Article Figures & Data
Figures
- Figure 1. TLR3 activation triggers morphological change in normal keratinocytes.
(A) Representative images of normal human epidermal keratinocytes 72 h after UVB (10 mJ/cm2) exposure or poly(I:C) (20 μg/ml) treatment. Cells exposed to UVB are given fresh media 24 h after exposure and left in culture for another 48 h; poly(I:C) is on the cells during the first 24 h, followed by a washout period of 48 h (72 h in total). Images were captured at 10X magnification; scale bar represents 100 μm. (B) Quantification of spindle-like cells 48 and 72 h after UVB (10 mJ/cm2) exposure, n ≥ 21 images quantified. (C) Quantification of spindle-like cells 48 and 72 h after poly(I:C) (20 μg/ml), n ≥ 24 images quantified. (D) IL-6 ELISA data from the media of keratinocytes exposed to UVB (10 mJ/cm2) or treated with poly(I:C) (20 μg/ml) 24 and 48 h after treatment, unpaired t test, two-tailed, P < 0.05, n ≥ 4. (E) TLR3 gene expression by qRT–PCR 48 h after UVB (10 mJ/cm2). Data were normalized to RPLP0, n ≥ 16. (F) TLR3 gene expression by qRT–PCR 48 h after poly(I:C) (20 μg/ml). Data were normalized to RPLP0, n ≥ 36. (G) Representative immunoblot from keratinocytes exposed to UVB (10 mJ/cm2) and poly(I:C) (20 μg/ml) shows increases in TLR3 protein (full-length and cleaved) by immunoblot, n = 3. (H) Number of differentially expressed genes in 72-h UVB-treated cells and poly(I:C)-treated cells and their overlap. The number of overlapping genes is significant using Fisher’s exact test, P = 0. (I) Hallmark gene sets of interest. NES, normalized enrichment score. (J, K) Volcano plots illustrating changes in the genes involved in epithelial-to-mesenchymal transition after (J) UVB and (K) poly(I:C) treatment. * denotes significance compared with the control, P < 0.05, Mann–Whitney t test, two-tailed, unless otherwise noted above.
- Figure 2. TLR3 activation increases key epithelial-to-mesenchymal transition (EMT) markers and confers migration and invasion properties on keratinocytes.
(A) Gene expression of EMT-associated transcription factors, ZEB1, SNAI1, and TWIST1, in keratinocytes treated with 10 mJ/cm2 UVB by qRT–PCR 72 h after exposure, normalized to RPLP0, Mann–Whitney t test, one-tailed, P < 0.05, n = 12. (B) Gene expression of EMT-associated transcription factors, ZEB1, SNAI1, and TWIST1, in keratinocytes treated with 20 μg/ml poly(I:C) by qRT–PCR 72 h after exposure, normalized to RPLP0, Mann–Whitney t test, one-tailed, P < 0.05, n ≥ 30. (C) Protein levels of mesenchymal proteins (fibronectin and vimentin) and epithelial protein (E-cadherin) analyzed at 48 and 72 h after 10 mJ/cm2 UVB or 20 μg/ml poly(I:C) exposure by Western blot, representative image. (D) Protein level after UVB treatment (10 mJ/cm2) or poly(I:C) (20 μg/ml) was semi-quantified by densitometric analysis. Samples were normalized to GAPDH, and then, fold change was calculated compared with the control, paired t test, one-tailed, P < 0.05, n ≥ 4. (E) Representative images of the migration assay after scratches at 0 h, 10x magnification; the scale bar represents 200 μm, n ≥ 5 technical replicate wells per group. (F) Graphical representation of relative wound density overtime for control and poly(I:C)-treated cells, n ≥ 29 total wells. Migration data were analyzed by multiple t tests, one per time-point, using the Holm–Sidak correction method, with alpha = 0.05, and a consistent SD was assumed. (G) Representative images of the invasion assay after scratches at 0 h, 10x magnification; the scale bar represents 200 μm, n ≥ 5 technical replicate wells per group. (H) Graphical representation of relative wound density overtime for control and poly(I:C)-treated cells, n ≥ 19. Invasion data were analyzed by multiple t tests, one per time-point, using the Holm–Sidak correction method, with alpha = 0.05, and a consistent SD was assumed. * denotes significance compared with the control, P < 0.05.
