Mal3 Masks Catastrophe Events in Schizosaccharomyces pombe Microtubules by Inhibiting Shrinkage and Promoting Rescue

Schizosaccharomyces pombe Mal3 is a member of the EB family of proteins, which are proposed to be core elements in a tip-tracking network that regulates microtubule dynamics in cells. How Mal3 itself influences microtubule dynamics is unclear. We tested the effects of full-length recombinant Mal3 on dynamic microtubules assembled in vitro from purified S. pombe tubulin, using dark field video microscopy to avoid fluorescent tagging and data-averaging techniques to improve spatiotemporal resolution. We find that catastrophe occurs stochastically as a fast (<2.2 s) transition from constant speed growth to constant speed shrinkage with a constant probability that is independent of the Mal3 concentration. This implies that Mal3 neither stabilizes nor destabilizes microtubule tips. Mal3 does, however, stabilize the main part of the microtubule lattice, inhibiting shrinkage and increasing the frequency of rescues, consistent with recent models in which Mal3 on the lattice forms stabilizing lateral links between neighboring protofilaments. At high concentrations, Mal3 can entirely block shrinkage and induce very rapid rescue, making catastrophes impossible to detect, which may account for the apparent suppression of catastrophe by Mal3 and other EBs in vivo. Overall, we find that Mal3 stabilizes microtubules not by preventing catastrophe at the microtubule tip but by inhibiting lattice depolymerization and enhancing rescue. We argue that this implies that Mal3 binds microtubules in different modes at the tip and on the lattice.Microtubules are intrinsically dynamic self-assembling structures of tubulin subunits (1) whose polymerization is subject to extensive spatial and temporal control in cells partly through the activity of microtubule-associated proteins (2). In cells, the EB family of microtubule plus end-tracking proteins (+TIPs)² localizes at the plus end of growing but not shrinking microtubules. EB depletion increases catastrophe frequency and reduces microtubule length in many species (3–5), suggesting that EBs suppress microtubule catastrophes. It is, however, unclear from these cellular studies whether this activity is direct or indirect because the dynamic binding of EBs to other +TIPs proteins enhances the localization of all EB complex proteins, including EB1, to microtubule ends (6).To determine the direct effect of EB family proteins on microtubule dynamics, in vitro experiments are necessary. These have established that microtubule end tracking is an intrinsic property of the EB proteins and that other +TIP proteins such as CLIP170 are dependent upon EBs for their microtubule end localization (7–9). However, EB1 binding also directly alters the structure of growing microtubule tips (10). In vitro studies show that Mal3, the EB1 homologue in Schizosaccharomyces pombe, can also affect the structure of microtubules. Sandblad et al. (11) found localization of Mal3 along the (A-lattice) seam of B-lattice microtubules and proposed this as a potential mechanism for direct microtubule stabilization by the EBs. Des Georges et al. (12) showed that Mal3 binds to and specifically stabilizes the A-lattice protofilament overlap, promoting nucleation and assembly of A-lattice-containing microtubules.Several studies in vitro have all shown that EBs can affect microtubule dynamics (4, 7, 10, 13) but conflict over which parameter is affected. Thus although Bieling et al. (7) and Manna et al. (13) observed no effect on microtubule growth rates, Komarova et al. (4) and Vitre et al. (10) found an acceleration of growth. Manna et al. (13) found that EB1 inhibits catastrophe, yet the other studies observed that EBs trigger catastrophe events. There is clearly a need to resolve these apparent conflicts, especially as the same proteins in vivo appear to suppress catastrophe.To try to elucidate the mechanism by which EB proteins influence microtubule assembly, we developed a minimalist approach in which the potential for confounding factors to affect the data is reduced or eliminated. Our assay uses proteins from a single organism, S. pombe, and GMPCPP-stabilized microtubule seeds assembled from purified tubulin with only the seeds attached to the chamber surface. We used this system to measure the effects of unlabeled full-length Mal3 on the polymerization dynamics of unlabeled S. pombe microtubules. Microtubules were imaged using dark field microscopy to avoid fluorescent labeling (see Fig. 1A). We also developed a semiautomated analysis system that allows us to digitize a large number of events, which can then be processed by data averaging and filtering. This reduces noise, allowing us to examine the detailed kinetics of the catastrophic switch from growth to shrinkage. Using this system, we find that Mal3 has no direct effect upon the frequency or kinetics of catastrophe events but that it does reduce shrinkage rates and increase rescue frequency in a dose-dependent manner.Open in a separate windowFIGURE 1.In vitro S. pombe microtubule dynamics assay. A, schematic diagram of S. pombe microtubule dynamics assay. GMPCPP stabilized polarity-marked microtubule seed assembled from Alexa Fluor 488- and Alexa Fluor 680-labeled pig brain tubulin. Only the center of the seed is attached to the surface by anti-Alexa Fluor 488 antibody. Dynamic non-fluorescently labeled S. pombe microtubules grown from seeds were observed by dark field illumination. B, merged fluorescence images of GMPCPP stabilized, polarity-marked pig microtubule seed (pig Alexa-MT). Green, Alexa Fluor 488; red, Alexa Fluor 680. Polarity is indicated by − or +. The plus end of the seed has a longer Alexa Fluor 680-labeled region (upper panel), a dark field image showing pig microtubule seed plus elongated S. pombe microtubules (middle panel), and the merged images (lower panel). Red broken lines show the ends of the seed, and yellow broken lines show the ends of the elongated S. pombe microtubules. Arrows indicate the dynamic S. pombe microtubule elongated from the stabilized microtubule seed. Scale bar: 10 μm. C, kymographs of microtubule length change over time. The left panel shows a diagram of a typical example. Time is indicated by the vertical axis, and length is indicated by the horizontal axis. Rescue (r) and catastrophe (c) events are labeled. Regrowth of shrinking microtubules from the seed (yellow arrow) were not counted as rescues. Scale bars: vertical, 5 min; horizontal, 20 μm. + and − ends of microtubule are indicated. D, enlargement of catastrophe events from the yellow rectangle in C. Scale bars: vertical, 30 s; horizontal, 5 μm.