Function and regulation of Maskin, a TACC family protein, in microtubule growth during mitosis

Abstract

The Xenopus protein Maskin has been previously identified and characterized in the context of its role in translational control during oocyte maturation. Maskin belongs to the TACC protein family. In other systems, members of this family have been shown to localize to centrosomes during mitosis and play a role in microtubule stabilization. Here we have examined the putative role of Maskin in spindle assembly and centrosome aster formation in the Xenopus egg extract system. Depletion and reconstitution experiments indicate that Maskin plays an essential role for microtubule assembly during M-phase. We show that Maskin interacts with XMAP215 and Eg2, the Xenopus Aurora A kinase in vitro and in the egg extract. We propose that Maskin and XMAP215 cooperate to oppose the destabilizing activity of XKCM1 therefore promoting microtubule growth from the centrosome and contributing to the determination of microtubule steady-state length. Further more, we show that Maskin localization and function is regulated by Eg2 phosphorylation.

Figure 1.

Localization of endogenous and recombinant Maskin in Xenopus XL177 cells and egg extract. (A) Western blot of XL177 and of mitotic (M) and interphase (I) egg extracts probed with the anti-Maskin antibodies used in this work. (B) Immunofluorescence with anti-Maskin antibodies on spindles and asters assembled in cells and in egg extract. (a and b) Maskin localizes all along spindle microtubules and accumulates at spindle poles. (c) In asters nucleated by centrosomes in egg extract, Maskin accumulates strongly at the center and aligns along the microtubules as small dots. (d) In taxol-induced asters, a small amount of Maskin can be detected at the center. Microtubules are red, Maskin is green, and DNA in the spindle is blue. Bars, 10 μm. (C) Schematic representation of the recombinant proteins added to the egg extract. (D) Representative images of spindles assembled in extracts containing the recombinant proteins (200 μM, final concentration). Anti-GST antibodies were used to detect GST or the GST-tagged recombinant proteins by immunofluorescence. Images from the three spindles were taken with the same camera settings. The GFP-TD protein was visualized directly through GFP fluorescence. Microtubules are red, Maskin and GFP-TD are green, and DNA is blue. Bar, 10 μm.

Figure 2.

Spindle formed in Maskin-depleted extract are smaller. (A) Western blot of mock- and Maskin-depleted egg extract probed with the anti-Maskin antibodies (left). The three successive round of mock and Maskin depletion are shown. After the third round more than 99% of Maskin was depleted. The blot was reprobed with anti-XMAP215 and anti-TPX2 antibodies. There were no visible changes of XMAP215 concentration in the Maskin-depleted extract. The TPX2 band was used as loading control. (Right) Western blot of a typical rescue experiment. M, mock depleted extract; Δ Maskin, Maskin-depleted extract; AdB1, Maskin-depleted extract containing purified GST-Maskin; Adb2, Maskin-depleted extract containing GFP-TD protein. The blot was probed with the anti-Maskin antibody directed against the TACC domain and reprobed with an anti-tubulin antibody. The tubulin band was used as a loading control. (B) Graphs representing the average length (left), width (middle), and tubulin fluorescence intensity (right) of spindles assembled in the four types of egg extract. White bars, mock; dark gray bar, Maskin-depleted; light gray, AdB1; black bar, AdB2. (C) Immunofluorescence of spindles assembled in mock, Maskin-depleted, and AdB1 extracts with the anti-Maskin antibody (green) and GFP fluorescence (green) on spindles assembled in AdB2 conditions. Microtubules are red and DNA is blue. Bar, 10 μm.

Figure 3.

Maskin depletion impairs centrosome aster formation in egg extract. (A) Representative images of asters by purified centrosomes in mock-depleted extract (Mock), Maskin-depleted extract (ΔMaskin), and Maskin-depleted extract containing purified GST-Maskin protein (AdB). The extract was sent to interphase and cycled back into mitosis by addition of cyclin B Δ90. Bar, 10 μM. (B) Graphs showing the average microtubule fluorescence intensity (left) and microtubule length (right) in asters assembled under the three conditions as in A: mock (white bars), Δ Maskin (dark gray bars), Add-Back (light gray bars). (C) Western blot of the egg extracts used in A and B probed with anti-Maskin and anti-tubulin antibodies.

Figure 4.

