The Activation of PPARγ by (2Z,4E,6E)-2-methoxyocta-2,4,6-trienoic Acid Counteracts the Epithelial-Mesenchymal Transition Process in Skin Carcinogenesis

Abstract

Cutaneous squamous cell carcinoma (cSCC) is the most common UV-induced keratinocyte-derived cancer, and its progression is characterized by the epithelial-mesenchymal transition (EMT) process. We previously demonstrated that PPARγ activation by 2,4,6-octatrienoic acid (Octa) prevents cutaneous UV damage. We investigated the possible role of the PPARγ activators Octa and the new compound (2Z,4E,6E)-2-methoxyocta-2,4,6-trienoic acid (A02) in targeting keratinocyte-derived skin cancer. Like Octa, A02 exerted a protective effect against UVB-induced oxidative stress and DNA damage in NHKs. In the squamous cell carcinoma A431 cells, A02 inhibited cell proliferation and increased differentiation markers' expression. Moreover, Octa and even more A02 counteracted the TGF-β1-dependent increase in mesenchymal markers, intracellular ROS, the activation of EMT-related signal transduction pathways, and cells' migratory capacity. Both compounds, especially A02, counterbalanced the TGF-β1-induced cell membrane lipid remodeling and the release of bioactive lipids involved in EMT. In vivo experiments on a murine model useful to study cell proliferation in adult animals showed the reduction of areas characterized by active cell proliferation in response to A02 topical treatment. In conclusion, targeting PPARγ may be useful for the prevention and treatment of keratinocyte-derived skin cancer.

Keywords: EMT; PPARγ; octatrienoic acid; skin carcinoma.

Figure 1

A02 induces PPARγ expression and activation, promotes antioxidant defense, and protects from UVB-induced damages in NHKs. (A) Schematic representation of the chemical structure of A02 and its 3D molecular model of docking to PPARγ human receptor. (B) Western blot analysis of PPARγ protein expression in NHKs treated with A02 (10–50–90 µM) for 24 h. Representative blots are shown. (C) Luciferase activity analysis of NHKs transfected with pGL3-(Jwt)3TKLuc reporter construct. After 24 h of transfection, cells were treated with A02 (10–50–90 µM) for 24 h. The variability of transfection was normalized with Renilla luciferase activity. The data are presented as the mean ± SD of three independent experiments and are expressed as the fold change with respect to untreated control cells (* p < 0.05, ** p < 0.01 vs. untreated control). (D) Real-time RT-PCR analysis of CAT, HO-1, and NQO1 in NHKs treated with A02 (10–50–90 µM) for 6 h. All mRNA values were normalized against the expression of GAPDH and were expressed relative to untreated control cells. The data in the graphs are the mean ± SD of three independent experiments (* p < 0.05, ** p < 0.01 vs. untreated control). (E) Western blot analysis of catalase protein expression in NHKs treated with A02 (10–50–90 µM) for 24 h. Representative blots are shown. (F) Real-time RT-PCR analysis of CAT, HO-1, and NQO1 in NHKs pre-incubated with A02 (10–50–90 µM) for 24 h and then irradiated with UVB 25 mJ/cm². mRNA was extracted 6 h after the irradiation. All mRNA values were normalized against the expression of GAPDH and were expressed relative to untreated control cells. The data in the graphs are mean ± SD of three independent experiments (* p < 0.05 vs. untreated control; ^$ p < 0.05, ^$$ p < 0.01 vs. UVB-treated cells). (G) Western blot analysis of catalase, p53, and phospho-γH2AX protein expression in NHKs pre-incubated with A02 (10–50–90 µM) for 24 h and then irradiated with UVB 25 mJ/cm². Proteins were extracted 24 h after the irradiation. Representative blots are shown. β-actin was used as an endogenous loading control for Western blot analysis. Densitometric scanning of band intensities was performed to quantify the change in protein expression. Data represent the mean ± SD of three independent experiments and are expressed as fold change with respect to untreated control cells (control value taken as 1-fold in each case).