- Figure 3. Activation of TLR4, TLR5, or TLR7 does not lead to an epithelial-to-mesenchymal transition (EMT)–like phenotype.
(A) Keratinocytes were treated for 1 h with LPS (1, 10 μg/ml), a TLR4 agonist, and underwent a 48-h washout treatment; representative images captured at 10X, scale bar = 100 μm. (B) qRT-PCR for EMT markers, VIM and ZEB1, and cytokine IL6 for keratinocytes treated with LPS (1, 10 μg/ml) for 1 h followed by a 48-h washout, normalized to RPLP0, n = 4. (C) Keratinocytes treated for 24 h with flagellin (100 ng/ml), a TLR5 agonist, with a 24-h washout; representative images captured at 10X, scale bar = 100 μm. (D) qRT-PCR for EMT markers, VIM and ZEB1, and cytokine IL6 for keratinocytes treated with flagellin (100 ng/ml) for 24 h followed by a 24-h or 48-h washout, respectively, normalized to RPLP0, n = 4. (E) Keratinocytes treated with imiquimod (20 μg/ml), a TLR7 agonist, for 24 h followed by a 48-h washout; representative images captured at 10X, scale bar = 100 μm. (F) qRT–PCR for EMT markers, VIM and ZEB1, and cytokine IL6 for keratinocytes treated for 24 h with imiquimod (20 μg/ml) followed by a 48- or 72-h washout, normalized to RPLP0, n ≥ 4. * denotes significance compared with the control, P < 0.05, Mann–Whitney t test, two-tailed.
- Figure S1. Pharmacological inhibition of TLR3 in normal human epidermal keratinocytes (NHEKs).
(A) TLR3 gene expression of NHEKs pre-treated with competitive TLR3 inhibitor CU CPT 4A (80 μM) for 1 h before UVB (10 mJ/cm2) exposure. Cells were collected 72 h after UVB. Data were normalized to RPLP0. * denotes significance compared with the control, # significance compared with UVB, Mann–Whitney t test, two-tailed, P < 0.05, n = 6. (B) TLR3 gene expression of NHEKs pre-treated with CU CPT 4A (80 μM) for 1 h before poly(I:C) (20 μg/ml) treatment. Cells were harvested 24 h after poly(I:C). Data were normalized to RPLP0, Mann–Whitney t test, two-tailed. * denotes significance compared with the control, # significance compared with poly(I:C), P < 0.05, n ≥ 10. (C) Chemical inhibition of TLR3 negates TLR3 protein expression in keratinocytes, representative image, n = 4. (D) Morphology of keratinocytes exposed to UVB (10 mJ/cm2) and CU CPT 4a (80 μM) 72 h after exposure, scale bar = 100 μm. (E) Morphology of keratinocytes treated with poly(I:C) (20 μg/ml) and CU CPT 4a (80 μM) 72 h after exposure, scale bar = 100 μm. (F) Chemical inhibition of TLR3 activation before UVB exposure attenuates epithelial-to-mesenchymal transition–associated marker gene expression 72 h after exposure. Data were normalized to RPLP0. * denotes significance compared with the control, # significance compared with UVB, Mann–Whitney t test, one-tailed, P < 0.05, n = 6. (G) Chemical inhibition of TLR3 activation before poly(I:C) treatment attenuates epithelial-to-mesenchymal transition–associated marker gene expression 72 h after treatment, normalized to RPLP0. * denotes significance compared with the control, # significance compared with poly(I:C), Mann–Whitney t test, one-tailed, P < 0.05, n ≥ 14.
- Figure 4. Inhibition of TLR3 attenuates the expression of epithelial-to-mesenchymal transition (EMT)–associated factors.