Maskin binds to microtubules and interacts with XMAP215 and Eg2 in vitro and in the egg extract. (A) Western blot of GST pull-downs in vitro with GST (lane 1) or GST-Maskin (lanes 2–5) in the presence of GFP-Eg2 (lane 3) or His-tagged XMAP215 (lane 4) or both (lane 5). Proteins recovered with the anti-GST antibody–coated beads were run on SDS-PAGE and blotted. The blot was probed with anti-XMAP215, anti-GST (to detect GST-Maskin), and anti-Eg2 antibodies to detect the corresponding proteins as indicated on the right. (B) Immunoprecipitations from CSF-arrested egg extract with different antibodies as indicated. (Lanes E) Total extract. (Lanes 1 and 3) Control immunoprecipitation with unspecific IgG. (Lanes 2 and 4) Immunoprecipitated proteins with specific antibodies as indicated. The blots were probed with anti-XMAP215, anti-Maskin, and anti-Eg2 antibodies to detect the presence of the corresponding proteins as indicated on the right. (C) Western blot analysis of a 0.5-μl total egg extract (lane 1), a microtubule pellet from 10 μl of egg extract incubated with taxol (lane 2), a pellet obtained from 10 μl of extract incubated with nocodazole (lane 3). The blot was probed with specific antibodies to detect the presence of the corresponding proteins as indicated on the right. Endogenous XMAP215, Maskin, and Eg2 copellet efficiently with microtubules in the extract. (D) Binding of Maskin or XMAP215 or both to microtubules in vitro. (D, top) Coomassie blue–stained gel from a representative microtubule pelleting experiment. GST-Maskin and XMAP215 at 1 μM final concentration were incubated with 3 μM taxol-stabilized microtubules either independently (left, Maskin; middle, XMAP215) or together (right, both). The supernatants (S) and pellets (P) obtained after centrifugation of the samples were run on SDS-PAGE and the gel stained with Coomassie blue. The positions of the XMAP215, Maskin, and tubulin bands are indicated on the right. Equivalent amount of the supernatants and pellets were run for direct comparison. (D, bottom) Quantification of the bands from the gel showing the proportion of proteins recovered in the supernatant and in the pellets (Maskin, gray bars; XMAP215, white bars) in each condition. The percentage of each protein in the pellet (P) and supernatant (S) was calculated over the total amount (S + P = 100%). This experiment was repeated five times and two gels were run for each experiment. Similar results were obtained in all independent experiments.

Figure 5.

Eg2 phophorylates Maskin in vitro and is required for its localization in extract. (A) Schematic representation of GST-Maskin with the position of the three consensus sites for phosphorylation by Eg2. (B) Autoradiography (top) and Coomassie blue–stained gel (bottom) of GST-Maskin proteins incubated with purified Eg2. Three proteins are shown: GST-Maskin (Maskin), GST-Maskin mutated on Ser620 and Ser626 (S620A, S626A), and GST-Maskin mutated on the three sites Ser33, Ser620, and Ser626 (Maskin3A). (C) Effect of Eg2 depletion on the localization of Maskin in egg extract. Eg2 was depleted from the egg extract to more than 95% as shown by Western blot analysis (right). Spindles were assembled in cycled extract, fixed, and processed for immunofluorescence with the anti-Maskin antibody. Two pairs of spindles are shown taken with the same camera settings for comparison. Microtubules are red, Maskin is green, and DNA is blue. Bar, 10 μm.

Figure 6.

Maskin-3A does not rescue the spindle and aster size in Maskin-depleted extract. Spindles and centrosome asters were assembled in four different extracts. Mock (white bars), Maskin-depleted extract (dark gray bars), AdB1, Maskin-depleted extract–containing purified GST-Maskin (light gray bars), and AdB2, Maskin-depleted extract containing purified GST-Maskin3A (black bars). (A) Graph showing the average length of the spindles is shown for the four conditions. (B) Western blot of the four extracts used in A probed with the anti-Maskin and anti-tubulin antibodies. The tubulin band was used as a loading control. (C) Graphs showing the average tubulin fluorescence intensity (left) and length of microtubules (right) of asters assembled in the four conditions. (D) Western blot of the four types used in C probed with the anti-Maskin and anti-tubulin antibodies.

Figure 7.

Inhibition of XKCM1 rescues aster size in Maskin-depleted extract. Microtubule asters were assembled as in Fig. 3 in mock-depleted extract (Mock, white bars), Maskin-depleted extract (ΔMaskin, dark gray bars), mock-depleted extract containing anti-XKCM1 antibodies (Mock + αXKCM1, light gray bar); Maskin-depleted extract containing anti-XKCM1 antibodies (ΔMaskin + αXKCM1, black bars). The average tubulin fluorescence intensity (left) and microtubule length (right) are shown.