Implication of RNA-binding protein La in proliferation, migration and invasion of lymph node-metastasized hypopharyngeal SCC cells

Abstract

The 5-year survival rate for oral cavity cancer is poorer than for breast, colon or prostate cancer, and has improved only slightly in the last three decades. Hence, new therapeutic strategies are urgently needed. Here we demonstrate by tissue micro array analysis for the first time that RNA-binding protein La is significantly overexpressed in oral squamous cell carcinoma (SCC). Within this study we therefore addressed the question whether siRNA-mediated depletion of the La protein may interfere with known tumor-promoting characteristics of head and neck SCC cells. Our studies demonstrate that the La protein promotes cell proliferation, migration and invasion of lymph node-metastasized hypopharyngeal SCC cells. We also reveal that La is required for the expression of β-catenin as well as matrix metalloproteinase type 2 (MMP-2) within these cells. Taken together these data suggest a so far unknown function of the RNA-binding protein La in promoting tumor progression of head and neck SCC.

Figure 1. RNA-binding protein La is overexpressed in squamous cell carcinoma (SCC) tissue.

(A) Human La-specific staining of normal epithelial and SCC tongue tissue by applying the monoclonal human La-specific 3B9 antibody. Scale bar represents 100 µm. (B) Scoring of oral tissue microarray (TMA) harboring 9 normal and 42 SCC tissue cores stained with human La-specific 3B9 antibody revealed highly significant overexpression of nuclear La in oral SCC. Percentage [%] of La-positive cells in normal versus SCC tissue (P value<0.001 (three asterisks)). (C) Intensity of nuclear La-staining in normal versus SCC tissue (P value<0.05 (one asterisks)). Two-tailed P values were determined by paired t-test applying Prism 4 (GraphPad Software).

Figure 2. La protein expression promotes proliferation and cell cycle progression of SCC 22B cells.

(A) Lower La expression in siRNA-mediated SCC 22B cells (see Western blot analysis in inlet) correlates with decreased proliferation rate (dotted line). (B) Reduced proliferation rate of La-depleted SCC 22B cells as determined as mean ±SD (error bars) of 5 independent experiments performed in triplicates. P value<0.01 (two asterisks). Two-tailed P value was determined by paired t-test applying Prism 4 (GraphPad Software). (C) SiRNA-mediated depletion of La does not reduce the number of viable cells as determined by counting of trypan blue-negative SCC 22B cells as determined as mean ±SD (error bars) of 3 independent experiments. (D) La depletion (La siRNA) in SCC 22B cells does not induce caspase-3 cleavage of PARP (under apoptosis PARP fragments into 116 and 85 kDa). Representative Western blot analysis of three independent experiments. As positive control PARP cleavage was induced by UV light (lane UV). GAPDH was applied as loading control. (E) SiRNA-mediated depletion of La correlates with a G1 cell cycle arrest in SCC 22B cells (P value<0.05 (one asterisks)). Two-tailed P value was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 3 independent experiments. (F) La depletion does not reduce cyclin D1 protein expression in SCC 22B cells. GAPDH was applied as loading control. Representative Western blot analysis of three independent experiments.

Figure 3. La supports cell motility of SCC 22B cells.

(A) Wound healing assay: La depletion significantly reduces migration of UM-SCC 22B cells into the denuded area. Phase contrast pictures of the scratched area were taken at time 0 and after 24 hours using a Zeiss Axio Observer microscope and x10 objective power, and (B) the denuded area was quantified at each time point using NIH image software (AU = arbitrary units). Two-tailed P value<0.001 (three asterisks) was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 3 independent experiments in triplicates. (C) Migration track assay: On a field of fluorescent microspheres gfp-transfected La-depleted and control-treated SCC 22B cells were seeded at low density (4 cells per mm²) and non-fluorescent tracks created by gfp-transfected cells were evaluated by fluorescent microscopy after 24 hours using a Zeiss Axio Observer microscope and x40 objective power. (D) Quantification of cell motility by counting the number of non-fluorescent tracks of gfp-transfected cells ( = migrating cells). La depletion significantly reduces migration of single SCC 22B cells. Two-tailed P value<0.01 (two asterisks) was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 2 independent experiments in triplicates.

Figure 4. Rescue of reduced cell motility by overexpression of siRNA-resistant gfp-tagged La in La-depleted SCC 22B cells.

(A) Western blot analysis demonstrates overexpression of gfp-tagged, La-siRNA resistant La in La-depleted (gfpLaR) SCC 22B cells. (D) Quantification of cell motility by counting the number of non-fluorescent tracks of gfp-transfected cells ( = migrating cells). La depletion significantly reduces migration of single SCC 22B cells and can be rescued to levels determined in control-treated cells by overexpression of La-siRNA resistant La. Two-tailed P value<0.01 (two asterisks) was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 2 independent experiments in triplicates.

Figure 5. Reduced β-catenin protein expression in La-depleted SCC 22B cells.

(A) Western blot analysis to determine FAK, β-catenin and cofilin protein expression levels in La-depleted SCC 22B cell extracts. GAPDH was applied as loading control. (B) For quantification of Western blot experiments chemiluminescent signals were recorded using an ImageQuant ECL systems and quantified using ImageQuant TL software. Two-tailed P value<0.001 (three asterisks) was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 4 independent experiments for β-catenin and 3 independent experiments for FAK and cofilin. (C) Overexpression of siRNA-resistant gfpLa protein (gfpLaR) rescues β-catenin expression in La-depleted SCC 22B cells. (D) Immunofluorescent staining shows localization of β-catenin within the cytoplasmic membrane in both La-depleted and control-treated SCC 22B cells.

Figure 6. La promotes in vitro invasion and MMP-2 protein expression of SCC 22B cells.

(A) In vitro invasion of La-depleted cells is reduced compared to control-treated SCC 22B cells. Two-tailed P value<0.001 (three asterisks) was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 3 independent experiments. (B) Reduced MMP-2 protein expression in La-depleted SCC 22B cells as determined by Western blot analysis. (C) For quantification of Western blot experiments chemiluminescent signals were recorded using an ImageQuant ECL systems and quantified using ImageQuant TL software. Two-tailed P value<0.01 (two asterisks) was determined by paired t-test applying Prism 4 (GraphPad Software). Error bars represent the mean ±SD of 4 independent experiments.