Pericentric chromatin is an elastic component of the mitotic spindle

Abstract

Background: Prior to chromosome segregation, the mitotic spindle bi-orients and aligns sister chromatids along the metaphase plate. During metaphase, spindle length remains constant, which suggests that spindle forces (inward and outward) are balanced. The contribution of microtubule motors, regulators of microtubule dynamics, and cohesin to spindle stability has been previously studied. In this study, we examine the contribution of chromatin structure on kinetochore positioning and spindle-length control. After nucleosome depletion, by either histone H3 or H4 repression, spindle organization was examined by live-cell fluorescence microscopy.

Results: Histone repression led to a 2-fold increase in sister-centromere separation and an equal increase in metaphase spindle length. Histone H3 repression does not impair kinetochores, whereas H4 repression disrupts proper kinetochore function. Deletion of outward force generators, kinesins Cin8p and Kip1p, shortens the long spindles observed in histone-repressed cells. Oscillatory movements of individual sister chromatid pairs are not altered after histone repression.

Conclusions: The increase in spindle length upon histone repression and restoration of wild-type spindle length by the loss of plus-end-directed motors suggests that during metaphase, centromere separation and spindle length are governed in part by the stretching of pericentric chromatin. Chromatin is an elastic molecule that is stretched in direct opposition to the outward force generators Cin8p and Kip1p. Thus, we assign a new role to chromatin packaging as an integral biophysical component of the mitotic apparatus.

Figure 1. Increased centromere separation following histone repression

(A) Schematic model of the budding yeast spindle showing the separation of sister centromeres during metaphase. Sister chromatid arms are held together by cohesin complexes (yellow); centromeres, bound to kinetochores (black), are pulled apart by kinetochore microtubules. (B) Predicted outcomes for centromere separation following the lowering of nucleosome concentration: (i) no change indicates that chromatin structure does not affect centromere separation, or (ii) increased centromere separation indicates role of chromatin in determining centromere separation. (C) Nuf2p-GFP kinetochore clusters in wild-type, GAL-H3, and GAL-H4 cells following 3 h growth in repressive media (YPD). Histone repression results in ∼2-fold increase in sister centromere separation. (Scale bar, 2 μm)

Figure 2. Spindle length increases following histone repression

(A) Predicted effects of increased centromere separation on spindle structure: (i) kinetochore microtubules shorten (no change in spindle length), or (ii) the entire spindle length increases with no change in kinetochore microtubule length. (B) Spc29p-CFP (spindle pole bodies) and Nuf2p-GFP in wild-type, H3-repressed, and H4-repressed cells. (Scale bar, 2 μm) (C) Histone repression results in increased separation of both kinetochore clusters and spindle pole bodies. Error bars represent standard deviation. (D) ChIP of Mcd1/Scc1p-6HA in GAL-H3 cells grown in permissive (YPG) or repressive (YPD) media. Centromere and arm loci were assayed for Mcd1/Scc1p association.

Figure 3. Pericentric chromatin is an elastic spindle component

(A) Theoretical force diagram of forces acting on centromere separation. Outward forces (green lines) are assumed to be constant regardless of centromere separation distance. Deletion of CIN8 or KIP1 is predicted to lower outward forces (dashed green line). Inward force (blue lines) is assigned to chromatin. Elastic chromatin (light blue line) is modeled with increasing force as centromere separation increases. Assuming chromatin behaves as a Hookean spring, the slope of this line is the spring constant of chromatin. Inelastic chromatin (dark blue line) is modeled to contribute inward force only when approaching nearly full extension. Intersection points of outward and inward force lines predict length of sister centromere separation. Thus, if chromatin is inelastic, motor deletion would not change centromere separation (compare arrow to filled arrowhead). However, if chromatin is elastic, lowered force (by motor deletion) would result in less stretching and therefore reduced centromere separation (hollow arrowhead). (B) Spc29p-CFP and Nuf2p-GFP in H3-repressed cells with either CIN8 or KIP1 deleted. (Scale bar, 2 μm) (C) Both spindle length and kinetochore separation are decreased in cin8Δ and kip1Δ cells, demonstrating that chromatin is elastic. Error bars represent standard deviation.

Figure 4. Single centromere dynamics following histone repression

(A) Kymographs of lacO arrays positioned 1.8kb from CEN15 in wild-type and H3-repressed cells shows dyanamics of centromeres. Images were acquired in one plane every 2 sec; approximately 3 min are shown. (B) Quantitation of centromere separation and movement shows increased centromere separation following H3 repression, but similar dynamics.

Figure 5. Modeling chromatin as a spring

(A) Model of metaphase spindle forces based on experimental data and simplified modeling of chromatin as an elastic element. Outward spindle force is decreased in kip1Δ and cin8Δ cells and is varied by the extent that centromere separation was affected by these motor deletions (see figure 3C). Inward chromatin-dependent force is shifted outward by histone repression, representing increased rest length. (B) Schematic spindle model including chromatin as a spring. Histone repression lowers the number of incorporated nucleosomes and primarily affects the chromatin spring by increasing rest length (decreasing number of “coils” in the spring), without affecting springiness (spring constant) of the remaining nucleosomes (coils). The inward, resistive force of the stretched spring contributes to balance of forces defining centromere separation and spindle length.