Loss of cell-cell adhesion triggers cell migration through Rac1-dependent ROS generation - PubMed
- ️Sat Jan 01 2022
Loss of cell-cell adhesion triggers cell migration through Rac1-dependent ROS generation
Yu-Hsuan Chen et al. Life Sci Alliance. 2022.
Abstract
Epithelial cells usually trigger their "migratory machinery" upon loss of adhesion to their neighbors. This default is important for both physiological (e.g., wound healing) and pathological (e.g., tumor metastasis) processes. However, the underlying mechanism for such a default remains unclear. In this study, we used the human head and neck squamous cell carcinoma (HNSCC) SAS cells as a model and found that loss of cell-cell adhesion induced reactive oxygen species (ROS) generation and vimentin expression, both of which were required for SAS cell migration upon loss of cell-cell adhesion. We demonstrated that Tiam1-mediated Rac1 activation was responsible for the ROS generation through NADPH-dependent oxidases. Moreover, the ROS-Src-STAT3 signaling pathway that led to vimentin expression was important for SAS cell migration. The activation of ROS, Src, and STAT3 was also detected in tumor biopsies from HNSCC patients. Notably, activated STAT3 was more abundant at the tumor invasive front and correlated with metastatic progression of HNSCC. Together, our results unveil a mechanism of how cells trigger their migration upon loss of cell-cell adhesion and highlight an important role of the ROS-Src-STAT3 signaling pathway in the progression of HNSCC.
© 2022 Chen et al.
Conflict of interest statement
The authors declare that they have no conflict of interest.
Figures
(A) SAS cells were seeded at low (5 × 104) and high (6 × 105) densities in a 3.5-cm culture dish. 24 h later, the cells were stained for E-cadherin and ZO-1. Representative images are shown. Scale bars, 10 µm. (A, B) SAS cells were grown as described in (A) and stained for ROS, vimentin, and nucleus. Representative images are shown. Scale bars, 10 µm. The fluorescence intensity of ROS and vimentin in the cell was measured and expressed as box-and-whisker plots. The P-values were calculated from at least 150 cells pooled from three independent experiments. ***P < 0.001. (C) SAS cells were grown to confluence and then treated with 2.5 mM EGTA in serum-free medium for 6 h (Ca2+ depletion). The cells were stained for E-cadherin and ZO-1. Representative images are shown. Scale bars, 10 µm. (C, D) SAS cells were grown as described in (C) and stained for ROS, vimentin, and nucleus. Representative images are shown. Scale bars, 10 µm. The fluorescence intensity of ROS and vimentin in the cell was measured and expressed as box-and-whisker plots. The P-values were calculated from at least 150 cells pooled from three independent experiments. ***P < 0.001. (E) SAS cells were infected with lentiviruses expressing shRNAs to E-cadherin (shE-cad) or luciferase (shLuc) as the control. Three shRNA target sequences to E-cadherin (shE-cad #1, #2, and #3) were used. An equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. The expression level of vimentin was quantified and expressed as –fold relative to the control. Values (mean ± SD) are from three experiments. *P < 0.05 and **P < 0.01. (E, F) Cells as described in (E) were seeded at high density (6 × 105) in a 3.5-cm culture dish. 24 h later, the cells were stained for ROS, vimentin, and nucleus. Representative images are shown. Scale bars, 40 µm. (E, G) Cells as described in (E) were grown at high density. The mRNA levels of E-cadherin and vimentin were measured by quantitative real-time PCR and expressed as –fold relative to the shLuc control. Values (mean ± SD) are from seven independent experiments. *P < 0.05 and ***P < 0.001. Source data are available for this figure.
Cell lines as indicated were seeded at low (5 × 104) and high (6 × 105) densities in a 3.5-cm culture dish. 24 h later, the cells were stained for ROS, vimentin, and nucleus. Representative images are shown. Scale bars, 10 µm. The fluorescence intensity of ROS and vimentin in the cell was measured and expressed as box-and-whisker plots. The P-values were calculated from at least 150 cells pooled from three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001. SCC-25, a human tongue squamous cancer cell line; SiHa and HeLa, human cervical cancer cell lines; Hs578T, a human breast cancer cell line; DU145, a human prostate cancer cell line; and MDCK, a canine renal epithelial cell line.
(A) SAS cells were grown at a sub-confluent condition and stained for vimentin and nucleus. (B) SAS cells (control) and those with E-cadherin depletion (shE-cad #1, #2, and #3) were grown at a confluent condition and stained for vimentin and nucleus. The insets show that the structure of vimentin expressed at a sub-confluent condition or induced by E-cadherin depletion was similar, which exhibited mainly as particles and/or squiggle forms.
