Intrinsic self-organization of integrin nanoclusters within focal adhesions is required for cellular mechanotransduction

Abstract

Upon interaction with the extracellular matrix, the integrin receptors form nanoclusters as a first biochemical response to ligand binding. Here, we uncover a critical biodesign principle where these nanoclusters are spatially self-organized, facilitating effective mechanotransduction. Mouse Embryonic Fibroblasts (MEFs) with integrin β3 nanoclusters organized themselves with an intercluster distance of ~550 nm on uniformly coated fibronectin substrates, leading to larger focal adhesions. We determined that this spatial organization was driven by cell-intrinsic factors since there was no pre-existing pattern on the substrates. Altering this spatial organization using cyclo-RGD functionalized Titanium nanodiscs (of 100 nm, corroborating to the integrin nanocluster size) spaced at intervals of 300 nm (almost half), 600 nm (normal) or 1000 nm (almost double) resulted in abrogation in mechanotransduction, indicating that a new parameter i.e., an optimal intercluster distance is necessary for downstream function. Overexpression of α-actinin, which induces a kink in the integrin tail, disrupted the establishment of the optimal intercluster distance, while simultaneous co-overexpression of talin head with α-actinin rescued it, indicating a concentration-dependent competition, and that cytoplasmic activation of integrin by talin head is required for the optimal intercluster organization. Additionally, talin head-mediated recruitment of FHOD1 that facilitates local actin polymerization at nanoclusters, and actomyosin contractility were also crucial for establishing the optimal intercluster distance and a robust mechanotransduction response. These findings demonstrate that cell-intrinsic machinery plays a vital role in organizing integrin receptor nanoclusters within focal adhesions, encoding essential information for downstream mechanotransduction signalling.

Figure 1.. Optimal inter-cluster spacing of 600 nm promotes cell spreading, adhesion maturation and YAP nuclear translocation.

a, Reconstructed PALM super-resolution image (left) of integrin β3 in nascent adhesions spread on fibronectin for 10–15 mins, and quantification of the integrin intercluster distance histogram (middle), and box plot (right). Scale bar, 500 nm. n=1702 from 6 cells. b, Representative super-resolution (SR) confocal image (left) of MEFs overexpressing integrin β3-GFP spread for 1 hr, along with zoom of white box (right). Relative frequency of the intercluster distance of integrin β3 nanoclusters. Scale bar (left) 10 µm. Scale bar (right) 2µm. c, Representative AFM images of the nanopatterned substrates, and linescans (middle) of the marked regions in the AFM images reporting the inter-disc distance. Scale bars, 500 nm. d, Confocal fluorescence images of the nanodisc substrates labelled with Dylight-405 labeled Neutravidin. Scale bars, 500 nm. e, Images of MEFs expressing EGFP-Paxillin with phalloidin staining for F-actin, along with the merge and zoom of the white box, spread on nanopatterns with different periods. Scale bar for Merge, 10 μm. Scale bar for zoomed inset, 3 μm. f, Quantifications of cell spread area (n ≥ 57 cells from three independent experiments, analyzed by the Kruskal–Wallis one-way ANOVA and Dunn’s post hoc test; P values are indicated in the figure) and g, percentage lamellipodia area with adhesions (n ≥ 14 cells from three independent experiments, analyzed by the Kruskal–Wallis test and Dunn’s post hoc test; P values are indicated in the figure). h, Confocal images showing YAP localization as a function of nanopattern periodicity. Nuclear boundary as marked by Hoechst staining indicated by whited dashed lines. Scale bar, 20 μm. i, Quantification of the nuclear to cytoplasmic YAP ratio (n ≥ 58 cells from three independent experiments, analyzed by the Kruskal–Wallis test and Dunn’s post hoc test; P values are indicated in the figure). Box plots with data overlay display the upper and lower quartiles and a median, the circle represents the mean, and the whiskers denote the standard deviation values, along with the individual points and P values. Detailed statistics are summarized in Supplementary Table 1.

Figure 2. Overexpression of α-actinin 1 integrin binding domain (SR12) disrupts integrin inter-cluster spacing establishment.

a, F-actin and paxillin immunofluorescence images of MEFs overexpressing mCherry-α-actinin 1 spread on different nanodisc periods, along with the merge and zoom of the white box in paxillin. Scale bar for merge, 15 µm. Scale bar for zoom, 2 µm. b, Quantifications of cell spread area (n ≥ 13 cells from three independent experiments, analyzed by one-way ANOVA and Tukey’s post hoc test; P values indicated in the figure) and average adhesion area/cell (n ≥ 14 cells from three independent experiments, analyzed by Kruskal–Wallis one-way ANOVA and Dunn’s post hoc test; P values indicated in the figure). c, Representative F-actin images of cells overexpressing EGFP-SR12 on different disc periods along with d, quantification of cell spread area (n ≥ 31 cells from three independent experiments, analyzed by Kruskal–Wallis test and Dunn’s post hoc test; P values indicated in the figure). Scale bar, 15 µm. e, Representative F-actin images of cells overexpressing mCherry-Myosin IIA-N93K mutant on different disc periods along with f, quantification of cell spread area (n ≥ 13 cells from three independent experiments, analyzed by one-way ANOVA and Tukey’s post hoc test; P values indicated in the figure). Scale bar, 15 µm. Box plots with data overlay display upper and lower quartiles and median, circle represents the mean, and whiskers denote the standard deviation, along with individual data points. Detailed statistics are summarized in Supplementary Table 2.

