Regional Activation of Myosin II in Cancer Cells Drives Tumor Progression via a Secretory Cross-Talk with the Immune Microenvironment

Abstract

ROCK-Myosin II drives fast rounded-amoeboid migration in cancer cells during metastatic dissemination. Analysis of human melanoma biopsies revealed that amoeboid melanoma cells with high Myosin II activity are predominant in the invasive fronts of primary tumors in proximity to CD206⁺CD163⁺ tumor-associated macrophages and vessels. Proteomic analysis shows that ROCK-Myosin II activity in amoeboid cancer cells controls an immunomodulatory secretome, enabling the recruitment of monocytes and their differentiation into tumor-promoting macrophages. Both amoeboid cancer cells and their associated macrophages support an abnormal vasculature, which ultimately facilitates tumor progression. Mechanistically, amoeboid cancer cells perpetuate their behavior via ROCK-Myosin II-driven IL-1α secretion and NF-κB activation. Using an array of tumor models, we show that high Myosin II activity in tumor cells reprograms the innate immune microenvironment to support tumor growth. We describe an unexpected role for Myosin II dynamics in cancer cells controlling myeloid function via secreted factors.

Keywords: NF-κB; ROCK-Myosin II; macrophages; protein secretion; rounded-amoeboid melanoma cells; tumor invasive front.

Graphical abstract

Figure 1

Invasive Fronts of Human Melanomas Are Enriched in Amoeboid Melanoma Cells Close to Macrophages and Blood Vessels (A) Melanoma cell-shape score in tumour body (TB) or invasive front (IF) of matched samples from human primary melanoma. Values range from 0 (all cells round) to 300 (all cells spindle). (B) H-score of p-MLC2 staining from patients in (A). Values range from 0 (no staining) to 400 (very intense staining). (C–F) (Left) Quantification and (right) representative images of (C) CD68⁺, (D) CD163⁺, (E) CD206⁺, and (F) CD31⁺ cells in TB and IF of primary melanomas. (G) Melanoma cell-shape score in primary and metastatic melanomas. (H and I) (Left) Quantification and (right) representative images of (H) CD206⁺ and (I) CD31⁺ cells in primary and metastatic melanomas. (J) mRNA levels of CD206 in primary (n = 68) and metastatic (n = 316) melanomas. Raw data were obtained from TCGA. (K) Schematic: IF and metastatic site of human melanoma. In (A)–(I), n = 24 primary and n = 16 metastatic melanomas. Scale bars, 200 μm for the tumor cores, 50 μm for all the focused images except (A) for the focused images showing the score where scale bar is 5 μm. All data are presented per patient. Average has been taken from 4 tumor cores per TB and 4 tumor cores per IF. In (A)–(F), matched TB and IF from same patients are presented. In (A)–(I), boxplots show min to max values. In (J), dot blot shows mean ± SEM. In (A)–(F), Wilcoxon matched-pairs signed-rank test is shown. In (G)–(J), t test is shown. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figure S1 and Tables S3 and S4.

Figure S1

Invasive Fronts of Human Melanomas Are Enriched in Amoeboid Melanoma Cells with High Myosin II Activity in the Vicinity of Macrophages and Blood Vessels, Related to Figure 1 (A) Melanoma cell shape score in tumor body (TB) or invasive front (IF) of human melanoma biopsies in Cohort B, values ranging from 0 (all cells round) to 300 (all cells spindle) (See also STAR methods) (n = 7). (B–D) Average number (top) and representative images (below) of (B) CD163+ macrophages, (C) CD206+ macrophages and (D) CD31+ vessels, per field of view (FOV), in TB or IF in Cohort B. Scale bar, 50 μm (n = 4 for CD163, n = 5 for CD206, n = 6 for vessels). (E) Average number of CD68+ macrophages and (F) CD163+ macrophages, in primary and metastatic melanoma lesions in Cohort A. Data are presented per patient. (G) Scatterplot for correlation of CD31 and CD206 mRNA levels in normal skin (black), nevi (cyan), primary melanoma (blue), metastatic melanoma (red). Pearson’s r. Raw data obtained from the publicly available database GEO. (A, E, and F) Boxplots show min to max values. (B–D) Graphs show mean ± SEM. (A–D) Paired t test. (E and F) t test. ns p > 0.05,^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001.

