Intracellular diffusion in the cytoplasm increases with cell size in fission yeast

Abstract

Diffusion in the cytoplasm can greatly impact cellular processes, yet regulation of macromolecular diffusion remains poorly understood. There is increasing evidence that cell size affects the density and macromolecular composition of the cytoplasm. Here, we studied whether cell size affects diffusion at the scale of macromolecules tens of microns in diameter. We analyzed the diffusive motions of intracellular genetically-encoded multimeric 40 nm nanoparticles (cytGEMs) in the cytoplasm of the fission yeast Schizosaccharomyces pombe. Using cell size mutants, we showed that cytGEMs diffusion coefficients decreased in smaller cells and increased in larger cells. This increase in diffusion in large cells may be due to a decrease in the DNA-to-cytoplasm ratio, as diffusion was not affected in large multinucleate cytokinesis mutant cells. In investigating the underlying causes of altered cytGEMs diffusion, we found that the proteomes of large and small cells exhibited size-specific changes, including the subscaling of ribosomal proteins in large cells. Comparison with a similar dataset from human cells revealed that features of size-dependent proteome remodeling were conserved. These studies demonstrate that cell size is an important parameter in determining the biophysical properties and the composition of the cytoplasm.

• Physical properties of the cytoplasm may modulate cellular processes by affecting in part the diffusion of macromolecules.

• Using small and large fission yeast mutants, we show that diffusion of nanoparticles in the cytoplasm increases with cell size.

• This change in cytoplasmic properties may reflect global cell size–dependent changes in the composition of the proteome.

INTRODUCTION

Cell size is an intrinsic physical property of all cells that can impact physiology from the cellular level to the organismal. Although cell size can vary over 6 orders of magnitude among diverse cell types, cell size varies within a much narrower range for a specific cell type due to homeostatic mechanisms (Ginzberg et al., 2015; Zatulovskiy and Skotheim, 2020). Cell size can impart different cellular functions and developmental potential (Hecht et al., 2016; Lengefeld et al., 2021). Aberrant cell size can also signify biological dysfunction and has been associated with aging, senescence, numerous cancers, and other human diseases (Lloyd, 2013). The mechanisms for how cell size impacts cellular physiology, however, remain poorly understood.

Recent studies have begun to implicate effects of cell size on the global properties of the cytoplasm. The cytoplasm can be regarded as a heterogenous, dynamic, and crowded viscoelastic matrix that exerts osmotic forces, and impacts nearly all biochemical reactions through effects on viscosity, macromolecular crowding, phase separation, and likely many other biophysical phenomena (Zhou et al., 2008; Mitchison, 2019). For instance, variations in density, which can be the result of complex dynamics between biosynthesis, degradation, and osmotic water fluxes, can cause significant changes in cellular physiology and/or function as seen during the cell cycle, differentiation, and stress (Neurohr and Amon, 2020).

Generally, the concentrations of cellular components are thought to be maintained at different cell sizes by scaling relationships. For instance, mRNA, protein, transcription, translation, and the volumes of various organelles can scale with cell size (Elliott et al., 1979; Creanor and Mitchison, 1982; Elliott, 1983; Neumann and Nurse, 2007; Zhurinsky et al., 2010; Padovan-Merhar et al., 2015; Chadwick et al., 2020; Marshall, 2020; Basier and Nurse, 2023; Swaffer et al., 2023). However, such scaling relationships have limitations, and so a breakdown in scaling mechanisms could explain cell size–dependent changes in cellular physiology. For example, it was observed in budding yeast cells that were arrested in G1 phase and grew to very large sizes (up to 10 times larger in volume than normal) exhibited defects in protein synthesis, and progressively became less dense (Neurohr et al., 2019; Terhorst et al., 2023). Similar dilution effects have been seen in senescent metazoan cells, indicating that a large cell size may be causal for aspects of senescent physiology, and is not merely a side effect of senescence (Demidenko and Blagosklonny, 2008; Neurohr et al., 2019; Lengefeld et al., 2021; Lanz et al., 2022). Similarly, in fission yeast, cells arrested in G2 phase grow excessively large leading to a gradual slowdown in rates of cell growth and protein translation, further illustrating that a large cell size is detrimental for proliferation and normal cell function (Basier and Nurse, 2023). Recent studies have demonstrated that cell size can also change the composition of the proteome even in cases where overall protein concentration remains relatively constant across different sizes (Schmoller et al., 2015; Keifenheim et al., 2017; Lanz et al., 2022, 2024).

One proposed explanation for certain cellular pathologies associated with overly large cells is that the DNA-to-cytoplasm ratio in these cells has dropped below a critical threshold necessary to scale biosynthetic processes (Zhurinsky et al., 2010; Marguerat and Bähler, 2012; Neurohr et al., 2019; Balachandra et al., 2022; Cadart and Heald, 2022; Xie et al., 2022). As cells grow larger without a concomitant increase in DNA, there may be insufficient gene copies or transcriptional or translational machinery to support biomass production for an exponentially-growing cell volume. This theory may explain why cells with increased ploidy can grow to larger sizes without exhibiting defects associated with large cell size (Neurohr et al., 2019; Mu et al., 2020; Lanz et al., 2022, 2024).

The fission yeast Schizosaccharomyces pombe is a leading model organism in defining cell size control mechanisms and scaling relationships (Nurse, 1985; Wood and Nurse, 2015). Many cell size scaling studies utilize well-characterized cell size mutants, such as wee1 and cdc25 mutants, which alter cell size by affecting the length of G2 phase through regulation of the CDK1 cell cycle–dependent kinase (Nurse, 1975; Fantes and Nurse, 1978; Neumann and Nurse, 2007; Zhurinsky et al., 2010; Knapp et al., 2019; Pickering et al., 2019; Sun et al., 2020). Thus, the molecular bases responsible for these perturbations in cell size and genomic copy number are generally well defined. Another feature of fission yeast is that although some effects of cell size in some cell types may be caused by changes in surface area-to-volume (SA/V) ratio (Harris et al., 2018), SA/V ratio varies little with size in fission yeast due to their characteristic rod cell shape (Shi et al., 2021). Recent studies in fission yeast demonstrate how the intracellular density of the cytoplasm fluctuates during the cell cycle, and how properties of the cytoplasm are altered in response to starvation and the breaking of dormancy in spores (Joyner et al., 2016; Munder et al., 2016; Heimlicher et al., 2019; Odermatt et al., 2021; Sakai et al., 2024).

Here, we investigated the effects of cell size on the biophysical properties of the fission yeast cytoplasm by assessing the diffusion of macromolecules within the cytoplasm. We measured diffusion by imaging and analyzing the diffusive motion of genetically-encoded multimeric cytoplasmic nanoparticles (cytGEMs), 40-nm-diameter fluorescent particles which inform on the diffusion of macromolecular complexes that are approximately the size of ribosomes (Delarue et al., 2018; Lemière et al., 2022; Molines et al., 2022). By analyzing cell size mutant strains, we found that diffusion within the cytoplasm decreased in a small cell mutant strain and increased in large cell mutant strains. Analyses of cytokinesis mutants demonstrated that this change in large cells was dependent on the DNA-to-cytoplasm ratio. To gain mechanistic insight on how cell size affects cytoplasmic properties, we discovered that proteome composition, including the concentration of ribosomal proteins, varies in cells of different sizes. Our studies reveal how cell size impacts the physical properties of the cytoplasm, providing new perspectives into how cell size affects cellular physiology and function.

