Huang WYC et al. (2022), Cytoplasmic organization promotes protein diffu...

XB-ART-59334

Nat Commun 2022 Sep 23;131:5599. doi: 10.1038/s41467-022-33339-0.

Show Gene links Show Anatomy links

Cytoplasmic organization promotes protein diffusion in Xenopus extracts.

Huang WYC , Cheng X , Ferrell JE .

???displayArticle.abstract???
The cytoplasm is highly organized. However, the extent to which this organization influences the dynamics of cytoplasmic proteins is not well understood. Here, we use Xenopus laevis egg extracts as a model system to study diffusion dynamics in organized versus disorganized cytoplasm. Such extracts are initially homogenized and disorganized, and self-organize into cell-like units over the course of tens of minutes. Using fluorescence correlation spectroscopy, we observe that as the cytoplasm organizes, protein diffusion speeds up by about a factor of two over a length scale of a few hundred nanometers, eventually approaching the diffusion time measured in organelle-depleted cytosol. Even though the ordered cytoplasm contained organelles and cytoskeletal elements that might interfere with diffusion, the convergence of protein diffusion in the cytoplasm toward that in organelle-depleted cytosol suggests that subcellular organization maximizes protein diffusivity. The effect of organization on diffusion varies with molecular size, with the effects being largest for protein-sized molecules, and with the time scale of the measurement. These results show that cytoplasmic organization promotes the efficient diffusion of protein molecules in a densely packed environment.

???displayArticle.pubmedLink??? 36151204
???displayArticle.pmcLink??? PMC9508076
???displayArticle.link??? Nat Commun
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: agxt sfpq
GO keywords: cytoplasm [+]

???attribute.lit??? ???displayArticles.show???