Figure 2

Myosin II Activity in Melanoma Cells Favors Secretion of Immunomodulatory Factors (A) (Top) Images and (bottom) immunoblotof p-MLC2 levels in A375M2 and A375P cells. (B) Heatmaps of secreted factors enriched in CM A375M2 with a >1.1 fold-increase compared to CM A375P, divided into 3 groups (0- to 300-, 0- to 50-, and 0- to 5-fold). Cyan and red indicate the lowest and highest expression levels, respectively. (C) Concentration of IL-1α, IL-10, TGF-β, and IL-8 in CM A375P or CM A375M2, by ELISA (n = 3). (D) After MLC2 knockdown in A375M2 cells, (left) representative immunoblot for p-MLC2 levels and (right) secreted levels of IL-1α, IL-10, TGF-β, and IL-8 in CM A375M2, by ELISA (n ≥ 3 for IL-1α, IL-8, and TGF-β, n = 2 for IL-10). (E) After ROCK1/2 knockdown in A375M2 cells, (left) representative immunoblots for ROCK1/2 and p-MLC2 levels and (right) secreted levels of IL-1α, IL-10, and TGF-β in CM A375M2 by ELISA (n ≥ 3 for IL-1α and TGF-β, n = 2 for IL-10). (F) After treatment with H1152 (5 μM) for 48 h in A375M2 cells, (left) representative immunoblot for p-MLC2 levels and (right) secreted levels of IL-1α, IL-10, TGF-β, and IL-8 in CM A375M2 by ELISA (n ≥ 3). (G) (Top) Images and immunoblot for p-MLC2 levels in WM983B and WM983A cells and (bottom) secreted levels of TGF-β and IL-8 in CM WM983B, CM WM983A, and CM WM88, by ELISA (n = 3 for all, n = 2 for IL-8 in CM WM88). (H) After MLC2 knockdown in WM983B cells, (top) representative immunoblot for p-MLC2 levels and (bottom) secreted levels of TGF-β and IL-8 in CM WM983B as tested by ELISA (n = 3). (I) After treatment with H1152 (5 μM) for 48 h in WM983B cells, (top) representative immunoblot for p-MLC2 levels and (bottom) secreted levels of TGF-β and IL-8 in CM WM983B (n ≥ 3). In (D)–(F), (H), and (I), data are presented as fold change versus the control. In (C)–(I), graphs show mean ± SEM. In (C)–(F), (H), and (I), t test is shown. In (G), one-way ANOVA with Tukey post hoc test is shown. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗∗p < 0.0001. See also Figure S3 and Table S1.

Figure S2

Myosin II Activity in Melanoma Cells Favors Secretion of Immunomodulatory Factors, Related to Figure 2 (A) Chart pie shows the factors influencing the balance toward tumor-promoting inflammation versus tumor-suppressive inflammation in CM A375M2. (B) (Top) Schematic illustrates secreted factors in melanoma progression and (bottom) heatmap shows fold change for mRNA levels of IL-4, IL-8, IL-1α, TGF-β and IL-10 in metastatic melanoma versus melanocyte and metastatic versus primary melanoma samples. Raw data were obtained from TCGA and GEO databases. (C) Relative p-MLC2 and MLC2 levels in A375M2 cells after MLC2 knockdown. (D) Relative ROCK1, ROCK2, p-MLC2 and MLC2 levels in A375M2 cells after ROCK1/2 knockdown. (E) Secreted levels of IL-1α, IL-10, TGF-β and IL-8 in CM A375M2 cells after treatment with Y27632 (10 μM) or GSK269962A (5 μM) for 48h (n ≥ 3). Data are presented as fold change versus the control. (F) (Left) Representative immunoblot for p-cofilin and (right) relative p-cofilin levels, in A375M2 cells after treatment with LIMKi 3 (1μM) for 48h (n = 3). (G) Roundness index of A375M2 cells seeded on top of collagen I, treated with H1152 (5μM), Blebbistatin (2.5μM) or LIMKi 3 for 48h (n = 3). (H) (Left) Representative immunoblots for p-MLC2 and (right) quantification of p-MLC2 levels, in A375M2 cells treated with H1152 or LIMKi 3 for 48h (n = 3). (I) Secreted levels of IL-1α, IL-10 and IL-8 in CM A375M2+H1152, CM A375M2+Blebbistatin or CM A375M2+LIMKi 3. Data are presented as fold change versus the control (n ≥ 3 for IL-10 and IL-8 and n ≥ 2 for IL-1α). (J) (Left) Representative immunoblot for p-MLC2 levels in WM793B cells and (right) secreted levels for TGF-β and IL-8 by WM793B cells, after treatment with H1152 (5 μM) for 48h (n ≥ 3).(C-J) Graphs and dot blots show mean ± SEM. (C, D, and F–J) t test. (E) One-way ANOVA with Tukey post hoc test. ns p > 0.05,^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001,^∗∗∗∗p < 0.0001.