Figure 2

Effects of A02 and Octa on viability, proliferation, and differentiation of A431 cells. (A) Cell number analysis was performed in A431 treated with Octa or A02 (90 µM) for 72 h. The data are presented as the mean ± SD of three independent experiments and are expressed as the fold change with respect to untreated control cells (* p < 0.05 vs. untreated control). (B) Cell cycle distribution evaluated by flow cytometric analysis on A431 treated with Octa or A02 (90 µM) for 48 h. The bar graph shows the distribution of cells among the different phases of the cell cycle. The data are expressed as the mean ± SD of three independent experiments (* p < 0.05 vs. untreated control). (C) Western blot analysis of p21 and cyclin D1 protein expression in A431 treated with Octa or A02 (90 µM) for 24 h and 48 h, respectively. Representative blots are shown. (D) Real-time RT-PCR analysis of FLG, IVL, and LOR in A431 treated with Octa or A02 (90 µM) for 48 h. All mRNA values were normalized against the expression of GAPDH and were expressed relative to untreated control cells. The data in the graphs are mean ± SD of three independent experiments (* p < 0.05, ** p < 0.01 vs. untreated control). (E) Western blot analysis of filaggrin, involucrin, and loricrin protein expression in A431 treated with Octa or A02 (90 µM) for 72 h. β-actin and GAPDH were used as endogenous loading control for Western blot analyses. Densitometric scanning of band intensities was performed to quantify the change in protein expression. Data represent the mean ± SD of three independent experiments and are expressed as fold change with respect to untreated control cells (control value taken as 1-fold in each case).

Figure 3

Effects of A02 and Octa on the TGF-β1-induced EMT in A431 cells. (A) Real-time RT-PCR analysis of NCAD, ECAD, SLUG, FIBRONECTIN, VIMENTIN, and MMP2 in A431 pre-incubated with Octa or A02 (90 µM) for 1 h and then treated with TGF-β1 (15 ng/mL). All mRNA values were normalized against the expression of GAPDH and were expressed relative to untreated control cells. The data in the graphs are the mean ± SD of three independent experiments (* p < 0.05, ** p < 0.01 vs. untreated control; ^$ p < 0.05, ^$$ p < 0.01 vs. TGF-β1-treated cells). (B) Western blot analysis of vimentin, E-cadherin, and fibronectin protein expression in A431 pre-incubated with Octa or A02 (90 µM) for 1 h and then treated with TGF-β1 (15 ng/mL) for 72 h. GAPDH was used as an endogenous loading control for Western blot analysis. Densitometric scanning of band intensities was performed to quantify the change in protein expression. Data represent the mean ± SD of three independent experiments and are expressed as fold change with respect to untreated control cells (control value taken as 1-fold in each case). (C) Immunofluorescence and quantitative analysis of vimentin, fibronectin, and E-cadherin in A431 cells. Results are expressed as fold change of positive cells or mean fluorescence intensity with respect to control. Nuclei were counterstained with DAPI. Bar: 20 μm. (* p < 0.05, ** p < 0.01 vs. control; ^$ p < 0.05 vs. TGF-β1; ^$$ p < 0.01 vs. TGF-β1).

Figure 4

Effects of A02 and Octa on ROS generation, inflammation, and migration in A431 cells. (A) ROS levels in A431 pre-treated with Octa and A02 (90 µM) for 1 h and then exposed to TGF-β1 (15 ng/mL) for 24 h. Results are expressed as the fold change with respect to untreated control cells (** p < 0.01 vs. untreated control; ^$$ p < 0.01 vs. TGF-β1-treated cells). Western blot analysis of phospho-p38, p38, phospho-AKT, AKT, phospho-ERK, and ERK protein expression in A431 pre-incubated with Octa or A02 (90 µM) for 1 h and then treated with TGF-β1 (15 ng/mL) for 15 min (B) and 24 h (C). β-actin was used as an endogenous loading control for Western blot analyses. Densitometric scanning of band intensities was performed to quantify the change in protein expression. Data represent the mean ± SD of three independent experiments and are expressed as fold change with respect to untreated control cells (control value taken as 1-fold in each case). (D) Representative images of the scratch assay of A431 cells treated with Octa or A02 alone or after TGF-β1 stimulation at T0 and 24 h. Bar: 200 μm. Quantitative analysis resulting from the measurements of the covered scratched area. The results are expressed as fold change with respect to T0 which was set as 1 by definition (** p < 0.01 vs. T0; ^$$ p < 0.01 vs. TGF-β1).

