Fascin overexpression promotes neoplastic progression in oral squamous cell carcinoma

Abstract

Background: Fascin is a globular actin cross-linking protein, which plays a major role in forming parallel actin bundles in cell protrusions and is found to be associated with tumor cell invasion and metastasis in various type of cancers including oral squamous cell carcinoma (OSCC). Previously, we have demonstrated that fascin regulates actin polymerization and thereby promotes cell motility in K8-depleted OSCC cells. In the present study we have investigated the role of fascin in tumor progression of OSCC.

Methods: To understand the role of fascin in OSCC development and/or progression, fascin was overexpressed along with vector control in OSCC derived cells AW13516. The phenotype was studied using wound healing, Boyden chamber, cell adhesion, Hanging drop, soft agar and tumorigenicity assays. Further, fascin expression was examined in human OSCC samples (N = 131) using immunohistochemistry and level of its expression was correlated with clinico-pathological parameters of the patients.

Results: Fascin overexpression in OSCC derived cells led to significant increase in cell migration, cell invasion and MMP-2 activity. In addition these cells demonstrated increased levels of phosphorylated AKT, ERK1/2 and JNK1/2. Our in vitro results were consistent with correlative studies of fascin expression with the clinico-pathological parameters of the OSCC patients. Fascin expression in OSCC showed statistically significant correlation with increased tumor stage (P = 0.041), increased lymph node metastasis (P = 0.001), less differentiation (P = 0.005), increased recurrence (P = 0.038) and shorter survival (P = 0.004) of the patients.

Conclusion: In conclusion, our results indicate that fascin promotes tumor progression and activates AKT and MAPK pathways in OSCC-derived cells. Further, our correlative studies of fascin expression in OSCC with clinico-pathological parameters of the patients indicate that fascin may prove to be useful in prognostication and treatment of OSCC.

Figure 1

Overexpression of fascin leads to reorganization of actin cytoskeleton. (A) Western blot and confocal analysis of stable overexpression of Fascin-GFP (AW-Fascin-1 and AW-Fascin-2) and empty vector control pEGFP (AW-GFP-Cont) clones derived from AW13516 cells with antibodies to GFP and fascin. β-actin was used as loading control. Scale bars: 10 μm. (B) Representative confocal images of stably expressed GFP tagged fascin and GFP alone in AW-Fascin-1, AW-Fascin-2 and AW-GFP-Cont clones respectively. Cells were counter stained with DAPI. Scale bars: 10 μm. Arrow heads and arrows indicate filopodia and lamellipodia respectively. (C) Representative confocal images of F-actin stained with phalloidin-TRITC and the co-localization of F-actin with Fascin-GFP in stable AW-Fascin-1, AW-Fascin-2 and AW-GFP-Cont clones. Cells were counter stained with DAPI. Scale bars: 10 μm. Arrows indicate colocalization of Fascin-GFP with F-actin at filopodia like structures. (D) Histogram showing number of filopodia formed by fascin overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells. Mean and standard deviation of 3 independent experiments is plotted (p < 0.0001). (E) Histogram showing number of lamellipodia formed by fascin overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells. Mean and standard deviation of 3 independent experiments is plotted (p < 0.0001)

Figure 2

Overexpression of fascin leads to increase in cell migration, invasion and MMPs activity. (A and B) Phase contrast images (magnification 10×) of wound closure at 0 h, 6 h and 12 h of the vector control (AW-GFP-Cont) and fascin overexpressed clones (AW-Fascin-1 and AW-Fascin-2) are shown in the figure. Scale bar: 100 μm. The data shown is the average migration rate from three independent experiments with the mean and standard deviation (p < 0.05). (C) Boyden chamber invasion assay of vector control (AW-GFP-Cont) and fascin overexpressed clones (AW-Fascin-1 and AW-Fascin-2). The data shown is the average from three independent experiments with the mean and standard deviation (p < 0.01). (D) Zymogram showing MMP-2 activity in vector control (AWGFP-Cont) and fascin overexpressed (AW-Fascin-1 and AW-Fascin-2) cells.

Figure 3

Fascin overexpressed cells demonstrated increase in cell-ECM adhesion and loss of cell-cell contacts. (A) Cell adhesion of fascin-overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells to different ECM substrates was measured as described. The data shown is the average from three experiments with the mean and standard deviation. (B) Western blot analysis of stable fascin-overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) clones with antibodies to α6 integrin, β4 integrin and phosphorylated FAK. β-actin and total FAK were used as internal loading controls respectively. (C) Representative images of aggregates formed by of fascin overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells in hanging drop assay. (D) Histogram showing size of aggregates formed by overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells. Mean and standard deviation of 3 independent experiments is plotted (p < 0.0001). (E) Confocal analysis of E-cadherin and β-catenin localization in fascin overexpressed cells. Scale bars; 10 μm. (F) Western blot analysis of fascin overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control clones (AW-GFP-Cont) with antibodies to E-cadherin and β-catenin. β-actin was used as a loading control. (G) RT-PCR analysis of snail, slug and vimentin overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells. GAPDH was used as internal control.

Figure 4

Fascin overexpression resulted in increase in cell proliferation and tumorigenicity. (A) Histogram showing the total number of colonies formed per plate by the indicated clones after 14 days. Mean and standard deviation of three independent experiments is plotted (p < 0.01). (B) Tumor growth of fascin-overexpressed clones (AW-Fascin-1, AW-Fascin-2) or the vector control (AW-GFP-Cont) was plotted against time interval. (C) Cell proliferation curves of fascin-overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) cells using MTT assay. The cell number was determined at the times indicated. Each point represents the mean and the standard errors of three independent experiments done in triplicate. (D and E) Western blot analysis of stable fascin-overexpressed (AW-Fascin-1 and AW-Fascin-2) and vector control (AW-GFP-Cont) clones with antibodies to PCNA and phosphorylated forms of Akt, ERK1/2, JNK1/2 and p38. Total Akt, ERK1/2, JNK1/2, p38 and β-actin were used as internal loading controls respectively.

Figure 5

IHC analysis of fascin levels in human OSCC tissues. (A) Representative images of immunohistochemical staining with antibodies against fascin of paraffin embedded sections of normal different lymph-node metastasis and stage of human OSCC tissues. Sections were counter stained with eosin (Magnification: 200×). (B) Histograms showing correlation of fascin expression with clinico-pathological parameters such as differentiation, lymph-node metastasis and tumor stage of OSCC patients. (C) Representative images of immunofluoroscence staining with antibodies against fascin of paraffin embedded sections of lymph-node metastasized tumor of human OSCC tissues. Sections were counter stained with DAPI. Scale bars: 50 μm.

Figure 6

Correlation of fascin expression with survival rate of OSCC patients using Kaplan-Meier analysis. (A and B) Analysis of overall survival in OSCC patients and fascin expression (fascin expression is either negative or positive; *, p < 0.025), (fascin expression is negative, low or high; *, p < 0.004) (C) Analysis of disease free survival times of OSCC patients and fascin (fascin expression is either negative or positive; **, p < 0.013) expression. (D) Histogram showing percentage of recurrence in fascin positive samples of OSCC patients.