RESULTS

Nanoparticle diffusion in the cytoplasm increases with cell size

To investigate the relationship between cell size and intracellular diffusion, we expressed and imaged 40 nm cytGEMs nanoparticles in S. pombe wild-type and cell size mutant cells, and through analyses of their motion, determined the effective diffusion coefficient in each strain (Delarue et al., 2018; Lemière et al., 2022; Molines et al., 2022). We grew wee1-50, wild-type, and cdc25-22 cells at the permissive temperature 25°C and then shifted the cultures to the nonpermissive temperature of 36°C for 6 h before imaging (Figure 1A). At this temperature, wee1-50 cells exhibit cell cycles with shorter G2 phases and enter mitosis at an abnormally short length, while cdc25-22 cells arrest in G2 phase and continue to grow in length (Nurse, 1975; Fantes and Nurse, 1978). As these rod-shaped cells grow by tip growth and maintain approximately similar cell widths, the length of the cell was used as a proxy of cellular volume (Mitchison, 1957; Facchetti et al., 2019; Knapp et al., 2019). Upon the 6 h temperature shift, wee1-50, wild-type, and cdc25-22 cells exhibited an average cell length of 5.61 ± 0.3 µm, 10.84 ± 1.39 µm, and 38.32 ± 1.36 µm, respectively (Figure 1B; mean ± STD of replicate experiments); the average cell widths were 4.28 ± 0.51 µm, 4.20 ± 0.37 µm, and 4.57 ± 0.39 µm, respectively. Measurement of the effective diffusion coefficient of cytGEMs in each cell population yielded average cytGEMs effective diffusion coefficients of 0.41 ± 0.04 µm²/s in wee1-50, 0.63 ± 0.07 µm²/s in wild-type, and 0.86 ± 0.04 µm²/s in cdc25-22 cells (Figure 1C; mean ± STD of replicate experiments). Thus, these data showed a positive correlation between cell size and nanoparticle diffusion in the cytoplasm at the population level.

FIGURE 1:

Nanoparticle diffusion increases with cell size

(A) Images (sum projection of three middle slices) of wee1-50, wild-type, and cdc25-22 cells with nuclear membrane marker Ish1-GFP (red) grown at the permissive temperature 25°C overnight and shifted to the nonpermissive temperature 36°C for 6 h before imaging. Scale bar is 5 µm. (B) Cell length (mean ± STD of replicate experiments; N_CELLS ≥ 100 per condition from at least four biological replicates) (one-way ANOVA, p < 0.0001) and (C) cytGEMs diffusion coefficients (mean ± STD of replicate experiments; N_GEMS ≥ 3183 per condition from at least four biological replicates) for wee1-50, wild-type, and cdc25-22 cells grown with the temperature shift protocol described in A (one-way ANOVA, p < 0.0001). (D) Cell length and cytGEMs diffusion coefficients (mean ± SEM of cytGEMs trajectories per cell from at least four biological replicates) plotted as function of cell length for individual wee1-50, wild-type, and cdc25-22 cells as described in A (N_CELLS = 160, 309, and 100, respectively). (E) Cell length (mean ± STD of replicate experiments; N_CELLS ≥ 113 per condition from three biological replicates) and (F) cytGEMs diffusion coefficients (mean ± STD of replicate experiments; N_GEMS ≥ 5709 per condition from three biological replicates) for cdc2-asM17 cells treated with 0.25% DMSO or 10 µM ATP analogue 1-NM-PP1 (one-way ANOVA, * - p < 0.05, ** - p < 0.01, *** - p < 0.001, **** - p < 0.0001).

A positive trend between diffusion coefficients and cell size was also evident in analyses of individual cells (Figure 1D). In analyses of data from each strain, positive correlations between cell length and cytGEMs effective diffusion coefficients (using a simple linear regression weighted by number of trajectories per cell) were apparent in the data of wee1-50 and cdc25-22 strains, but not in the range of sizes of wild-type cells, as noted previously (Supplemental Figure S1, A–C) (Garner et al., 2023).

To address concerns of possible effects of acute temperature shifts on diffusion and the regulation of cytoplasmic viscosity (Persson et al., 2020), we analyzed wee1-50, wild-type, and cdc25-22 cells grown at semipermissive temperatures of 25 and 28°C in steady-state conditions. At these temperatures, the mutants exhibited significant but more modest changes in cell size compared with cells shifted to 36°C. While cytGEMs diffusion coefficients for each strain were decreased at lower temperatures, we still observed modest positive trends with cell size across the three strains at the population level (Supplemental Figure S1, D–G).

To generate large G2-arrested cells using another approach, we inhibited the analogue-sensitive CDK1 allele cdc2-asM17 with the ATP analogue 1-NM-PP1 at 30°C (Aoi et al., 2014). These cells arrested in G2 phase and over 3–6 h grew into large mononucleate cells, similar to the cdc25 mutants (Figure 1E). While untreated cdc2-asM17 cells had an average cell length of 11.32 ± 0.57 µm, cdc2-asM17 cells treated with 1-NM-PP1 for 3 and 6 h had average cell lengths of 18.93 ± 0.84 µm and 30.05 ± 3.18 µm, respectively (mean ± STD of replicate experiments); the average cell widths were 3.84 ± 0.32 µm, 3.78 ± 0.39 µm, and 4.08 ± 0.48 µm, respectively. The average cytGEMs effective diffusion coefficients were 0.42 ± 0.04 µm²/s for control cells, 0.52 ± 0.02 µm²/s for cells treated with 3 h of 1-NM-PP1, and 0.57 ± 0.02 for cells treated with 6 h of 1-NM-PP1 (mean ± STD of replicate experiments) (Figure 1F; Supplemental Figure S1H). Control treatments did not alter cytGEMs diffusion (Supplemental Figure S1H). Overall, our results revealed a striking positive correlation between intracellular diffusion and cell size across various fission yeast strains and conditions.

