Dynamic positioning of mitotic spindles in yeast: role of microtubule motors and cortical determinants

Abstract

In the budding yeast Saccharomyces cerevisiae, movement of the mitotic spindle to a predetermined cleavage plane at the bud neck is essential for partitioning chromosomes into the mother and daughter cells. Astral microtubule dynamics are critical to the mechanism that ensures nuclear migration to the bud neck. The nucleus moves in the opposite direction of astral microtubule growth in the mother cell, apparently being "pushed" by microtubule contacts at the cortex. In contrast, microtubules growing toward the neck and within the bud promote nuclear movement in the same direction of microtubule growth, thus "pulling" the nucleus toward the bud neck. Failure of "pulling" is evident in cells lacking Bud6p, Bni1p, Kar9p, or the kinesin homolog, Kip3p. As a consequence, there is a loss of asymmetry in spindle pole body segregation into the bud. The cytoplasmic motor protein, dynein, is not required for nuclear movement to the neck; rather, it has been postulated to contribute to spindle elongation through the neck. In the absence of KAR9, dynein-dependent spindle oscillations are evident before anaphase onset, as are postanaphase dynein-dependent pulling forces that exceed the velocity of wild-type spindle elongation threefold. In addition, dynein-mediated forces on astral microtubules are sufficient to segregate a 2N chromosome set through the neck in the absence of spindle elongation, but cytoplasmic kinesins are not. These observations support a model in which spindle polarity determinants (BUD6, BNI1, KAR9) and cytoplasmic kinesin (KIP3) provide directional cues for spindle orientation to the bud while restraining the spindle to the neck. Cytoplasmic dynein is attenuated by these spindle polarity determinants and kinesin until anaphase onset, when dynein directs spindle elongation to distal points in the mother and bud.

Figure 1

Segregation of the “late acquiring” SPB into the bud in kar9 mutants. kar9 mutants expressing dynein-GFP were grown to midlogarithmic phase and were examined by fluorescence microscopy. (A) Images from a time-lapse series are displayed (times in min indicated in the upper right of each panel). The nucleus lies distal to the neck of a budded cell before spindle elongation. Microtubules emanate from a single spindle pole body (0 min). The poles have duplicated by 8 min, and the 2nd pole (indicated with white arrowhead) acquires cytoplasmic dynein. At 28 min, late acquiring pole nucleates an astral microtubule(s) directed toward the bud. At 78 min, the spindle has aligned along the mother-bud axis, and a single microtubule from the late-acquiring pole penetrates the bud. The late-acquiring SPB migrates into the bud. (B) Quantitative fluorescence accumulation as a function of time and distance between the spindle pole bodies. The time points in (A) are plotted. Note spindle pole separation as increased distance from peak to peak (one pole at 0 min, two at 8, 28, and 78,min). The delayed acquisition of dynein-GFP is evident in the fluorescent intensity of the two peaks (8 and 28 min) until spindle elongation, whereupon the two poles are equally fluorescent (78 min).

Figure 2

SPB nucleating microtubules that first penetrate the bud are not the SPB destined for the bud in bni1 mutants. bni1 mutants expressing dynein-GFP were grown to midlogarithmic phase and examined by fluorescence microscopy. Images from a time-lapse series are displayed (right, times in min indicated in the upper right of each panel). Corresponding DIC images are shown to the left. Two spindle pole bodies can be seen in the upper cell (4 o'clock and 6 o'clock, respectively). Astral microtubules from the pole at 6 o'clock penetrate the bud (0,1 min). At 2 and 3 min, the bud is devoid of astral microtubules. At 4 and 5 min, the pole at 4 o'clock nucleates an astral microtubule that penetrates the bud. This pole leads the nucleus into the bud at later time points.

Figure 3

Simultaneous penetration of astral microtubules in the bud in bud6 mutants. Bud6 mutants expressing dynein-GFP were visualized by fluorescence microscopy. The mother cell is in the bottom right quadrant, the daughter to the top left. Astral microtubules emanating from spindle pole bodies in the mother (8 and 10 o'clock respectively) cross over the neck and penetrate the bud simultaneously. These defects were observed in 3/18 bud6 mutants for a total of 3.7 percentage of the time before anaphase onset (8 min of alternating cycles of microtubule polymerization/depolymerization in the bud emanating from each pole, 30 s where microtubules from both poles penetrate the bud; 230 total minutes). 5/20 bni1 mutants exhibited dual penetration from both spindle poles for 8.1% of the time (50/617 total min) and 2/11 kar9 mutants for a total duration of 1.6% of the time before anaphase onset (7.5/464 total min).

Figure 4

Distance from SPB proximal to neck 1 min before spindle pole elongation. Wild-type, bud6, bni1, kar9, kip3, and dhc1 mutants containing Tub1-GFP were grown to midlogarithmic growth and examined by fluorescence microscopy. Cells containing a 1–1.5 μm spindle were examined and followed by time-lapse microscopy through spindle elongation. At 1 min before anaphase spindle elongation, a point at the center of the neck was determined in DIC and was used as the origin when measuring the distance to the center of SPB fluorescence. Track Points software in Metamorph (Universal Imaging Corp.) recorded the SPB position at each time point. The distance was determined by converting pixels to microns using the image of the stage micrometer. Values are the average of ten to twenty cells. Wild-type spindles are statistically closer to the neck than spindles in bud6, bni1, kar9, and kip3 mutants as determined by 2-tailed unequal variance Students t test (95% confidence). kar9 spindles are significantly further from the neck relative to bud6, bni1, kip3, and wild-type. Spindle position in bud6, bni1, kip3, and dhc1 are statistically indistinguishable.

