Insulin-like growth factor 2 expression modulates Taxol resistance and is a candidate biomarker for reduced disease-free survival in ovarian cancer

Abstract

Purpose: This study was undertaken to examine the role of the insulin-like growth factor (IGF) signaling pathway in the response of ovarian cancer cells to Taxol and to evaluate the significance of this pathway in human epithelial ovarian tumors.

Experimental design: The effect of Taxol treatment on AKT activation in A2780 ovarian carcinoma cells was evaluated using antibodies specific for phospho-AKT. To study the drug-resistant phenotype, we developed a Taxol-resistant cell line, HEY-T30, derived from HEY ovarian carcinoma cells. IGF2 expression was measured by real-time PCR. A type 1 IGF receptor (IGF1R) inhibitor, NVP-AEW541, and IGF2 small interfering RNA were used to evaluate the effect of IGF pathway inhibition on proliferation and Taxol sensitivity. IGF2 protein expression was evaluated by immunohistochemistry in 115 epithelial ovarian tumors and analyzed in relation to clinical/pathologic factors using the chi(2) or Fisher's exact tests. The influence of IGF2 expression on survival was studied with Cox regression.

Results: Taxol-induced AKT phosphorylation required IGF1R tyrosine kinase activity and was associated with upregulation of IGF2. Resistant cells had higher IGF2 expression compared with sensitive cells, and IGF pathway inhibition restored sensitivity to Taxol. High IGF2 tumor expression correlated with advanced stage (P < 0.001) and tumor grade (P < 0.01) and reduced disease-free survival (P < 0.05).

Conclusions: IGF2 modulates Taxol resistance, and tumor IGF2 expression is a candidate prognostic biomarker in epithelial ovarian tumors. IGF pathway inhibition sensitizes drug-resistant ovarian carcinoma cells to Taxol. Such novel findings suggest that IGF2 represents a therapeutic target in ovarian cancer, particularly in the setting of Taxol resistance.

Figure 1. Effect of Taxol treatment on AKT phosphorylation

A2780 cells were maintained in complete media containing 10% FBS and treated for 24 hours with Taxol (5 nmol/L), NVP-AEW541 (1 µmol/L), or both drugs concurrently. Immunoblotting was performed to determine the effect of Taxol treatment on AKT phosphorylation. A. Representative immunoblots are shown for phospho-AKT (Threonine 308) and phospho-AKT (Serine 473), along with the corresponding total AKT and GAPDH immunoblots. B. The relative level of AKT phosphorylation at Threonine 308, as determined by densitometry, is shown in the graph (mean ±SE; 4 independent experiments). *p<0.05 (One-way ANOVA, Dunnett post-test) C. The relative level of AKT phosphorylation at Serine 473, as determined by densitometry, is shown in the graph (mean ±SE; 7 independent experiments). **p<0.01 (One-way ANOVA, Dunnett post-test)

Figure 2. Taxol-induced IGF2 mRNA expression

A. The time course of IGF2 expression following Taxol treatment (5 nmol/L) in A2780 cells is depicted with squares and a solid connecting line. The IGF2 mRNA level after Baccatin treatment (5 nmol/L) for 24 hours is also shown on the graph, marked by an X. Results shown are the mean ±SE of two independent experiments, each performed in triplicate. B. A2780 cells were treated for 24 hours with vehicle alone (DMSO) or with cytotoxic drugs, at their equipotent molar concentrations. The IGF2 mRNA level is expressed as fold-change relative to untreated cells. Results shown are the mean ±SE of two independent experiments, each performed in triplicate. *p<0.05; **p<0.01 (One-way ANOVA, Dunnett post-test) C. HEY cells were treated with vehicle alone (DMSO) or Taxol 5 nmol/L for 24 hours. The IGF2 mRNA level, determined by reverse transcription quantitative real-time PCR, is expressed as fold-change relative to untreated cells in complete growth medium. Results shown are the mean ±SE of three independent experiments, each performed in triplicate. **p< 0.01 (t-test) D. IGF2 mRNA expression in Taxol-resistant HEY-T30 cells, determined by reverse transcription quantitative real-time PCR, is expressed as fold-change relative to HEY parental cells. Results shown are the mean ±SE of three independent experiments, each performed in triplicate. *p<0.05 (t-test)