Nanoparticle diffusion is maintained in large multinucleate cells

To determine whether the increase in cytGEMs diffusion in larger cell sizes was due to a decrease in the DNA-to-cytoplasm ratio, we analyzed large multinucleated fission yeast cells in which this ratio remains unchanged. We generated these multinucleate cells using the well-established mutants sid2 and cdc11, both of which are defective in the SIN regulatory pathway of cytokinesis (Nurse et al., 1976; Balasubramanian et al., 1998; Krapp et al., 2004; Grallert et al., 2012). These conditional mutants continue to grow in length and undergo nuclear division cycles in the absence of septation. sid2-as cells treated with the ATP analogue 1-NM-PP1 formed progressively larger cells with multiple nuclei at 3 and 6 h of treatment (Figure 2, A–C). Control sid2-as cells had an average length of 10.36 ± 0.34 µm, while sid2-as cells treated with 1-NM-PP1 for 3 and 6 h had lengths of 16.55 ± 0.91 µm and 25.95 ± 1.6 µm, respectively (mean ± STD of replicate experiments). Based on the cell length and number of nuclei per condition, we estimated that the DNA-to-cytoplasm ratio of 1-NM-PP1-treated cells did not decrease compared with the control. Despite being larger in cell size, we found that cytGEMs diffusion coefficients in treated sid2-as cells (3 h: 0.48 ± 0.6 µm²/s; 6 h: 0.56 ± 0.01 µm²/s) were comparable with those in control cells (0.52 ± 0.04 µm²/s) (mean ± STD of replicate experiments) (Figure 2D; Supplemental Figure S2A).

FIGURE 2:

Nanoparticle diffusion does not change with cell size in large multinucleate cells

(A) Images (sum projection of three middle slices) of S. pombe sid2-as cells with nuclear membrane marker Ish1-mScarlet (red). Cells were grown at steady state 30°C and treated with 0.25% DMSO or 10 µM ATP analogue 1-NM-PP1. Left to right: 6 h DMSO, 3 h 1-NM-PP1, and 6 h 1-NM-PP1. Scale bar is 5 µm. (B) Cell length, (C) number of nuclei (mean ± STD of replicate experiments; N_CELLS ≥ 140 per condition from three biological replicates), and (D) cytGEMs diffusion coefficients (mean ± STD of replicate experiments; N_GEMS ≥ 5546 per condition from four biological replicates) for sid2-as cells (one-way ANOVA, *** - p < 0.001, **** - p < 0.0001).

Next, we inhibited cytokinesis by using a temperature-sensitive mutant cdc11-119 (Nurse et al., 1976). Wild-type cells and cdc11-119 cells were grown at the permissive temperature 25°C overnight before being shifted to the nonpermissive temperature 36°C for 3 h. We observed comparable cytGEMs diffusion coefficients in the cdc11-119 cells compared with control populations (Supplemental Figure S2, B and C). Overall, these results suggest that a decrease in the DNA-to-cytoplasm ratio, rather than an increase in cell size alone, underlies the observed increase in intracellular diffusion in the large cells in Figure 1.

Ribosomal and total protein concentrations decrease in large cells

We hypothesized that the cell size–dependent changes in cytGEMs diffusion reflect alterations in cytoplasmic composition or concentration. Previous studies indicate that an increase in diffusion can correlate with a decrease in ribosome concentration or overall protein concentration (Delarue et al., 2018; Neurohr et al., 2019; Molines et al., 2022). To assess ribosomal concentration, we measured the fluorescence intensity of Rps2-GFP, a functional fusion of the essential small subunit ribosomal protein expressed at the native locus (Knapp et al., 2019; Lemière et al., 2022). wee1-50, wild-type, and cdc25-22 cells expressing Rps2-GFP were grown at the permissive temperature 25°C overnight and then shifted to the nonpermissive temperature 36°C for 6 h before imaging (Figure 3A). To ensure consistent processing, cells from the three strains were mixed and imaged in the same field. We found a distinct inverse relationship of Rps2 intensity with cell size (Figure 3B). In binned data, the average intensity of Rps2-GFP was significantly lower in larger cells (cell length ≥ 18 µm) compared with medium-sized cells (cell length between 9 and 18 µm); however, no significant differences were detected between medium and smaller cells (cell length ≤ 9 µm) (Figure 3, B and C).

FIGURE 3:

Large cells have decreased concentrations of ribosomal protein

(A) Image (sum projection of three middle slices) of a mixture of wee1-50, wild-type, and cdc25-22 live cells with ribosomal protein marker Rps2-GFP. Scale bar is 5 µm. (B) Rps2-GFP intensity and length per cell. (C) Rps2-GFP intensities (mean ± STD per cell length category; N_CELLS ≥ 72 per condition from three biological replicates) measured in a mixture of wee1-50, wild-type, and cdc25-22 cells and categorized by cell length. (D) Image (sum projection of three middle Z-slices) of a mixture of wee1-50, wild-type, and cdc25-22 fixed cells, treated with RNase A, and stained with FITC. (E) FITC intensity and length per cell. (F) FITC intensities (mean ± STD per length category; N_CELLS ≥ 73 per condition from three biological replicates) measured in a mixture of wee1-50, wild-type, and cdc25-22 fixed cells and categorized by cell length. Intensity values for Rps2-GFP and FITC are normalized to the mean intensity of the 9 ≤ L <18 category (one-way ANOVA test, * - p < 0.05, **** - p < 0.0001).

To assess overall protein concentration, we measured the intensity of fluorescent dye fluorescein isothiocyanate (FITC) staining (Knapp et al., 2019). wee1-50, wild-type, and cdc25-22 cells were shifted to 36°C for 6 h, mixed, fixed, stained with FITC, and then imaged and analyzed for fluorescence intensity (Figure 3, D and E). In data binned by cell size, compared with medium-sized cells, mean FITC intensity was ∼ 6% lower in bigger cells (one-way ANOVA, p = 0.04) and 5% lower in smaller cells (one-way ANOVA, p = 0.1) (Figure 3F). These results were consistent with an overall decrease in dry mass density seen previously in cdc25-25 cells (Odermatt et al., 2021). Overall, our results showed that larger cells exhibited a decrease in ribosomal protein concentration, and to a lesser extent, overall protein concentration, which begin to provide an explanation for the increase in cytGEMs mobility with increasing cell size.

Proteome composition varies with cell size

Finally, to gain further insight into the mechanisms underlying changes in diffusion, we investigated how the molecular composition of the cytoplasm changes with cell size by comparing the proteomes of S. pombe wee1-50, wild-type, and cdc25-22 cells. Mass spectrometry analysis was performed using stable isotope labeling by amino acids in cell culture (SILAC) in pairwise comparisons. First, wee1-50 and cdc25-22 SILAC strains were labeled with different lysine and arginine isotopes overnight at the permissive temperature of 25°C and then shifted to the nonpermissive temperature of 36°C for 6.5 h. This produced similar size distributions similar as those observed in the wee1-50 and cdc25-22 cell populations analyzed in Figure 1. Proteomic analyses detected 3353 proteins out of 5,117 identified S. pombe proteins (∼65% coverage) (The UniProt Consortium, 2022), Relative protein concentration ratios, plotted on log₂-transformed graphs, showed consistent results across two experimental repetitions (R = 0.83, Pearson) (Figure 4A). We categorized proteins by their subcellular location or macromolecular complex such as histones (magenta), ribosomes (orange), and endoplasmic reticulum (ER) (cyan) (Figure 4A). Finally, we grouped proteins by their subcellular location or macromolecular complex and averaged their collective ratios (Figure 4B). Because the relative concentrations of each protein were calculated and normalized within each strain, we note that these analyses cannot reveal alterations in real protein concentrations but only the changes relative to other proteins.

