Nuclear size and physical properties of the nucleoplasm are determined by colloid osmotic pressure at the nuclear envelope
Abstract
The size of the nucleus scales with cell size, suggesting a universal scaling rule. Yet the biophysical determinants of nuclear size and the significance and consequences of altered nuclear-to-cell (N/C) ratios, which are observed across diverse pathological states and cell-fate transitions, remain poorly understood. Recent theoretical models propose that nuclear size arises from a balance of colloid osmotic pressures generated by macromolecules in the nucleoplasm and cytoplasm. Here we demonstrate that altering this osmotic balance through massive overexpression of an exogenous protein targeted to either the nucleoplasm or cytoplasm produces predictable changes in the N/C ratio in S. pombe. These quantitative perturbations show that nuclear size is set primarily by the number of proteins in the nucleus and cytoplasm, providing strong support for a pure osmotic pressure mechanism. Furthermore, cells with altered N/C ratios display tunable changes in nucleoplasmic crowding, nuclear condensate formation, nucleolar scaling and heterochromatin organization, establishing a causal link between nuclear size and gene regulatory processes. These findings reveal how cells exploit osmotic forces to set organelle dimensions, with broad implications for understanding how nuclear size shapes gene expression and cell identity in health and disease.
02 /Specification & Classification
Click any tag to browse matching papers- Citations
- 1
- Organism
- Assay
- Research Focus
- Approach
In-depth summary
Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (23 Jul 2026) · CC BY 4.0
Overexpressing mCherry-GST targeted to the nucleus or cytoplasm in fission yeast shifts the nuclear-to-cell volume ratio as a parameter-free colloid osmotic model predicts, retuning nucleoplasm diffusivity, condensates and heterochromatin.
Altered nuclear-to-cell (N/C) volume ratios accompany disease and cell-fate transitions, but the biophysics of nuclear size remains poorly understood, and recent theories tracing it to balanced colloid osmotic pressures needed direct experimental tests. By changing protein numbers in each compartment, this work tests that model quantitatively and uses the resulting N/C perturbations to probe nucleoplasm physics and chromatin organization.
Key results
ShownLoading the nucleus with a large protein that cannot diffuse through nuclear pores (mCherry-GST-NLS) enlarges nuclei, and loading the cytoplasm (mCherry-GST-NES) shrinks and indents them, with dose-dependent effects. Untargeted mCherry and the empty vector leave the N/C ratio unchanged.
NLS raised N/C up to 4-fold above normal. NES lowered it up to 2-fold, often giving bi-lobed nuclei. Exogenous proteins reached 10^6 to 10^7 copies per cell (Hta1-mCherry calibration). N≥112 cells per condition, Welch's t-tests. GFP, tandem-mCherry, promoter and vector swaps gave qualitatively similar N/C shifts. · Addition of exogenous proteins into the nucleus…; Fluorescent intensities quantification; Figure legends
ShownA dilute, colligative van't Hoff model with negligible nuclear envelope tension, and no free parameters, reproduces the N/C dose-response to both nuclear and cytoplasmic protein loads. This implies that protein numbers in each compartment set nuclear size.
Fits to Fig. 1d data (970 cells) gave 4.4×10^7 to 5.4×10^7 (5±0.5×10^7) osmotically active macromolecules per cell, vs 6×10^7 proteins from proteomics. Fitted nuclear count was 3.2×10^6 to 3.8×10^6 vs 4.5×10^6 predicted. Implied colloid pressure: about 5 mM or 12 kPa, about 1% of turgor. · Quantitative validation of an osmotic model for…; Modeling the effect of adding osmotically active…; Discussion; Figure legends
ShownThe N/C ratio tunes nucleoplasmic but not cytoplasmic physics. 40 nm GEM nanoparticles move faster in enlarged nuclei and slower in shrunken ones, and a synthetic SUMO10-SIM6 condensate forms more readily in small nuclei, where its nuclear concentration is higher.
