Relevance of nuclear IRS-1 in breast cancer

Abstract

Insulin receptor substrate 1 (IRS-1), one of the major molecules transmitting signals from the insulin and insulin-like growth factor 1 receptors, has been implicated in breast cancer. Recently, data obtained in different cell models, suggested that in addition to its conventional role as a cytoplasmic signal transducer, IRS-1 can function in the nuclear compartment. However, the role of nuclear IRS-1 in breast cancer has never been addressed. Experiments undertaken in our laboratory showed that in estrogen receptor α (ERα)-positive MCF-7 cells 1) a fraction of IRS-1 is translocated to the nucleus upon 17-β-estradiol (E2) treatment; 2) E2-dependent nuclear translocation of IRS-1 is blocked with the antiestrogen ICI 182,780; 3) nuclear IRS-1 colocalizes and coprecipitates with ERα; 4) the nuclear IRS- 1:ERα complex is recruited to the E2-sensitive pS2 gene promoter. Furthermore, transfection reporter assays with E2-sensitive promoters suggested that the presence of IRS-1 modulates ERα activity at estrogen response element (ERE)-containing DNA. Furthermore, since the expression of nuclear IRS-1 in breast cancer biopsies has never been examined, we wanted to assess whether nuclear IRS-1 is present in breast cancer and non-cancer mammary epithelium and if it correlates with other markers, especially ERα. Parallel studies were done for cytoplasmic IRS-1. IRS-1 and ERα expression was assessed by immunohistochemistry. Data were evaluated using Pearson correlation, linear regression, and ROC analysis. Nuclear IRS-1 was expressed at low levels in normal mammary epithelial cells and at higher levels in benign tumors, ductal carcinoma, and lobular carcinoma. Similarly, ERα expression was low in normal cells and benign tumors, but high in ductal and lobular cancer. Nuclear IRS-1 and ERα positively correlated in ductal breast cancer and benign tumors, but were not associated in lobular cancer and normal mammary epithelium. In ductal carcinoma, both nuclear IRS-1 and ERα negatively correlated with tumor grade, size, mitotic index, and lymph node involvement. Cytoplasmic IRS-1 was expressed in all specimens and positively correlated with ERα in ductal cancer. A positive association between nuclear IRS- 1 and ERα is a characteristic for ductal breast cancer and marks a more differentiated, nonmetastatic phenotype. In summary, our data suggest the existence of interactions between IRS-1 and ERα occurring in the nucleus. These interactions might represent a novel aspect of ER/IGF-I crosstalk in breast cancer.