Functional overlap of microtubule assembly factors in chromatin-promoted spindle assembly

Abstract

Distinct pathways from centrosomes and chromatin are thought to contribute in parallel to microtubule nucleation and stabilization during animal cell mitotic spindle assembly, but their full mechanisms are not known. We investigated the function of three proposed nucleation/stabilization factors, TPX2, gamma-tubulin and XMAP215, in chromatin-promoted assembly of anastral spindles in Xenopus laevis egg extract. In addition to conventional depletion-add back experiments, we tested whether factors could substitute for each other, indicative of functional redundancy. All three factors were required for microtubule polymerization and bipolar spindle assembly around chromatin beads. Depletion of TPX2 was partially rescued by the addition of excess XMAP215 or EB1, or inhibiting MCAK (a Kinesin-13). Depletion of either gamma-tubulin or XMAP215 was partially rescued by adding back XMAP215, but not by adding any of the other factors. These data reveal functional redundancy between specific assembly factors in the chromatin pathway, suggesting individual proteins or pathways commonly viewed to be essential may not have entirely unique functions.

Figure 1.

Time-Lapse Fluorescence Imaging of Anastal Spindle Assembly. (A) Selected frames from a time-lapse movie of spindle assembly around DNA-coated beads highlighting the following transitions (for A, C, and D): i. Initial tubulin polymerization (cloud), ii. Radial array of MTs, iii. Appearance of a pole, iv. Extension/extrusion of poles, and v. Bipole structure. The timing of each highlighted transition varied for each spindle assembly reaction (compare time stamps on lower right corner of frames for A, C, and D. (B) Quantification of tubulin intensity over time for 3 different bead spindle assembly reactions from 3 different extracts. The dark blue trace is from the assembly reaction shown in 1A. Note that maximum fluorescence intensity (max-FI) is reached ∼25–30 min after onset of polymerization. Average tubulin fluorescence intensity (N = 11) of bead spindle assembly over time with transitions i-v highlighted. (Samples were averaged after onset of tubulin polymerization because polymerization onset varied for each spindle). (C) Selected frames from a two-color time-lapse movie of x-rhodamine tubulin (top row), Alexa-488 labeled gamma-tubulin antibody (middle row) and color combine (lower row) during DNA bead spindle assembly with transitions i-v highlighted. (D) Frames from a two-color time-lapse movie of x-rhodamine tubulin (top row), TPX2-GFP (middle row), and color combine (lower row) during spindle assembly around DNA beads with transitions i-v highlighted. Scale bars: 10 μm. Time is shown in minutes:seconds.

Figure 2.

TPX2, γ-tubulin and XMAP215 are Essential for Anastral Spindle Assembly. (A) Anastral spindle assembly requires TPX2 and can be rescued by adding back purified TPX2. (B) Western blot analysis of TPX2 reveals efficient immunodepletion and addition of purified TPX2 to depleted extract at endogenous concentration (100 nM). The shift in molecular weight for the purified TPX2 is due to the presence of a GST tag. (C) γ-tubulin is required for spindle assembly around DNA beads. (D) γ-tubulin can be significantly (>90%) immunodepleted from assembly reactions. (E) XMAP215 is essential for bead spindle assembly and its depletion can be rescued by the addition of purified recombinant XMAP215. (F) XMAP215 is efficiently immunodepleted from extract and the purified protein is added back at endogenous levels (300 nM). (G) Quantification of the structures assembled around DNA-coated beads for conditions shown in A–F. The three types of structures, examples of which are shown in the inset, are naked beads, MT arrays, and bipolar spindles. Greater than 70% of structures in mock-depleted extracts are bipolar spindles while 78%, 85%, and 97% of bead structures lacked any associated microtubules in TPX2-, γ-tubulin- and XMAP215-depleted structures, respectively. Microtubule polymerization around beads (39% MT arrays and 61% bipolar spindles) was rescued by the addition of purified TPX2 to TPX2-depleted extracts. Microtubule polymerization around beads (46% MT arrays and 45% bipolar spindles) was rescued by the addition of purified XMAP215 to XMAP215-depleted reactions. (Scale bars: 10 μm; Error bars: SD for 4 different extracts; 100 structures counted per experiment).