Nuclear GEM Deff was up to 47% higher with NLS and fell with NES; a four-fold N/C range spanned a four-fold Deff range; cytoplasmic GEMs were unchanged. With two probe copies, foci formed in ~90% of control and NES cells vs 20% of NLS cells (N = 534). With one copy, only NES cells formed foci (N = 608). · N/C ratio tunes the rheology of the nucleoplasm; Measurement of effective diffusivity and…; Synthetic condensate formation; Figure legends
ShownNuclear size dilutes or concentrates the HP1 orthologue Swi6 without homeostatic compensation, modestly tuning centromeric heterochromatin foci. A Lever-rule fit suggests that a minority of focal Swi6 is in a condensate phase.
Doubling N/C gave about a two-fold lower nuclear Swi6 concentration and a 20% drop in centromeric focus intensity (N = 512 puncta; N = 1741 cells). Focus intensity rose linearly with concentration (y-intercept 0.82), implying ~80% chromatin-bound and ~20% condensate Swi6, which matches single-molecule estimates. · N/C ratio tunes heterochromatin condensates; Figure legends
ShownNucleolar volume scales with cell volume rather than nuclear volume. Perturbing N/C changed the nucleolus-to-nucleus ratio but preserved the nucleolus-to-cell ratio. Nucleoli turned rounded and less envelope-associated in enlarged nuclei and often filled one lobe of shrunken bi-lobed nuclei.
Wild-type controls: Nu/N 38±6% and Nu/C 3±1% (N=104 cells). Across conditions (N≥97 cells each), Nu/N differed significantly from control only for the nucleus- and cytoplasm-targeted constructs, while Nu/C did not differ (one-way ANOVA on replicate means). Gar1-GFP expression stayed constant. · Nucleolus size scales with cell size not nuclear…; Figure legends
Methods, data and reproducibility
Multi-copy plasmids driven by the Ptdh1 promoter overexpressed NLS- or NES-tagged mCherry-GST in S. pombe. An automated imaging pipeline (2D cell segmentation, LimeSeg 3D nuclear segmentation, Hta1-mCherry copy-number calibration) measured N/C ratios, which were compared with a van't Hoff colloid osmotic model. GEM tracking and condensate, Swi6 and Gar1 reporters then probed effects on the nucleoplasm.
- Strains expressing mCherry-GST-NLS, mCherry-GST-NES, mCherry or empty vector, imaged for N/C ratio vs exogenous protein dose · Schizosaccharomyces pombe (fission yeast) · n = 970 cells (Fig. 1d); N≥112 cells per condition
- Control constructs swapping fluorophore (GFP), GST (tandem mCherry), promoter and vector · Schizosaccharomyces pombe (fission yeast) · n = N≥162 cells per condition
- Strains with genome-integrated nuclear or cytoplasmic 40 nm GEMs (mSapphire) for microrheology · Schizosaccharomyces pombe (fission yeast)
- Strains carrying one or two genomic copies of an NLS-sfGFP-SUMO10-SIM6 synthetic condensate probe · Schizosaccharomyces pombe (fission yeast) · n = 608 cells (one copy); 534 cells (two copies)
- Swi6-GFP heterochromatin reporter strains across N/C perturbations · Schizosaccharomyces pombe (fission yeast) · n = 1741 cells; 512 centromeric puncta
- Gar1-GFP nucleolar reporter strains across N/C perturbations · Schizosaccharomyces pombe (fission yeast) · n = N≥97 cells per condition; 505 cells for Gar1 concentration
- Calibration strains with mCherry-tagged Hta1 (chromosomal locus), Eno1 and Tdh1 for absolute protein counts · Schizosaccharomyces pombe (fission yeast)
- Published quantitative proteome of exponentially growing cells (total protein count 6×10^7 per cell; Hta1 copy number) · Schizosaccharomyces pombe (fission yeast)
- Methods & tools
- pTOWsp-M multi-copy plasmid with Ptdh1 promoter, Morphometrics (2D cell contour segmentation), LimeSeg (ImageJ 3D nuclear segmentation), MosaicSuite Particle Tracker 2D/3D (ImageJ), TrackMate (LoG spot detection), MATLAB R2022 (MSD, Deff and anomalous exponent fitting), µManager v. 1.41, Genetically encoded multimeric nanoparticles (GEMs, 40 nm), GFP-SUMO10-SIM6 synthetic condensate probe, Spinning-disk confocal and TIRF microscopy, Van't Hoff colloid osmotic model with nonlinear least-squares fitting, Linear mixed-effects model, Welch's t-test, one-way ANOVA
- Reagents & environment
- Microscopes controlled with µManager v. 1.41; MSD analysis in MATLAB_R2022; ImageJ plugins LimeSeg, MosaicSuite Particle Tracker 2D/3D and TrackMate; custom Python Jupyter notebook for Swi6 puncta analysis.