Figure 3

Amoeboid Melanoma Cells Induce Tumor-Promoting Macrophages (A) Percentage of migrated human PBMC-derived monocytes, THP-1 and U937 toward media (–), CM A375P, or CM A375M2 (n = 3). (B) (Left) Schematic: in-vitro-polarized macrophages or melanoma-conditioned macrophages and (right) %CD163⁺CD206⁺ or %HLA-DR⁺CD86⁺ macrophages after treatment with M-CSF, IL-4, IL-10, or IFN-γ&LPS or media only (–) (n = 5;5 different healthy donors). (C) Fluorescence-activated cell sorting (FACS) dot plots from one donor showing (left) %CD163⁺CD206⁺ and (right) %HLA-DR⁺CD86⁺ macrophages after treatment with CM A375P, CM A375M2 or culture media only (–). (D–F) %CD163⁺CD206⁺ macrophages (D), mean fluorescence intensity (MFI) for CD206 (E), and %HLA-DR⁺CD86⁺ macrophages (F), after treatment with CM A375P, CM A375M2, or culture media only (–) (n = 5;5 different healthy donors). (G) Quantification of macrophage morphology (see also Figure S3D) (n = 3;3 different healthy donors). (H) Schematic shows macrophage cytotoxicity assay. (I) Fold change of dead tumor cells (A375M2 or WM88) in co-cultures with PBMC-derived monocytes treated with CM A375M2 or IFN-γ&LPS. Data are presented as fold change versus the control untreated monocytes. Log2 scale is presented in y axis (n = 6;6 different healthy donors for A375M2 co-cultures and n = 2;2 different healthy donors for WM88 co-cultures). (J) (Top) Schematic shows TAMs induction in vitro with serum from melanoma patients and (bottom) representative FACS dot plots from one donor showing %CD163⁺CD206⁺ and %HLA-DR⁺CD86⁺ macrophages after treatment with healthy volunteer’s serum or melanoma patient’s serum. (K and L) Number of (K) CD163⁺CD206⁺ and (L) HLA-DR⁺CD86⁺ macrophages after treatment with healthy volunteer’s serum or melanoma patient’s serum (n = 3;3 different healthy donors. Sera from n = 10 healthy volunteers, n = 23 melanoma patients, each dot represents a different treatment). In (A), (B), (D)–(G), (I), (K), and (L), graphs and dot blots show mean ± SEM. In (A), (B), and (D)–(G), one-way ANOVA with Tukey or Bonferroni (for G) post hoc tests are shown. In (I), t test is shown. In (K) and (L), t test with Welch’s correction is shown. Nonsignificant p > 0.05, ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figure S3 and Tables S1 and S2.

