INTRODUCTION

TOP ABSTRACT INTRODUCTION CELLULAR AND BIOCHEMICAL... VASOPROTECTIVE EFFECTS OF... ENDOGENOUS ESTROGEN: BIOLOGICAL... VASCULAR PROTECTIVE EFFECTS OF... PROTECTIVE EFFECTS OF ESTRADIOL... CONCLUSION REFERENCES

PREMENOPAUSAL WOMEN HAVE A decreased incidence of cardiovascular disease and a decreased rate of progression of renal disease. With the onset of menopause, however, decreased synthesis of 17-estradiol (estradiol) is accompanied by an increased incidence of cardiovascular disorders and accelerated progression of renal diseases.

The glomerulus and the vascular wall are not static, and components of these structures dynamically increase, decrease, or reorganize in response to physiological and pathological stimuli. Although multiple cellular and biochemical processes are involved in glomerular and vascular remodeling, glomerular mesangial cells (GMCs) in the kidney and smooth muscle cells (SMCs) in the vasculature are the final common pathway for dynamic changes in glomerular and vascular wall structure. In glomerular and vascular remodeling, GMCs and SMCs undergo one or more of four basic processes: cell growth involving hypertrophy or hyperplasia; cell migration involving the immigration of GMCs or SMCs from one locale to another in the glomerular tuft or vascular wall, respectively; modulation of the amount and types of extracellular matrix (ECM); and apoptosis, which provides an important means of population control for GMCs and SMCs. Glomerular and vascular endothelial cells also play a critical role in maintaining homeostasis by generating a battery of both growth-inhibitory and growth-stimulatory factors, as well as relaxing and contracting factors. Consequently, endothelial damage or dysfunction often leads to increased SMC and GMC migration, proliferation, and ECM synthesis.

Estradiol may induce protective effects on the renal and cardiovascular system by altering SMC, GMC, or endothelial cell biology so as to prevent glomerular and vascular remodeling, and the main purpose of this review is to provide an overview of the participating mechanisms in this regard. A prerequisite for comprehending these mechanisms is an understanding of the processes by which estradiol induces its cellular, biochemical, and molecular effects.

	CELLULAR AND BIOCHEMICAL EFFECTS OF ESTRADIOL

TOP ABSTRACT INTRODUCTION CELLULAR AND BIOCHEMICAL... VASOPROTECTIVE EFFECTS OF... ENDOGENOUS ESTROGEN: BIOLOGICAL... VASCULAR PROTECTIVE EFFECTS OF... PROTECTIVE EFFECTS OF ESTRADIOL... CONCLUSION REFERENCES

Estradiol influences cell growth and differentiation of the male and female reproductive tissues. For example, estradiol regulates the development of mammary glands, uterus, vagina, ovary, testes, epididymis, and prostate (130) and also plays an important role in the vascular system, a system that is essential to reproductive processes. (107, 130, 173, 243). Findings in the last decade indicate that estradiol induces its biological effects via genomic and nongenomic mechanisms and that the effects of estradiol are triggered by estrogen receptor-dependent as well as estrogen receptor-independent mechanisms.

Role of Estrogen Receptors

Changes in the generation of intracellular proteins triggers a cascade of events that influences metabolic processes and translates into cell growth and differentiation. It is well documented that estradiol induces uterine growth (hypertrophy and hyperplasia) and the expression of protooncogenes, which may play a role in estradiol-induced cell growth and proliferation (230). In uterine cells, treatment with estradiol rapidly increases the steady-state expression of N-myc, c-myc, c-ras, and c-fos mRNA (230). In addition, ERs increase the synthesis of growth factors such as epidermal growth factor (EGF) and insulin-like growth factor (IGF) and stimulate the levels of growth-promoting peptides that can act in an autocrine fashion to induce cell growth (see reviews in Refs. 230 and 237).

In MCF-7 cell lines, estradiol increases EGF-induced activator protein-1 (AP-1), suggesting that there may be signaling connections between ERs and EGF-induced AP-1 activity (216). Thus estradiol not only induces growth factor synthesis and growth factor receptor synthesis (230) but also facilitates growth factor receptor signal transduction mechanisms. Estradiol, by inducing c-fos, synergies with low concentrations of insulin, which is able to induce c-jun but not c-fos (282). Even when c-fos and c-jun synthesis and AP-1 activity are maximally induced by growth factors, estradiol can still enhance AP-1-dependent transcriptional activity (248). ER mutants lacking the DNA binding domain interact with fos and jun to increase transcription (75). This demonstrates that the ER, a classic DNA binding protein, can act through protein-protein interactions without binding directly to DNA. Consequently, anything that alters the ER to fos or jun ratio will alter the rate of transcription of specific target genes. This observation provides the basis for crosstalk between multiple pathways. Via the above mechanism the AP-1 DNA binding sequence in the chicken ovalbumin gene promoter can be a target for both estradiol activation and cooperation for AP-1 (75). Moreover, in vivo studies in ovariectomized rats show that EGF enhances the nuclear localization of ERs, suggesting that EGF activates the ER to bind to ERE sequences (108).