(A) SAS cells were grown into a monolayer, and a cell-free gap (i.e., wound) of ∼500 µm in width was created. The cells were stained for E-cadherin and nucleus at 0 and 6 h after the wound was created. Representative images are shown. Scale bars, 20 µm. The percentage of the cells with the adherens junction at the proximal (<200 µm) and distal (>400 µm) areas from the wound was measured (n ≥ 192). Values (mean ± SD) are from three independent experiments. **P < 0.01. (B) Wound healing assay was performed, and the cells were stained for ROS, vimentin, and nucleus at 9 h after the wound was created. Representative images are shown. Scale bars, 100 µm. The fluorescence intensity of ROS and vimentin at the proximal (<200 µm) and distal (>400 µm) areas from the wound was measured (n ≥ 450). Values (mean ± SD) are from three independent experiments. *P < 0.05 and **P < 0.01. (C) Wound healing assay was performed in the presence (+) or absence (−) of 10 mM NAC (ROS scavenger) for 12 h. The cells were stained for ROS, vimentin, and nucleus. Representative images taken at 0, 3, 6, 9, and 12 h are shown. Scale bars, 250 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure, as described in the Materials and Methods section. Values (mean ± SD) are from three independent experiments. ***P < 0.001.
CAL-27 cells and SCC-25 cells were grown into a monolayer, and a cell-free gap (i.e., wound) of ∼500 µm in width was created. The wound healing assay was performed in the presence (+) or absence (−) of 1 mM NAC (ROS scavenger) for 9 h. The cells were stained for ROS, vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. *P < 0.05 and **P < 0.01.
(A) SAS cells (control) were infected with lentiviruses expressing shRNAs to vimentin (shVIM) or luciferase (shLuc). Three shRNA target sequences to vimentin (shVIM #1, #2, and #3) were used. The expression levels of vimentin and actin were analyzed by immunoblotting. (B) SAS cells (control) and those infected with lentiviruses expressing shRNAs to vimentin (shVIM #1, #2, and #3) or luciferase (shLuc) were subjected to the wound healing assay. 9 h later, the cells were stained for vimentin and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. ***P < 0.001. (A, C) Cells as described in (A) were subjected to the random cell motility assay, as described in the Materials and Methods section. Cell migration trajectory and speed from the 8th to 10th h (total 2 h) were analyzed. The trajectories of 120 cells for each group are shown. Cell migration speed was analyzed, and the P-values were calculated from at least 150 cells pooled from three independent experiments. Values (mean ± SD) are from three experiments. ***P < 0.001. Source data are available for this figure.
SAS cells were grown at a confluent condition and treated with hydrogen peroxide at various concentrations as indicated. 9 h later, the cells were stained for ROS and vimentin. The fluorescence intensity of ROS and vimentin in the cell was measured (n = 150). Values (mean ± SD) are from three independent experiments. ***P < 0.001.
(A) SAS cells were grown into a monolayer, and a cell-free gap (i.e., wound) of ∼500 µm in width was created. The wound healing assay was performed in the presence (+) or absence (−) of 10 µM NSC23766 (Rac1 inhibitor) for 9 h. The cells were stained for GTP-bound Rac1 (Rac-GTP), vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. (B) SAS cells were subjected to the wound healing assay in the presence (+) or absence (−) of 10 µM NSC23766 for 9 h. The cells were stained for ROS, vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. (C) SAS cells were subjected to the wound healing assay in the presence (+) or absence (−) of 10 µM NSC23766 for 9 h. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. ***P < 0.001. (D) SAS cells were infected with lentiviruses expressing shRNAs to Tiam1 (shTiam1) or luciferase (shLuc) as the control. Three shRNA target sequences to Tiam1 (shTiam1 #1, #2, and #3) were used. An equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. (D, E) Cells as described in (D) were subjected to the wound healing assay. 9 h later, the cells were stained for Rac-GTP, vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. (D, F) Cells as described in (D) were subjected to the wound healing assay. 9 h later, the cells were stained for ROS, vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. (D, G) Cells as described in (D) were subjected to the wound healing assay for 9 h. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. **P < 0.01. (H) SAS cells infected with lentiviruses expressing shRNAs to E-cadherin (shE-cad #1, #2, and #3) or luciferase (shLuc) were lysed, and an equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. The expression level of Tiam1 was quantified and expressed as –fold relative to the shLuc control. (I) SAS cells were subjected to the wound healing assay in the presence (+) or absence (−) of 1 µM DPI (NOX inhibitor) for 9 h. The cells were stained for ROS, vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. **P < 0.01. (J) SAS cells were infected with lentiviruses expressing shRNAs to NOX1 (shNOX1 #1 and #2) or luciferase (shLuc) as the control. The expression levels of NOX1 and actin were analyzed by immunoblotting. (J, K) The cells as described in (J) were subjected to the wound healing assay for 9 h. The cells were stained for ROS, vimentin, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. ***P < 0.001. Source data are available for this figure.
The cell lines as indicated were grown at a sub-confluent condition and treated with 10 µM NSC23766 (25 µM for Du145) or 1 µM DPI for 9 h. The cells were stained for ROS, vimentin, and nucleus. Representative images are shown. Scale bars, 10 µm.
(A) SAS cells (control) and those infected with lentiviruses expressing shRNAs to E-cadherin (shE-cad #1, #2, and #3) or luciferase (shLuc) were lysed, and an equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. The levels of phospho-Src (p-Src) and phospho-STAT3 (p-STAT3) were quantified and expressed as –fold relative to the shLuc. Values (mean ± SD) are from three independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001. (A, B) Cells as described in (A) were treated with or without 10 mM NAC (ROS scavenger) for 24 h. An equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. (A, C) Cells as described in (A) were treated with or without 200 nM dasatinib (Src inhibitor) for 24 h. An equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. (A, D) Cells as described in (A) were treated with or without 5 µM Stattic (STAT3 inhibitor) for 24 h. An equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. (E) SAS cells were subjected to the wound healing assay in the presence or absence (−) of 200 nM dasatinib or 5 µM Stattic for 9 h. The cells were stained for vimentin, pSTAT3, and nucleus. Representative images taken at 0 and 9 h are shown. Scale bars, 250 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. ***P < 0.001. Source data are available for this figure.
(A) SAS cells and SiHa cells were grown at low and high densities. 24 h later, the cells were stained for phospho-Src (p-Src), vimentin, and nucleus. (B) Cells were grown at low density, treated with (+) or without (−) 250 nM dasatinib (Src inhibitor) for 9 h, and stained for pSrc, vimentin, and nucleus. (C) Cells were grown at low density, treated with (+) or without (−) 250 nM dasatinib (Src inhibitor) for 9 h, and stained for p-STAT3, STAT3, and nucleus. Scale bars, 10 µm.
(A) Src C245A mutant was generated and confirmed by dideoxy DNA sequencing. (B) SAS cells stably expressing GFP-Src WT, the C245A mutant, or GFP as the control were established. The cells were treated with (+) or without (−) hydrogen peroxide for 30 min and lysed. An equal amount of whole-cell lysates was analyzed by immunoblotting with the antibodies as indicated. The p-SrcY416 levels of GFP-Src WT and the C245A mutant were measured and expressed as –fold relative to the level of control without hydrogen peroxide treatment. (B, C) Cells as described in (B) were grown at low and high densities. 24 h later, the cells were stained for vimentin. Representative images are shown. Scale bars, 10 µm. The average of the fluorescence intensity of vimentin was measured (n = 150). (B, D) Cells as described in (B) were grown into a monolayer, and a cell-free gap of ∼500 µm in width was created. The wound healing assay was performed for 6 h. Representative images taken at 0 and 6 h are shown. Scale bars, 200 µm. The width of the cell-free gap was measured and expressed as a percentage of wound closure. Values (mean ± SD) are from three independent experiments. **P < 0.01. Source data are available for this figure.
(A) Formalin-fixed, paraffin-embedded tumor slides from HNSCC patients were subjected to the multiplex immunofluorescence staining using the Opal 7-Color manual IHC kit (Akoya Biosciences), as described in the Materials and Methods section. Representative composite and single-color images from three HNSCC patients (#001T, #008T, and #009T) are shown. Scale bars, 50 or 100 µm as indicated. The colors used are as follows: pSTAT3 (green), pSrc (yellow), 4-HNE (red), vimentin (blue), nucleus (hyacinth), and pan-cytokeratin (PanCK; gray). The boundary between tumor (T) and stroma (S) is indicated by white dash lines. The fluorescence intensities of individual colors at the internal (I; >100 µm from the tumor boundary) and marginal (M; <40 µm from the tumor boundary) region of the tumors were quantified and expressed as box-and-whisker plots. Three tumor foci were selected from a patient, and the fluorescence intensity of 10 selected areas (20 × 20 µm) at the internal and marginal region of each tumor focus was measured, respectively. The P-values were calculated from 30 data points. ***P < 0.001. (B) Expression of pSTAT3 in the tumor slides from HNSCC patients was examined by immunohistochemistry. Representative images from a patient (#008T) are shown. Scale bars, 50 µm. Note that pSTAT3-positive cells are more abundant at the tumor invasive front. (C) Expression of pSTAT3 in the tumor slides from 62 HNSCC patients was examined by immunohistochemistry and scored (5 score levels: 0, 0.5, 1, 2, and 3), as described in the Materials and Methods section. The corresponding clinical stage (cTNM) of HNSCC patients was classified by the pathologists of Taipei Veterans General Hospital. Note that the level of pSTAT3 with a score ≥2 was detected more frequently in stage IV (58%, 18 of 31 cases) than in stage III (38%, 5 of 13 cases), stage II (47%, 7 of 15 cases), and stage I (0 of 3 cases). Values (mean ± SD) are presented.
The expression of pSTAT3 in the tumor slides from HNSCC patients was examined by immunohistochemistry and scored. The score of pSTAT3 was measured by counting the number of pSTAT3-positive cells (≥5 fields at 20× magnification) for each case. The scores (0, 0.5, 1, 2, and 3) were defined as follows: score 0, <5 positive cells/field; score 0.5, less than five fields that contain >10 positive cells/field; score 1, more than five fields that contain >10 positive cells/field; score 2, more than five fields that contain >30 positive cells/field; and score 3, more than five fields that contain >50 positive cells/field. Representative images for each score are shown.
Tiam1 is localized at cell–cell junctions and important for establishment and maintenance of these structures. Upon loss of cell–cell adhesion, Tiam1 is released from cell–cell junctions and activates Rac1, which then serves as a major molecular switch to turn on the cellular “migratory machinery” through NOX-mediated ROS generation. The ROS–Src–STAT3 signaling pathway that leads to vimentin expression is important for cell migration.
Similar articles
-
RAP1-RAC1 Signaling Has an Important Role in Adhesion and Migration in HNSCC.
Liu M, Banerjee R, Rossa C Jr, D'Silva NJ. Liu M, et al. J Dent Res. 2020 Jul;99(8):959-968. doi: 10.1177/0022034520917058. Epub 2020 May 13. J Dent Res. 2020. PMID: 32401565 Free PMC article.
-
NADPH-oxidase-derived ROS alters cell migration by modulating adhesions dynamics.
Tamborindeguy MT, Matte BF, Ramos GO, Alves AM, Bernardi L, Lamers ML. Tamborindeguy MT, et al. Biol Cell. 2018 Oct;110(10):225-236. doi: 10.1111/boc.201800011. Epub 2018 Sep 16. Biol Cell. 2018. PMID: 30157294
-
Shen CJ, Chan SH, Lee CT, Huang WC, Tsai JP, Chen BK. Shen CJ, et al. Cancer Lett. 2017 Feb 1;386:110-122. doi: 10.1016/j.canlet.2016.11.012. Epub 2016 Nov 16. Cancer Lett. 2017. PMID: 27865799
-
Romick-Rosendale LE, Hoskins EE, Privette Vinnedge LM, Foglesong GD, Brusadelli MG, Potter SS, Komurov K, Brugmann SA, Lambert PF, Kimple RJ, Virts EL, Hanenberg H, Gillison ML, Wells SI. Romick-Rosendale LE, et al. Clin Cancer Res. 2016 Apr 15;22(8):2062-73. doi: 10.1158/1078-0432.CCR-15-2209. Epub 2015 Nov 24. Clin Cancer Res. 2016. PMID: 26603260 Free PMC article.
-
Crosstalk between Rac1-mediated actin regulation and ROS production.
Acevedo A, González-Billault C. Acevedo A, et al. Free Radic Biol Med. 2018 Feb 20;116:101-113. doi: 10.1016/j.freeradbiomed.2018.01.008. Epub 2018 Jan 10. Free Radic Biol Med. 2018. PMID: 29330095 Review.
Cited by
-
Baba RA, Mir HA, Mokhdomi TA, Bhat HF, Ahmad A, Khanday FA. Baba RA, et al. Front Pharmacol. 2024 Jan 31;15:1318797. doi: 10.3389/fphar.2024.1318797. eCollection 2024. Front Pharmacol. 2024. PMID: 38362155 Free PMC article.
-
TRP Channels in Tumoral Processes Mediated by Oxidative Stress and Inflammation.
Piciu F, Balas M, Badea MA, Cucu D. Piciu F, et al. Antioxidants (Basel). 2023 Jun 23;12(7):1327. doi: 10.3390/antiox12071327. Antioxidants (Basel). 2023. PMID: 37507867 Free PMC article. Review.
-
Arfin S, Kumar D, Lomagno A, Mauri PL, Di Silvestre D. Arfin S, et al. Cancers (Basel). 2023 Sep 4;15(17):4420. doi: 10.3390/cancers15174420. Cancers (Basel). 2023. PMID: 37686696 Free PMC article.
References
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Medical
Research Materials
Miscellaneous