Figure S3

Amoeboid Melanoma Cells with High Myosin II Activity Induce Tumor-Promoting Macrophages, Related to Figure 3 (A) Expression levels (geometric mean of fluorescent intensity, gMFI) of HLA-DR, CD86, CD163 and CD206 in macrophages after treatment with M-CSF (50ng/ml), IL-4 (20ng/ml), IL-10 (20ng/ml), IFN-γ & LPS (20ng/ml and 100ng/ml) or culture media only (-) (n = 5; 5 different healthy donors). (B and C) Expression levels (gMFI) of (B) CD163, (C) HLA-DR and CD86, in macrophages treated with CM A375P or CM A375M2 or culture media only (-) (n = 5; 5 different healthy donors). (D) Representative bright-field images of macrophages treated with M-CSF, IL-4, IL-10, IFN-γ & LPS or culture media only (-). Blue and red arrows show ‘fried-egg’ and elongated shapes, respectively. Scale bar, 50 μm. (E) Dead tumor targets (WM1366, WM793B, WM3854 and WM983A) upon co-culture with CM A375M2- or IFN-γ&LPS- stimulated macrophages. Data are presented as fold change versus the control untreated monocytes. Log2 scale is presented in y axis. (F and G) Tumor volume in (F) A375M2-xenografts and (G) WM983B-xenografts, upon depletion of macrophages via clodronate in SCID mice (n = 6 mice PBS group and n = 5 mice clodronate group). (H) Representative IHC images showing F4/80+ and CD206+ macrophage depletion upon clodronate administration in A375M2 and WM983B xenografts. Scale bar, 100 μm. (A–C and E–G) Graphs show mean ± SEM. (A–C) One-way ANOVA with Tukey post hoc test. (F and G) Two-way ANOVA with Bonferroni’s multiple comparison test. ns p > 0.05,^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001,^∗∗∗∗p < 0.0001.

Figure 4

AATME Composition Is a Conserved Feature in Melanoma In Vivo (A) Confocal images for p-MLC2 (cyan) and F-actin (red) in EGFP-A375P and EGFP-A375M2 cells. (B) Tumor volume of xenografts post-injection of EGFP-A375P and EGFP-A375M2 cells (n = 8 mice/group). (C) Immunohistochemistry (IHC) images of p-MLC2 levels (scale bar, 50 μm; insert, 10 μm). (D) (Top) IHC quantification for p-MLC2 levels showing percentage of melanoma cells with the highest score (4) at different distances from IF (0–2 mm) and (bottom) H-score for p-MLC2 staining for A375P and A375M2 tumors (n = 8 mice/group). (E) Representative IHC images of CD206⁺ macrophages (scale bar, 30 μm and insert: 10 μm). (F–H) Quantification of (F) CD206⁺, (G) F4/80⁺, and (H) CD31⁺ cells in A375P and A375M2 tumors (n = 8 mice/group). (C)–(H) correspond to TB and IF of A375P and A375M2 xenografts, as tested by IHC (n = 8 mice/group). (I) Confocal images for p-MLC2 (cyan) and F-actin (red) in EGFP-WM983A and EGFP-WM983B cells. (J) Tumor volume of xenografts over time (35 days) post-injection of EGFP-WM983A and EGFP-WM983B cells (n = 8 mice/group). (K–M) H-score for p-MLC2 staining (K), quantification of CD206⁺ (L), and F4/80⁺ macrophages (M) in TB and IF of WM983A and WM983B xenografts, as tested by IHC (n = 8 mice/group). (N and O) H-score for p-MLC2 (N) and quantification (O) of CD206⁺ macrophages in TB and IF of B16F10 tumors, as tested by IHC (n = 8 mice). (P and Q) (Left) IHC images and (right) H-score for p-MLC2 staining (P), (left) IHC images and (right) quantification of CD206⁺macrophages (Q) in TB and IF of 5555 tumors (n = 5 tumors). Scale bar, 30 μm. In (B), (D, top), and (J), graphs show mean ± SEM. In (D, bottom), (F)–(H), and (K)–(Q), boxplots show 10–90 percentile. In (B), (D, top), and (J), two-way ANOVA with Bonferroni post hoc test is shown. In (D, bottom), (F)–(H), and (K)–(M), one-way ANOVA with Tukey post hoc test is shown. In (N)–(Q), t test is shown. Nonsignificant p > 0.05, ^∗p < 0.05^∗∗, p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figure S4.