	Figure 1. a Schematic of the experimental setup used in this study: Xenopus laevis eggs were fractionated to obtain the undiluted cytoplasmic extract, which was then deposited on the surface of an imaging dish and covered with mineral oil. b Time-lapse images of the self-organizing cytoplasm, visualized with 1 µM ER-Tracker Red dye. c Image (left) of dilute 100-nm (diameter) microspheres in extracts. Trajectories (right) are examples from single-particle tracking in the organized cytoplasm (interior regions in Fig. 2a). The relative positions between the three tracked particles were adjusted for display (they were tens of microns apart in the extracts). d Ensemble mean squared displacement (MSD) analysis derived from the single-particle tracking. Shaded area, standard error of the mean (SEM). Dashed lines are fitting to MSD(t)=Γtα, where Γ is the generalized diffusion coefficient34. The fitted {Γ, α} for disorganized and organized cytoplasm are {1.12, 0.75}, and {0.65, 0.78}, respectively. The localization error of these SPT experiments, quantified with immobilized microspheres, was below 0.01 µm2 (Supplementary Fig. 1a). e, f Representative FCS data of BSA-Alexa Flour 488 in cytoplasmic extracts. The full intensity trajectory (60 s) is shown in Supplementary Fig. 2c. The inset depicts a molecule diffusing through a confocal spot. G(τ) is the autocorrelation function, where τ is the lag time; δI(t)=I(t)−⟨I(t)⟩, where I(t) is the fluorescence intensity and the brackets denote time averages. The autocorrelation curve was fitted by an anomalous diffusion model (dashed line). In this example, the number of particles was 42, and the average intensity (shown in e) was 32 kHz, corresponding to a molecular brightness of 0.76 kHz/molecule. Source data are provided as a Source Data file.
	Figure 2. a, Spatial regions of self-organization (color-coded), classified by the ER morphology. The dark sphere near the center of the inner interior is the nuclei. The image was stitched from multiple confocal images. b Representative confocal images of FCS foci, which are marked by the crosshairs. The annotated numbers refer to snapshots taken prior to FCS measurements in c. c Tracing protein diffusion during self-organization by FCS. The plots show the characteristic diffusion times τD and the diffusion mode α obtained from fitting the FCS curves. Black and gray dashed lines correspond to averages from disordered cytoplasmic extracts (in this example, 740 µs, Deff = 11.9 µm2/s) and cytosolic extracts (360 µs, Deff = 24.7 µm2/s), respectively. d Benchmark measurements from cytosolic extracts and buffers. ELB, egg lysis buffer; PBS, Phosphate-buffered saline. Insets show zoom-in of the data. Each condition was measured five times; all data presented by error bars are mean ± SEM. n = 5 measurements of different positions in a sample. e, Biological repeat of c, prepared from a separate batch of eggs. f, g are isolated traces from c and e. Error bars, estimated errors (standard deviation) from curve fitting (each fitting was performed over n = 107 data points). Scale bars, 100 µm. Source data are provided as a Source Data file.
	Figure 3. a FCS data from Fig. 2c–e consolidated in a parameter space of {α, τD}. b Nocodazole (NOC, 33 µM) abolishes self-organization of cytoplasm, visualized by ER-Red Tracker. c Nocodazole-treated cytoplasmic extracts show little changes in diffusion time (diffusion time ratio close to 1), compared with the sample without nocodazole at similar timepoints (<5 min). The y-axis shows τD normalized by the average τD from early timepoints (<20 min). Error bars, SEM. n = 4 and three measurements of different positions in an extract with and without nocodazole, respectively. Source data are provided as a Source Data file.
	Figure 4. a Molecular sizes of probes used in this study. b–d Normalized diffusion times of Alexa Fluor 488 dye (AF488), dextran-MW70,000 (Dex-70K), and dextran-MW2,000,000 (Dex-2M) in organized cytoplasmic extracts. These experiments focused on the border regions of the organized cytoplasm such that data collected <10 min could be better compared. Each panel shows data from two droplets of extracts prepared from a single batch of eggs; BSA diffusion was taken as a standard for comparison. Error bars, SEM. n = 4, 6, 4, 5, 6, and 4 measurements of different positions in an extract for data from left to right, respectively. Source data are provided as a Source Data file.
	Figure 5. a, b Normalized diffusion times of BSA in actin-intact extracts, prepared without the addition of cytochalasin B (CyB). Note that the diffusion rate did not increase more during the self-organization of actin-intact extracts (open circles) than it did in actin-inhibited extracts (solid circles). The two panels are two independent experiments. c, d Degree of subdiffusion in actin-intact and actin-inhibited extracts. Data shown in panel c and d are from the same experiment shown in panel a and b, respectively. Error bars, SEM. n = {3, 3, 4, 5, 6; 3, 3, 3, 4, 5, 6, 3} measurements of different positions in an extract for data from left to right and top to bottom, respectively. Source data are provided as a Source Data file.
	Figure 6. a Schematic of the model: molecular and organelle reorganization optimizes molecular diffusion by creating longer mean free paths for diffusing molecules. The green object represents a protein of interest diffusing in the cytoplasm; the blue objects represent obstacles in the cytoplasm, such as organelles, cytoskeleton elements, and macromolecular complexes. In this model, proteins in organized cytoplasm remain free to diffuse in all directions. b Diffusion effects of molecular reorganization illustrated with mean squared displacement, MSD (t) = Γtα. The parameters {Γ, α} estimated for disorganized cytoplasm, organized cytoplasm and the cytosol are {8, 0.75}, {8.25, 0.70}, and {30, 0.85}, respectively. Note that based on the tradeoff between diffusion times and subdiffusivity in the organizing cytoplasm (Fig. 3a), we have assumed that the acceleration of protein movements over these distance scales comes at a cost of slower dynamics over longer distance scales. We also limit the comparison around the length scale of FCS measurements (~0.1–0.5 µm) since protein dynamics may show distinct characteristics at shorter and longer distance scales. Source data are provided as a Source Data file.
