The Physical Biology of Nucleolus Disassembly

Abstract

During cell division, precise and regulated distribution of cellular material between daughter cells is a critical step and is governed by complex biochemical and biophysical mechanisms. To achieve this, membraneless organelles and condensates often require complete disassembly during mitosis. The biophysical principles governing the disassembly of condensates remain poorly understood. Here, we used a physical biology approach to study how physical and material properties of the nucleolus, a prominent nuclear membraneless organelle in eukaryotic cells, change during mitosis and across different scales. We found that nucleolus disassembly proceeds continuously through two distinct phases with a slow and reversible preparatory phase followed by a rapid irreversible phase that was concurrent with the nuclear envelope breakdown. We measured microscopic properties of nucleolar material including effective diffusion rates and binding affinities as well as key macroscopic properties of surface tension and bending rigidity. By incorporating these measurements into the framework of critical phenomena, we found evidence that near mitosis surface tension displays a power-law behavior as a function of biochemically modulated interaction strength. This two-step disassembly mechanism, which maintains structural and functional stability of nucleolus while allowing for its rapid and efficient disassembly in response to cell cycle cues, may be a universal design principle for the disassembly of other biomolecular condensates.

Figure 1.. Morphology changes of nucleolus at mitosis entry.

(A) Live HeLa cells expressing mClover3-H2B and mRuby3-NPM1 in interphase and mitosis. White dash-dot and solid lines outline segmented nucleus and nucleolus regions for feature quantification. White dashed line shows cell membrane boundary after nuclear membrane and nucleolus disassemble at prometaphase. (B) Relative volume (volume of nucleolus/volume of nucleus) and (C) relative mean intensity (mean intensity of nucleolus/mean intensity of nucleus) in interphase (triangle) and prophase (circle) cells. Live HeLa cells expressing mRuby3-NPM1 in interphase (D) and prophase (E). White solid lines represent interfacial boundaries between nucleolus and nucleoplasm. (F) Distribution of the relative radial distance in interphase and prophase cells. (G) Entropy of relative radial distance measurement for interphase (triangle) and prophase (circle). For all boxplots (B, C, G), red bars indicate the mean. Significance was tested by t-test (****P<0.0001). Biological replicates: n = 15 cells (B, C). n = 20 cells (G); Scale bar, 5 μm (A); 3 μm (D, E).

Figure 2.. Dynamics dissociation of nucleolus.

(A) Time-lapse of HeLa cell expressing mRuby3-NPM1 progressing from prophase to prometaphase. Dash-dot lines denote nucleus, dashed lines illustrate cell membrane boundary. t = 0 denotes nuclear envelope breakdown. (B) Probability density function of pixel intensities at different time points at 1 minute time intervals. t = 0 denotes the time of nuclear envelope breakdown. Dashed lines represent the power law exponent coefficients $(\alpha )$ of dense phase in log-log scale. (C) Power law exponent coefficients $(\alpha )$ as a function of time for cells in interphase and mitosis. n=12 cells. Slow and fast slopes (dashed lines) highlight two phases of the nucleolus disassembly. (D) Slow and fast disassembly rates for cells mitosis and interphase. Red bars indicate the mean and box extend from the first to third quartile. Significance was tested by t-test (****P<0.0001). (E) Left, NPM1 free energy of transfer $(\Delta G)$ from nucleoplasm to nucleolus as a function of time for cells in mitosis and interphase. n = 12 cells. Dash vertical line indicates time when the free energy becomes positive. Right, schematic showing changes of free energy landscape as NPM1 relocated from dense to dilute phases through mitosis. (F) Top, representative images of HeLa cells expressing EGFP-NPM1 and mRuby2-Lamin A/C before (t=0 min) and after (t=40 min) Triton-X 100 treatment at t = 5 min. Bottom, quantification of mean intensity of NPM1 in dense and dilute phases. (G) Top, representative images of HeLa cells expressing mClover3-H2B and mRuby3-NPM1 right before (t=20 min) and after (t=50 min) RO-3306 treatment. Bottom, quantification of the nucleolus disassembly before and after treatment. Dash vertical line indicates time adding Triton-X 100 (F) and RO-3306 (G). Data points and shaded areas (C, E) represent mean ± s.d. Scale bar, 8 μm (A); 5 μm (F, G).

Figure 3.. Nucleolus disassembly during prophase results in decreased of NPM1 binding kinetics.

(A) Top, FRAP time course of nucleolus in HeLa cell expressing mRuby3-NPM1. Dashed circles indicate the bleached spot. Scale bar, 2μm. Bottom, quantification of the mean mRuby3-NPM1 fluorescence intensity recovery after FRAP at different time points of nucleoli through the prophase, and for cells in interphase (blue). (B) The fluorescence half-recovery times t_1/2 quantified at different time points. (C) The effective diffusion coefficient of mRuby3-NPM1 at different time points. (D) Relative association/dissociation rate of NPM1 calculated from FRAP experiments. Red solid lines represent a single exponential decay function (B, D) and growth (C). Plot markers represent mean value, shaded areas (a, n=8 cells) and error bars (B-D, n=20 cells) represent mean ± s.d.

Figure 4.. Reduction of nucleolar material properties during the disassembly process.

(A) Representative image of a HeLa cell nucleolus. The white contour denotes the nucleolar shape. The white dot is nucleolar center of mass and the origin of polar coordinate. Scale bar, 2 μm. (B) Contours of nucleolus plotted in the polar coordinates at different time points for a cell in prophase denoting the shape fluctuation around a mean contour (red dashed line). Inset represents a zoom-in view. Color bar shows the time indicator for contours in the time series. (C) Shape fluctuation spectra as a function of mode number at different time points before nuclear membrane breakdown, and for cells in interphase (blue). Plot markers represent mean value, and shaded areas represent 95% bootstrap confidence intervals of standard deviation from n=10 cells. Dashed lines are the best fit data for experimental results from the theoretical model. Dash-dot lines represent surface tension dominant regime $({\mathrm{q}}^{\text{-}1})$ and bending rigidity dominant regime $({\mathrm{q}}^{\text{-}3})$, respectively. Surface tension (D) and bending rigidity (E) obtained from fitting experimental results with theoretical model as shown in (C). Red solid lines represent a single exponential decay function fit. Plot markers represent mean value, and error bars represent 95% bootstrap confidence intervals of standard deviation with n=22 cells.

Figure 5.. Range of the critical regime.

(A) A schematic describing the emergence of nucleolar morphologies and surface tension changes at macroscopic scale as cells regulate the NPM1 interaction strength at microscopic scale throughout nucleolus disassembly. (B) Surface tension fitted to power law function with different critical exponent $\mu$ resulting in the critical parameter ${\epsilon }_{\mathrm{c}}$ in a range from 0.0215 to 0.0245 (red bar). (C) Reduced surface tension $(\sigma ∕{\sigma }_{\mathrm{o}})$ as a function of the normalized interaction strength difference $(({\epsilon }_{\mathrm{c}}\text{-}\epsilon )∕{\epsilon }_{\mathrm{c}})$ in a log-log scale. Solid line is power-law function with the critical exponent of μ=1.14. Vertical dashed line indicates estimated point where the system deviates from critical scaling law. error bars represent 95% bootstrap confidence intervals of standard deviation with n=22 cells.