Figure S4

AATME Composition Is a Conserved Feature in Melanoma In Vivo, Related to Figure 4 (A) (Left) Roundness index and (right) p-MLC2 levels/area, in EGFP-A375P and EGFP-A375M2 cells seeded on top of collagen I. Quantification corresponds to the area occupied by p-MLC2 staining normalized by the total area of the cell. (B) Melanoma cell shape score in TB and IF of A375P and A375M2 tumors (n = 8 mice/group). (C) Percentage of melanoma cells with score 0-4 for p-MLC2 at different distance from IF (0-2 mm) for A375P and A375M2 tumors, as tested by IHC (n = 8 mice/group). (D) (Left) Roundness index and (right) p-MLC2 levels/area, in EGFP-WM983A and EGFP-WM983B cells seeded on top of collagen I. (E) Melanoma cell shape score in TB and IF of WM983A and WM983B tumors (n = 8 mice/group). (F) Percentage of melanoma cells with score 0-4 for p-MLC2 at different distance from IF (0-2 mm) for WM983A and WM983B tumors (n = 8 mice/group). (G) Percentage of melanoma cells with score 4 for p-MLC2 at different distance from IF (0-2 mm) for WM983A and WM983B tumors, as tested by IHC (n = 8 mice/group). (H) Cell proliferation rates of (left) A375P versus A375M2 and (right) WM983A versus WM983B cells. Data are presented as fold change versus day 0. (I) Schematic shows the isolation of melanoma cells from TB or IF of A375M2 tumors. (J) Concentration of secreted IL-10, as measured by ELISA, in melanoma cells isolated from TB or IF of A375M2 tumors (n = 3). (K) Percentage of CD163+CD206+ macrophages after treatment with CM from melanoma cells isolated from TB or IF of A375M2 tumors (n = 2; 2 different healthy donors; 2 matched TB/IF samples; 1 pair used twice with 2 different donors). (L and M) (L) Melanoma cell shape and (M) F4/80+ macrophages, in TB and IF of B16F10 tumors (n = 8 mice). (N) Melanoma cell shape score and (O) F4/80+ macrophages, in TB and IF of tumors generated 8 days post-intradermal injection of Venus- 5555 cells (n = 5 tumors). (A left, D left, G, and H) Graphs and dot blots show mean ± SEM. (A right, B, D right, E, and L–O) Boxplots show 10-90 percentile. (A, D, H, J, L–O) t test. (K) paired t- test. (B and E) One-way ANOVA with Tukey post hoc test. (G) Two-way ANOVA with Bonferroni post hoc test. ns p > 0.05,^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001,^∗∗∗∗p < 0.0001

Figure 5

Blocking Myosin II Activity in Melanoma Cells Reprograms Macrophages (A) Schematic: in vitro treatment of PBMC-derived monocytes with CM from ROCK inhibited or MLC2 depleted A375M2 cells and subsequent assays. (B) CD163⁺CD206⁺ macrophages after treatment of PBMC-derived monocytes with CM A375M2, CM A375M2+ROCKi, or media only (–) (n = 8;8 different healthy donors; each dot is a different donor). (C) PBMC-derived monocytes treated with CM A375M2 depleted from MLC2 and quantification of CD163⁺CD206⁺ macrophages. Data are presented as fold changes versus control (n = 4). (D) Quantification of fold change of absorbance (O.D.): viability of endothelial cells (HMVECs and HUVECs) after treatment with macrophage-derived supernatants (50%). Media were derived from monocytes ± CM A375M2+ROCKi (n ≥ 3 for HMVECs and n = 6 for HUVECs). Endothelial cells were treated with macrophage-derived supernatants for 72 h. (E) Schematic: in vivo experiment with GSK269962A ROCKi pre-treated 5555 cells. (F) Representative H&E image (top) and IHC image for p-MLC2 showing amoeboid melanoma cells with high p-MLC2 invading the dermis in the IF of tumors 14 days post-intradermal injection of DMSO (vehicle) pre-treated 5555 cells. Scale bar, 50 μm; insert, 10 μm. (G–I) Number of invading melanoma cells (G), roundness index (H), and H-score (I) for p-MLC2 for (G). (J) (Left) CD206⁺ and (right) F4/80⁺ macrophages in the IF. (G–J) Tumors from DMSO (vehicle) pre-treated or ROCKi-pretreated Venus-5555 cells 14 days post-intradermal injection (n = 14 control and n = 12 ROCKi). (K) Tumor volume of A375M2 xenografts treated with PBS or Y27632 ROCKi. (L–N) Melanoma cell-shape score (L) and quantification (M) of (left) CD206⁺ and (right) F4/80⁺ and (N) CD31⁺ cells, in A375M2 xenografts treated with PBS or Y27632 ROCKi (n ≥ 4). In (B)–(D), (G), (H), and (K), graphs and dot blots show mean ± SEM. In (I), (J), and (L)–(N), boxplots show 10–90 percentile. In (B) and (L)–(N), one-way ANOVA with Tukey post hoc test is shown. In (C) and (G)–(J), t test is shown. In (D), Kruskal-Wallis test followed by Dunn’s multiple comparisons test is shown. In (K), two-way ANOVA with Bonferroni post hoc test is shown. Nonsignificant p > 0.05, ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figure S5.

