Figure S5.
Tensin3 is required for fibronectin fibrillogenesis. (A) Generation of TNS3 KO cells using CRISPR; Western blot shows endogenous tensin3 levels in WT and 3 clonal populations generated following CRISPR. Clone #2 was a complete KO line (all cells were depleted of TNS3); clones #1 and #3 were a mixed population of KO and WT cells. (B) Analysis of (normalized) distance of adhesion structures from the cell periphery: a region of interest (ROI) was drawn around the edge of the cell to be analyzed. The ROI was filled with the Fill command, then the threshold function was used to select the pixels within the cell ROI, and converted to a binary image. The Distance Map function was used to create a Euclidean distance map (EDM), which was multiplied by the pixel size to convert to distance in µm. The resulting image was divided by the maximum distance value, then multiplied by 100 to convert to normalized distance between 0 and 100% (displayed here in a heat-map). Adhesion structures in the original image were segmented, and the resulting adhesion ROIs were applied to the normalized distance image. The mean intensity of each adhesion ROI was measured, which gives the normalized distance from the cell periphery, where 100 is the centre and 0 is the cell edge. Scale bar indicates 10 µm. (C) Western blot of endogenous tensin3 in TIFs after 2 rounds of transfection with either scrambled siRNA (Ctrl) or two different oligos targeting tensin3 separately (siRNA #1 or siRNA #2). Strongest knockdown was achieved with siRNA #1 alone, which was used for subsequent experiments. (D) Representative images of TIFs after siRNA-mediated knockdown of tensin3 as described above. Cells were cultured on fibronectin-coated glass overnight prior to fixation, then immunostained for α5 integrin. Quantification shows that tensin3 knockdown reduced the number of α5 integrin-positive adhesions. Errors bars are SEM; n = 14 (Control) or 16 (siRNA #1) cells; *** indicates P < 0.001 (t test). (E) Histograms and accompanying gaussian curve fit for the normalized distance (percent of maximum) of each a5-integrin positive adhesion structure from the cell edge to the cell centre; bar graphs show the mean and SEM of the normalized distance of all the adhesions; n = 3,913 (Control) and 2,450 (siTensin3) adhesions from 21 to 21 cells, respectively; **** indicates P < 0.0001, Mann-Whitney t test. Results are representative of two independent repeats. (F) Representative images of endogenous tensin3 co-localization with fibronectin in U2OS WT and TNS3 KO or TIFF control and TNS3 siRNA-treated cells. (G) Representative images (background subtracted) of fibronectin fibrils produced by U2OS WT or U2OS TNS3 KO cells spread overnight on fibronectin-coated soft (5 kPa) or stiff (50 kPa) polyacrylamide gels. Quantification of fibril formation (as performed for Fig. 7, A–C) normalized to WT cells on 5 kPa polyacrylamide gel shows that absence of tensin3 reduces significantly fibronectin fibril production on all substrates. Error bars are SEM, n = 63–77 squares from 30 to 35 images, pooled from three independent experiments; **** indiciates P < 0.0001 (Kruskal-Wallis ANOVA with Dunn’s multiple comparison test). (H) Representative (background subtracted) images of fibronectin fibrils produced by WT U2OS cells spread overnight on fibronectin-coated soft (5 kPa) or stiff (50 kPa) polyacrylamide gels in the presence of either DMSO (−) or Y-27632 (+). Quantification of fibril formation (as above) normalized to DMSO-treated cells on 5 kPa polyacrylamide gels shows that Y-27632 treatment dramatically reduces fibronectin fibril formation on all substrates. Error bars are SEM, n = 60–86 squares from 30 to 35 images, pooled from three independent experiments; **** indiciates P < 0.0001 (Kruskal-Wallis ANOVA with Dunn’s multiple comparison test). (I) Histograms and accompanying gaussian curve fit (related to Fig. 7 D) for the normalized distance (percent of maximum) of the distance of fibronectin (647-FN) fibers from the cell periphery formed by vinculinKO MEFs with or without expression of mCh-vinFL. Note that cells without vinculin have fewer centrally-located FN fibrils compared to cells expressing mCh-vinFL. Scale bars in E and F indicate 10 µm. Source data are available for this figure: SourceData FS5.

