Decisive role of mDia-family formins in cell cortex function of highly adherent cells

Abstract

Cortical formins, pivotal for the assembly of linear actin filaments beneath the membrane, exert only minor effects on unconfined cell migration of weakly and moderately adherent cells. However, their impact on migration and mechanostability of highly adherent cells remains poorly understood. Here, we demonstrate that loss of cortical actin filaments generated by the formins mDia1 and mDia3 drastically compromises cell migration and mechanics in highly adherent fibroblasts. Biophysical analysis of the mechanical properties of the mutant cells revealed a markedly softened cell cortex in the poorly adherent state. Unexpectedly, in the highly adherent state, associated with a hyperstretched morphology with exaggerated focal adhesions and prominent high-strain stress fibers, they exhibited even higher cortical tension compared to control. Notably, misguidance of intracellular forces, frequently accompanied by stress-fiber rupture, culminated in the formation of tension- and contractility-induced macroapertures, which was instantly followed by excessive lamellipodial protrusion at the periphery, providing critical insights into mechanotransduction of mechanically stressed and highly adherent cells.

Fig. 1.. Loss of mDia1 and mDia3 impairs cell migration.

(A) Loss of mDia1 and mDia3 by CRISPR-Cas9 in NIH 3T3 cells was confirmed by immunoblotting using specific antibodies. GAPDH was used as loading control. (B) Consecutive elimination of mDia1 and mDia3 increasingly decreased 2D cell migration on FN. (C) Analyses of mean square displacement of NIH 3T3 control versus mutant cells. Data represent means ± SEM. (D) Representative images from wound scratch assays of control and mutant cells as indicated. Scale bar, 200 μm. (E) Reduction of wound scratch area over time. Data indicate mean ± SD. (F) Quantification of wound closure rate. (G) Scheme of experimental setup for analysis of 3D cell migration. (H) Representative images from 3D migration movies of control and mutant cells in hydrogel. Scale bar, 200 μm. (I) Quantification of relative migration distance over time. (J) Quantification of cell speed in 3D. [(B), (F), and (J)] Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. Kruskal-Wallis test with Dunn’s multiple comparison test. Percentages are shown to better illustrate the differences between cell lines. [(B) and (C)] n, number of cells. [(E), (F), (I), and (J)] N, number of movies analyzed. [(B), (C), (E), (F), (I), and (J)] Results are pooled data from at least four biologically independent replicates. ***P < 0.001. n.s., not significant.

Fig. 2.. Elimination of mDia1 and mDia3 disrupts the contractile actin cortex.

(A) Schematic and representative images from micropipette aspiration assays illustrating the projection lengths of probed cells (black arrowheads) as indicated. Scale bar, 10 μm. (B) Initial indentation length (L_p) of control and mutant cells, as indicated, was determined by micropipette aspiration using a constant suction pressure of 50 Pa from time-lapse movies; data correspond to movie S1. (C) Representative phase-contrast images of control and mutant cells migrating on FN. Insets, enlarged images of dashed boxed regions depicting cell protrusions. Scale bar, 50 μm. (D and E) Schematic and quantitative analysis of cortical tension by AFM of control and mutant cells that adhere weakly or strongly to the underlying substrate. Note increased cortical tension of mDia1/3-KO cells on FN as compared to NIH 3T3 control. [(B), (D), and (E)] Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. Kruskal-Wallis test with Dunn’s multiple comparison test. Results are pooled data from at least three biologically independent replicates. n, number of analyzed cells. ***P < 0.001. n.s., not significant.

Fig. 3.. Loss of mDia1 and mDia3 promotes FA formation.

(A) Representative images of control and mutant cells seeded on FN and stained for vinculin and the F-actin cytoskeleton. Scale bar, 20 μm. (B) Representative images of control and mutant cells seeded on FN and stained for paxillin and the F-actin cytoskeleton. Scale bar, 20 μm. (C) Representative micrographs of control and mutant cells stained for vinculin before (top) and after binarization (bottom). Scale bar, 10 μm. (D) Quantification of relative mean vinculin intensities in FA. (E) Quantification of FA size. (F) Quantification of FA length. (G) Quantification of FA number per cell. [(D) to (G)] Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. Kruskal-Wallis test with Dunn’s multiple comparison test. n, number of cells. Results are pooled data from six biologically independent replicates. Percentages are shown to better illustrate the differences between cell lines. ***P < 0.001. n.s., not significant.