Figure S5

Remodeling of the AATME after Manipulation of Myosin II Activity In Vivo, Related to Figure 5 (A–D) Expression levels (gMFI) of (A) CD163, (B) CD206, (C) HLA-DR and (D) CD86, in macrophages after treatment with CM A375M2, CM A375M2+H1152, CM A375M2+Y27632, CM A375M2+GSK269962A or culture media only (-) (n ≥ 5 different heathy donors). (E) Relative p-MLC2 levels in A375M2 cells treated with H1152, Y27632 or GSK269962A ROCK inhibitors (n = 4). (F) HLA-DR+CD86+ macrophages after treatment with CM A375M2 depleted from MLC2. Data are presented as fold changes versus the control (n = 4). (G and H) (G) Macrophage secretion of TGF-β (pg/ml) and (H) IL-10 (pg/ml) after treatment with CM A375M2, CM A375M2+H1152, CM A375M2+Y27632, CM A375M2+GSK269962A or culture media only (-) (n = 8; 8 different healthy donors). (I) Concentration of IL-10 in macrophages induced by melanoma patient-derived sera or healthy volunteer-derived sera (n = 2; 2 different healthy donors for macrophages; n = 5 volunteers-derived sera and n = 9 melanoma patients-derived sera). (J) MFI for phagocytosed zymosan particles by macrophages after treatment with CM A375M2, CM A375M2+H1152 or CM A375M2+GSK269962A, with or without cytochalasin D (5 μM) (n = 2). (K) H-score for p-MLC2 expression of melanoma cells in TB and IF. (L) (Left) CD206+ macrophages and (right) F4/80+ macrophages in the IF. (K and L) Tumors 8 days post-intradermal injection of DMSO (vehicle)-pre-treated and ROCKi-pre-treated Venus- 5555 cells (n = 5 mice/group). (M) Percentage of viable DMSO (vehicle)-pre-treated and ROCKi-pre-treated Venus- 5555 cells in vitro as measured by IncuCyte (n = 3). Viability was measured after drug removal. (N) Tumor volume (mm³) in C57BL/6J mice after intradermal injection of DMSO (vehicle)-pre-treated and ROCKi-pre-treated Venus- 5555 cells (days: 0-14) (n = 13 tumors for DMSO group and n = 14 tumors for ROCKi group; n = 7 mice/group). (A–J, M, and N) Graphs show mean ± SEM. (K–L) Boxplots show 10-90 percentile. (A-E, J, and K) One-way ANOVA with Tukey post hoc test. (F–H and L) t test. (I) t test with Welch’s correction. (N) Two-way ANOVA with Bonferroni post hoc test. ns > 0.05,^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001,^∗∗∗∗p < 0.0001.

