Background Since estrogen is known to increase vascular endothelial cell growth, elevated estrogen exposure from hormone replacement therapy or oral contraceptives has the potential to contribute in the development of abnormal proliferative vascular lesions and subsequent thickening of the vasculature. How estrogen may support or promote vascular lesions is not clear. We have examined in this study whether estrogen exposure to vascular endothelial cells increase the formation of reactive oxygen species (ROS), and estrogen-induced ROS is involved in the growth of endothelial cells. Methods The effect of estrogen on the production of intracellular oxidants and the role of estrogen-induced ROS on cell growth was studied in human umbilical vein endothelial cells. ROS were measured by monitoring the oxidation of 2'7'-dichlorofluorescin by spectrofluorometry. Endothelial cell growth was measured by a colorimetric immunoassay based on BrdU incorporation into DNA. Results Physiological concentrations of estrogen (367 fmol and 3.67 pmol) triggered a rapid 2-fold increase in intracellular oxidants in endothelial cells. E2-induced ROS formation was inhibited to basal levels by cotreatment with the mitochondrial inhibitor rotenone (2 μM) and xanthine oxidase inhibitor allopurinol (50 μM). Inhibitors of NAD(P)H oxidase, apocynin and DPI, did not block E2-induced ROS formation. Furthermore, the NOS inhibitor, L-NAME, did not prevent the increase in E2-induced ROS. These findings indicate both mitochondria and xanthine oxidase are the source of ROS in estrogen treated vascular endothelial cells. E2 treated cells showed a 2-fold induction of BrdU incorporation at 18 h which was not observed in cells exposed to vehicle alone. Cotreatment with ebselen (20 μM) and NAC (1 mM) inhibited E2-induced BrdU incorporation without affecting the basal levels of DNA synthesis. The observed inhibitory effect of NAC and ebselen on E2-induced DNA synthesis was also shown to be dose dependent. Conclusion We have shown that estrogen exposure stimulates the rapid production of intracellular ROS and they are involved in growth signaling of endothelial cells. It appears that the early estrogen signaling does not require estrogen receptor genomic signaling because we can inhibit estrogen-induced DNA synthesis by antioxidants. Findings of this study may further expand research defining the underlying mechanism of how estrogen may promote vascular lesions. It also provides important information for the design of new antioxidant-based drugs or new antioxidant gene therapy to protect the cardiovascular health of individuals sensitive to estrogen.