Ginseng has been used as a medicine for over two thousand years. Studies have showed that ginseng could relieve menopausal symptoms, bleeding disorders, sleeping disorders, depression, anxiety etc. Ginsenoside Rg1 is the major pharmacologically active compound of ginseng. Our previous studies clearly demonstrated the estrogenic activities of Rg1 in human breast cancer MCF-7 cells. However, it is unclear if Rg1 exert selective estrogenic effects in other cell types in vitro. The present study was designed to determine the selective estrogenic effects of ginsenoside Rg1 by using endometrial, neuroblastoma and preosteoblast cell lines. The endometrial Ishikawa cells, neuroblastoma SK-N-SH cells and preosteoblast MC3T3-E1 cells were cultured in phenol-red free DMEM supplemented with 5% sFBS and were treated with different dosage of Rg1(0.0001-1μM, n=3) and 17β-estradiol (0.01μ M, n=3). MTT method was used to detect the cell viability. The western blot was used to detect the protein expressions of mitogen-activated protein kinase kinase (MEK) and estrogen receptor α (ERα). The activity of estrogen responsive element (ERE) promoter was detected by Dual Luciferase Assay. Treatment of three kinds of cells with various concentrations of Rg1 (0.0001-1μM) had no stimulatory effect on cell number. Rg1 significantly increased estrogen receptor-dependent alkaline phosphatase activity in Ishikawa cells (0.0001μM: 1.34±0.081, 0.01μM: 1.28±0.049, 1μM: 1.22±0.023 vs control 1.00±0.018, P<0.01) but not in MC3T3-E1 cells. Similar to estrogen, Rg1 treatment induced the phosphorylation of MEK (0.0001μM: 1.42±0.075, 0.01μM: 1.42±0.103, 1μM: 1.65±0.117 vs control 1.00±0.030, P<0.05) and ERα (0.0001μM: 1.47±0.049, 0.01μM: 1.72±0.035, 1μM: 2.02±0.100 vs control 1.00±0.029, P<0.05) in a dose-dependent manner in Ishikawa cells. Similar effects were obtained in neuroblastoma SK-N-SH cells. Cells transfected with the ERE-luciferase reporter construct exhibited significantly increased ERE-luciferase activity in the Rg1 presence in Ishikawa cells (0.0001μM: 1.69±0.094, 0.01μM: 1.79±0.108, 1μM: 2.03±0.194 vs control 1.00±0.134, P<0.05) and neuroblastoma SK-N-SH cells (0.01μM: 2.70±0.235 vs control 1.00±0.231, P<0.01), suggesting that the estrogenic effects of Rg1 were mediated through the endogenous ERs. In contrast, Rg1 did not induce any estrogenic responses in preosteoblast MC3T3-E1 cells. Our results suggest that the estrogenic effects of Rg1 are distinct from those of estradiol and are cell type selective. Further study is needed to characterize the action by which Rg1 exerts tissue-selective estrogenic effects in vivo.