Figure 6

Myosin II Activity in Melanoma Cells Is Self-Perpetuated via Secreted IL-1α-Induced NF-κB Activation (A and B) (Top) Images, (center) roundness index, and (bottom) relative p-MLC2 levels of (A) A375P cells on top of collagen I upon treatment with CM A375P or CM A375M2 and (B) A375M2 cells upon treatment with BFA for 6 h. Data are presented as fold change versus the control (n = 3). (C) MetaCore enrichment network of factors upregulated in A375M2 cells is centered on NF-κB. (D) (Top) Immunoblot and (bottom) quantification of p-IKBα levels in A375P cells treated with CM A375P or CM A375M2. Data are presented as fold change versus CM A375M2 treatment (n = 3). (E and F) Immunoblot (E) and quantification (F) of p-IKBα levels in A375M2 cells after MLC2 knockdown. Data are presented as fold change versus control (n = 2). (G) (Top) Confocal images and (bottom) quantification of nuclear p65, as percentage versus the total p65, in A375P and A375M2 cells. Each dot represents a different cell. Scale bar, 10 μm. (H) (Top) Confocal images for p65 in A375P cells after treatment with CM A375P or CM A375M2. Scale bar, 10 μm. (bottom) Percentage of A375P cells with p65 in the cytoplasm (green), nucleus (red), or in both (orange) after indicated treatments. Blocking was for 1 h at 37°C (n = 3; 3 pictures per experiment; total 9 pictures per condition). (I) Roundness index of A375P cells on collagen I after same treatments as in (H) (n = 3). (J) Confocal images of p-MLC2 and F-actin in NFKB1 depleted A375M2 cells. Scale bar, 20 μm. (K) (Left) Roundness index and (right) relative p-MLC2 levels in A375M2 cells after NFKB1 knockdown, from confocal images (30 cells/condition, each dot represents a single cell). (K bottom right) Representative immunoblot showing NFKB1 knockdown in A375M2 cells. (L) Schematic: cross-talk between ROCK-Myosin II, secreted IL-1α, and NF-κB activation in amoeboid melanoma cells. In (A, center) and (B, center), boxplots show 10–90 percentile. In (A, bottom) and (B, bottom), (D), (F)–(I), and (K), graphs and dot blots show mean ± SEM. In (A), (B), (D), (F), and (G), t test is shown. In (H), (I), and (K), one-way ANOVA with Tukey post hoc test is shown. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. See also Figure S6.

Figure S6

Myosin II Activity in Cancer Cells Self-Perpetuates via Secreted IL-1α-Induced NF-κB Activation, Related to Figure 6 (A) (Top) Schematic illustrates the treatment of A375P cells with CM A375M2 and (bottom) representative immunoblot for p-MLC2 in A375P cells after treatment with CM A375M2. (B) Percentage of migrated A375P cells after treatment with CM A375P, CM A375M2 or media only (-). Data are presented versus CM A375M2 treatment (n ≥ 3). (C) Representative immunoblot for p-MLC2 in A375M2 cells after treatment with BFA for 6h. (D) (Top) Representative confocal images and (bottom) bright-field images, showing p65 localization and cell morphology, respectively, in A375P cells treated with CM A375P, CM A375M2, IgG2α-blocked CM A375M2, IL-1α-blocked CM A375M2, IgG1-blocked CM A375M2 or IL-8-blocked CM A375M2. Blocking was for 1h at 37°C. Scale bar, 10μm for confocal images and 20μm for bright-field images. (E) Relative NFKB1 levels in A375M2 cells after NFKB1 knockdown. (F) Representative immunoblot for p-IκBα after treatment of A375M2 cells with IKKβ inhibitor. (G) Roundness index and (H) relative p-MLC2 levels, in A375M2 cells after treatment with IKKβ inhibitor. (I) Roundness index (n = 3), (J) representative bright-field images and (K) representative immunoblots for p-MLC2, in A375M2 cells seeded on bovine collagen I upon treatment with the ROCKi H1152 (5 μM) for 1h, 2h, 4h or 24h. (L) Roundness index (n = 3), (M) representative bright-field images and (N) representative immunoblots for p-MLC2, in A375M2 cells seeded on bovine collagen I upon treatment with the IKKβ inhibitor IKKβ III (0.5 μM) for 1h, 2h, 4h or 24h. (B, E, and H) Graphs and dot blots show mean ± SEM. (G, I, and L) Boxplots show 10-90 percentile. (B, E, I, and L) One-way ANOVA with Tukey post hoc test. (G and H) t test. ^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001,^∗∗∗∗p < 0.0001.