Figure 3. Potentiation of Taxol treatment by the IGF1R inhibitor NVP-AEW541

Cytotoxicity assays for HEY and HEY-T30 ovarian carcinoma cells were performed using the sulforhodamine B method to determine the effect of Taxol treatment alone or with concurrent NVP-AEW541. A. HEY parental cells: the cell number relative to untreated cells, expressed as cell growth (%), is plotted over a range of Taxol concentrations. Treatment with Taxol alone is depicted with squares and solid lines, compared with combination treatment with Taxol plus NVP-AEW541 1 µmol/L, depicted with triangles and dotted lines. The cell growth (%) resulting from NVP-AEW541 (1 µmol/L) alone is plotted on the Y-axis (triangle). p<0.05 (Two-way ANOVA) B. HEY-T30 resistant cells: the cell number relative to untreated cells, expressed as cell growth (%), is plotted over a range of Taxol concentrations, as in panel A. The cell growth (%) resulting from NVP-AEW541 (1 µmol/L) alone is plotted on the Y-axis (triangle). p<0.001 (Two-way ANOVA) C. For each cell line, the IC50 concentration of Taxol alone or with concurrent NVP-AEW541 1 µmol/L is plotted in the bar graph (mean ±SE; at least 3 independent experiments with 6 replicates each). p<0.0001 (One-way ANOVA, Bonferroni post-test)

Figure 4. Effect of IGF2 depletion by siRNA transfection in HEY and HEY-T30 cells

IGF2 siRNA transfection was used to evaluate the effect of IGF2 depletion on cell proliferation and Taxol sensitivity. Cells were transfected with IGF2 siRNA (IGF2-SI) or control siRNA (c-SI). Twenty-four hours after transfection, cells were counted, reseeded, and incubated for an additional 72 hours, with or without the addition of Taxol at the IC50 concentration of Taxol for each cell line, followed by cell counting. Cell numbers are expressed as % cell growth relative to DMSO-treated, c-SI-transfected cells. Results shown are the mean ±SEM of two independent experiments, each performed in triplicate for HEY parental cells (Panel A) and HEY-T30 cells (Panel B). *p< 0.05, **p<0.01 (one-way ANOVA, Bonferroni post-test)

Figure 5. Representative images of IGF2 immunohistochemical staining

A) Absence of staining in the surface epithelium of a normal ovary (40X) B) Extremely low staining in a Stage 1 mucinous borderline tumor; H-score=0.5 C) Low staining in a Stage 3, grade 1 serous carcinoma; H-score=75 D) Medium staining in a Stage 3, grade 3 serous carcinoma; H-score=100 E) High staining in a Stage 3, grade 3 adenocarcinoma, not otherwise specified; H-score=250 F) High staining in a Stage 3, grade 3 serous carcinoma; H-score 300

Figure 6. Correlation between IGF2 expression and clinicopathologic factors in ovarian tumors

A) IGF2 staining is correlated with FIGO stage, with higher frequency of IGF2 overexpression in advanced stage tumors; p<0.001 (Fisher’s exact test) B) Serous (Ser) and endometrioid (Endo) histologic types are correlated with high IGF2 staining, compared with mucinous (Muc) tumors; p=0.011 (Chi-square test) C) IGF2 staining is correlated with tumor grade, with low malignant potential (LMP; borderline) tumors demonstrating the lowest frequency of IGF2 overexpression; p=0.003 (Fisher’s exact test) D) Patients with high IGF2 expression (red line) had a significant reduction in progression-free survival (PFS) compared with patients with low IGF2 expression (black line); p=0.03 (log-rank test). E) When LMP tumors are excluded from the analysis, the association of IGF2 expression and progression-free survival (PFS) is not statistically significant.