Fig. 4.. Loss of mDia1 and mDia3 affects FA turnover and dynamics.

(A) Representative images from time-lapse, TIRF movies of control and mDia1/3-KO cells migrating on FN and displaying EGFP-paxillin accumulation in FAs (left); data correspond to movie S2. Temporal color-coded time stacks of the same cells over a 60-min period (right). Bar on far right shows color-coded time stamp. Scale bar, 25 μm. (B) Quantification of FA turnover. Normalized autocorrelation of EGFP-paxillin fluorescence. (C) Quantification of FA lifetime originating from protrusions or the cell body. (D and E) Quantification of FA assembly and disassembly rates. (F and G) Representative examples of merging and splitting FAs in control and mDia1/3-KO cells color coded as in (A). Time is indicated in minutes. Scale bars, 3 μm. (H) Quantification of FA splitting and merging. Boxes indicate mean ± SD. Student’s t test. (I) Representative examples of sliding FAs in control and mDia1/3-KO cells color coded as in (A). Time is indicated in minutes. Scale bars, 3 μm. (J and K) Quantification of FA sliding velocity and distance. [(B) and (H)] n, number of cells. [(C), (D), (E), (J), and (K)] n, number of FAs. Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. Mann-Whitney U test. Results are pooled data from six biologically independent replicates. Percentages or calculated values are shown to better illustrate the differences between cell lines. [(C), (J), and (K)] Blue and red curves depict normal distribution curves of data points in box plots, corresponding to FAs originating from protrusions or the cell body, respectively. The values above the curves correspond to the respective ratios of lifetimes, sliding velocity or sliding distance of the FA populations originating from the cell body and lamellipodium. [(C), (D), (E), (H), (J), and (K)] **P < 0.01 and ***P < 0.001. n.s., not significant.

Fig. 5.. mDia1/3-deficient fibroblasts exhibit marked changes in F-actin cytoskeleton architecture.

(A) Representative STED images of an NIH 3T3 control and mDia1/3-KO cell stained with STAR635-phalloidin for the F-actin cytoskeleton. Scale bars, 10 μm. (B) Enlarged image of boxed regions shown in (A). Scale bar, 1.5 μm. (C) Quantification of relative phalloidin intensity in regions lacking SFs mainly representing the cortical F-actin cytoskeleton. (D) Quantification of relative phalloidin intensity in SFs. (E) Representative SEM images of ultrastructural cortex architecture in a control and an mDia1/3-KO cell after detergent extraction. Enlarged images of boxed regions are shown on the right. Scale bars, 5 μm at low magnification or 1 μm at higher magnification. (F) Quantification of cortical actin network pore size. [(C), (D), and (F)] Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. Mann-Whitney U test. Results are pooled data from at least three biologically independent replicates. Percentages are shown to better illustrate the differences between cell lines. n, number of cells analyzed. ***P < 0.001.

Fig. 6.. Loss of mDia1 and mDia3 amplifies force development.

(A) Representative images of an NIH 3T3 control and an mDia1/3-KO cell on FN treated with 50 μM blebbistatin (bleb) for 30 min as indicated and stained for the four and a half LIM domains 3 protein (FHL3) and phalloidin. Insets, enlarged images of boxed regions. Scale bar, 20 μm. (B) Quantification of FHL3 intensity at SFs in the absence or presence of blebbistatin. Kruskal-Wallis test with Dunn’s multiple comparison test. Results are pooled data from three biologically independent replicates. (C) Comparable expression of FHL3 in control and mDia1/3-KO cells was confirmed by immunoblotting using FHL3-specific antibodies. GAPDH was used as loading control. (D) Representative images of a fixed control and an mDia1/3-KO cell on FN costained for the F-actin cytoskeleton, zyxin and VASP. Insets, enlarged images of boxed regions depicting an SF damage site. Scale bar, 20 μm. (E) Quantification of zyxin intensity on intact SFs. (F) Quantification of VASP intensity on intact SFs. (G) Images depicting heat map and traction force field representations of cells indicated. Scale bar, 10 μm. Force scale bar is in pascals and arrows represent the local force magnitude and orientation. (H) Quantification of contractile forces. (I) Quantification of FRET efficiencies in control and mDia1/3-KO mutants expressing the internal talin tension sensor (iTS) or the C-terminal no-force control talin probe (CTS). [(B), (E), (F), (H), and (I)] Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. Mean values and percentages are shown to better illustrate the differences between cell lines. n, number of cells analyzed. [(E), (F), (H), and (I)] Mann-Whitney U test. Results are pooled data from at least three biologically independent replicates. ***P < 0.001. n.s., not significant.