Figure 5

A02 and Octa effects against the alteration of cellular and extracellular lipid composition following the TGF-β1-induced EMT in A431 cells. Quantification by GCMS of (A) cell membrane MUFA/SFA ratio, (B) extracellular C16:1/C16:0 ratio, (C) extracellular C18:1/C18:0 ratio, and (D) cholesterol amount in A431 cells pre-treated with Octa and A02 (90 µM) for 1 h and then exposed to TGF-β1 (15 ng/mL) alone or in combination with Octa and A02 for 24 h and 72 h. Quantification by LC-MS/MS of (E) 5HETE, (F) PGE2, (G) PGD2, and (H) PGF2 released in the medium by A431 cells pre-treated with Octa and A02 (90 µM) for 1 h and then exposed to TGF-β1 (15 ng/mL) alone or in combination with Octa and A02 for 24 h and 72 h. The results are expressed as a ratio or as fold change with respect to control. (* p < 0.05 vs. control; ** p < 0.01 vs. control; ^$ p < 0.05 vs. TGF-β1; ^$$ p < 0.01 vs. TGF-β1).

Figure 6

PPARγ activation is needed to promote A02 biological effects. (A) A431 cells were transfected with siRNA specific for PPARγ (siPPARγ) or non-specific siRNA (siCtr). PPARγ level was evaluated by real-time RT-PCR and Western blot analysis 24 h after transfection. The mRNA values were normalized against the expression of GAPDH. The data in the graphs are the mean ± SD of three independent experiments (** p < 0.01 vs. siCtr cells). Representative blots are shown. Western blot analysis of (B) p21 and (C) phospho-p38, p38, phospho-ERK, and ERK protein expression in A431 cells transfected with siPPARγ and siCtr exposed to A02 (90 µM) for 1 h and then treated with TGF-β1 (15 ng/mL) for 24 h. Representative blots are shown. Western blot analysis of (D) involucrin and loricrin and (E) vimentin and fibronectin protein expression in A431 pre-incubated with GW9662 (3 µM) for 1 h, exposed to A02 (90 µM) for 1 h, and finally treated with TGF-β1 (15 ng/mL) for 72 h. GAPDH was used as an endogenous loading control for Western blot analysis. Densitometric scanning of band intensities was performed to quantify the change in protein expression. Data represent the mean ± SD of three independent experiments and are expressed as fold change with respect to siCtr cells (B,C) and untreated control cells (D,E) (control value taken as 1-fold in each case).

Figure 7

A02 impact on DMBA-TPA effects in vivo. (A) Luciferase activity in MITO-Luc mouse with induced papillomas. BLI of MITO-Luc mice before treatment (pre-imaging), after treatment with DMBA-TPA on the ventral abdomen (t0), and after 15 treatments with placebo or A02 (t1). Images were collected for 5 animals for each treatment, and a representative animal is shown. Photon emission from the lesions was measured as photons per second per square centimeter per steradian (photons/s/cm²/sr). (B) The graph illustrates BLI signal intensities relative to DMBA-TPA drug treatment control (t0) after 15 treatments with placebo or A02 (t1). Each bar represents the mean value ± SEM of five animals. The number of lesions (C) and tumor area (cm²) (D) have been measured at t0 and t1 by automatically drawing the regions of interest upon BLI analysis. The graph represents the mean value ± SEM of five animals. (E) Real-time RT-PCR analysis of NCAD, ECAD, SLUG, FIBRONECTIN, VIMENTIN, α-SMA, and MMP2 in samples of mouse skin collected in lesional and nonlesional areas at the end of the 15 treatments with placebo or A02 (3 different mice per treatment group). All mRNA values were normalized against the expression of GAPDH and were expressed relative to nonlesional skin (* p < 0.05, ** p < 0.01 vs. placebo).