Prostacyclin, a pleotropic prostaglandin with multiple atheroprotective effects, is produced from arachidonic acid via the catalytic activity of cyclooxygenase-2 (COX-2). Estrogen is known to increase COX-2 expression in vascular endothelial cells. Here, Egan et al (Science 306:1954, 2004) examine the putative effects of estrogen on the regulation of COX-2 and subsequent production of atheroprotective prostacyclin.
Methods and Results:
LDL receptor knock-out (LDLR KO) mice, a strain known to exhibit an increased burden of atherosclerosis compared to wild-type mice, were studied. Male LDLR KO mice exhibited a greater extent of aortic atherosclerosis at 3 and 6 months of age than female LDLR KO mice. However, female LDLR KO mice lacking the prostacyclin receptor developed significantly greater aortic atherosclerosis compared to female LDLR KO mice with intact prostacyclin receptors. This effect was not observed in male LDLR KO mice lacking the prostacyclin receptor.
Male and female LDLR KO mice lacking prostacyclin receptors exhibited increased platelet activation (measured by the level of urinary excretion of a metabolite of thromboxane), and oxidative stress (measured by the level of urinary excretion of isoprostanes, reflecting lipid peroxidation) compared to LDLR KO mice with functioning prostacyclin receptors.
Cultured mouse aortic smooth muscle cells (MASMCs) exposed to oxidative stress (via H2O2) demonstrated increased prostacyclin synthesis, lipid peroxidation, and production of reactive oxygen species. COX-2 expression was not affected by oxidative stress. MASMCs lacking prostacyclin receptors demonstrated comparatively greater levels of lipid peroxidation and production of reactive oxygen species. Addition of a prostacyclin receptor agonist attenuated lipid peroxidation and production of reactive oxygen species in MASMCs exposed to oxidative stress, all suggesting a role for prostacyclin receptors in mediating the response of aortic smooth muscle cells to oxidative stress. Long-term exposure of MASMCs to estrogen resulted in increased production of COX-2 and prostacyclin, and an attenuated response to oxidative stress. This effect was found to be mediated through estrogen receptor.
Treatment of ovariectomized LDLR KO and wild-type mice with estrogen resulted in increased prostacyclin synthesis and decreased lipid peroxidation. Ovariectomized LDLR KO mice lacking prostacyclin receptors demonstrated increased aortic atherosclerosis compared to LDLR KO mice with functioning prostacyclin receptors. Treatment with estrogen reduced aortic atherosclerosis by 80% in LDLR KO mice with functioning prostacyclin receptors, but only by 32% in LDLR KO mice lacking prostacyclin receptors.
Atheroprotective effects of prostacyclin in female mice are mediated, at least in part, through estrogen induced upregulation of COX-2.
This well designed series of in vitro and in vivo experiments illustrates an important novel association between estrogen and COX-2 in protection from atherosclerosis. In this context, states leading to reduced activity of either COX-2 (via selective inhibition with COX-2 inhibitors) or estrogen (via menopause) might be expected to confer an increased susceptibility to atherosclerosis. In addition, these results suggest that chronic COX-2 inhibition in pre-menopausal women may serve to cancel out or reduce the atheroprotective effects of estrogen in younger women. Though this relationship has not been confirmed in humans, recent high profile evidence of increased cardiovascular events associated with several commercially available COX-2 inhibitors is compelling, though sub-analyses by gender and hormone status have not been made public to date. However, humans are complex, and many pro- and anti-atherogenic mechanisms are likely at play. To add additional complexity, some mediators may play both pro- and anti-atherogenic roles, depending on the context. An example is particularly pertinent in the discussion of the present investigation: Though estrogen would appear to play an unequivocally anti-atherogenic role based on the model presented, evidence from several large epidemiologic studies in women (HERS II and the Womens Health Initiative) have identified no role for estrogen replacement therapy for the primary or secondary prevention of atherosclerosis. In some circumstances, estrogen replacement was associated with a greater risk of cardiac events, perhaps through the promotion of a pro-thrombotic state. As always, extrapolation of data from animal studies requires a certain level of caution. However, a fresh look at the HERS II and Womens Health Initiative data considering concomitant COX-2 inhibitor use might yield evidence to sway opinion on this subject.