Figure S7

Metastatic Colonization via Amoeboid Melanoma Cell Secretion, Related to Figure 7 (A) Migrated THP-1 cells toward CM A375M2 or CM from NFKB1-depleted A375M2 cells (n = 4). (B) gMFI for CD206 in macrophages after treatment with CM from MLC2-depleted or NFKB1-depleted A375M2 cells (n = 4). (C) (Top) Schematic showing treatment of endothelial cells with CM A375P or CM A375M2. (Bottom) Representative confocal images of VE-cadherin (green), F-actin (red) and DAPI (blue) immunostaining in HMVECs after treatment with CM A375P or CM A375M2. Dashed white lines represent gaps. Scale bar, 40 μm. (D) Quantification of monolayer disruption area (top) and VE-cadherin junctional index (bottom) in HMVECs after treatment with CM A375P or CM A375M2 (n ≥ 3). (E) Tables show the most upregulated phospho-proteins found in the cytoskeleton phospho-antibody array. HMVECs treated with CM A375M2 (s1) were compared to HMVEC cells treated with CM from ROCK1/2-depleted A375M2 cells (s2). Values are represented as ratio changes s1/s2. Ratio = (Signal Intensity of Phospho Site-Specific Antibody) / (Signal Intensity of Site-Specific Antibody). Results are highlighted in different shades of red which shows the highest expression levels. Fold-change increase is considered significant when the values are > 2. (F) Percentage of permeability of a confluent monolayer of HMVECs (left) treated with CM A375P or CM A375M2 and (right) treated with CM A375M2+H1152 or CM A375M2+Y27632 (n ≥ 3). (G) Relative p-MLC2 levels in A375P cells, A375M2 ± H1152 or ± Y27632 cells (n ≥ 3). (H) Percentage of permeability of a confluent monolayer of HUVECs treated with CM A375M2 (-) or CM from ROCK1/2-depleted A375M2. Data are presented as fold-change versus the control (n = 3). (A, B, D, and F–H) Graphs and dot blots show mean ± SEM. (A, D, F left, and H) t test. (F right) Kruskal-Wallis and Dunn’s multiple comparison. (B and G) One-way ANOVA with Tukey post hoc test. ^∗p < 0.05,^∗∗p < 0.01,^∗∗∗p < 0.001.

Figure 7

NF-κB Cross-Talk with ROCK-Myosin II in Amoeboid Melanoma Cells Educates the Tumor Microenvironment (A) (Top) Schematic: macrophage phenotypes after indicated treatments. (Bottom left) Immunoblots for p-MLC2 after MLC2 or NFKB1 knockdown in A375M2 cells. (Bottom right) Percentage of CD163⁺CD206⁺ macrophages upon treatment of PMBC-derived monocytes with CM from NFKB1-depleted or MLC2-depleted A375M2 cells, ±IL-4 (n = 4). (B) (Top) Schematic: treatment of endothelial cells with indicated conditions and downstream assays. (Bottom) Confocal images of VE-cadherin (green), F-actin (red), and DAPI (blue) immunostaining in HMVECs after indicated treatments. Dashed white lines represent gaps. Scale bar, 40 μm. (C) (Left) Relative p-MLC2 levels in indicated conditions (n ≥ 3). (Top right) Quantification of monolayer disruption area and (bottom right) VE-cadherin junctional index in HMVECs after indicated treatments (n ≥ 2). (D) Heatmap shows fold change in expression of proteins regulating endothelial permeability in HMVECs treated with CM A375M2 ± siROCK1/2. Blue and red show the highest and the lowest expression levels, respectively (6 replicates/antibody). (E) Permeability (versus the control) in HMVECs treated with indicated conditions (n ≥ 3). (F) Immunoblots for p-MLC2, ROCK1, and ROCK2 of A375M2 ± siROCK1/2. (G) (Left) Confocal images of mouse lungs after tail vein injection of 5-chloromethylfluorescein diacetate (CMFDA)-Green labeled A375M2 ± siROCK1/2 and dextran (purple) and (right) percentage of field area covered by dextran (20 fields/mouse/condition). (H) (Left) Confocal images of mouse lungs from (G) and (right) percentage of field area covered by cells (20 fields/mouse/condition). Scale bar in (G) and (H), 100 μm. In (G) and (H), n = 5 mice/condition for each experiment, n = 2 independent experiments. In (A), (C), and (E), graphs and dot blots show mean ± SEM. In (G) and (H), boxplots show min to max values. In (A), (C), and (E), one-way ANOVA and Tukey post hoc test are shown. In (G) and (H), t test is shown. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗∗p < 0.0001. See also Figure S7.