	Supplementary Figure 1: Localization error in SPT and comparison with FCS. a, Data from Fig. 1d overlaid with MSD of immobilized microspheres on glass. This shows that the localization error was below 0.01 µm2, which was lower than the mobile particles in extracts. b, Data from Fig. 1d overlaid with the dextran-2M FCS data from Fig. 4. The SPT data was analyzed for 100-nm microspheres and the FCS data was obtained for dextran-2M (diameter of ~90 nm). Error bars, SEM (standard error of the mean).
	Supplementary Figure 2: FCS calibration. a, Estimated point-spread function (PSF) imaged by scanning a 100-nm immobilized microsphere on a glass substrate. These microspheres were chosen because of their exceptionally high molecular brightness (~300-fold brighter than BSA-AF488), which makes them ideal for characterizing the PSF and ��. Scale bar, 1 µm. b, FCS curve of the identical microsphere but in water. Dashed line, fitting to Brownian diffusion. Fitting to an anomalous diffusion yielded �� =0.97, suggesting that imperfection in optics did not notably distort the type of diffusion. c, Same as a but measured in a disorganized extract. d, e, FCS curves of Alexa Fluor 488 in disorganized and organized extracts, respectively. Dashed lines, fitting to Brownian diffusion. Fitting to an anomalous diffusion yielded �� = 0.98 ± 0.06 (±SEM, 3 data points) and 0.94±0.03 (4 data points), respectively. f, Full intensity trajectory of data shown in Fig. 1e.
	Supplementary Figure 3: Diffusion analysis of FCS autocorrelation functions. a, c, e, Fitting a Brownian model to data for BSA diffusion in PBS buffers, cytosolic extracts, and disordered cytoplasmic extracts. b, d, f, Fitting an anomalous diffusion model to the same data. Note that Brownian motion ideally yields 𝛼 = 1 even with anomalous diffusion fits (also see Supplementary Fig. 2b)
	Supplementary Figure 4: Dynamical transition in cytoplasmic extracts: an example of extracts self-organized slightly earlier. A biological repeat of Fig. 2 experiments. In this sample, the cytoplasm formed organized structures earlier than the other experiments; as a result, the diffusion changes between the disorganized and organized states were especially apparent. Note that this extract formed cell-like pattern earlier than other examples, leading to earlier transition to faster diffusion.
	Supplementary Figure 5: Measurement error in cytoplasmic extracts. Estimation of the upper bound of measurement error in cytoplasmic extracts. The data show n =10 repetitive measurements at the same position in a mature border; each measurement was acquired identically as Fig. 2c and e. The error bar shows the standard deviation (50 µs) around the mean. Note that since cytoplasmic extracts are highly dynamic and non-stationary (unlike homogeneous samples), this error includes measurement error and fluctuations of cytoplasmic organization. Source data are provided as a Source Data file.
	Supplementary Figure 6: Statistical tests of BSA diffusion in microtubule-perturbed cytoplasmic extracts. Mann–Whitney U tests for data in Fig. 3c. Each panel are data measured on the same day. Asterisks () denotes the significance levels calculated as two-tailed p-values: , �� ≤ 0.15 ; n.s., not significant, �� > 0.15. The p-values for a and b are 0.686, 0.100, respectively. Error bars, SEM. n= 4, 4, 3, and 3 measurements of different positions in an extract for data from left to right, respectively. Source data are provided as a Source Data file.
	Supplementary Figure 7: Size estimation of probes used in this study. The relative sizes of the diffusion probes were measured in PBS buffers using FCS. The autocorrelation functions were fitted by a Brownian model (𝜶 = 𝟏) (dashed curves). The diffusion times yield the relative Stokes radii by the Stokes-Einstein equation, 𝑫 ∝ 𝟏/𝑹. The relative Stokes radii for AF488, BSA, dextran-70K, and dextran-2M are 0.12, 1.00, 0.88, and 13.3, respectively. The relative values were converted to absolute lengths by taking the Stokes radius of BSA to be 3.5 nm. As a note, the polydisperse nature of dextran-2M could also be effectively fitted by the anomalous diffusion model, though the diffusion was expected to be Brownian11. Nevertheless, the fitted diffusion times between the two models were similar.
	Supplementary Figure 8: Statistical tests of particle diffusion in cytoplasmic extracts. Mann–Whitney U tests for data in Fig. 4. Each panel are data measured on the same day. Asterisks () denotes the significance levels calculated as two-tailed p-values: , �� ≤ 0.01; , �� ≤ 0.05; *, �� ≤ 0.15; n.s., not significant, �� > 0.15. The p-values for a-f are 0.686, 0.004, 0.114, 0.016, 0.310, and 0.057, respectively. Error bars, SEM. n = {3, 4, 5, 6; 4, 4, 3, 5; 6, 6, 3, 4} measurements of different positions in an extract for data from left to right and top to bottom, respectively. Source data are provided as a Source Data file.
	Supplementary Figure 9: Statistical tests of BSA diffusion in actin-intact cytoplasmic extracts. Mann–Whitney U tests for data in Fig. 5. Asterisks () denotes the significance levels calculated as two-tailed p-values: , �� ≤ 0.05; , �� ≤ 0.15; n.s., not significant, �� > 0.15. The p-values for a and b are {0.34, 1.00} (from left to right), and {0.036, 0.032, 0.071, 0.057} (from top left to top right to bottom), respectively. The two panels show data from two independent experiments. Error bars, SEM. n = 4, 3, 3, 3, 3, 4, 5, and 3 measurements of different positions in an extract for data from left to right, respectively. Source data are provided as a Source Data file.

References [+] :

Afanzar, The nucleus serves as the pacemaker for the cell cycle. 2020, Pubmed, Xenbase