Serine-Threonine Kinase Receptor Associate Protein (STRAP) confers an aggressive phenotype in neuroblastoma via regulation of Focal Adhesion Kinase (FAK)

Abstract

Background: Serine-threonine kinase receptor associated protein (STRAP), a scaffolding protein, is upregulated in many solid tumors. As such, we hypothesized that STRAP may be overexpressed in neuroblastoma tumors and may play a role in neuroblastoma tumor progression.

Methods: We examined two publicly available neuroblastoma patient databases, GSE49710 (n = 498) and GSE49711 (n = 498), to investigate STRAP expression in human specimens. SK-N-AS and SK-N-BE(2) human neuroblastoma cell lines were stably transfected with STRAP overexpression (OE) plasmid, and their resulting phenotype studied. PamChip® kinomic peptide microarray evaluated the effects of STRAP overexpression on kinase activation.

Results: In human specimens, higher STRAP expression correlated with high-risk disease, unfavorable histology, and decreased overall neuroblastoma patient survival. STRAP OE in neuroblastoma cell lines led to increased proliferation, growth, supported a stem-like phenotype and activated downstream FAK targets. When FAK was targeted with the small molecule FAK inhibitor, PF-573,228, STRAP OE neuroblastoma cells had significantly decreased growth compared to control empty vector cells.

Conclusion: Increased STRAP expression in neuroblastoma was associated with unfavorable tumor characteristics. STRAP OE resulted in increased kinomic activity of FAK. These findings suggest that the poorer outcomes in neuroblastoma tumors associated with STRAP overexpression may be secondary to FAK activation.

Keywords: FAK; Focal adhesion kinase; Neuroblastoma; STRAP; Serine-threonine kinase receptor-associated protein.

Figure 1:. STRAP expression is associated with poor patient prognosis.

Using R2, a publicly available genomic analysis platform, we examined the neuroblastoma patient database, GSE49711 (n=498) consisting of RNA-Seq data and investigated STRAP expression in human patient tumors. Increased STRAP mRNA was significantly associated with (A) unfavorable tumor histology, (B) high-risk disease, (C) disease progression, (D) decreased event free survival, (E) decreased overall survival, and (F) increased risk of death from disease.

Figure 2:. Overexpression (OE) of STRAP increased proliferation and growth in SK-N-AS and SK-N-BE(2) cells.

SK-N-AS and SK-N-BE(2) neuroblastoma cells were stably transfected with pcDNA3-Flag (empty vector, EV) or pcDNA3-STRAP-Flag (STRAP overexpression, OE) plasmids. Immunoblotting confirmed the expression of Flag protein, a marker of successful transfection, and overexpression of STRAP protein in STRAP OE cells compared to control EV or wild type (WT) cells (A, D). (B, E) Proliferation was compared between EV and STRAP OE cells using CellTiter 96^® assay. Cells (5 × 10³) were plated onto 96-well plates. After 24 hours, CellTiter 96^® dye (10 μL) was added to each well and the absorbance was measured at 490 nm using a microplate reader. At 24 hours, AS STRAP OE (B) and BE STRAP OE cells (E) demonstrated significantly increased proliferation compared to control EV cells. (C, F) EV and STRAP OE cells (5 × 10⁴) were plated in 12-well plates. To measure cell growth over time, cells were lifted, stained with trypan blue and live cells counted at 24, 48, and 72 hours. AS STRAP OE (C) and BE STRAP OE (F) cells had significantly increased cell growth over time compared to respective control EV cells. Data reported as mean fold change ± SEM. Experiments were repeated with at least three biologic replicates. *p≤0.05, **p≤0.01, ***p≤0.001.

Figure 3:. STRAP OE resulted in progression of the cell cycle and increased stemness.

(A) AS EV and AS STRAP OE cells were plated and serum starved for 24 hours to synchronize the cells. Cell cycle was analyzed using flow cytometry. AS STRAP OE cells had significantly decreased percentage of cells in G1 phase and increased percentage in G2 phase compared to AS EV cells, indicating increased progression through the cell cycle. (B) Representative histograms from a single cell cycle experiment are shown for AS EV and AS STRAP OE cells. (C) Cumulative cell cycle data are presented in tabular form and represent data from three biologic replicates. (D) Tumorsphere formation was examined using an extreme limiting dilution assay. AS EV and AS STRAP OE cells were plated at increasing cell concentrations per row ranging from 10 to 5000 cells per well in a 96 well plate. After 1 week, tumorspheres were counted. AS STRAP OE cells had an increased ability to form tumorspheres at lower cell numbers, indicating increased cell stemness. (E) AS EV and AS STRAP OE cells were stained with CD133 antibody, and percent CD133 expression was analyzed by flow cytometry. AS STRAP OE cells had significantly increased percentage of CD133 positive cells compared to EV cells. (F) Abundance of mRNA for stemness markers, Oct4, Nanog, and Sox2, was evaluated using real-time PCR. AS STRAP OE cells had significantly increased abundance of mRNA of stemness markers compared to AS EV cells. Data reported as mean ± SEM. Experiments were repeated with at least three biologic replicates. *p≤0.05, **p≤0.01, ***p≤0.001.

Figure 4:. STRAP is associated with FAK activation.

(A) AS EV and AS STRAP OE cells were examined using PamChip^® kinomic peptide microarray, and the resulting phosphorylation curves shown (far right column). AS STRAP OE cells had an increase in the phosphorylation of FAK downstream targets compared to AS EV cells, suggesting an increase in FAK activity in the OE cells. (B) Immunoblotting was used to detect target proteins. CRISPR-Cas9 was used to generate STRAP knockout (KO) cells. The AS STRAP KO cells had decreased FAK phosphorylation compared to the control AS wildtype (WT) cells. (C, D) AS EV and AS STRAP OE cells were treated with increasing concentrations (0, 7.5, 15 μM) of the small molecule FAK inhibitor, PF-573,228 (PF), for 24 hours. STRAP OE cells had decreased (C) proliferation and (D) viability compared to EV. Respective IC₅₀ was 19.4 ± 3.4 μM vs 11.8 ± 1.4μM (EV vs OE, p=0.03) and LD₅₀ 18.6 ± 3.1 μM vs 13.0 ± 1.0 μM (EV vs OE, p=0.05). Data reported as mean fold change ± SEM. Experiments were repeated with at least three biologic replicates. **p≤0.01, ***p≤0.001.