Although assembly of a mitotic spindle is not well understood, the process requires dynamic MTs, plus-end and minus-end directed MT motors, and a plethora of accessory proteins, including static crosslinking molecules [29
]. Kinesin-5 motors localize within the nucleus until nuclear envelope breakdown (NEB) [14
] when they are released into the cytoplasm and accumulate at the midzone and poles of the forming spindle. They are required for maintenance of bipolar spindle integrity and poleward flux of MTs [30
]. In the midzone, Kinesin-5 motors interact with MTs from opposite poles and move at ~20 nm s-1
toward MT plus-ends, effectively pushing spindle poles apart [16
]. Inhibition of Kinesin-5 function by gene knockout, RNAi-mediated depletion, or treatment with pharmacological agents results in monaster formation during spindle assembly [10
]. For example, mutations in Drosophila
KLP61F, a Kinesin-5 ortholog, have been shown to disrupt maintenance of spindle pole separation [31
]. Kinesin-5 proteins act in concert with other spindle associated proteins to maintain proper spindle length. For example, increasing the levels of NuMA, a MT crosslinking protein, restores spindle bipolarity in mitotic extracts lacking Kinesin-5 activity [32
]. In sum, animal Kinesin-5 motors function with other mitotic proteins to generate and maintain pole-to-pole separation by pushing apart antiparallel MTs at the spindle midzone.
The roles of kinesins in the stramenopile lineage have not been functionally investigated. We examined the roles of Kinesin-5 motors during the first cell division cycle in S. compressa
by employing Kinesin-5-specific inhibitors to brown algal zygotes. S. compressa
zygotes treated with monastrol (or STLC) appeared normal and had normal MT arrays in interphase of the first cell cycle, consistent with nuclear localization prior to NEB. Unfortunately, we were unable to confirm nuclear localization since available antibodies to animal Kinesin-5-proteins did not label fucoid zygotes. Importantly, monasters were formed upon entry into mitosis. These findings are similar to reports in animals [10
] and indicate that monastrol binds and inhibits Kinesin-5 motors in S. compressa
. However, monastrol had additional effects on zygotes not observed in animal cells; in addition to monasters, multipolar spindles and cytasters were formed at mitotic entry.
Multipolar spindles were formed at a much higher frequency than monasters, indicating that spindle poles do not fully collapse when Kinesin-5 motors are inhibited. This is unrelated to monastrol dosage since increasing concentration did not significantly increase monaster frequency. Instead, zygotes probably possess other mechanisms that work in concert with Kinesin-5 to maintain pole separation. There may be MT-based motors with overlapping function or MT crosslinking proteins, such as NuMA, that continue to function in the presence of monastrol, maintaining some pole separation. Centrosome position at the onset of mitosis may also contribute to the relatively low monaster frequency. Brown algal centrosomes are fully separated on the nuclear envelope prior to entry into mitosis, residing about 15 μm apart [33
]. Animal centrosomes, however, separate concurrent with NEB so spindle poles are still close together when Kinesin-5 motors become active [34
]. The greater separation of algal spindle poles may reduce the likelihood of complete spindle collapse to a monaster during drug treatment.
The presence of multipolar spindles also implies that supernumerary spindle poles are formed at entry into mitosis, and the extra poles may be derived from cytasters. Although the composition of fucoid cytasters is unknown, they are distinct small astral-like radial bursts of MTs. These are commonly associated with centrin labeling in other systems, and are likely MTOCs [22
]. The observation that nearly all zygotes with multipolar spindles additionally displayed cytasters, while cytasters were only present in about half of cells with monasters, suggests a causal a link between cytasters and supernumerary spindle poles. We speculate that monastrol treatment leads to spindle pole breakup in S. compressa
and fragments of spindle poles nucleate MTs and become cytasters. Numerous cytasters are located throughout the cytoplasm of treated cells, and occasionally one residing close to condensed chromatin captures chromosomes, thereby becoming a supernumerary spindle pole. In this model, Kinesin-5 motors must organize and maintain the integrity of spindle poles. Kinesin-5 members residing at spindle poles have been postulated to bundle long MTs [35
], sort numerous short MTs in the cloud of spindle pole proteins [16
], and/or mediate attachment to a hypothesized scaffold-like spindle matrix [35
]. In fucoid zygotes, one or more of these putative functions may be needed to hold pole components together.
Interestingly, the vast majority of zygotes treated with monastrol had aberrant spindles that were displaced from a central cellular position toward the rhizoid. Proper alignment of the mitotic spindle in brown algae has been shown to be a MT-dependent process [27
], and we observed that condensed chromatin was displaced in zygotes treated with paclitaxel or oryzalin (data not shown). MTs that position the fucoid spindle are thought to do so by interacting with the cell cortex before and after metaphase [26
]. Perhaps the dramatically prolonged metaphase during monastrol treatment permits unregulated spindle movement, resulting in aberrant spindle position. Even so, it is not clear why the spindles preferentially drift in the rhizoid direction.
Cytokinesis bisects the spindle in fucoid zygotes [27
], so it was not surprising to find abnormal divisions following monastrol treatment. The preponderance of multipolar spindles likely resulted in many of the abnormal divisions. Some zygotes with multipolar spindles must still progress through mitosis because the vast majority of cells treated with 25 μM monastrol display multipolar spindles at 24 h AF but completed cell division by 48 h AF. The spindle assembly checkpoint apparently does not monitor spindle pole number. Likewise, cytokinesis proceeds in sea urchin zygotes and in vertebrate somatic cells possessing supernumerary spindle poles [36
]. Since it is unlikely that monaster spindles could achieve equivalent kinetochore tension on chromosomes, these zygotes probably remained arrested in metaphase and failed to divide. Therefore, most abnormal division planes in monastrol-treated S. compressa
zygotes are likely due to multipolar or abnormally positioned spindles that achieve balanced kinetochore attachments, permitting cell cycle progression and aberrant placement of the cytokinetic plane.
Future studies will be aided by an ongoing genomic sequencing project of closely related Ectocarpus siliculosis  and by creation of an EST database for fucoid algae. This information will permit isolation of S. compressa Kinesin-5 sequences for use in molecular investigations and antibody production, and may provide insight into what elements, beyond MTs, are associated with cytasters.