Long non-coding RNA PICSAR decreases adhesion and promotes migration of squamous carcinoma cells by downregulating α2β1 and α5β1 integrin expression

Abstract

Long non-coding RNAs (lncRNAs) regulate various cellular processes, and they have emerged as potential biomarkers and therapeutic targets in cancer. We have previously characterized the oncogenic role of lncRNA PICSAR (p38 inhibited cutaneous squamous cell carcinoma associated lincRNA) in cutaneous squamous cell carcinoma (cSCC), the most common metastatic skin cancer. In this study, we show that knockdown of PICSAR in cSCC cells upregulates expression of α2, α5 and β1 integrins, resulting in increased cell adhesion and decreased cell migration on collagen I and fibronectin. In contrast, overexpression of PICSAR in cSCC cells downregulates expression of α2, α5 and β1 integrins on cell surface, resulting in decreased cell adhesion on collagen I and fibronectin and increased cell migration. These results demonstrate a novel mechanism for regulation of the expression of collagen and fibronectin binding integrins by lncRNA PICSAR, leading to altered adhesion and migration of cSCC cells.This article has an associated First Person interview with the first author of the paper.

Keywords: Adhesion; Cancer; Cutaneous squamous cell carcinoma; Integrin; Long non-coding RNA; PICSAR.

Fig. 1.

Knockdown of PICSAR inhibits cSCC cell migration and increases cSCC cell adhesion and formation of lamellipodia. cSCC cells (UT-SCC12A) were transfected with PICSAR siRNA (siRNA1) or control siRNA. (A) Cells were plated on collagen I or fibronectin 72 h after transfection and migration of individual cells was imaged using the IncuCyte ZOOM^® real-time cell imaging system. Cell tracking (n=15) was quantitated using ImageJ software. Median±s.d. is shown; *P<0.05, Mann–Whitney two-way U-test. Representative images of cell tracking are shown. Scale bars: 100 µm. (B) Cells were plated on collagen I or fibronectin coated 96-well electronic microtiter plate 72 h after transfection and cell adhesion was measured using the xCELLigence system (n=3). The cell index indicates the readout of the microelectrode impedance which corresponds to the strength of cell adhesion. Mean±s.d. is shown; **P<0.01, ***P<0.001; two-tailed t-test. (C) Cells were plated on collagen I or fibronectin 72 h after transfection and fixed 4 h after plating. The number of lamellipodia containing cells was quantitated from microscopic images with a 20x objective (n=3). Representative images of the quantification are shown. Scale bars: 10 µm. Mean±s.d. is shown; **P<0.01, ***P<0.001; two-tailed t-test.

Fig. 2.

Knockdown of PICSAR increases integrin expression in cSCC cells. (A) Three cSCC cell lines (UT-SCC12A, UT-SCC59A, and UT-SCC118) were transfected with PICSAR siRNA (siRNA1) or control siRNA. After 72 h of transfection, whole transcriptome analysis was performed with RNA-seq. Heatmap showing several cell adhesion related genes regulated by PICSAR knockdown (FC log 2>0.5). GEF, guanine nucleotide exchange factor; PIPase, phosphatidylinositol phosphatase; STK, serine/threonine specific protein kinase; TK, tyrosine kinase; PTP, protein tyrosine phosphatase; *P<0.05. (B) Levels of PICSAR and α2, α5 and β1 integrin mRNAs were determined with qPCR in PICSAR knockdown UT-SCC12A and UT-SCC59A cells (n=3). Mean±s.d. is shown. *P<0.05, ***P<0.001; two-tailed t-test. (C) Flow cytometry was performed to measure α2, α5 and β1 integrin expression on the cell surface in PICSAR knockdown UT-SCC12A and UT-SCC59A cells. (D) UT-SCC12A cells were transfected with PICSAR siRNA or control siRNA and plated on collagen I or fibronectin. Cells were fixed and stained with primary antibodies against α2 or α5 integrins (red), Alexa Fluor^® 488 conjugated phalloidin (green) and Hoechst (blue). Scale bars: 10 µm.

Fig. 3.

PICSAR overexpression decreases cell adhesion and spreading, and increases migration of cSCC cells by regulating integrin expression. UT-SCC59A cells were transfected with PICSAR expression construct (pcDNA3.1_PICSAR) or empty vector (pcDNA3.1) and selective pressure of cell pools was maintained by Geneticin. (A) Expression of PICSAR and α2, α5 and β1 integrin mRNAs was measured using qPCR (n=3). Mean±s.d. is shown. (B) Flow cytometry was used to measure α2, α5 and β1 integrin expression on the cell surface. (C) PICSAR overexpressing and control UT-SCC59A cells were plated on collagen I and fibronectin and the number of spread cells was quantitated from microscopic images with a 20x objective (n=3). Scale bars: 10 µm. Mean±s.d. is shown. Representative images of the quantification are shown. (D) PICSAR overexpressing and control UT-SCC59A cells were plated on collagen I or fibronectin and stained with primary antibodies against α2 or α5 integrin (red), Alexa Fluor^® 488 conjugated phalloidin (green) and Hoechst (blue). Scale bars: 10 µm. (E) Adhesion of PICSAR overexpressing and control UT-SCC59A cells on collagen I and fibronectin was measured using the xCELLigence system (n=3). Mean±s.d. is shown. (F) PICSAR overexpressing and control UT-SCC59A cells were plated on a 96-well plate and incubated for 6 h with 1 mM hydroxyurea to prevent cell proliferation. Cell monolayer was scratched and cells were imaged using the IncuCyte ZOOM^® real-time cell imaging system (n=5). Mean±s.e.m. is shown. (G) Protein levels for phosphorylated and total Src were determined by western blotting analysis of PICSAR siRNA (siRNA1) transfected and stably PICSAR overexpressing UT-SCC59A cells. The representative image and quantification shown are from one individual experiment. β-Actin was used as a reference. *P<0.05, **P<0.01, ***P<0.001; two-tailed t-test.