Figure 3.

TPX2-dependent Microtubule Polymerization Requires γ-tubulin and XMAP215. (A) Addition of purified TPX2 (3X final extract concentration) rescues anastral spindle assembly in TPX2-depleted extracts but not γ-tubulin- or XMAP215-depleted reactions. (B) Quantification of the bead structures assembled in the conditions shown in 3B. While addition of purified TPX2 rescues TPX2 depletion, TPX2 addition did not rescue γ-tubulin- or XMAP215-depleted reactions (>80% naked beads in each case). (Scale bars: 10 μm; Error bars: SD for 4 different extracts; 100 structures counted per experiment).

Figure 4.

EB1 Rescues the Microtubule Assembly Defects of TPX2 Depletion but Not γ-tubulin or XMAP215 Depletions. (A) Addition of purified EB1 (3X final extract concentration) restores bead-associated microtubule polymerization to TPX2-depleted reactions but not γ-tubulin- or XMAP215-depleted extracts. Both bipolar spindles and microtubule arrays were commonly observed for the ΔTPX2 + EB1 condition. (B) Quantification of the structures assembled in the conditions shown in 4B. Addition of purified EB1 to TPX2-depleted extracts yielded a fourfold reduction in the percentage of naked beads relative to ΔTPX2 (46% microtubule arrays and 35% bipolar spindles). EB1 addition had no affect on the types of structures formed in either γ-tubulin or XMAP215-depleted extracts (>80% naked beads). (Scale bars: 10 μm; Error bars: SD for 4 different extracts; 100 structures counted per experiment).

Figure 5.

Microtubule Polymerization Stimulated by MCAK inhibition Requires γ-tubulin and XMAP215 But Not TPX2. (A) Inhibition of MCAK (with α-MCAK) stimulates the assembly of large microtubule arrays around beads in both mock- and TPX2-depleted extracts but not in Δγ-tubulin or ΔXMAP215 extracts. (B) The number of α-MCAK arrays were counted in 2 μl samples for mock-, TPX2, γ-tubulin- and XMAP215-depleted extracts. Each condition is reported as a percentage of the number of microtubule arrays in ΔMock + α-MCAK reactions. ΔTPX2 supported the assembly of 58% of control α-MCAK arrays while both γ-tubulin and XMAP215 depleted extracts supported assembly of ∼1% the number of control structures. (Scale bars: 10 μm; Error bars: SD of for 3 different extracts; all the microtubule arrays in 2 μl of each reaction sample were counted per experiment).

Figure 6.

XMAP215 Rescues the Microtubule Assembly Defects of TPX2 and XMAP215 Depletion. (A) Addition of purified XMAP215 (4X final extract concentration) restores localized microtubule polymerization around DNA beads in TPX2-, γ-tubulin- and XMAP215 depleted extracts. Representative images of microtubule arrays and bipolar spindles observed in all the depletion + XMAP215 conditions. (B) Quantification of the structures assembled in the conditions shown in 6A. Addition of XMAP215 to ΔTPX2 extracts led to a ∼fourfold reduction in the prevalence of naked beads compared with ΔTPX2. The ΔTPX2 + XMAP215 treatment resulted in 64% microtubule arrays and 15% bipolar spindles. XMAP215 also rescued γ-tubulin depletion with nearly a sixfold reduction in the percentage of naked beads (55% microtubule arrays and 29% bipolar spindles). Purified XMAP215 addition to XMAP215-depleted extract resulted in a nearly 11-fold reduction in the percentage of naked beads (46% microtubule arrays and 45% bipolar spindles). (Scale bars: 10 μm; Error bars: SD for 3 different extracts; 100 structures counted per experiment).