(A) Treatment with chemical inhibitors limits EMT-associated protein changes induced by UVB treatment at 48 and 72 h, representative image. (B) Protein level was semi-quantified by densitometric analysis. Samples were normalized to GAPDH; fold change was calculated compared with the control, n ≥ 3. (C) Treatment with chemical inhibitors limits EMT-associated protein changes induced by poly(I:C) treatment at 48 and 72 h, representative image. (D) Protein level was semi-quantified by densitometric analysis. Samples were normalized to GAPDH; fold change was calculated compared with the control, n ≥ 3. (E) TLR3 KD CRISPR keratinocytes have decreased EMT-associated protein changes induced by UVB treatment at 48 and 72 h compared with normal controls, representative image. (F) Relative density of fibronectin, vimentin, and E-cadherin was normalized to GAPDH. Fold change was calculated compared with the control, n = 3. (G) TLR3 KD CRISPR keratinocytes have decreased EMT-associated protein changes induced by poly(I:C) at 48 and 72 h compared with normal controls, representative image. (H) Relative density of fibronectin, vimentin, and E-cadherin was normalized to GAPDH. Fold change was calculated compared with the control, n = 3. * denotes significance compared with the control, # significance compared with UVB or poly(I:C), paired t test, one-tailed, P < 0.05.
- Figure S2. Validation of CRISPR/Cas9-mediated knockdown of active cleaved TLR3 in keratinocytes.
(A) Schematic showing the CRISPR gDNA cut site upstream of the essential TLR3 ubiquitination site (K831) required for TLR3 transport from ER to endolysosomes where it is activated by proteolytic cleavage (Li et al, 2020). (B) TLR3 CRISPR/Cas9 cells have less activated cleaved TLR3 protein expression than normal human epidermal keratinocytes 24 h after poly(I:C) (20 μg/ml) stimulation. TLR3 CRISPR B, C, and D represent three independent pools of primary KD keratinocytes. (C) Images of TLR3 CRISPR keratinocytes, representative of three independent experiments; the scale bar represents 100 μm.
- Figure 5. IL-6 and IFN-λ1 are not sufficient for epithelial-to-mesenchymal transition (EMT)–like change.
(A) IL6 (16-fold) mRNA compared with controls by qRT–PCR after poly(I:C) treatment, 24 h of treatment with a 48-h washout, normalized to RPLP0, n ≥ 30. (B) Cell morphology after rhIL-6 (50 ng/ml) treatment for 72 h. Representative images were captured at 10x magnification; the scale bar represents 100 μm. Additional IL-6 concentrations (10, 50, 100, and 200 ng/ml) tested did not alter cell morphology. (C) qRT–PCR of EMT-associated genes (VIM, ZEB1, SNAI1, TWIST1) after recombinant (rh) rhIL-6 (50 ng/ml) treatment. Data were normalized to RPLP0, Mann–Whitney t test, one-tailed, P < 0.05, n ≥ 8. (D) IFNL1 (27-fold) mRNA compared with controls by qRT–PCR after poly(I:C) treatment, 24 h of treatment with a 48-h washout, normalized to RPLP0, n ≥ 4. (E) Cell morphology after rhIFN-λ1 (50 ng/ml) treatment for 72 h. Representative images were captured at 10x magnification; the scale bar represents 100 μm. Additional IFN-λ1 concentrations (50 and 100 ng/ml) tested did not alter cell morphology. (F) qRT–PCR of EMT-associated genes (VIM, ZEB1, SNAI1, TWIST1) after recombinant rhIFN-λ1 (50 ng/ml) treatment. Data were normalized to RPLP0, Mann–Whitney t test, one-tailed, P < 0.05, n ≥ 6. * denotes significance compared with the control, P < 0.05, Mann–Whitney t test, two-tailed, unless otherwise noted above.
- Figure 6. NF-κB signaling is required for epithelial-to-mesenchymal transition after TLR3 activation.