Fig. 7.. mDia1/3-deficient fibroblasts form tension and contractility-induced macroapertures.

(A) Still images from a time-lapse, phase-contrast movie of an mDia1/3-KO cell migrating on FN illustrating the formation and aperture-like closure of a TCM; data correspond to movie S5. Scale bar, 50 μm. Time is indicated in hours and minutes. (B) Representative confocal z-stack projections of a control and an mDia1/3-KO cell forming multiple TCMs stained with phalloidin. Scale bar, 20 μm. (C) Close-up STED images of representative TCMs in mDia1/3-KO mutants stained with phalloidin. Insets, enlarged images of boxed regions. Scale bar, 10 μm. (D) Quantification of TCM size in mDia1/3-KO cells. (E) Quantification of TCM aspect ratio in mDia1/3-KO cells. (F to H) Representative confocal z-stack projections of TCMs stained with phalloidin and costained for α-actinin-1, nonmuscle myosin 2A (NM2A) or phosphorylated myosin light chain 2 (pMLC2). Scale bars, 10 μm. (I) Still images from a time-lapse TIRF movie illustrating TCM closure in an mDia1/3-KO cell expressing EGFP-tagged α-actinin-1; data correspond to movie S6. (J) Analysis of TCM closure in boxed, dashed region depicted in (I). Red dashed lines represent linear fits of data points. Red arrowhead indicates completion of TCM closure. (K) Representative epifluorescence and binarized images of unstretched and stretched control and mDia1/3-KO cells on elastic substrate derivatized with FN after phalloidin staining. Scale bar, 50 μm. (L) Quantification of TCM area versus total cell area in stretched and unstretched cells. Each data point represents the mean of an independent experiment in which at least 10 images were analyzed. Student’s t test. N, number of independent biological replicates. [(D) and (E)] Boxes in box plots indicate 50% (25 to 75%) and whiskers 90% (5 to 95%) of all measurements, with dashed red lines depicting the means. Medians are highlighted by indentation of boxes. n, number of cells analyzed. **P < 0.01 and ***P < 0.001.

Fig. 8.. TCM formation is preceded by SF rupture and promotes lamellipodial protrusion.

(A) Close-up images from a time-lapse, phase-contrast movie of an mDia1/3-KO cell on FN; data correspond to movie S8. Scale bar, 20 μm. Time is indicated in minutes and seconds. (B) Close-up images from a time-lapse TIRF movie of an mDia1/3-KO cell on FN expressing mScarlet-LifeAct (top) and EGFP-CAAX (bottom) showing the rupture of prominent SFs as indicated by the red arrowhead, followed by local expansion of this area and subsequent rupture of the plasma membrane leading to TCM formation; data correspond to movie S9. Scale bar, 20 μm. Time is indicated in minutes and seconds. (C) Still images from a time-lapse, phase-contrast movie of an mDia1/3-KO cell on FN exemplifying that formation of a prominent TCM is followed by a burst of protruding lamellipodia; data correspond to movie S10. Scale bar, 20 μm. Time is indicated in hours and minutes. (D) Kymographs along the two depicted lines as marked in (C) are shown. Arrowheads illustrate an increase in lamellipodial protrusion following the formation of the TCM. (E) Quantification of distance traveled by lamellipodia (green and red lines) versus TCM area (blue dashed line) over time. (F) Quantification of lamellipodial protrusion rates versus TCM area over time. Note marked decrease in protrusion rate shortly after the TCM begins to close again. (G) Still images from representative time-lapse, phase-contrast movies of a control and an mDia1/3-KO cell illustrating TCM formation after hypotonic shock exclusively in the mutant cell; data correspond to movie S11. Scale bar, 25 μm. Time is indicated in minutes and seconds. (H) Quantification of the fraction of cells forming TCMs directly after hypotonic shock. Boxes indicate fraction of cells. n, number of cells. Results are pooled data from three biologically independent replicates.