Live-cell fluorescence microscopy and SMdM unveil fast disassembly and even faster reassembly of the vimentin cytoskeleton under osmotic pressure drop and recovery. (A) Fluorescence micrographs of vimentin-mEos3.2 expressed in a living COS-7 cell, (i) initially in an isotonic cell medium, (ii-iii) under hypotonic stress in water for 140 s (ii) and 180 s (iii), and then (iv-vi) after returning to the isotonic medium for 8 s (iv), 2 min (v), and 15 min (vi). (B) Color-coded SMdM maps of the local diffusion coefficient D for vimentin-mEos3.2 expressed in a living COS-7 cell, (i) initially in an isotonic medium, (ii) under hypotonic treatment for 5-22 min, so that the vimentin cytoskeleton had disassembled, and (iii) after next reverting to the isotonic medium for 0.5-5 min, so that the vimentin cytoskeleton had reassembled. (C) Distribution of the SMdM-measured displacements of single vimentin-mEos3.2 molecules in 1 ms time windows, for the boxed regions marked as a-d in (B). Blue curves: Fits to the SMdM diffusion model, with resultant D values marked in each plot.

Fluorescence microscopy of vimentin-mEos3.2 in COS-7 cells permeabilized under different ionic strengths and pHs unveils hypersensitivity of the vimentin cytoskeleton stability to protein net charges. (A) Fluorescence micrographs of a cell (i) initially in an isotonic cell medium, (ii) 6 min after replacing the medium with a 5 mM phosphate buffer (pH = 7.3) with 350 mM glucose added, and (iii,iv) 12 s and 45 s after next adding 0.2% Triton X-100 into the medium. (B) Fluorescence micrographs for another cell before and 6 min after changing the cell medium to DPBS with 0.2% Triton X-100. (C) Model: The vimentin filament is stabilized when the accumulated negative net charges (red) are ionically screened or otherwise reduced, and destabilized when the charges are less screened or otherwise intensified. (D) The expected net charge of vimentin as a function of pH, estimated with Protein Calculator v3.4 (http://protcalc.sourceforge.net). (E) Fluorescence micrographs for a cell before and 15 min after changing the cell medium to a 5 mM acetate buffer (pH = 5.0) with 0.2% Triton X-100. (F) Fluorescence micrographs for another cell before and after replacing the medium with a 5 mM CAPSO buffer (pH = 9.0) with 0.2% Triton X-100 at 10 s and 40 s. (G) Fluorescence micrographs for another cell before and after replacing the cell medium with a 5 mM CASPO buffer (pH = 10.0) with 0.2% Triton X-100 at 1 s and 2 s.

Live-cell fluorescence microscopy of COS-7 cells expressing vimentin-mEos3.2 constructs with differently charged linkers further underscores protein net charge as a key factor in vimentin cytoskeleton stability. (A) Representative fluorescence micrographs with vimentin-(-5)-mEos3.2, in which a (-5)-charged 11 amino-acid linker was inserted between vimentin and mEos3.2 sequences, after hypotonic treatment (5 mM phosphate buffer, pH = 7.3) of different durations. (B) Representative fluorescence micrographs with vimentin-(+6)-mEos3.2, in which a (+6)-charged 11 amino-acid linker was inserted between vimentin and mEos3.2 sequences, after the same hypotonic treatment of different durations.

Comparison with cytokeratin and GFAP further generalizes the net-charge mechanism of intermediate-filament intracellular stability. (A) Comparison of immunofluorescence micrographs of endogenous cytokeratin and vimentin in fixed wild-type PtK2 cells subjected to hypotonic treatment of 5 mM phosphate buffer (pH = 7.3) for 5 min. (i) Anti-pan cytokeratin, (ii) Anti-vimentin, (iii) Overlaid image, (iv) 3D-STORM super-resolution image of anti-vimentin for the boxed region in (ii). (B) Two-color live-cell fluorescence microscopy of a PtK2 cell co-transfected with keratin-mEmerald and vimentin-mCherry, shown as separate and merged images, before (i) and after (ii,iii) hypotonic treatment in a 5 mM phosphate buffer for 120 s (ii) and 240 s (iii). (C) 3D-STORM super-resolution images of anti-GFAP for primary astrocytes isolated from the rat hippocampus, for samples without (i) and with (ii) hypotonic treatments in a 5 mM phosphate buffer for 5 min. (iii,iv) Zoom-in of the boxes in (i,ii). Colors in (A iv) and (C) encode axial (depth) position, based on the color scale shown in (C i).

List of protein amino acid (AA) sequences and estimated net charges at pH = 7.3 per Protein Calculator v3.4 (http://protcalc.sourceforge.net

Representative 3D-STORM images of immunolabeled vimentin in COS-7 cells. (A) An untreated cell. The bottom figure is a zoom-in of the boxed region in the top figure. (B) Another cell, fixed after hypotonic treatment with water for 5 min. The zoom-in image shows the complete disassembly of filaments. (C) Another cell, after hypotonic treatment with water for 5 min, but then allowed to recover for 30 min in the regular cell culture medium in the incubator, before being fixed and labeled. Color encodes axial (depth) position, according to the color scale shown in (A).

Additional fluorescence micrographs of vimentin-mEos3.2 in COS-7 cells after permeabilization under different ionic strengths and pHs. (A) Fluorescence micrographs of a cell (i) 6 min in a 5 mM phosphate buffer (pH = 7.3) with 350 mM sorbitol added and (ii) 150 s after next adding 50 µg/mL saponin into the medium. (B) Fluorescence micrographs of another cell before and after replacing the cell medium with a 5 mM CAPSO buffer (pH = 9.0) with the addition of 200 mM KCl and 0.2% Triton X-100, at 30 s and 80 s. (C) Fluorescence micrographs for another cell before and after replacing the cell medium with a 5 mM CAPSO buffer (pH = 10.0) with the addition of 150 mM KCl and 0.2% Triton X-100, at 2 s and 8 s.