Fig 2: Myosin-II dynamics in nocodazole-treated cells. a Upregulation of phosphorylated MLC2 (p-MLC2) after nocodazole treatment, assessed by western blotting. Densitometric comparison for the electrophoretic bands is also shown. n = 4 independent experiments, mean ± SEM. b Immunostaining for p-MLC2, F-actin, and E-cadherin in HT29 cells. c Lateral views of HT29 cells immunostained as in b. d Densitometric scans of immunosignals for p-MLC2 and E-cadherin along the yellow line drawn in b. e Time-lapse images of MLC2-EGFP expressed in HT29 cells during nocodazole treatment. Images at the right are a close-up view of the cell indicated by arrow at the left panel. Images are identical to Supplementary Movie 3. f Effects of aE-catenin knockdown on MHC-IIA distribution, as assessed by immunostaining of HT29 cells transfected with control or aE-catenin specific siRNA. Images were obtained with Airyscan. g Effects of nocodazole treatment on the distribution of p-MLC2 and E-cadherin in Colo205, LS180, and SW620 cells. Scale bars, 10 µm b, c, f left and g; 2 µm (f right). See also Supplementary Fig. 5

Fig 3: ROCK1 deletion enhances anti-detachment effects of NAC.A. Representative images of rhodamine-phalloidin staining for F- actin (red), p-MLC staining (green) and DAPI (blue) of WT and ROCK1 deficient cells treated with 3 µM doxorubicin and/or 2 mM NAC for 8 h. Cells showing disruption of central stress fibers are indicated with white arrowheads. ROCK1 deletion, but not NAC treatment, prevents DOX-induced disruption of central stress fibers. Bar, 25 µm. B and C. Representative image (B) of Western blot of p-MLC2 and MLC2, or p-cofilin and cofilin in cell lysates from attached WT and ROCK1-/- MEFs treated with 3 µM doxorubicin and/or 2 mM NAC for 30 min. Quantitative analysis (C) showing reduction of both p-MLC and p-cofilin in WT MEFs after NAC/DOX treatment compared to DOX treatment alone. The ratio of p-MLC to MLC or p-cofilin to cofilin was expressed as fold change relative to WT control. D and E. Both floating and attached cells were collected after treatment for 16 h with 3 µM doxorubicin and/or 2 mM NAC. Floating cell ratio was expressed as percentage of total cells (floating plus attached cells) under each treatment condition (D). Attached cell number was expressed as percentage of attached cells under control condition without treatment (E). *P<0.05 vs. control of the same genotype. # P<0.05 vs. WT under the same treatment condition. ¶ P<0.05 vs. the same genotype under doxorubicin only condition.

Fig 4: Hypoxia upregulates transcription of HIF target and early chondrogenic genes as well as immunodetection of collagen type II on soft substrates, which can be impacted by inhibitors of cytoskeletal tension. (A–B) Expression of mRNA encoding conserved HIF targets VEGFA and EGLN (A) and the chondrogenic transcription factor SOX9 (B) following 24 h of culture at 2% O2 on soft and stiff substrates. (C–H) Expression of mRNA for SOX9 (C), COL2A1 (D) and VEGFA (E) following culture at 2% O2 on soft substrates without and with the ROCK inhibitor Y-27632 (-/+Y). (F–G) Expression of mRNA for SOX9 (F), COL2A1 (G) and VEGFA (H) following culture at 2% O2 on soft substrates without and with the inhibitor of MLC2 polymerisation Blebbistatin (-/+Blebb). Values plotted are from 4 independent experiments and are fold change compared to expression at 20% O2 represented by the horizontal dotted line. Mean values are represented by grey/red lines. (I–K) Collagen Type II immunofluorescence staining at day 14 of chondrogenesis of hMSC on soft substrates at 20% O2 (I), 2% O2 (J) and 2% O2 with Blebbistatin (K). Brightness and contrast were adjusted to an equal degree between all conditions. (L) Quantification of Collagen Type II immunofluorescence on a per cell basis (Bl = Blebbistatin). Black horizontal lines represent means for each condition. *p < 0.05 compared to the 20% O2 condition. # represents a significant change (p < 0.05) between +/-Y conditions 2% O2 and ? represents a significant change (p < 0.05) between +/-Blebb conditions at 2% O2. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig 5: CLU interacts with ROCK1 to suppress metastatic traits. (A,B) IP–MS analysis was performed using 293FT cells with ectopic expression of CLU-Flag. Metastasis-related regulator ROCK1 was identified as a CLU binding protein. (C) Co-IP assay was performed in 293FT cells to confirm the interaction between precursor CLU and ROCK1. (D) Representative images of immunofluorescent co-staining of intracellular CLU and ROCK1 in CLU-overexpressing and vector-control cells. (E) PLA assay was performed to confirm the interaction of CLU and ROCK1 in H460 and A549 cells. (F,G) The effect of Y27632 treatment in CLU-silenced H460 cells and vector-control cells on their migration and invasion abilities. (H) Cell morphology had no difference in CLU-overexpressing H1299 cells and CLU-silenced H460 cells. F-actin was used to detect the cytoskeleton in those cells. (I) The effect of CLU on ROCK1, p-MLC2, Cofilin, p-LIMK1 and LIMK2 in CLU-overexpressing H1299 and H1975 cells and CLU-silenced H460 cells. GAPDH was used as a loading control. Results are presented as mean ± SD, *** p < 0.001, ns means no significance.