Tensin3 is required for fibronectin fibrillogenesis. (A) Generation of TNS3 KO cells using CRISPR; Western blot shows endogenous tensin3 levels in WT and 3 clonal populations generated following CRISPR. Clone #2 was a complete KO line (all cells were depleted of TNS3); clones #1 and #3 were a mixed population of KO and WT cells. (B) Analysis of (normalized) distance of adhesion structures from the cell periphery: a region of interest (ROI) was drawn around the edge of the cell to be analyzed. The ROI was filled with the Fill command, then the threshold function was used to select the pixels within the cell ROI, and converted to a binary image. The Distance Map function was used to create a Euclidean distance map (EDM), which was multiplied by the pixel size to convert to distance in µm. The resulting image was divided by the maximum distance value, then multiplied by 100 to convert to normalized distance between 0 and 100% (displayed here in a heat-map). Adhesion structures in the original image were segmented, and the resulting adhesion ROIs were applied to the normalized distance image. The mean intensity of each adhesion ROI was measured, which gives the normalized distance from the cell periphery, where 100 is the centre and 0 is the cell edge. Scale bar indicates 10 µm. (C) Western blot of endogenous tensin3 in TIFs after 2 rounds of transfection with either scrambled siRNA (Ctrl) or two different oligos targeting tensin3 separately (siRNA #1 or siRNA #2). Strongest knockdown was achieved with siRNA #1 alone, which was used for subsequent experiments. (D) Representative images of TIFs after siRNA-mediated knockdown of tensin3 as described above. Cells were cultured on fibronectin-coated glass overnight prior to fixation, then immunostained for α5 integrin. Quantification shows that tensin3 knockdown reduced the number of α5 integrin-positive adhesions. Errors bars are SEM; n = 14 (Control) or 16 (siRNA #1) cells; *** indicates P < 0.001 (t test). (E) Histograms and accompanying gaussian curve fit for the normalized distance (percent of maximum) of each a5-integrin positive adhesion structure from the cell edge to the cell centre; bar graphs show the mean and SEM of the normalized distance of all the adhesions; n = 3,913 (Control) and 2,450 (siTensin3) adhesions from 21 to 21 cells, respectively; **** indicates P < 0.0001, Mann-Whitney t test. Results are representative of two independent repeats. (F) Representative images of endogenous tensin3 co-localization with fibronectin in U2OS WT and TNS3 KO or TIFF control and TNS3 siRNA-treated cells. (G) Representative images (background subtracted) of fibronectin fibrils produced by U2OS WT or U2OS TNS3 KO cells spread overnight on fibronectin-coated soft (5 kPa) or stiff (50 kPa) polyacrylamide gels. Quantification of fibril formation (as performed for Fig. 7, A–C) normalized to WT cells on 5 kPa polyacrylamide gel shows that absence of tensin3 reduces significantly fibronectin fibril production on all substrates. Error bars are SEM, n = 63–77 squares from 30 to 35 images, pooled from three independent experiments; **** indiciates P < 0.0001 (Kruskal-Wallis ANOVA with Dunn’s multiple comparison test). (H) Representative (background subtracted) images of fibronectin fibrils produced by WT U2OS cells spread overnight on fibronectin-coated soft (5 kPa) or stiff (50 kPa) polyacrylamide gels in the presence of either DMSO (−) or Y-27632 (+). Quantification of fibril formation (as above) normalized to DMSO-treated cells on 5 kPa polyacrylamide gels shows that Y-27632 treatment dramatically reduces fibronectin fibril formation on all substrates. Error bars are SEM, n = 60–86 squares from 30 to 35 images, pooled from three independent experiments; **** indiciates P < 0.0001 (Kruskal-Wallis ANOVA with Dunn’s multiple comparison test). (I) Histograms and accompanying gaussian curve fit (related to Fig. 7 D) for the normalized distance (percent of maximum) of the distance of fibronectin (647-FN) fibers from the cell periphery formed by vinculinKO MEFs with or without expression of mCh-vinFL. Note that cells without vinculin have fewer centrally-located FN fibrils compared to cells expressing mCh-vinFL. Scale bars in E and F indicate 10 µm. Source data are available for this figure: SourceData FS5.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal