Skip to main content
. 2017 Oct 2;214(10):2889–2900. doi: 10.1084/jem.20170354

Figure 2.

Figure 2.

The defective switch is proinvasive. (A) Immunostaining of DIAPH1 on normal tongue (n = 11) and HNSCC (n = 64) tissue sections. In comparison with normal tongue, a significant increase in DIAPH1 expression is observed in HNSCC tissue sections. Bars, 100 µm. (B) Expression pattern of miR-198 and DIAPH1 across HNSCC tissue samples indicates a clear inverse correlation. (C and D) Boyden chamber invasion assay shows a significant decrease in the number of cells invading the chamber matrix in siDIAPH1 or siFSTL1 compared with control cells. Representative images of migrated cells detected with Giemsa staining (representative of five independent experiments). Error bars represent SD. Bars, 100 μm. (E) Morphology of SCC cells transduced with control shRNA or shRNA against FSTL1 or DIAPH1 (top) visualized using phalloidin-conjugated TRITC. Knockdown of FSTL1 or DIAPH1 showed a significant difference in morphology compared with control cells (top). Cell trajectory displacement over 24 h (bottom; n = 3). Bars, 10 μm. (F) Analysis of cell displacement in control, shFSTL1, and shDIAPH1 cells (representative of three independent experiments). (G) Organotypic invasion assay with shFSTL1 and shDIAPH1 independently or in combination or control SCC12 cells. SCC12 cells were detected by KRT14 staining (in red). Bars, 100 μm. (H) Histogram represents relative percentage of cells that invade the matrix (representative of three independent experiments; *, P < 0.05; **, P < 0.01). Error bars denote mean ± SEM.