Recent studies by Kuiper et al. (134) demonstrate that, in addition to classic ER receptors cloned more than a decade ago (84) and now classified as ER-, cells from rats, mice, and humans also express another ER, termed ER- (59, 134, 178, 277). Rat ER- cDNA encodes a protein of 485 amino acid residues with a molecular mass of 54,200. In the DNA binding domain, this ER protein is highly homologous to rat ER-, with 95% amino acid identity, whereas in the COOH-terminal ligand binding domain, it has 55% homology. Similar homologies between ER- and ER- have been found in ERs from mice (215, 277) and humans (59). ER- is expressed in prostate, ovary, epididymis, testis, bladder, uterus, kidney, lung, thymus, colon, small intestine, blood vessels, pituitary, hypothalamus, cerebellum, and brain cortex (59, 107, 133, 215, 277).

Whether ER- and ER- play a similar or different role in mediating the physiological effects of estradiol is unclear. However, differential expression of ER- and ER- is observed in some tissues (131, 133, 134), which suggests different physiological roles for these receptors. In this regard, compared with ER-, high amounts of ER- mRNA are in fetal ovaries, testes, adrenals, and spleen of the midgestational human fetus (24). Moreover, differential activation by liganded ER- vs. ER- occurs at the AP-1 site (208), and xenoestrogens differentially activate EREs when liganded to ER- vs. ER- (213).

In addition to the classic ERs, another ER, termed type II ER, exists (160). Some ligands for type II ER, such as bioflavonoids, have no affinity for ER- or ER- yet abrogate the effects of estradiol on cell growth (161), suggesting that, in addition to ER- and ER-, the biological effects of estradiol may be modulated via type II ER. The type II ER may also be involved in inducing the effects of estradiol in the vasculature and the kidney.

Apart from the cytosolic/nuclear ERs, estradiol also binds with high affinity to membrane fractions prepared from isolated pituitary and uterine cells (209, 217, 230). The functional role of the membrane receptors for estradiol is evident from the findings that estradiol stimulates adenylyl cyclase activity in membranes prepared from secretory human endometrium (230). Moreover, estradiol induces rapid changes in intracellular calcium levels/flux, K⁺ conductance, and cAMP levels (9, 183, 217).

Role of Estradiol Metabolism

Estradiol is largely metabolized within the liver via oxidative metabolism (to form hydroxylated metabolites such as 2- and 4-hydroxyestradiol) (164); glucuronidation (to form glucuronide conjugates) (307); sulfatase action (to form sulfates) (307); esterase action (to form fatty acid esters) (307); and O-methylation of catechol estradiols (to form O-methylated catechols; 12; for details, see reviews in Refs. 230 and 307).

Even though estradiol is largely metabolized within the liver, cells in several other tissues, including the kidney and the vasculature, contain CYP450 and metabolize estradiol and its metabolites (307). The metabolism of estradiol locally within a tissue may be of immense importance in mediating several of its physiological as well as pathophysiological effects. Several studies provide evidence that the catechol estradiols can induce biological effects, and 2-hydroxyestradiol is a weak ligand for ER and may regulate multiple mechanisms in reproductive tissues, including growth of cancer cells and generation of prostaglandins in the uterus during pregnancy (12, 230, 307). 2-Hydroxyestradiol attenuates catabolism of catecholamines by inhibiting catechol O-methyltransferase activity, and this may modulate the neurophysiological and pharmacological effects of catecholamines within the kidneys and vasculature (12, 230, 307). 2-Hydroxyestradiol also modulates the interaction of dopamine with its receptors (230, 307). Additionally, 2-hydroxyestradiol is a potent antioxidant and thereby protects membrane phospholipids and cells against peroxidation (56).

Similar to 2-hydroxyestradiol, 4-hydroxyestradiol induces several important biological effects. Even though it is not the dominant metabolite formed by the liver, 4-hydroxyestradiol is a major metabolite formed in some extrahepatic tissues such as rat pituitary and human myometrial, myoma, and breast tissue, as well as kidney and vasculature tissue (307). In contrast to estradiol, 4-hydroxyestradiol binds with low affinity to ER; however, its dissociation rate from the receptor is much lower than that observed for estradiol (230). Within the renal system, 4-hydroxyestradiol has been shown to stimulate tumor growth in Syrian hamsters (307) and uterotrophic effects in rats (164, 225). 4-Hydroxyestradiol is more efficacious than estradiol in inducing progesterone receptor expression in the rat pituitary (230). Similar to 2-hydroxyestradiol, 4-hydroxyestradiol also acts as a cooxidant and increases the formation of prostaglandins from arachidonic acid within the uterus during pregnancy (230). 4-Hydroxyestradiol prevents inactivation of catecholamines by inhibiting catechol O-methyltransferase activity and thereby regulates the neuropsychological/pharmacological effects of catecholamines on the central nervous system (230, 307).

