on our recent findings in budding yeast.     

Gradients in the self-organization of the mitotic spindle

Recent evidence points at a role of protein interaction gradients around chromatin in mitotic spindle morphogenesis in large eukaryotic cells. Here, we explain how gradients can arise over distances of tens of microns around supramolecular structures from mixtures of soluble molecules. We discuss how coupled sets of such reaction diffusion processes generate the spatial information that determines the local dynamics of microtubules required to form a bipolar spindle. We argue that such reaction diffusion processes are involved in the self-organization of supramolecular structures in the cell.     

Centrosome-independent mitotic spindle formation in vertebrates

BACKGROUND: In cells lacking centrosomes, the microtubule-organizing activity of the centrosome is substituted for by the combined action of chromatin and molecular motors. The question of whether a centrosome-independent pathway for spindle formation exists in vertebrate somatic cells, which always contain centrosomes, remains unanswered, however. By a combination of labeling with green fluorescent protein (GFP) and laser microsurgery we have been able to selectively destroy centrosomes in living mammalian cells as they enter mitosis. RESULTS: We have established a mammalian cell line in which the boundaries of the centrosome are defined by the constitutive expression of gamma-tubulin-GFP. This feature allows us to use laser microsurgery to selectively destroy the centrosomes in living cells. Here we show that this method can be used to reproducibly ablate the centrosome as a functional entity, and that after destruction the microtubules associated with the ablated centrosome disassemble. Depolymerization-repolymerization experiments reveal that microtubules form in acentrosomal cells randomly within the cytoplasm. When both centrosomes are destroyed during prophase these cells form a functional bipolar spindle. Surprisingly, when just one centrosome is destroyed, bipolar spindles are also formed that contain one centrosomal and one acentrosomal pole. Both the polar regions in these spindles are well focused and contain the nuclear structural protein NuMA. The acentrosomal pole lacks pericentrin, gamma-tubulin, and centrioles, however. CONCLUSIONS: These results reveal, for the first time, that somatic cells can use a centrosome-independent pathway for spindle formation that is normally masked by the presence of the centrosome. Furthermore, this mechanism is strong enough to drive bipolar spindle assembly even in the presence of a single functional centrosome.     

Dynactin function in mitotic spindle positioning

Dynactin is a multisubunit protein complex necessary for dynein function. Here, we investigated the function of dynactin in budding yeast. Loss of dynactin impaired movement and positioning of the mitotic spindle, similar to loss of dynein. Dynactin subunits required for function included p150^Glued, dynamitin, actin-related protein (Arp) 1 and p24. Arp10 and capping protein were dispensable, even in combination. All dynactin subunits tested localized to dynamic plus ends of cytoplasmic microtubules, to stationary foci on the cell cortex and to spindle pole bodies. The number of molecules of dynactin in those locations was small, less than five. In the absence of dynactin, dynein accumulated at plus ends and did not appear at the cell cortex, consistent with a role for dynactin in offloading dynein from the plus end to the cortex. Dynein at the plus end was necessary for dynactin plus-end targeting. p150^Glued was the only dynactin subunit sufficient for plus-end targeting. Interactions among the subunits support a molecular model that resembles the current model for brain dynactin in many respects; however, three subunits at the pointed end of brain dynactin appear to be absent from yeast.     

The mitotic spindle and actin tails

To segregate their chromosomes, eukaryotic cells rely on a dynamic structure made of microtubules: the mitotic spindle. This structure can form in cells lacking centrosomes, because their chromosomes also nucleate microtubules. This second assembly pathway is observed even in some cells that naturally have centrosomes, for example when the centrosomes are ablated by laser surgery. Recent results have started to address the complementary question of whether centrosome-nucleated microtubules alone could sustain the formation of a functional mitotic spindle. We wonder in this respect whether lower eukaryotes such as yeasts are different from higher eukaryotes such as vertebrates.     

Kinetochore capture and bi-orientation on the mitotic spindle

Kinetochores are large protein complexes that are formed on chromosome regions known as centromeres. For high-fidelity chromosome segregation, kinetochores must be correctly captured on the mitotic spindle before anaphase onset. During prometaphase, kinetochores are initially captured by a single microtubule that extends from a spindle pole and are then transported poleward along the microtubule. Subsequently, microtubules that extend from the other spindle pole also interact with kinetochores and, eventually, each sister kinetochore attaches to microtubules that extend from opposite poles - this is known as bi-orientation. Here we discuss the molecular mechanisms of these processes, by focusing on budding yeast and drawing comparisons with other organisms.     

Polarity of microtubules of the mitotic spindle

Microtubules are polar structures that grow preferentially at one end. Measurement of their rate of directional growth can be used as a polarity indicator to determine their orientation with respect to a nucleation site. The results are interpreted to signify that the microtubules originating from the centrosomes and chromosomes of the mitotic spindle are antiparallel to each other.     

Cell division: AAAtacking the mitotic spindle

Cell division requires the assembly of a microtubule-based spindle which captures and segregates sister chromatids. But how is this spindle broken down once chromosome segregation is complete? New evidence implicates a highly conserved AAA-ATPase in spindle disassembly at the end of mitosis.