Differences in Intrinsic Tubulin Dynamic Properties Contribute to Spindle Length Control in Xenopus Species: Current Biology | Current Biology
William G. Hirst, Abin Biswas, Kishore K. Mahalingan, Simone Reber
To show that indeed intrinsic tubulin properties contribute to setting spindle length, the classic experiment would be a depletion/add-back experiment, i.e., deplete tubulin from the extract of one Xenopus species and add back tubulin from the other species. Therefore, we tried to deplete endogenous tubulin from X. laevis egg extracts, add back X. tropicalis tubulin, and assemble spindles. However, we and others (Heald lab and Good lab, personal communication) did not succeed in significantly depleting tubulin from extracts. Although we successfully pulled down tubulin (Figure S4A), we were not able to significantly reduce the extract tubulin concentration (Figure S4B). Therefore, we decided to perform the depletion and add-back experiment in silico using the cytoskeletal simulation software Cytosim (http://www.cytosim.org) [
A computational model predicts Xenopus meiotic spindle organization.
]. Next, we performed an in silico depletion and add-back experiment: we modeled either the X. laevis (Figure S4C) or X. tropicalis (Figure S4D) extract environment and provided our experimentally determined microtubule polymerization velocities for X. laevis and tropicalis microtubules (for parameters, see STAR Methods and Tables S2 and S3). The increase in X. laevis polymerization velocity was sufficient to significantly increase spindle length (Figures S4E and S4G) and spindle mass (Figures S4F and S4H). An alternative way to test the contribution of tubulins to setting spindle length would be the addition of different tubulin species to preassembled, steady-state spindles. We therefore modified the simulation in that we increased the basic microtubule polymerization velocity by the different polymerization velocities measured in vitro (for parameters, see Table S4). Plugging in X. laevis polymerization velocity led to a greater increase in spindle length than X. tropicalis polymerization velocity (Figure 4B). These findings support the idea that an increased microtubule growth velocity is sufficient to increase spindle mass. This is consistent with data showing that an increase in microtubule growth velocity increases Xenopus spindle length [
Microtubule dynamics scale with cell size to set spindle length and assembly timing.
], where the growth rate of spindle microtubules is the primary parameter that decreases proportionally to cell volume and spindle size during early development. Next, to experimentally test whether Xenopus tubulins would affect spindle length and mass, we preassembled spindles in X. laevis egg extracts and added increasing amounts of tubulin from B. taurus, X. tropicalis, and X. laevis (Figure 4C). Although the addition of B. taurus tubulin had no significant effect on spindle length (−0.08%, as previously reported in [
Cytoplasmic volume modulates spindle size during embryogenesis.
]) and mass (Figure S4I), the addition of X. tropicalis and X. laevis tubulin increased spindle length by 21% and by 31%, respectively (Figure 4C). Similarly, we assembled spindles in X. tropicalis egg extracts and added 8 μM of X. tropicalis or X. laevis tubulin (Figure 4D). The addition of X. laevis tubulin led to a 70% increase in spindle length (wild-type spindles: 18.9 ± 3.9 μm; X. laevis spindles: 32.3 ± 5.6 μm) although the addition of X. tropicalis tubulin lead to a 23% increase in spindle length (23.2 ± 3.3 μm). These data show that additional tubulin can increase spindle mass and length. Importantly, tubulin biochemistry and its inherent consequences for dynamic instability are essential to setting steady-state microtubule mass and thus spindle length.