FIGURE 4:

Proteome composition varies with cell size

The proteomes of cdc25-22 and wee1-50 cells grown at 25°C and shifted to 36°C for 6.5 h were compared using SILAC mass spectrometry. Concentrations of each protein were determined per strain and normalized to the respective strain's proteome. (A) Relative protein concentration ratios (cdc25-22/wee1-50) for each detected protein; log₂ ratios from two replicates are plotted. Upper right quadrant (blue) indicates superscaling proteins that have relative protein concentration ratios that are more than 1. These proteins are relatively more abundant in cdc25-22 compared with wee1-50. By contrast, the lower left quadrant (red) indicates subscaling proteins that have relative protein concentration ratios that are less than 1. These proteins are relatively less abundant in cdc25-22 compared with wee1-50. Histone proteins, ribosomal proteins and ER proteins (with the number of proteins represented in parentheses) are highlighted to demonstrate different scaling relationships at different subcellular locations. (B) Average relative concentration log₂ ratios of proteins grouped by subcellular localization in comparison of two replicates. (C) 2D annotation enrichment analysis using the Protein Slope values paired by sequence orthology  between fission yeast and cultured human cells (data from Lanz et al. 2022). The identity line is dashed. Each dot is an annotation group, and the position of the dot is determined by the mean slope value of the proteins in the group (rank-based). Positive and negative enrichment scores indicate groups of superscaling and subscaling proteins, respectively.

Overall, we observed differential scaling of proteins when comparing the proteomes of large and small cells. Proteins associated with the nucleus including the nucleolus, histones, and other chromosome-associated proteins subscaled with cell size: they were underrepresented in the large cdc25-22 cells compared with small wee1-50 cells (Figure 4, A and B, red quadrant). This subscaling behavior was expected, as chromosome-associated proteins such as histones, are known to scale with DNA content rather than cell size (Amodeo et al., 2015; Claude et al., 2021). Notably, ribosomal proteins also exhibited subscaling, in accordance with our observation that Rps2 concentration was decreased in these larger cells (Figure 4, A and B). Overall cytoplasmic proteins were also subscaling with cell size. In contrast, proteins associated with the ER, mitochondria, and vacuoles superscaled with cell size: they were overrepresented in the large cdc25-22 cells compared with small wee1-50 cells (Figure 4, A and B, blue quadrant).

Next, we examined the proteome data for scaling of cellular processes and signaling pathways previously implicated in the regulation of cytoplasmic properties. One candidate signaling pathway that regulates ribosome concentration and stress responses is the TORC1 pathway (Delarue et al., 2018). Proteins associated with TORC complexes and ribosome biogenesis exhibited subscaling with cell size (Supplemental Figure S3A). To test whether TORC1 activity was decreased in large cells, we found that factors downstream of the TORC1-regulated Sfp1 transcription factor also subscaled with size (Tai et al., 2023). However, in contrast to cell size–dependent proteome changes in other cell types (Neurohr et al., 2019; Terhorst et al., 2023; Lanz et al., 2022, 2024), we detected no significant superscaling effects on stress-associated pathways such as the core environmental stress response (Chen et al., 2003). Given that small viscogens such as trehalose and glycerol also regulate cytoplasmic viscosity, we noted that proteins involved in trehalose biosynthesis subscaled with size and those associated with trehalose breakdown superscaled with size (Supplemental Figure S3A) (Sakai et al., 2024).

To characterize the top hits for subscaling and superscaling proteins in our dataset, we performed a gene ontology enrichment analysis (PANTHER overrepresentation test) (Supplemental Figure S3B). The top superscaling proteins were generally involved in metabolic pathways associated with membrane-bound organelles, whereas top subscaling proteins were associated with cell polarity regulation at cell tips, and mRNA regulation and gene expression in the nucleus. Notably, among the subscaling cell tip proteins were the DYRK family protein kinase Pom1 and its regulators Tea4, Tea1, and Mod5, which had mean relative protein concentration log₂ ratios of −1.10, −0.84, −0.58, and −0.82, respectively. These factors localize to cell tips and contribute to cell size sensing for cell size regulation (Martin and Berthelot-Grosjean, 2009; Moseley et al., 2009; Hachet et al., 2011; Wood and Nurse, 2015).

Additional proteomic comparisons between wee1-50, wild-type, and cdc25-22 cells grown under other conditions supported these results. First, ratios of the proteomes of cdc25-22 and wild-type cells that were shifted to 36°C for 6.5 h (as described in Figure 4) showed similar trends as our comparison between cdc25-22 and wee1-50 cells (Supplemental Figure S4A). Second, we compared cdc25-22 and wee1-50 strains grown at steady state at 28°C (similar as Supplemental Figure S1, D and E). Here, we observe the same general trends in the proteome, with the notable exception of ribosome proteins which scaled with cell size in these conditions (Supplemental Figure S4B).

Our fission yeast results showed striking similarities with the effects of cell size on human proteomes. An annotation-based analyses comparing subscaling and superscaling proteins in H. sapiens RPE-1 cells (Lanz et al., 2022, 2024) with our data on cdc25-22/ wee1-50 in S. pombe revealed notable conservation (Figure 4C). For instance, the subscaling of ribosomal and chromatin-associated proteins, as well as superscaling of ER and mitochondria proteins with cell size, were conserved. Moreover, a comparison of cell size–dependent changes in the proteomes of RPE-1 cells and Saccharomyces cerevisiae also revealed similar relationships (Lanz et al., 2022, 2024). Thus, the composition of the proteome exhibits characteristic changes that are likely generally conserved signatures of cell size in eukaryotes.

DISCUSSION

Here we show that intracellular diffusion coefficients of macromolecular complex-sized particles exhibit a significant positive correlation with increasing cell size. Relative to wild-type cells, cytGEMs diffusion decreased in small wee1-50 mutant cells and increased in large cdc25-22 mutant cells (Figure 1). However, cytGEMs diffusion was not changed in large multinucleate cells, suggesting that DNA-to-cytoplasm ratio may be a critical parameter influencing diffusion rates rather than cell size alone (Figure 2). In investigating the mechanism underlying these changes in diffusion, we found that small and large cells exhibited distinct proteomic compositions, with large cells showing decreased concentrations of ribosomal and nuclear proteins relative to other proteomic components (Figures 3 and 4). These results support a model in which diffusion increases in larger cells due to a decrease in ribosome concentration and alterations in the concentrations of various cytoplasmic components. In proliferating cells, the effect of the DNA-to-cytoplasm ratio suggests that a limiting factor may be the number of gene copies necessary to maintain gene expression levels needed to support the exponential growth of the cytoplasm (Zhurinsky et al., 2010; Marguerat and Bähler, 2012; Neurohr et al., 2019; Balachandra et al., 2022; Cadart and Heald, 2022; Xie et al., 2022).

