Fission yeast Alp14 is a dose-dependent plus end-tracking microtubule polymerase

Abstract

XMAP215/Dis1 proteins are conserved tubulin-binding TOG-domain proteins that regulate microtubule (MT) plus-end dynamics. Here we show that Alp14, a XMAP215 orthologue in fission yeast, Schizosaccharomyces pombe, has properties of a MT polymerase. In vivo, Alp14 localizes to growing MT plus ends in a manner independent of Mal3 (EB1). alp14-null mutants display short interphase MTs with twofold slower assembly rate and frequent pauses. Alp14 is a homodimer that binds a single tubulin dimer. In vitro, purified Alp14 molecules track growing MT plus ends and accelerate MT assembly threefold. TOG-domain mutants demonstrate that tubulin binding is critical for function and plus end localization. Overexpression of Alp14 or only its TOG domains causes complete MT loss in vivo, and high Alp14 concentration inhibits MT assembly in vitro. These inhibitory effects may arise from Alp14 sequestration of tubulin and effects on the MT. Our studies suggest that Alp14 regulates the polymerization state of tubulin by cycling between a tubulin dimer-bound cytoplasmic state and a MT polymerase state that promotes rapid MT assembly.

FIGURE 1:

Alp14 tracks MT plus ends in S. pombe cells. (A) Images of fission yeast cells expressing Alp14-GFP fusion protein (FC1907). Alp14 is seen as cytoplasmic dots aligned in linear tracks in interphase cells. In mitotic cells, it is present in bright dots on the mitotic spindle. (B) Alp14 colocalization with other +TIP proteins. Images of cells (FC2347, FC2336) coexpressing Alp14-RFP with Mal3-GFP (EB1) or Tip1-GFP (CLIP-170). (C) Images of cells expressing Alp14-GFP and mRFP-Atb2 (tubulin) in wild-type (FC2493), tea2∆ (FC2471), and mal3∆ (FC2472) backgrounds. Kymographs were derived from images of MT bundles marked by arrowheads. Streaks of Alp14-GFP show plus end tracking of Alp14-GFP in these strains. Scale bar, 5 μm. Timescale bar, 60 s.

FIGURE 2:

alp14-null mutant cells display impaired microtubule assembly. (A) Wild-type and alp14∆ mutant cells expressing GFP-tubulin (FC1234, FC2332). Note the decreased number of MT bundles in alp14∆ cells compared with wild-type cells. Scale bar, 5 μm. (B) Histogram of number of interphase MT bundles in interphase cells in wild-type (n = 159) and alp14∆ (n = 228) cells. (C) Kymographs derived from time-lapse images of a MT bundle in wild-type and alp14∆ cells expressing GFP-tubulin. Scale bar, 5 μm. Timescale bar, 30 s. (D) Dynamics of individual MT plus ends. Rates of MT assembly and disassembly in wild-type (n = 38) and alp14∆ (n = 40) cells. (E) Percentage of time spent in each MT dynamic phase. Pause was defined as <0.5 mm/min change. Note that these measurements do not account for very small, transient changes seen in the ragged patterns in the alp14∆ kymographs. Twenty-two microtubules were measured in 18 wild-type cells. Twenty-six microtubules were measured in. (F) Images and kymographs of cells expressing Mal3-GFP as a marker of plus end dynamics in wild-type and alp14∆ background (FC1439, FC2328). Scale bar, 5 μm. Timescale bar, 30 s.

FIGURE 3:

