Fig 2: Regulation of cPLA2/PGE/EP4 signaling by Th17 cytokines(A) Representative images showing immunostaining with anti-Ki-67 (DAB+) and anti-cytokeratin 10 (Permanent Red) antibodies. Scale bar, 10 µM.(B) Quantification of (i) the proportion of CK10 positive cells, (ii) the proportion of proliferating cells, and (iii) the thickness of the cultures. The data shown are the mean ± SEM for three replicates. One-way ANOVA followed by Dunnett's multiple comparison test was used to make statistical comparisons with either the untreated control (CTRL) or stimulated control (Vehicle), where ### indicates p < 0.001 versus CTRL(C) Eicosanoid levels measured by ELISA for (i) PGE2, (ii) 12s-HETE, and (iii) LTB4. Data shown are the mean ± SEM of three replicates, measured in duplicate. One-way ANOVA followed by Dunnett's multiple comparison test was used to compare the PGE2 levels with either the untreated control (CTRL) or the stimulated control (Vehicle), where ### indicates p < 0.001 versus CTRL and *** indicates p < 0.001 versus Vehicle.(D) Heatmap of Log10 fold changes in expression relative to unstimulated controls (CTRL) for genes involved in PGE2 synthesis, degradation, and signaling. Data are the mean of three replicates. X indicates that expression was undetectable.(E) Schematic summary showing how Th17 cytokines could regulate PGE2 release and activate downstream signaling events via the EP4 receptor. The gene expression of the components in red was increased and decreased for components in blue in cultures treated with Th17 cytokines.

Fig 3: Cytokeratin-10 (K10) expression of co-cultured spheroids under 1 G and taSMG.a K10 and DAPI staining showing the cytokeratin expression in keratinocyte layer and both nuclei of the keratinocyte layer and fibroblast core. Scale bar: 100 µm. b qRT-PCR analysis of K10. Data represent the mean ± SEM (n = 4, *p < 0.05).

Fig 4: Transcriptomic and histological comparison between AM and OKC. A, B Heatmap of differentially expressed genes (DEG) between AM and OKC tissues. Significantly different genes (adj. p < 0.01, |Fold change| > 2) were labeled in red boxes. A L-type and P/Q type VGCCs were upregulated in AM, and N-type, R-type and T-type VGCCs showed a high deviation within a group. B LGR5 was enriched in AM, and KRT 5, 10, 14, 17, 20, and 27 were upregulated in OKC. C, D Hematoxylin and eosin staining performed on AM and OKC patient samples. AM epithelial island shows peripheral palisade cells surrounding a central area of stellate-shaped cells. The high magnification images demonstrate the nuclei displaced away from the basement membrane. The thin, regular lining of parakeratinized stratified squamous epithelium with palisading hyperchromatic basal cells (high magnification) was observed in the OKC sample. E In the immunofluorescence staining, CACNA1C was intensely expressed in CK14 positive peripheral cells of AM. F Moderate expression of CACNA1C was observed in the CK14-labeled basal layer of OKC. G The terminal differentiation marker CK10 is negatively expressed in the CACNA1C positive cell enriched region of AM. H Intensive expression of CK10 was observed in the supra-basal cell layer of OKC. I, J E-cadherin showed moderate expression in AM, and intensive expression in OKC. MMP-9 showed intensive expression in basal cell layer of AM, and negative expression in OKC. Nuclei were stained with TO-PRO-3 (TB3). K, L The stem cell marker LGR5 showed intensive expression in the peripheral cell layer of AM and showed negative expression in OKC. M, N Intense expression of ß-catenin and location in nucleus (Arrow and arrowhead) were observed in peripheral layer of AM and it expressed in the adherence junctions in OKC. O, P Moderate expression of Axin2 in epithelial tumor mass of AM, and not observed in OKC. The white or black dotted lines represented the interface of the epithelium and mesenchyme tissues. Scale bar: C, D, 100 µm; E-J, 50 µm; K-P, 100 µm. ep epithelium, st stroma.

Fig 5: Activation of actomyosin contractility induces epidermal differentiation of cSCC cells. (A) Whole-cell lysates of confluent A431 cells treated with or without 5 µg/ml RhoA activator (RhoA act.) for 24 h were immunoblotted for K5, K10, and ß-actin. Bar graphs show the result of the densitometric quantification of K5 and K10. Values were normalized with the mean values in control cells. Each bar represents mean ± SD. n = 3 independent experiments. *p < 0.05; **p < 0.01. (B) Confluent A431 cells treated with 5 µg/ml RhoA activator (RhoA act.), 100 µM p-aminoblebbistatin (Blebb) or combination of both drugs (RhoA act. + Blebb) for 24 h were stained for markers of basal (undifferentiated, K5, magenta) and suprabasal (differentiated, K10, cyan) keratinocytes. Cell nuclei were labeled with Hoechst 33342 (green). Scale bar, 50 µm. (C, D) Quantification of (B). Fluorescence intensities of K5 (C) and K10 (D) in confluent A431 cells treated with indicated compounds for 24 h. Values were normalized with the mean values in control cells. Box-and-whisker plots show the median, the interquartile range, and the tenth and ninetieth percentiles. n = 60 fields of view from three independent experiments were pooled for the analysis. *p < 0.05; ***p < 0.001; NS, p > 0.05. (E, F) Quantification of fluorescence intensities of K5 (E) and K10 (F) in confluent A431 cells expressing control-shRNA or a-catenin-shRNA treated with indicated compounds for 24 h. Values were normalized with the mean values in non-treated controls. Box-and-whisker plots show the median, the interquartile range, and the tenth and ninetieth percentiles. n = 60 fields of view from three independent experiments were pooled for the analysis. **p < 0.01; ***p < 0.001; NS, p > 0.05.