Probing short-range protein Brownian motion in the cytoplasm of living cells

Abstract

The translational motion of molecules in cells deviates from what is observed in dilute solutions. Theoretical models provide explanations for this effect but with predictions that drastically depend on the nanoscale organization assumed for macromolecular crowding agents. A conclusive test of the nature of the translational motion in cells is missing owing to the lack of techniques capable of probing crowding with the required temporal and spatial resolution. Here we show that fluorescence-fluctuation analysis of raster scans at variable timescales can provide this information. By using green fluorescent proteins in cells, we measure protein motion at the unprecedented timescale of 1 μs, unveiling unobstructed Brownian motion from 25 to 100 nm, and partially suppressed diffusion above 100 nm. Furthermore, experiments on model systems attribute this effect to the presence of relatively immobile structures rather than to diffusing crowding agents. We discuss the implications of these results for intracellular processes.

Figure 1. iMSD analysis of molecular motion in 3D: GFP Brownian diffusion in dilute solution.

(a) Pictured experiment: ‘fluorescent’ molecules freely diffuse distributing in space and time. Scanning the samples with decreasing speed allows measuring the average particle displacements over a wide spatiotemporal scale. (b) When the scanning is faster than the particle dynamics, the particle image corresponds to the autoconvolution of the instrumental point spread function, which is well approximated by a Gaussian profile. On the other hand, when scanning speed decreases, the particle starts to move significantly, that is, the correlation function squeezes in space becoming much larger than the PSF. Equation 15 in Supplementary Information describes this deformation effect as a function of the particle average displacement, and allows recovering the iMSD as a function of time delay. (c) Schematic representation of the iMSD for a molecule diffusing in 3D.

Figure 2. iMSD in dilute solution.

Experimental iMSD values at the different timescales for differently sized molecules in dilute solution at 37 °C. Monomeric GFP (N=7 measurements, green dots) shows a linear behaviour in time, as expected for Brownian motion: fit by a free diffusion model (equation 13 in Supplementary Information) yields D_w=134±4 μm² s⁻¹ (, H_r=2.5 nm); Alexa488 (D_w=428±15 μm² s⁻¹, H_r=0.75 nm, , N=7 measurements; blue dots) and 30-nm-diameter fluorescent beads (D_w=22±0.5 μm² s⁻¹, , H_r=15 nm, N=7 measurements; black dots) are acquired under the same conditions. Data are mean values±s.d.

Figure 3. iMSD analysis of GFP diffusing in the cell cytoplasm.

(a) GFP is transiently transfected into living CHO cells and analysed under physiological conditions. Arbitrary micrometre-sized areas in the cell cytoplasm are imaged sequentially at tunable timescales (inset). The iMSD values at the different timescales in the cytoplasm are reported in a double-logarithmic representation, as average of N=3 experiments, n=24 cells (black dots) and compared with the iMSD values obtained in solution (dashed green line, taken from Fig. 2). GFP translational motion below 2 × 10⁻⁵ s (corresponding to displacements in the 25–100 nm range) matches that observed in solution, indication of Brownian motion. Thus, it can be well described by the diffusion equation (χ²=0.42, P<0.01). However, the latter description does not apply to iMSD values above 2 × 10⁻⁵ s (χ²>>25, P>0.995) where a strong suppression in the GFP translational mobility is detected. (b) iMSD of GFP dimer in cytoplasm: double-logarithmic representation as average of N=2 experiments, n=9 cells (black dots). D₀ (95±5 μm² s⁻¹) and D_inf (12±3 μm² s⁻¹) are estimated by a linear fit of the iMSD below 5 × 10⁻⁵ s and above 1 × 10⁻³ s, respectively. The green bars represent the GFP dimer iMSD measured in cell lysate and the green line represents the linear fit corresponding to D_w=105±4 μm² s⁻¹. Scale bar, 5 μm. Data are mean values±s.d.

Figure 4. iMSD analysis of GFP diffusing in the cell nucleus.

(a) GFP is transiently transfected into living CHO cells and analysed under physiological conditions (inset). Arbitrary micrometre-sized areas in the cell nucleus are imaged sequentially at tunable timescales. The iMSD values at the different timescales in the nucleoplasm are reported in a double-logarithmic representation, as average of N=3 experiments, n=22 cells (black squares) and compared with the iMSD in solution (dashed line, taken from Fig. 2). The GFP motion in this compartment is never coincident with that in dilute solution. Conversely, it is suppressed over the entire spatiotemporal scale observed. Also, no clear crossover spatial scale is visible over time. Scale bar, 5 μm. Data are mean values±s.d.

Figure 5. In cuvette validation of the protein motion model.

(a) (Upper panel) Pictured GFP dynamics in the presence of colliding crowders. (a) (Lower panel) Experimental iMSD of GFP diffusing in BSA solution with different excluded volume (ϕ). For both tested excluded volume, the iMSD is not distinguishable from free diffusion and increasing excluded volume decrease molecular diffusivity (for ϕ=0.1, D=46±2 μm² s⁻¹ and N=7 measurements; for ϕ=0.2, D=24±2 μm² s⁻¹ and N=7 measurements). (b) (Upper panel) Pictured GFP dynamics in the presence of spatially organized obstacles. Fast scanning allows measuring GFP dynamics in the free space between obstacles. Decreasing scanning speed allows measuring the effect of boundaries on the GFP dynamics. (b) (Lower panel) The complete iMSD of GFP in Sephacryl beads is reported, as average of N=5 beads (black dots), and compared with the iMSD in solution outside the beads (dashed green line). The local dynamics (below 2 × 10⁻⁵ s) can be described well by GFP diffusion outside the Sephacryl beads (χ²=0.8, P<0.025). However, above 2 × 10⁻⁵ s, the latter is not representative for the whole iMSD (χ²>25, P>0.995) because of the reduction of long-range GFP mobility by the solid structure of the polymeric beads. The inset shows a representative scanning electron microscopy image of a Sephacryl beads, in which many submicrometer cavities are visible. Scale bar, 200 nm. Data are mean values±s.d.