(A, B) Volcano plots illustrating changes in NF-κB-1–regulated genes after (A) UVB and (B) poly(I:C) treatment. (C) Normal human epidermal keratinocytes treated with 100 μM SC-514 (IKK2 inhibitor) for 1 h before 4 h of poly(I:C) treatment (20 μg/ml) were stained for NF-κB(p65) localization. Representative images were captured at 10X, insets to highlight + or – nuclear location of p65. Scale bars represent 10 μm. (D) Graphical representation of image quantification (n = 10 per group), unpaired t test, two-tailed. * denotes significance compared with the control, P < 0.05, # significance compared with poly(I:C), P < 0.05. (E) Treatment with SC-514 IKK2 (100 μM) limits epithelial-to-mesenchymal transition–associated protein changes induced by poly(I:C) treatment at 48 h and 72 h, representative image. (F) Relative density of fibronectin, vimentin, and E-cadherin was normalized to GAPDH. Fold change was calculated compared with the control, paired t test, one-tailed, n = 4. (G) Representative invasion assay with normal human epidermal keratinocytes treated with SC-514, n ≥ 5 technical replicate wells per group. Data were analyzed by multiple t tests, one per time-point, using the Holm–Sidak correction method, with alpha = 0.05, and a consistent SD was assumed. * denotes significance compared with the control, # significance compared with poly(I:C), P < 0.05.
- Figure S3. Poly (I:C) induces nuclear localization of p65 NF-κB protein in normal human epidermal keratinocytes.
Normal human epidermal keratinocytes treated with 100 μM SC-514 (IKK2 inhibitor) for 1 h before 4 h of poly(I:C) treatment (20 μg/ml) were stained for NF-κB(p65) localization. Isotype IgG antibody served as a control. Representative images were captured at 10X. Scale bars represent 10 μm.
- Figure 7. Patient tumors express TLR3.
(A) Gene expression of TLR3 and non-melanoma skin cancer (NMSC)–associated genes in clinically normal skin from both unexposed and sun-exposed body areas, qRT–PCR, normalized to RPLP0, n ≥ 4 patient samples per group. (B) Gene expression of NMSC-associated genes in normal human epidermal keratinocytes treated with TLR3 agonist poly(I:C) (20 μg/ml) for 24 h followed by a 24-h washout. Data were normalized to RPLP0, n ≥ 11. (C) Gene expression of TLR3- and NMSC-associated genes in normal human epidermal keratinocytes pre-treated with CU CPT 4A (80 μM) for 1 h before poly(I:C) (20 μg/ml) treatment. Cells were harvested 48 h after poly(I:C). Data were normalized to RPLP0, # significance compared with poly(I:C), P < 0.05, n ≥ 20. (D) TLR3 protein expression in well- and moderately differentiated cutaneous squamous cell carcinomas. Tumors were graded by H&E and stained for TLR3 and IgG by IHC (n = 13 individual tumors), images were captured at 20X, and the scale bar represents 50 μm. (E) TLR3 protein expression in graded basal cell carcinomas. Tumors were graded by H&E and stained for TLR3 and IgG by IHC (n = 8 individual tumors), images were captured at 20X magnification, and the scale bar represents 50 μm. * denotes significance compared with the control, P < 0.05, Mann–Whitney t test, two-tailed.
Tables
Group Condition Replicate Total sequences (million) % aligned M aligned (million) 72-h UVB Control A 9.2 86.10% 8 B 9.4 86.50% 8.2 C 11.8 84.40% 10 10 mJ/cm2 A 7.8 82.40% 6.4 B 9.3 85.60% 8 C 10.8 85.70% 9.3 72-h poly(I:C) Control A 11.4 84.80% 9.7 B 8.9 86.80% 7.7 C 11.8 87.00% 10.2 20 μg/ml A 9.5 87.50% 8.3 B 12.2 88.20% 10.8 C 7.6 86.80% 6.6 Average 9.98 85.98% 8.60
Supplemental Data 1.
Biological and technical replicates of experimental assays.[LSA-2024-02777_Supplemental_Data_1.xlsx]
In this Issue
Subjects
Related Articles
Cited By...
- No citing articles found.