Alp14 is moderate MT polymerase at low concentrations in vitro. (A) Scheme for TIRF microscopy assay used to study MT dynamics. Surface-attached anti-biotin antibodies (blue) specifically bind biotin-labeled, GMPCPP-stabilized MT seeds (red). The MT seeds near the surface nucleate Alexa Fluor 488–labeled (green) dynamic MT from 6 μM soluble tubulin dimers in the presence of GTP. TIRF illumination by 564- and 488-nm lasers. Image adapted from Al-Bassam et al. (2010). (B) TIRF image of growing MTs. Seeds (red) initiate the assembly of dynamic MTs (green) only at their plus ends (+). In any of the experiments, no assembly occurred at the minus ends (–) in the presence of 6 μM tubulin dimer concentration. (C) Kymographs showing the assembly and disassembly of single dynamic MTs. At 6 μM tubulin, dynamic MTs show slow assembly, rapid disassembly, and frequent catastrophes. At 20–100 nM Alp14, MT assembly rate is incrementally increased, as seen in the slopes of dynamic MT kymographs (broken arrows). The MT assembly rate (at 300–1000 nM) becomes slower, indicating that Alp14 promotes inefficient MT polymerase at the higher concentrations. (D) Effect of Alp14 on the dynamic MT assembly rate. Right, Alp14 accelerates MT assembly by threefold at low (0–100 nM) concentration. At higher concentrations, MT assembly is dramatically decreased. Each point (see also Supplemental Table 1) represents the mean of a Gaussian fit to a distribution of a large number of assembly events measured for each Alp14 concentration (distributions shown in Supplemental Figure S4). (E) Increasing Alp14 concentration causes a moderate increase in MT disassembly rates (Supplemental Table 1). Right, Alp14 increases MT disassembly rate moderately by 25% at the low concentration. Each point (Supplemental Table 1) represents an average Gaussian distribution fit of a large number of events for different Alp14 concentrations (Supplemental Figure S4). (F) Increasing Alp14 concentration does not affect MT catastrophe frequency (Supplemental Table 1). At 1000 nM two catastrophe frequencies are observed. Each point (Supplemental Table 1) represents an average from a Gaussian fit to the number of catastrophes measured in separate kymographs for each Alp14 concentration (Supplemental Figure S4). (G) Increasing Alp14 concentration increases the MT dynamic length (Supplemental Table 1). Right, Alp14 increases the dynamic length by twofold at low (0–100 nM) concentration. However at the higher concentrations, Alp14 loses polymerase activity. Each point (Supplemental Table 1) represents the mean of a Gaussian fit to a distribution of a large number of assembly events measured for each Alp14 concentration (distributions shown in Supplemental Figure S4). (H) The effect of different Alp14 concentrations on MT assembly rates at 6 (cyan), 8 (blue), and 10 μM (deep blue) tubulin. In the absence of Alp14, MT assembly increased with tubulin concentration. Alp14, 200 nM, induced a twofold increase in MT assembly rate. At higher Alp14 concentrations, Alp14 inhibited MT assembly completely or decreased MT assembly efficiency. Increasing Alp14 concentration (250–500 nM) progressively decreased MT assembly rate to values lower than in the absence of Alp14. See also Supplemental Table S2. (I) The tubulin association rate calculated from the effect of tubulin concentration on MT assembly rate in the absence of Alp14 (blue), 200 nM Alp14 (red), and 300 nM Alp14 (purple). The tubulin association rate was calculated as described (Brouhard et al., 2008). Alp14, 200 nM, increased the association rate from 4.0 (blue) to 6.4 μM/min (red), whereas at 300 nM Alp14 the association rate was decreased to 4.8 μM/min (purple).

FIGURE 4:

Alp14-GFP tracks growing MT plus ends in vitro. (A) Schematic of the TIRF microscopy assay used for imaging MT dynamics and for simultaneous imaging of dynamic MTs and Alp14-GFP localization. Anti-biotin antibodies (blue) bind biotin- and Texas red–labeled, GMPCPP-polymerized MT seeds (red) near the silanized glass surface. The densely labeled Texas red MT seeds nucleate dynamic MT assembly from 6 μM tubulin dimers, less densely labeled with Texas red, in the presence of GTP and 25 nM Alp14-GFP (see Materials and Methods). Image adapted from Al-Bassam et al. (2010). (B) Raw TIRF image of dynamic MTs and Alp14-GFP. Top, dynamic MTs growing at plus ends of MT seeds. Note that MT seeds have a high fluorescence intensity compared with the dynamic portion of the MTs, which are less intensely labeled with Texas red. Middle, Alp14-GFP. Bottom, overlay showing Alp14-GFP at MT plus ends; note that only some MTs display plus end localization. (C) Montage of Alp14-GFP tracking a growing MT plus end. In successive (10 s/frame), the MT plus end assembles while Alp14-GFP molecules track MT plus ends (broken arrow). Weaker GFP signals attributed to single Alp14-GFP molecules (white arrowheads) diffuse along MT lattices and may accumulate at the plus end. Plus end–tracking Alp14-GFP dissociates and disappears upon MT catastrophe and disassembly (white broken arrow). (D) Kymographs of Alp14-tracking dynamic MT plus ends. Left kymograph, Alp14 tracks a growing MT plus end and then dissociates upon activation of MT catastrophe. As MT plus end assembly reinitiates, Alp14 signal reappears midway through the MT assembly event, and then disappears with MT catastrophe. Middle, MT growing with faint Alp14 signal at the growing MT plus end (top, arrowhead), which then suddenly becomes more intense (lower, arrowhead) and disappears upon the occurrence of MT catastrophes. Right kymograph, three successive MT assembly events from a single MT seed. In the first and third MT assembly events, little or no Alp14-GFP tracking localization is observed at MT plus ends (broken arrow). The second event shows MT plus ends with strong Alp14-GFP localization (white arrowheads). Note that Alp14-GFP molecules bind along the MT seed and diffuse along the lattice over time (arrowhead). (E) Twofold acceleration in MT assembly rate correlates with Alp14-GFP tracking along plus end MTs. Top, Histogram and a Gaussian fit (blue) of all MT assembly events observed with 25 nM Alp14-GFP, showing an intermediate, half-maximal MT assembly rate similar to the distribution observed at 20 nM Alp14 (Supplemental Figure S5). Middle, histogram distribution (green bars) and Gaussian fitting (green line) of Alp14-GFP tracking MT assembly events have a high MT assembly rate. Bottom, histogram distribution (red bars) and Gaussian fitting (red line) of MT assembly events without Alp14 tracking shows lower MT assembly rates. The full MT dynamic parameters for each sets of events are shown in Supplemental Figure S5.