Figure 5

Metaphase and anaphase spindle movements in wild-type and mutant cells. Spindle movements in wild-type, bud6, bni1, kar9, kip3, and dhc1 mutants. The distances between each spindle pole body and neck were measured (μm) at 30-s intervals using Track Points software in Metamorph (Universal Imaging Inc.). The position of the SPB relative to the mother is plotted as positive numbers, the distance between the neck and the bud is plotted as a negative number. The position of the neck is set as zero. Two examples of each cell type are shown. The spindle pole that is initially distal to the neck at t = 0 is indicated by a circle, the pole proximal to the neck by a diamond. Preanaphase spindle oscillations were observed in 5/7 wild-type cells, 5/5 kar9, 10/13 kip3, 9/16 bni1, 8/15 bud6, and 0/8 dhc1 mutants, respectively.

Figure 6

Average microtubule lengths during mitosis. Microtubule lengths were determined in wild-type and the indicated mutants expressing Tub1-GFP. The three classes include those extending into the mother cell and nucleated from the spindle pole distal to the neck (SPB_{d mother}), those extending into the mother cell and nucleated from the spindle pole proximal to the neck (SPB_{p mother}), and those extending into the bud and nucleated from the spindle pole proximal to the neck (SPB_{p bud}). Lengths were determined by measuring the distance from the center of SPB fluorescence to the microtubule end. Pixels were converted to microns using the image of the stage micrometer. Time-lapsed image sequences from five to ten metaphase and anaphase cells were used. The average lengths, plus or minus the SD, are indicated. Microtubule numbers are for wild-type (Spb_pb = 39, Spb_pm = 8, Spbd = 25); bud6 (Spb_pb = 36, Spb_pm = 7, Spbd = 23); bni1 (Spb_pb = 30, Spb_pm = 3, Spbd = 25), kar9 (Spb_pb = 25, Spb_pm = 24, Spbd = 30), kip3 (Spb_pb = 32, Spb_pm = 6, Spbd = 42), and dhc1 (Spb_pb = 57, Spb_pm = 10, Spbd = 43). All three classes of microtubules were significantly longer in kip3 mutants versus other mutants and wild-type (>95% confidence). Statistical analysis (2-tailed unequal variance Students t test) indicate the following significant differences in microtubule length relationships: SPB_{d mother} kip3> kar9 = bud6 = bni1= dhc1>wild-type; SPB_{p mother} kip3 = kar9 = dhc1 > bud6 = bni1 = wild-type; SPB_{p bud} kip3 = kar9= bni1 > dhc1 = bud6 = wild-type.

Figure 7

Spindle translocation in the absence of spindle elongation. Astral microtubules pull the nucleus into the bud in the absence of Kar9p (n = 9). (A). The spindle has elongated in the mother perpendicular to the mother-bud axis. N denotes the position of the neck, an arrow marks the end of the microtubule, and the spindle pole indicated by an asterisk. An astral microtubule from the SPB at 12 o'clock grows into the bud in the top three panels (0–60 s). Once the astral microtubule interacts with the bud cortex, there is net movement of the spindle toward the bud (90 s). (B). Graph of the microtubule length (open circles), spindle length (closed diamonds), and SPB movement (open triangles) as a function of time (sec). Movement of the SPB_p into the bud is concomitant with astral microtubule interaction with the bud cortex. SPB movement is independent of spindle elongation.

Figure 8

Spindle movement in dhc1,bud6, and dhc1, kar9 mutants. dhc1, bud6 (A), and dhc1, kar9 (B) double mutants containing GFP-Tub1 were grown at 24°C. GFP-Tub1 was used to visualize the spindle and astral microtubules. Dynein-GFP could not be used to image astral microtubules in the double mutants because dynein-GFP complements the heavy chain deletion. The spindle in bud6, dhc1 (A) is visible as the bright bar and remains in the body of the mother cell. The astral microtubules originate from the spindle pole body and extend into the bud. The spindle is aligned along the mother/bud axis at t = 0. The spindle is 1.91 μm in length and lies 1.34 μm from the neck of the budded cell. At t = 31 min, the spindle elongates slightly (2.82 μm) and migrates away from the neck (2.01 μm from the neck). By 60 min, spindle length = 5.74 μm and the spindle lies 2.11 μm from the neck, and at 81 min the spindle has disassembled. There is no evidence for nuclear migration toward the bud even though an astral microtubule penetrates the bud throughout 81 min of continuous observation. The spindle in kar9, dhc1 (B) is aligned along the mother-bud axis at t = 0. An astral microtubule extends into the bud at t = 0 min. The spindle does not migrate into the bud; in contrast, spindle movement away from the neck is apparent at 42 min.

Figure 9

Chromatin segregation in the absence of a bipolar spindle and cytoplasmic motor proteins. Ndc1; ndc1, kip3 Δ and ndc1, dhc1Δ mutants were grown to early logarithmic phase at 25°C before shifting to 14°C, the nonpermissive growth temperature for ndc1–1, for 24 h. Cells arrest as large budded cells with a single nucleus. The cells were stained with Calcofluor to visualize the birth scar and with DAPI to visualize the chromatin DNA. An example of DNA in the bud is displayed to the left, DNA at the neck in the middle panel, and DNA in the mother to the right. Percentages of each phenotype in the different mutants are indicated below. Greater then 100 cells were counted for each determination.