Differential insulin-like growth factor II (IGF-II) expression: A potential role for breast cancer survival disparity

Abstract

Objective: Increased risk of cancer and other adult diseases have been associated with perinatal exposure to adverse conditions such as stress and famine. Recently, Insulin-like growth factor II (IGF-II) was identified as the first gene associated with altered expression caused by fetal exposure to poor nutrition. IGF-II regulates fetal development and breast cancer cell survival, in part, by regulating anti-apoptotic proteins through activation of the IGF-I and insulin receptors. African-American (AA) women have a lower overall breast cancer (BC) incidence, however, they present with advanced disease at diagnosis, poorer prognosis and lower survival than Caucasian (CA) women. The reasons for the BC survival disparity are not well understood. We hypothesize that IGF-II plays a role in the survival disparity observed among AA breast cancer patients by stimulating rapid tumor growth, inhibiting apoptosis, and promoting metastasis.

Design: This study examines IGF-II expression and regulation of the anti-apoptotic proteins Bcl-2, Bcl-X(L), and survivin in Hs578t (ER-), CRL 2335 (ER-), and CRL 2329 (ER+) breast cancer cells and compares with the expression of these proteins in paired breast tissue samples from AA and CA women by qRT-PCR and Western blotting.

Results: IGF-II expression was significantly higher in AA cell lines and tissue samples when compared to Caucasians. IGF-II siRNA treatment decreased anti-apoptotic protein levels in all cell lines (regardless of ER status). These effects were blocked by the addition of recombinant IGF-II. Of significance, IGF-II expression and regulation of Bcl-X(L) and survivin in cell lines correlated with their expression in paired breast tissues.

Conclusions: IGF-II and the anti-apoptotic proteins differential expression among AA and CA patients may contribute to the breast cancer survival disparities observed between these ethnic groups.

Figure 1

IGF-II, Bcl-2, Bcl-X_L and survivin protein expression in African-American and Caucasian breast cancer cell lines (CRL-2335, CRL-2329 and Hs578t) assessed by Western blot analysis. Figure 1A shows IGF-II Western blot in AA and CA breast cancer cells for comparison purposes. Although the monoclonal IGF-II antibody used (Amano) recognizes both forms (mature and precursor), proIGF-II was the predominant form expressed (11–17 kDa, according to glycosylation levels). Figures 1B–D shows IGF-II siRNA transfection effect on IGF-II levels for each cell line (top panel). Lower panels show Bcl-2, Bcl-X_L and survivin protein expression in IGF-II transfected cells and cells treated with either mature or proIGF-II. β-actin was used as a loading control.

Figure 2

IGF-II gene expression in African-American and Caucasian paired breast tissue samples assessed by Real Time-PCR (RT-PCR). Figure 2 (A–C) shows IGF-II gene expression represented as fold change after normalization using GAPDH as an internal control. Two (2) or more fold change was considered significant (*). Figure 2A shows IGF-II mRNA fold change between AAM and CAM for all samples. Figure 2B shows IGF-II mRNA comparison between AAN and CAN for all samples, while Figure 2C represents IGF-II mRNA fold change for AAN and CAN only in samples from women younger than 45 years of age. AAN=African-American normal tissue, AAM=African-American malignant tissue, CAN=Caucasian normal tissue and CAM=Caucasian malignant tissue. Total number of patients (n) analyzed per group was as follows: AAN= 23, AAM=27, CAN=20 and CAM=24.

Figure 3

Bar graph representation of percentage of total number of samples that were positive for IGF-II, Bcl-2, Bcl-X_L and survivin protein expression from African-American (AA) and Caucasian (CA) paired breast tissue samples. Data was assessed by western blot analysis and presented as percentage of cases that showed positive bands for the corresponding proteins (panels A–D). Total number of patients (n) analyzed per group was as follows: AAN= 23, AAM=27, CAN=20 and CAM=24.

Figure 4

Representative Western blot analyses of IGF-II, Bcl-X_L, Bcl-2 and survivin in paired tissue samples from African-American (AA) and Caucasian (CA) women (<45 and >45 years of age, figures 4A and 4B respectively). The samples were separated into these two age groups based on the higher incidence of breast cancer in AA women younger than 45 years (but not after 45 years). Immunoreactive bands for IGF-II, Bcl-X_L, Bcl-2, survivin and cytokeratin 18 were identified using ECL, scanned by densitometry and normalized to cytokeratin 18.The 17 kDa band represents pro-IGF-II and represents the main form of IGF-II produced by breast epithelial and malignant cells. Cytokeratin 18 was used as an epithelial cell marker (45 kDa). Lower panels (A–D) show bar graphs of IGF-II, Bcl-X_L, Bcl-2 and survivin data normalized to cytokeratin 18 and presented as the mean ± SE of all samples per group. Asterisks indicate values statistically different (*p<0.05). Total number of patients (n) analyzed per group was as follows: AAN= 23, AAM=27, CAN=20 and CAM=24.

Figure 4

Representative Western blot analyses of IGF-II, Bcl-X_L, Bcl-2 and survivin in paired tissue samples from African-American (AA) and Caucasian (CA) women (<45 and >45 years of age, figures 4A and 4B respectively). The samples were separated into these two age groups based on the higher incidence of breast cancer in AA women younger than 45 years (but not after 45 years). Immunoreactive bands for IGF-II, Bcl-X_L, Bcl-2, survivin and cytokeratin 18 were identified using ECL, scanned by densitometry and normalized to cytokeratin 18.The 17 kDa band represents pro-IGF-II and represents the main form of IGF-II produced by breast epithelial and malignant cells. Cytokeratin 18 was used as an epithelial cell marker (45 kDa). Lower panels (A–D) show bar graphs of IGF-II, Bcl-X_L, Bcl-2 and survivin data normalized to cytokeratin 18 and presented as the mean ± SE of all samples per group. Asterisks indicate values statistically different (*p<0.05). Total number of patients (n) analyzed per group was as follows: AAN= 23, AAM=27, CAN=20 and CAM=24.

Figure 5

Immunohistochemistry of IGF-II, Bcl-2, Bcl-X_L and survivin in human normal and malignant breast tissue samples. Malignant AA and CA samples correspond to ER (+) invasive ductal carcinomas (IDC) Bloom and Richardson’s grade III. Panels A–D and E–H corresponds to IGF-II, Bcl-2, Bcl-X_L and survivin immunostaining, in AAN and AAM samples respectively. Panels I–L and M–P corresponds to IGF-II, Bcl-2, Bcl-X_L and survivin immunostaining, in CAN and CAM samples respectively. Original magnifications 20× (insert represent a 40× section). Total number of patients (n) analyzed per group was as follows: AAN= 7, AAM=7, CAN=5 and CAM=5.