Estrogen‐induced Superoxide Production in Coronary Artery Endothelial Cells via Uncoupled Nitric Oxide Synthase in Diabetes

Abstract

Premenopausal women exhibit substantially lower risk for cardiovascular disease (CVD) than men of similar age; however, diabetes mellitus eliminates this sex‐related protection such that diabetic women have nearly double the risk of CVD as diabetic men. In fact, diabetes increases a woman's risk for coronary heart disease (CHD) by approximately 10‐fold. Although estrogens are believed to protect cardiovascular function in premenopausal women, potential effects of diabetes on how estrogen influences coronary arteries are largely unknown. To address this important gap in our knowledge, the present study has investigated the effects of estrogen on human coronary artery endothelial cells (HCAEC) obtained from normal female donors or donors with type 1 (HCAEC‐D1) diabetes. Human umbilical vein endothelial cells (HUVECS) were also employed as non‐diabetic controls. Production of nitric oxide was detected via fluorescence microscopy using 4‐amino‐5‐methylamino‐2′,7′‐difluorescein (DAF‐FM; 2 μM), whereas superoxide production was detected with dihydroethidium (DHE; 20μM). We found that treating control cells with 100nM 17®‐estradiol (17®‐E; 20–30 min) markedly increased DAF fluorescence by 13‐fold (n=42 cells; p<0.0001). In contrast, 17β‐E decreased DAF fluorescence in HCAEC‐D1 by 75% (n=38 cells; p<0.0001). In the presence of 100 μM L‐NAME, an inhibitor of nitric oxide synthase (NOS), 17®‐E had little effect on NO production. As expected, HCAEC‐D1 exhibited more than double the level of DHE fluorescence (n=16 cells; p<0.001) compared to control cells, consistent with diabetes‐induced oxidative stress. Moreover, 17®‐E further increased DHE fluorescence by an average of 20% (n=55 cells; p<0.0001) in HCAEC‐D1, but did not alter DHE fluorescence in control cells. Interestingly, L‐NAME inhibited the effect of 17®‐E on superoxide production in diabetic cells strongly suggesting uncoupled NOS as the primary source of estrogen‐stimulated superoxide. Finally, estrogen‐stimulated oxidant production in HCAEC‐D1 was prevented by “recoupling” NOS activity with 100μM sapropterin, an analog of tetrahydrobiopterin (BH4). Estrogen had no effect on DHE fluorescence in HCAEC‐D1 pretreated with sapropterin (n=16 cells). Taken together, these findings suggest that estrogens stimulate the release of NO from normal coronary artery endothelial cells, but in diabetes the primary effect of estrogen is enhanced superoxide production from these cells. Further, the fact that L‐NAME inhibits both NO and superoxide production in endothelial cells strongly suggests that NOS is the primary target of estrogen in these cells. Lastly, restoration of BH4 levels prevented estrogen from increasing oxidative stress in HCAEC‐D1. Thus, we conclude that diabetes uncouples NOS activity in coronary arteries, likely through depletion of BH4. The higher estrogen levels in premenopausal women would then stimulate NOS to produce predominately superoxide in diabetes to increase oxidative stress. We propose that these abnormal responses to estrogen contribute to the greater risk for CVD experienced by female diabetics. These findings now suggests a novel therapeutic target to alleviate endothelial dysfunction in women afflicted by this disease: recoupling of NOS activity via restoration of BH4 levels (which are known to be depleted by diabetes).Support or Funding InformationSupported by the Biomedical Sciences Program at GA‐PCOM and the PCOM Center for Chronic Disorders of Aging.