Figure 3.. Integrin β3 but not integrin β1, through its talin binding function, is involved in establishing inter-cluster spacing.

a, Representative images of cells spread on the 300, 600, and 1000 nm disc diameter patterns treated with the β3 blocking peptide- G-Pen, or the β1 blocking antibody, HMβ1–1 clone, stained for F-actin. Scale bar (top), 10 µm. Scale bar (lower), 20 µm. b, Quantification of cell spread area (n ≥ 26 cells from 3 independent experiments, analyzed by one-way ANOVA test and Tukey’s post hoc test) for cells treated with G-Pen. c, Quantification of cell spread area (n ≥ 27 cells from three independent experiments, analyzed by one-way ANOVA test and Tukey’s post hoc test) for cells treated with β1 blocking antibody. d, Representative F-actin images of CHO cells overexpressing integrin β3 wt or β3 A711P mutant on different disc periods. Scale bar (top), 15 µm. Scale bar (bottom), 15 µm. Quantifications of cell spread area for e, β3-wt (n ≥ 19 cells from three independent experiments, analyzed by one-way ANOVA and Tukey’s post hoc test) and f, β3-A711P (n ≥ 36 cells from three independent experiments, analyzed by Kruskal-Wallis test). Box plots with data overlay display upper and lower quartiles and median, circle represents the mean, and whiskers denote standard deviation, along with individual data points. Detailed statistics are summarized in Supplementary Table 3.

Figure 4.. Cell establishment of inter-cluster spacing is fine-tuned by competition between talin head domain and α-actinin SR12 domain for integrin β3 binding.

a, Control, SR12 overexpressing, SR12 and Talin full length (TFL) co-overexpressing, SR12 and Talin rod (TR) co-overexpressing, SR12 and Talin head (TH) co-overexpressing, SR12 and Talin head_K272A, K274Q co-overexpressing, SR12 and Talin head_K322A, K324A co-overexpressing cells were spread on fibronectin-coated glass dishes for 15–20 mins and immunostained for FAK and actin (phalloidin). Representative images showing F-actin, FAK, merge, and zoom of the white box in FAK. Scale bar for Merge, 10 µm. Scale bar for Zoom, 3 µm. b, Quantification of the cell spread area (n >= 107 cells from three independent experiments, analyzed by Kruskal-Wallis test and Dunn’s post hoc test; P values are indicated), c, total adhesion area normalized to cell area (analyzed by Kruskal-Wallis test and Dunn’s post hoc test; P values are indicated in the figure). d, Representative images of cells spread on the 300, 600, and 1000 nm disc diameter patterns co-overexpressing SR12 and talin head, stained for F-actin. Scale bar, 20 µm. e, Quantification of cell spread area (n ≥ 18 cells from three independent experiments, analyzed by Kruskal-Wallis test and Dunn’s post hoc test; P values are indicated). All graphs show individual points along with box plots of interquartile distance, line represents median, circle represents the mean, whiskers represent standard deviation. Detailed statistics are summarized in Supplementary Table 4.

Figure 5. FHOD1 formin is required for establishing inter-cluster spacing and adhesion maturation.

a, Control, FHOD1 shRNA expressing, SR12 expressing, SR12 and Talin head (TH) co-overexpressing, and SR12, TH and FHOD1 shRNA co-overexpressing cells were spread on fibronectin-coated glass dishes for 15–20 mins and immunostained for FAK and actin (phalloidin). Representative images showing F-actin, FAK, merge, and zoom of the white box in FAK. Scale bar for Merge, 10 µm. Scale bar for Zoom, 3 µm. b, Quantification of the cell spread area (n >= 97 cells from three independent experiments, analyzed by Kruskal-Wallis test and Dunn’s post hoc test; P values are indicated), c, total adhesion area normalized to the cell area (analyzed by Kruskal-Wallis test and Dunn’s post hoc test; P values are indicated). d, Representative images of cells spread on the 300, 600, and 1000 nm disc diameter patterns expressing FHOD1 shRNA, stained for F-actin. Scale bar, 20 µm. e, Quantification of the cell spread area (n ≥ 18 cells from three independent experiments, analyzed by One-way ANOVA test and Tukey’s post hoc test; P values are indicated). All graphs show individual points along with box plots of interquartile distance, the line represents the median, the circle represents the mean and the whiskers represent the standard deviation. Detailed statistics are summarized in Supplementary Table 5.

Figure 6.. Proposed model for how integrin nanocluster distance regulate cell spreading and adhesion maturation.

Cells establish an optimal 600 nm inter-cluster spacing and show a decrease in cell spreading, adhesion formation, and YAP signalling when inter-cluster spacing is changed to 300 nm or 1000 nm. Establishment of inter-cluster spacing depends on FHOD1 formin recruited to the integrin clusters in the presence of talin head domain, to promote local actin polymerization. Conversely, FHOD1 depletion or displacement of talin head from integrin by α-actinin overexpression disrupts the establishment of integrin inter-cluster spacing.