Overall, our study supports the premise that the properties of the cytoplasm vary at different cell sizes. While previous studies have focused on the apparent dilution of the cytoplasm and changes in biochemical composition in larger cells (Neurohr et al., 2019; Lanz et al., 2022), our findings show that cell size impacts diffusion and crowding in the cytoplasm in a range of cell sizes. Given that physical properties of the cytoplasm have broad range of effects on the inner workings of the cell, including the rates of most biochemical reactions, our results introduce a critical component in our understanding of the effects of cell size on cellular physiology.

One way in which cell size affects GEMs diffusion could be through alterations in cytoplasmic density (Molines et al., 2022). Quantitative phase imaging in fission yeast reveals fluctuations in intracellular dry mass density during the cell cycle, as well as a progressive drop in density in cdc25-22 arrested cells (Odermatt et al., 2021). This work suggests that density fluctuations arise from variations in growth rate accompanied by a constant mass biosynthesis. Therefore, one mechanism for cytoplasmic dilution and increased diffusion in cdc25-22 arrested cells may be the extended duration in G2 phase, during which the rate of volume growth slightly exceeds the rate of mass biosynthesis. In contrast, decreased diffusion in wee1-50 cells may be attributed to a higher intracellular density in dividing cells, which are enriched in the wee1-50 population (Hagan et al., 1990; Rowley et al., 1992; Odermatt et al., 2021).Thus, cell cycle–dependent alterations in the coordination between biosynthesis and volume growth may be contributors to the observed size-dependent changes in diffusion.

Our studies contribute to a growing body of evidence indicating that the cytoplasm not only becomes more dilute, but also that its composition remodels with increasing cell size. Comparison of our results with recent data in human cells and budding yeast suggests that this remodeling of the proteome is largely conserved across these eukaryotic organisms (Figure 4C) (Lanz et al., 2022, 2024). For example, data from these three organisms reveal a consistent pattern in which subscaling proteins are enriched in nuclear and ribosomal proteins while superscaling proteins are enriched in ER and mitochondrial proteins, as well as metabolic proteins (Figure 4; Supplemental Figure S4). Another class of subscaling proteins includes factors involved in sensing cell size for cell size control through dilution-based mechanisms. Examples include Whi5 in S. cerevisiae and Rb in mammalian cells (Schmoller et al., 2015; Zatulovskiy et al., 2020). In S. pombe, it is noteworthy that a component of its cell size sensing system—the DYRK kinase Pom1, along with its regulators Tea4, Tea1, and Mod5—also exhibited subscaling behavior. This property may contribute to cell size sensing through local dilution of cortical Pom1 at the midcell (Martin and Berthelot-Grosjean, 2009; Moseley et al., 2009; Wood and Nurse, 2015; Gerganova et al., 2019). This analysis identifies a new cellular location enriched for subscaling proteins—at the cell tips—consistent with the idea that, similar to DNA, the physical size of the cell tips does not scale proportionally with cell size in these rod-shaped cells.

In the cell, changes in diffusive-like movements may be due to multiple factors, including intracellular density, viscogen concentration, ATP-dependent processes, and cytoplasmic organization (Brangwynne et al., 2011; Munder et al., 2016; Persson et al., 2020; Huang et al., 2022). Ribosomes and mRNA are not only central for the biosynthesis of proteins, but also may directly impact cytoplasmic diffusion as significant macromolecular crowding agents (Delarue et al., 2018; Gade et al., 2024; Xie et al., 2024). Our studies (Figure 3; Supplemental Figure S3) revealed a decrease in ribosomal protein concentration in large fission yeast cells, which may be mediated through regulation by the TORC1 pathway. However, in addition to ribosomes, our data suggest that other factors also contribute to effects of cell size. We found that variations in ribosomal protein concentration could not fully account for all the diffusion data; for instance, we identified several cases where cells exhibited changes in diffusion without detectable changes in ribosomal protein concentration—for instance in wee1-50 cells (Figures 1 and 3) or lack of subscaling of ribosomal proteins in wee1-50 and cdc25-22 cells grown at 28°C (Supplemental Figures S1 and S4). It is possible that the methods to assess ribosome concentration were not sensitive enough to detect subtle changes, or that they serve as an imperfect measure of the more relevant polysome concentration. Additionally, there are likely to be other factors that contribute to diffusion changes, including the concentration of small viscogens like trehalose and glycerol (Supplemental Figure S3) and aspects of cytoplasmic organization.

Changes in macromolecular crowding and diffusion are predicted to have significant impacts on the biochemistry and mechanobiology within the cell. For instance, these physical properties of the cytoplasm not only affect rates of biochemical reactions, dynamics of molecular conformational changes, and protein expression, but they also impact organelle size, cytoskeletal dynamics and the organization of condensates (Rivas and Minton, 2016; Mitchison, 2019; Marshall, 2020). One reason why cell size is maintained in a homeostatic manner is to maintain the state of the cytoplasm. When homeostatic mechanisms fail in abnormally large cells seen in senescence, aging and disease states, altered cytoplasmic properties may contribute to slower growth rates, changes in cell regulatory states, abnormal cellular function, and cell death (Neurohr and Amon, 2020; Xie et al., 2022). Future studies are likely to reveal the diverse ways cell size impacts the intracellular environment and its implications for cellular functions.

MATERIALS AND METHODS

Yeast strains and media

S. pombe strains were constructed and maintained using standard methods (Forsburg, 2003). The strains used in this study are listed in Table 1. For expression of 40 nm cytGEMs, yeast cells were transformed with the plasmid pREP41X-PfV-mSapphire for expression of the protein fusion PfV encapsulin-mSapphire (Delarue et al., 2018; Lemière et al., 2022; Garner et al., 2023). These cells were grown in EMM3S (250 mg/ml adenine, 250 mg/ml histidine, 250 mg/ml uracil, and no leucine) media with 0.1 µg/ml thiamine for an intermediate level of expression from the nmt1* promoter to optimize the appropriate numbers of cytGEMs in each cell (Maundrell, 1993; Molines et al., 2022). In other experiments, cells were grown in rich YES (Figure 3) or SILAC adjusted EMM (Figure 4) (Swaffer et al., 2016).

TABLE 1.

Key resources used in this study.