FIGURE 5:

An Alp14 TOG-domain mutant is defective in tubulin binding and MT polymerase activity in vitro. (A) Tubulin binding. Wild-type Alp14 or TOG12-Alp14 proteins were mixed with soluble tubulin and assayed for complex formation using size exclusion chromatography. Bottom right, SDS–PAGE of the fractions shows that wild-type Alp14 (blue) and tubulin dimer comigrate in a single peak. TOG12-Alp14 and tubulin dimer migrate as separate peaks, showing that this mutant is defective in tubulin binding (see Supplemental Figure S1). (B–E) Effects of purified TOG1,2 Alp14 on MT dynamics in TIRF in vitro assays (as in Figure 3). Effects of increasing TOG12-Alp14 (red) are shown compared with similar concentrations of wild-type Alp14 (blue; data from Figure 3) on MT assembly (B), MT disassembly (D), MT length (E), and MT catastrophe (F).

FIGURE 6:

Effect of Alp14 domains on alp14 complementation and localization in vivo. (A) mCherry-Alp4 fusion constructs were expressed from an nmt42 promoter (medium strength) on multicopy plasmids in alp14∆ GFP-tubulin fission yeast cells (FC2484-FC2489). Cells were shifted to inducing conditions (media without thiamine) and examined by time-lapse imaging at multiple time points from 16 to 24 h. Cells with low but detectable level of mCherry proteins were assayed. Table shows a summary of results. Red bars show location of point mutations in the TOG domains. Mutations are: W23A, R109A in TOG1; W300A, K381A in TOG2; Alp14 1-509 in N-term; Alp14 510-809 in C-term. (B) Images of representative fields of alp14∆ cells expressing the indicated construct. mCherry images show protein localization to MT lattice, MT plus ends, and/or cytoplasm. MT images show phenotypes in which interphase cells with an alp14∆-like phenotype (not rescued) have a small number of dim bundles, whereas cells with a wild type–like phenotype (rescued) have robust MT bundles. Scale bar, 5 µm. (C) Numbers of interphase MT bundles/cell were counted as a measure of Alp14 rescue by these Alp14 constructs (n > 30). These were consistent with other indications of Alp14 activity, including fluorescence intensity of GFP-tubulin in MT bundles and MT dynamics. * p < 0.01, *** p < 0.0001 on t-test compared to empty vector.

FIGURE 7:

Overexpression of Alp14 leads to a TOG-dependent loss of microtubules in vivo. (A) Wild-type GFP-tubulin cells expressing mCherry-Alp14 driven by the thiamine-repressible nmt42 promoter on multicopy plasmid (FC2477). Image shows a field of fission yeast cells with different levels of Alp14 mCherry expression. Cell expressing high levels lack MTs, those expressing intermediate levels have weakly fluorescing, shorter MTs, and those with no detectable expression have robust MTs. See also Supplemental Figure S6. (B) Images of cells expressing the different mutant constructs (see Figure 6A) in wild-type cells expressing GFP-tubulin (FC2477-2482). Asterisks denote cells lacking MTs. Scale bar, 5 μm. (C) Correlation of Alp14-mCherry expression levels with MT phenotype. Protein concentrations (μM) were estimated by Alp14-mCherry fluorescence intensities over the whole cell. Each dot represents measurement of one cell.