Megakaryocyte lineage-specific class VI β-tubulin suppresses microtubule dynamics, fragments microtubules, and blocks cell division

Abstract

Class VI β-tubulin (β6) is the most divergent tubulin produced in mammals and is found only in platelets and mature megakaryocytes. To determine how this unique tubulin isotype affects microtubule assembly and organization, we expressed the cDNA in tissue culture cells under the control of a tetracycline regulated promoter. The β6 coassembled with other endogenous β-tubulin isotypes into a normal microtubule array; but once the cells entered mitosis it caused extensive fragmentation of the microtubules, disrupted the formation of the spindle apparatus, and allowed entry into G1 phase without cytokinesis to produce large multinucleated cells. The microtubule fragments persisted into subsequent cell cycles and accumulated around the membrane in a marginal band-like appearance. The persistence of the fragments could be traced to a pronounced suppression of microtubule dynamic instability. Impairment of centrosomal nucleation also contributed to the loss of a normal microtubule cytoskeleton. Incorporation of β6 allowed microtubules to resist the effects of colcemid and maytansine, but not vinblastine or paclitaxel; however, cellular resistance to colcemid or maytansine did not occur because expression of β6 prevented cell division. The results indicate that many of the morphological features of megakaryocyte differentiation can be recapitulated in non-hematopoietic cells by β6 expression and they provide a mechanistic basis for understanding these changes.

Fig. 1. Expression of β6-tubulin

Clonal lines, derived from CHO cells transfected with human β6-tubulin cDNA in a tetracycline (tet)-regulated vector, were incubated in the presence (+) or absence (−) of tet for 24 h and then examined by western blots stained with an N-terminal antibody that recognizes both ectopic β6-tubulin (β6) and other endogenous isoforms of β-tubulin (β). The blot was also stained for actin as a loading control. Note that tet inhibited the expression of ectopic β6-tubulin. The numbers at the bottom of four lanes indicate the percentage of total β-tubulin contributed by β6 expression.

Fig. 2. Tubulin immunofluorescence in cells expressing β6

Nontransfected wild-type (WT) CHO cells and stably transfected clones expressing β6 tubulin, as labeled in the panels, were pre-extracted with microtubule stabilizing buffer, fixed, and then stained with an antibody to α-tubulin. For the cells in panels A and B, their corresponding nuclear staining is also shown (A’ and B’). Before fixation, the transfected cells were induced to express β6 by incubating without tetracycline (Tet) for 1 or 4 d as indicated on the figure. Arrows in panel C point to microtubule bundles near the cell periphery. The arrow in panel F points to a cell with tubulin aggregates but no microtubule cytoskeleton. Bar = 50 μm.

Fig. 3. Effects of β6 on mitosis

A: Tubulin immunofluorescence of CHO and β6-7 cells 24 h after removing tetracycline. Note that α-tubulin is green, γ-tubulin is white, and DNA is red in the merged pictures. Expression of β6 caused the cells to block in mitosis, leading to an increase in the mitotic index (B), and to accumulate predominantly in prometaphase (C). Unlike wild-type (WT) CHO cells, β6-7 cells exhibited extensive microtubule fragmentation in prophase (D). Arrows in panel D indicate the location of the spindle poles. Bars = 10 μm.

Fig. 4. Effects of β6 on microtubule acetylation

A stable CHO cell line expressing HAβ1-tubulin (A) and clone β6-7 (B) were induced for 12, 24, 36, and 48 h by removing tetracycline. Cells were stained with antibodies against α-tubulin and acetylated α-tubulin for immunofluorescence (A and B) and lysed for western blot analysis using antibodies to α-tubulin, β-tubulin, and acetylated α-tubulin (C). Time-dependent changes in the levels of α-tubulin and acetylated α-tubulin relative to the zero time point are shown in panel D, *p < 0.05 compared to zero time point, n=5. Only the 48 h images are shown in panels A and B. Bar = 10 μm.

Fig. 5. Effects of β6 on sensitivity to microtubule targeted drugs

Wild type CHO cells were mixed with clone β6-7 that was induced to express β6-tubulin by removing tetracycline for 24 h. The indicated nM drug concentrations were added for an additional 24 h. Because the β6 expressing cells were more sensitive to vinblastine, this drug was added for just the last 3 h of the 48 h incubation. The cells were fixed and stained with antibodies against β6 and α-tubulin to identify β6 expressing and non-expressing cells, but only the α-tubulin staining is shown. Arrows indicate the β6 expressing cells and arrowheads indicate cells that do not express β6. Bar = 10 μm.

Fig. 6. Microtubule life history plots

Wild-type CHO (WT), clone β6-47, and clone β6-7 were transfected with EGFP-MAP4 to visualize microtubules and incubated without tetracycline for 48 h. Microtubules were imaged 5 s apart by live cell fluorescence microscopy, their lengths were measured from an arbitrary reference point, and the lengths were plotted against time. Each line represents a separate microtubule followed for up to 250 s. Note that the position of a line on the graph is arbitrary and does not represent the actual total length of the microtubule. Graphs similar to these were used to calculate growth rates, shortening rates, catastrophes, rescues, and other parameters characteristic of dynamic instability. The results from those calculations are summarized in Table I.

Figure 7. Microtubule nucleation

A: Wild-type (WT) CHO and clone β6-7 cells were treated with 2.5 μM colcemid for 2 h to depolymerize microtubules. After washing out the drug, the cells were fixed and stained at the indicated times (min) with an antibody to α-tubulin. Bar = 10 μm. B: WT and β6-7 cells were transfected with EB1-GFP and imaged by live cell fluorescence microscopy. Representative images are shown. Arrows point to the centrosome. Bar = 10 μm.