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Figure	Additional Information
Genetic reagent (S. pombe)	Ish1-mScarlet	Chang Lab collection	FC 3318	1	h- ade6<<mCherry-psy1 ish1-GFP:kanMX ura4-D18
Genetic reagent (S. pombe)	Ish1-mScarlet, cdc25 mutant	This manuscript	FC 3339	1	cdc25-22 ade6<<mCherry-psy1 ish1-GFP:kanMX
Genetic reagent (S. pombe)	Ish1-mScarlet, wee1 mutant	This manuscript	FC 3340	1	wee1-50 ish1-GFP:kanMX
Genetic reagent (S. pombe)	cytGEMs	Chang Lab collection	FC 287	1	h- pREP41X-PfV-Sapphire leu1-32
Genetic reagent (S. pombe)	cytGEMs, cdc25 mutant	This manuscript	FC 3341	1	h+ cdc25-22, pREP41X-PfV-Sapphire leu1-32
Genetic reagent (S. pombe)	cytGEMs, wee1 mutant	This manuscript	FC 3342	1	h- wee1-50, pREP41X-PfV-Sapphire leu1-32 hist?
Genetic reagent (S. pombe)	cytGEMs, cdc2 mutant	This manuscript	FC 3343	1	h90 cdc2-asM17, pREP41X-PfV-Sapphire, leu1-32, ura4-D18
Genetic reagent (S. pombe)	cytGEMs, sid2 mutant, ish1-mScarlet	This manuscript	FC 3344	2	h+ sid2-as ish1:mScarlet-I:hphMX6 ade6-M210, pREP41X-PfV-Sapphire, leu1-32 ura4-D18
Genetic reagent (S. pombe)	cytGEMs, cdc11 mutant	This manuscript	FC 3345	2	h- cdc11-119, pREP41X-PfV-Sapphire leu1-32
Genetic reagent (S. pombe)	Rps2-GFP	Chang Lab collection	FC3209	3	h- rps2-GFP::kanR leu1-32 ura4-D18 ade6-210
Genetic reagent (S. pombe)	Rps2-GFP, cdc25 mutant	This manuscript	FC 3346	3	cdc25-22 rps2-GFP::kanR leu1-32 ura4-D18
Genetic reagent (S. pombe)	Rps2-GFP, wee1 mutant	This manuscript	FC 3347	3	wee1-50 rps2-GFP::kanR leu1-32 ade6-210
Genetic reagent (S. pombe)	car2 mutant	This manuscript	FC 3348	4	car2∆::kanMX4 arg1-230 lys3-37
Genetic reagent (S. pombe)	car2 mutant, cdc25 mutant	This manuscript	FC 3349	4	cdc25-22 car2∆::kanMX4 arg1-230 lys3-37
Genetic reagent (S. pombe)	car2 mutant, wee1 mutant	This manuscript	FC 3350	4	wee1-50 car2∆::kanMX4 arg1-230 lys3-37
Chemical compound/drug	YES 225 Media	Sunrise Science Production	#2011	3
Chemical compound/drug	Edinburgh Minimum Media (EMM)	MP Biomedicals	#4110-32	1-4
Chemical compound/drug	Agar
Chemical compound/drug	Histidine	Sigma-Aldrich	#H8000	1-3
Chemical compound/drug	Uracil	Sigma-Aldrich	#U0750	1-3
Chemical compound/drug	Adenine	Sigma-Aldrich	#A9126	1-3
Chemical compound/drug	Thiamine	Sigma-Aldrich	#T4625	1, 2
Chemical compound/drug	DMSO	Thermo Fisher Scientific	#67-68-5	1, 2
Chemical compound/drug	1-NM-PP1	Thermo Fisher Scientific	#50-203-0494	1, 2
Chemical compound/drug	Agarose	Invitrogen	#16500500	3
Chemical compound/drug	Dulbecco's Phosphate Buffer Saline	Thermo Fisher Scientific	14190144	3
Chemical compound/drug	4% formaldehyde (methanol-free)	Thermo Fisher Scientific	#28,906	3
Chemical compound/drug	RNAse A	Thermo Fisher Scientific	#EN0531	3
Chemical compound/drug	Fluorescin isothiocyanate isomer I	Sigma	#F7250	3
Chemical compound/drug	Light arginine (L-ARGININE:HCL–Unlabeled)	Cambridge Isotope Laboratories	#ULM-8347-PK	4
Chemical compound/drug	Heavy arginine (L-ARGININE:HCL(13C6,99%))	Cambridge Isotope Laboratories	#CNLM-2265-H-0.25	4
Chemical compound/drug	Light lysine (L-LYSINE:2HCL–Unlabeled)	Cambridge Isotope Laboratories	#ULM-8766-PK	4
Chemical compound/drug	Heavy lysine (L-LYSINE:2HCL(4,4,5,5-D4,96-98%))	Cambridge Isotope Laboratories	#DLM-2640-0.5	4
Chemical compound/drug	Iodoacetamide	Sigma	#I1149-5G	4
Chemical compound/drug	TPCK-treated trypsin	Worthington	#LS003740	4
Chemical compound/drug	Sep-Pak 50 mg C18 column	Waters	##054955	4
Software, algorithm	µManager v. 1.41	Edelstein et al., 2010; Edelstein et al., 2014		1-3
Software, algorithm	Mathworks	Mathworks	2018a	1, 2
Software, algorithm	Python	Drake Jr. and Van Rossum, 1995	3.8.8	3, 4
Software, algorithm	Prism	GraphPad	Version 9.4.1	1-3
Software, algorithm	FIJI ImageJ	Schindelin et al., 2012		1-3
Software, algorithm	MaxQuant (v2.4.2)			4
Other	µ-Slide VI 0.4 channel slide	Ibidi	#80606	1, 2

Temperature shift and inhibitor treatments

Fission yeast cells of different cell sizes and ploidy were generated using conditional cell cycle mutants. For temperature-shift experiments (Figures 1 and 2) wild-type and temperature-sensitive mutant cells were inoculated from colonies freshly grown from the frozen stocks on EMM3S (minus leucine) agar plates grown at 25°C for 3 d and stored at room temperature for less than 7 d. Cells were inoculated in liquid EMM3S medium and grown at 25°C with shaking for over 12 h to exponential phase in the range of OD₆₀₀ 0.2 to 0.6. The flasks were then transferred to a 36°C shaking incubator for the indicated period (3–6 h). The cells were then harvested and mounted in chambers for imaging on the lab bench and promptly returned to 36°C in the prewarmed microscope system incubator. No significant differences in cytGEMs diffusion were found when mounting cells on the bench at room temperature (∼5 min of preparation time) versus preparing cells inside the temperature-controlled cage installed on the microscope. For experiments at permissive or semipermissive temperatures (Supplemental Figure S1), cells were maintained at a steady temperature (25 or 28°C) for ∼18 h and imaged at the indicated temperature in the incubator. For inhibition of cdc2-as and sid2-as alleles, cells were grown in liquid EMM3S at 30°C with shaking and treated with 10 µM 1-NM-PP1 (100-fold dilution of a 4 mM stock in DMSO) (#50-203-0494, #67-68-5, Thermo Fisher Scientific) for 3–6 h. Cells were harvested and imaged as above.

Preparation of cells for live cell microscopy

Cells were mounted just before imaging in µSlide VI 0.4 channel slides (#80606, Ibidi – 6 channels slide, channel height 0.4 mm, length 17 mm, and width 3.8 mm, tissue culture treated and sterilized). The µSlide was first precoated by incubation with 100 µg/ml lectin (#L1395, Sigma) for at least 15 min at room temperature and then removed from the chamber. For mounting cells, 1 ml of liquid yeast culture was centrifuged for 2 min in a microcentrifuge tube at 400 × g at room temperature. Most supernatant was removed, and the cell pellet was gently resuspended in the remaining ∼100 µl media. A total of 50 µl of this concentrated cell mixture was added to the precoated chamber and allowed to adhere for 2 min and then washed three times with prewarmed media to remove nonadhered cells.