- Reproducibility gaps
- No code repository, data deposition or accession is given. The custom segmentation pipeline and Python script are described but not linked, and the strain list (Table S1) has no content in the provided text.
Limitations (6)
- Despite controls, off-target or indirect effects of the artificial solutes on cell physiology (e.g., on transport, stress or biosynthesis) cannot be fully excluded. · stated by the authors
- The match between fitted osmolyte counts and proteomic protein totals may be fortuitous. Protein counts are uncertain, some proteins may be sequestered in complexes such as chromatin, and GST dimerization could inflate mCherry-GST counts. · stated by the authors
- Charge-based osmotic alternatives (pump-leak models that include counterions) are not ruled out. The effects of varying protein charge and assembly state remain to be tested. · stated by the authors
- The findings come from fission yeast, where nuclear envelope tension is negligible. Mammalian nuclei are partly constrained by tension from the lamina, perinuclear actin or chromatin, so extending these results requires further work. · stated by the authors
- Absolute counts are calibrated to one proteomic Hta1 value plus a nucleosome estimate. Measured Eno1 and Tdh1 (3.1 ± 0.6 ×10^6 and 3.7 ± 0.9 ×10^6) sit above published point estimates (1.4 and 1 ×10^6); only Eno1 lies inside the published range. · noted in this summary
- Transcription, silencing and gene expression were never measured. The proposed link from nuclear size to gene regulation rests on modest Swi6 focus-intensity changes and on a Lever-rule interpretation of the condensate fraction. · noted in this summary
Who should read:Read this if you study nuclear size scaling, intracellular osmotic pressure or macromolecular crowding and want a quantitative, parameter-free experimental test of colloid osmotic models of nuclear size in fission yeast.
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Control of nuclear size by osmotic forces in Schizosaccharomyces pombe
J. Lemiere, P. Real-Calderon, … F. Chang·bioRxiv·Dec 5, 2021
- Protein-driven colloid-osmotic pressure controls nuclear size, organization, and function
Shin Ohsawa, Gloria Sáncho-Andrés, … Gabriel E. Neurohr·bioRxiv·Sep 7, 2026
- Balance of osmotic pressures determines the volume of the cell nucleus
D. Deviri, S. Safran·Proceedings of the National Academy of SciencesPeer-reviewed·May 17, 2022
- Force-biased nuclear import sets nuclear-cytoplasmic volumetric coupling by osmosis
F. A. Pennacchio, A. Poli, … P. Maiuri·bioRxiv·Jun 7, 2022
- Nuclear size is genetically controlled and influences cell fate
Hisashi Moriizumi, Ruyi Shi, … Evgeny A. Zatulovskiy·bioRxiv·Aug 31, 2026
- Cell size scaling laws: a unified theory
R. Rollin, J.-F. Joanny, P. Sens·bioRxiv·Aug 1, 2022