Microscopy

For imaging of cytGEMs, live cells were imaged with a TIRF Diskovery system (Andor) with a Ti-Eclipse 2 inverted microscope stand (Nikon Instruments), a 488 nm laser illumination, a 60X TIRF oil objective (NA: 1:49, oil DIC N2) (#MRD01691, Nikon), and a sCMOS camera (Zyla, Andor). These components were controlled with µManager v. 1.41 (Edelstein et al., 2010; Edelstein et al., 2014). Temperature was maintained by a black panel cage incubation system (#748-3040, OkoLab). Cells were mounted in µSlide VI 0.4 channel slides (#80606, Ibidi – 6 channels slide, channel height 0.4 mm, length 17 mm, and width 3.8 mm, tissue culture treated and sterilized).

For imaging of nuclei and fluorescence intensity quantification, cells were imaged on a Ti-Eclipse inverted microscope (Nikon Instruments) with a spinning-disk confocal system (Yokogawa CSU-10) that includes 488 nm and 541 nm laser illumination (with Borealis) and emission filters 525±25 nm and 600±25 nm respectively, 40X (NA: 0.6) and 60X (NA: 1.4) objectives, and an EM-CCD camera (Hamamatsu, C9100-13). These components were controlled with µManager v. 1.41 (Edelstein et al., 2010, 2014). Temperature was maintained by a black panel cage incubation system (#748-3040, OkoLab).

Imaging and analysis of cytGEMs

Cells expressing cytGEMs nanoparticles were imaged in fields of 1K x 1.2K pixels or smaller using highly inclined laser beam illumination at 100 Hz for 5 s. Cells generally exhibited 10–20 of cytGEMs nanoparticles/cell. CytGEMs were tracked with the ImageJ (Schindelin et al., 2012) particle Tracker 2D-3D tracking algorithm from MosaicSuite (Sbalzarini and Koumoutsakos, 2005) with the following parameters: run(“Particle Tracker 2D/3D,” “radius = 3 cutoff = 0 per/abs = 0.03 link = 1 displacement = 6 dynamics = Brownian”). The analyses of the cytGEMs tracks were as described in Delarue et al., 2018, with methods to compute mean square displacement (MSD) using MATLAB (MATLAB_R2018, MathWorks). The effective diffusion coefficient D_eff was obtained by fitting the first 10 timepoints of the MSD curve (MSD_truncated) to the canonical two-dimensional (2D) diffusion law for Brownian motion: MSD_truncated(τ) = 4 ⋅ Deff ⋅ τ. cytGEMs were generally analyzed collectively in multiple cells in the whole field of view. For analyses of individual cells (Figure 1D; Supplemental Figure S1, A–C), cells were individually cropped from field images, and cytGEMs were tracked with the same MosaiSuite parameters with the exception of per/abs = 0.03. In correlation analyses in Supplemental Figure S1, A–C, size outliers (wee1-50 ≥ 10 µm, N_CELLS = 5; cdc25-22 ≤ 10 µm, N_CELLS = 6) were excluded.

Measurement of cellular dimensions and nuclei count

Cellular dimensions were measured manually from brightfield images using ImageJ Line Selection tool. “Straight Line” or “Segmented Line” was used depending on cell morphology. For determination of the number of nuclei, strains with the nuclear envelope marker Ish1 tagged with a fluorescent protein were grown in EMM3S (minus leucine) media, and number of nuclei were counted manually. Septated cells were excluded from analysis.

Ribosomal concentration quantification

Ribosomal concentration was measured in individual cells using the fluorescence intensity of ribosomal protein Rps2-GFP, as described (Knapp et al., 2019; Lemière et al., 2022). Cells expressing Rps2-GFP were grown in rich YES liquid media at 25°C overnight and shifted to 36°C for 6 h before imaging. Cells were mounted on a 2% agarose (#16500500, Invitrogen) in YES 225 (#2011, Sunrise Science Production) pad and imaged with 488 nm laser illumination via spinning disk confocal microscopy. The Rps2-GFP signal was acquired in 500 nm z-step stacks, and a sum of stack of the middle three z-slices was used for intensity quantification. For each selected cell, the Rps2-GFP signal intensities were measured along the long cell axis (averaged over 4 µm in width) and normalized by cell length. The signal was corrected for background intensity and uneven illumination of the field. Rps2-GFP signals were defined as the average of the mean signal between 0.2 and 0.3 and the mean signal between 0.7 and 0.8 (peak signals in the cytoplasm, avoiding the nucleus) along the normalized cell length. Finally, all Rps2-GFP signals were normalized to the mean of the cell length (L) category 9 ≤ L < 18 µm.

FITC staining to quantify cellular protein concentration

Total protein was measured in individual fission yeast cells using FITC staining, similar as described (Knapp et al., 2019; Odermatt et al., 2021; Lemière et al., 2022). Cells were grown in YES liquid media at 25°C overnight and shifted to 36°C for 6 h until fixation. A total of 1 ml of exponential-phase (OD₆₀₀ = 0.2–0.6) cell culture was fixed with 4% final concentration of formaldehyde (methanol-free 37% solution, #28906, Thermo Fisher Scientific, Waltham) and incubated at 4°C overnight. Fixed cells were washed three times with PBS (#14190, Thermo Fisher Scientific) and resuspended in 100 µl of PBS. A total of 100 µl of fixed cells were treated with 0.1 mg/ml RNAse A (#EN0531, Thermo Fisher Scientific) and incubated in a rotator for 2 h at 37°C. Next, cells were washed and resuspended in PBS and stained with 50 ng/ml FITC (#F7250, Sigma) for 30 min, washed three times with PBS, and resuspended in PBS. Cells were mounted on a 2% agarose (#16500500, Invitrogen) in Dulbecco's Phosphate Buffer Saline (Thermo Fisher Scientific, 14190144) pad and imaged with 488 nm laser illumination via spinning disk confocal microscopy. FITC signal was acquired and analyzed using similar methods as the Rps2-GFP experiments described above.

LC-MS/MS sample preparation

Proteomic experiments were performed using SILAC (Ong et al., 2002). SILAC-compatible fission yeast strains containing car2∆ were grown in SILAC adjusted media (Edinburgh Minimal Media [#4110712, MP Biomedicals] + 6 mM ammonium chloride + 0.04 mg/ml arginine and 0.03 mg/ml lysine) using either light or ‘‘heavy’’ versions of Lysine and Arginine (Swaffer et al., 2016). The ‘‘light’’ (Agr0 Lys0) version of the media contained L-Arginine and L-Lysine built with normal 12C and 14N isotopes; the ‘‘heavy” (Arg6 Lys4) version had L-Arginine containing six 13C atoms and L-Lysine containing four deuterium atoms. For SILAC experiments, cells were grown for at least eight generations at the indicated temperatures 25–36°C with shaking before collection, diluted in the morning and evening so they are always below OD₆₀₀ = 0.3. The mean cell volume for proteomics samples was determined by Z2 Coulter Counter (Beckman Coulter), and the mean cell volumes of these samples matched those of the corresponding samples used in the cytGEMs experiments.

A total of 10 ml of fission yeast cultures were pelleted by centrifugation at 3000 × g for 2 min at 4°C. The supernatant was removed, and cell pellets were snap frozen in liquid nitrogen and stored at −80°C. Frozen pellets were resuspended in 300 µl of yeast lysis buffer (50 mM Tris, 150 mM NaCl, 5 mM EDTA, 0.2% Tergitol, pH 7.5; + a cOmplete ULTRA Tablet) with 700 µl of glass beads. Lysis was performed at 4°C in a MPBio Fastprep24 (four cycles with the following settings: 6.0 m/s, 40 s). Cell lysates were cleared by centrifugation at 12,000 × g for 5 min at 4°C. Protein concentration was quantified using a Pierce BCA Protein Assay Kit (Prod# 23255). Equal amounts of protein from each SILAC-labeled lysate were mixed. The mixed lysates were then denatured/reduced in 1% SDS and 10 mM DTT (15 min at 65°C), alkylated with 5 mM iodoacetamide (15 min at room temperature), and then precipitated with three volumes of a solution containing 50% acetone and 50% ethanol (on ice for 10 min). Proteins were resolubilized in 2 M urea, 50 mM Tris-HCl, pH 8.0, and 150 mM NaCl, and then digested with TPCK-treated trypsin (50:1) overnight at 37°C. Trifluoroacetic acid and formic acid were added to the digested peptides for a final concentration of 0.2% (pH ∼3). Peptides were desalted with a Sep-Pak 50 mg C18 column (Waters). The C18 column was conditioned with 500 µl of 80% acetonitrile and 0.1% acetic acid and then washed with 1mL of 0.1% trifluoroacetic acid. After samples were loaded, the column was washed with 2 ml of 0.1% acetic acid followed by elution with 400 µl of 80% acetonitrile and 0.1% acetic acid. The elution was dried in a Concentrator at 45°C.

LC-MS/MS data acquisition

Desalted SILAC-labeled peptides were analyzed on a Fusion Lumos mass spectrometer (Thermo Fisher Scientific, San Jose, CA) equipped with a Thermo EASY-nLC 1200 LC system (Thermo Fisher Scientific, San Jose, CA). Peptides were separated by capillary reverse phase chromatography on a 25 cm column (75 µm inner diameter, packed with 1.6 µm C18 resin, AUR2-25075C18A, Ionopticks, Victoria Australia). Peptides were introduced into the Fusion Lumos mass spectrometer using a 125-min stepped linear gradient at a flow rate of 300 nl/min. The steps of the gradient are as follows: 3–27% buffer B (0.1% (vol/vol) formic acid in 80% acetonitrile) for 105 min, 27–40% buffer B for 15 min, 40–95% buffer B for 5 min, and finally maintained at 90% buffer B for 5 min. Column temperature was maintained at 50°C throughout the procedure. Xcalibur software (Thermo Fisher Scientific) was used for the data acquisition and the instrument was operated in data-dependent mode. Advanced peak detection was enabled. Survey scans were acquired in the Orbitrap mass analyzer (Profile mode) over the range of 375 to 1500 m/z with a mass resolution of 240,000 (at 200 m/z). For MS1, the Normalized AGC Target (%) was set at 250 and max injection time was set to “Auto.” Selected ions were fragmented by Higher-energy Collisional Dissociation (HCD) with normalized collision energies set to 31, and the fragmentation mass spectra were acquired in the Ion trap mass analyzer with the scan rate set to “Turbo.” The isolation window was set to 0.7 m/z window. For MS2, the Normalized AGC Target (%) was set to “Standard” and max injection time was set to “Auto.” Repeated sequencing of peptides was kept to a minimum by dynamic exclusion of the sequenced peptides for 30 s. Maximum duty cycle length was set to 1 s.

Spectral searches

All raw files were searched using the Andromeda engine (Cox et al., 2011) embedded in MaxQuant (v2.4.2) (Cox and Mann, 2008). In brief, two-label SILAC search was conducted using MaxQuant's default Arg6/10 and Lys4/8. Variable modifications included oxidation (M) and protein N-terminal acetylation, and carbamidomthyl (C) was a fixed modification. The number of modifications per peptide was capped at 5. Digestion was set to tryptic (proline-blocked). Peptides were ‘‘Requantified,’’ and maxquant's match-between-runs feature was not enabled. Database search was conducted using the UniProt proteome - UP000002485. Minimum peptide length was 7 amino acids. FDR was determined using a reverse decoy proteome (Elias and Gygi, 2007).

Peptide quantitation

Our SILAC analysis utilized MaxQuant's “proteinGroups.txt” output file. Contaminant and decoy peptide identifications were discarded. When applicable, the “Leading Razor Protein” designation was used to assign non-unique peptides to individual proteins. Normalized SILAC ratios were used to determine changes in the relative concentrations of individual proteins.

Protein annotations

Protein annotations in Figure 4 were sourced from UniProt columns named “Gene Ontology IDs” “Subcellular localization [CC]” or PomBase “Complex Annotations” unless otherwise noted (The Uniprot Consortium, 2022; Rutherford et al., 2024). Protein localization was strictly parsed so that each annotated protein belongs to only one of the designated groups. Proteins with two or more annotations were ignored (except for the “Cytoplasm/Nucleus” category which required a nuclear and cytoplasmic annotation and for categories, e.g., Histone, Chromosome, Nucleolus, which also contained a “Nucleus” annotation).

2D Annotation Enrichment analysis

Annotation enrichment analysis was performed as described previously (Cox and Mann, 2012). The protein annotation groups were deemed significantly enriched and plotted if the Benjamini–Hochberg FDR was smaller than 0.02. The position of each annotation group on the plot is determined by the enrichment score (S). The enrichment score is calculated from the rank ordered distribution of Protein Slope values:

S = 2(R_group − R_remaining proteins)/n

Where R_group and R_remaining proteins are the average ranks for the proteins within an annotation group and all remaining proteins in the experiment, respectively, and n is the total number of proteins. Highlighted annotation groups were manually curated.

Gene ontology enrichment analysis

Relative protein concentration ratios were averaged between the two repetitions of proteomics experiments. Underrepresented and overrepresented proteins were defined as having a minimum of a 10% change in their mean relative protein concentration ratio. Gene ontology (GO) process characterization of protein lists was performed using Protein Analysis Through Evolutionary Relationships (PANTHER) overrepresentation analysis version PANTHER 18.0 (Thomas et al., 2022).

Abbreviations used:

cytGEMs

cytoplasmic genetically-encoded multimeric nanoparticles

Deff

effective diffusion coefficient

DMSO

dimethyl sulfoxide

EMM

Edinburgh Minimal Media

GO

gene ontology

LC-MS/MS

liquid chromatography with tandem mass spectrometry

MSD

mean squared displacement

SA/V

surface area to volume ratio

SILAC

stable isotope labeling by amino acids in cell culture

YES

yeast extract supplemented.