In contrast to 2-hydroxyestradiol, 4-hydroxyestradiol induces carcinogenic effects (230, 307). In fact, recent studies provide evidence for reduced 2-hydroxylation and increased 4-hydroxylation of estradiol in subjects with cancer, suggesting that 2-hydroxyestradiol or its methylated metabolite (2-methoxyestradiol) may be anticarcinogenic, whereas 4-hydroxyestradiol and its metabolite 4-methoxyestradiol may be carcinogenic (148, 230, 307). Because abnormal growth of cells is a hallmark for both atherosclerosis, glomerulosclerosis, and cancer, it is feasible that the catechol metabolites of estradiol may also play an important role in regulating cell growth within the vessel wall and the glomeruli.

Role of Binding Proteins

Some metabolites of estradiol, e.g., 2- and 4-methoxyestradiol, have low binding affinity to classic ERs but have a higher binding affinity to SHBG than estradiol (230, 307). This suggests that the methoxy metabolites of estradiol may have a longer half-life and may be present within the circulation at much higher levels.

Role of Aryl Hydrocarbon Receptors in Regulating the Biological Effects of Estradiol

The inactive cytosolic AhR is found complexed with two heat shock protein 90 (hsp90) molecules and a 43-kDa protein (p43). Binding of halogenated aromatic hydrocarbons and environmental estrogens to the AhR initiates disassociation of the hsp90 and p43 molecules and formation of an AhR-ARNT dimer (231). The liganded AhR-ARNT complex is active, translocates to the nucleus, binds to DNA at xenobiotic regulatory elements, and induces the expression of several genes including CYP450 (231, 242). Because CYP450 is involved in estradiol metabolism, the activation of AhR results in increased metabolism of estradiol and influences its biological activity. Moreover, the profile of estradiol metabolites is influenced depending on the types of CYP450 isozymes induced. In MCF-7 cell lines, ligand-activated AhR induces CYP1A1 activity and estradiol metabolism via increases in oxidative metabolism by activating 2- and 4-hydroxylation as well as the 15- and 16-hydroxylases (230), suggesting that AhR may influence the biological effects of estradiol by increasing its metabolism.

Crosstalk between ERs and AhRs may also play a role in modulating the effects of estradiol (230, 237). In both MCF-7 and T47D cell lines, ligand-induced activation of AhR inhibits several ER-induced responses. In this regard, AhR decreases ER-induced secretion/production of tissue plasminogen activator (230, 237), postconfluent focus proteins (230, 237), 52- and 160-kDa proteins (237), cathepsin D mRNA, cathepsin D protein (237), pS2 mRNA (237), progesterone receptors and progesterone receptor mRNA levels (230). Moreover, AhR activation alters ER-induced changes in cell proliferation (237) and glycolysis (230, 237). Also, AhR activation downregulates ER mRNA and protein. Safe (237) proposes additional mechanisms via which ligand-activated AhR may induce antiestrogenic effects including interactions between AhR and the ER-induced pathways via generation of intermediary metabolites; direct interactions between the nuclear AhR and the cis-acting genomic sequences in the promoter regions of ER-regulated or growth factor-regulated genes; and induction of trans-acting factors or proteins that facilitate degradation of the nuclear ER (for details, see reviews in Refs. 230, 231, 237).

	VASOPROTECTIVE EFFECTS OF ESTRADIOL: ROLE OF GENOMIC VS. NONGENOMIC AND RECEPTOR VS. NONRECEPTOR MECHANISMS

TOP ABSTRACT INTRODUCTION CELLULAR AND BIOCHEMICAL... VASOPROTECTIVE EFFECTS OF... ENDOGENOUS ESTROGEN: BIOLOGICAL... VASCULAR PROTECTIVE EFFECTS OF... PROTECTIVE EFFECTS OF ESTRADIOL... CONCLUSION REFERENCES

Estradiol most likely induces vasoprotective effects; however, the mechanisms by which these effects are induced remain unclear. Alterations in plasma concentrations of lipoproteins [decrease in low-density lipoprotein (LDL) levels, decrease in oxidized LDL formation, increase in high-density lipoprotein (HDL) levels], hemostatic factors, glucose, insulin, and endothelium-derived factors (decrease in endothelin, increase in nitric oxide and prostaglandins) and inhibition of smooth muscle cell migration and proliferation induced by various mitogens [i.e., platelet-derived growth factor (PDGF), basic fibroblast growth factor (bFGF), insulin-like growth factor I (IGF-I), and ANG II] are hypothesized to contribute to the vasoprotective effects of estradiol (62, 67, 166, 203, 229). Regardless of the mechanisms, the two most important effects of estradiol in the cardiovascular system are modulation of vascular tone and inhibition of vascular growth.

Effects on Vascular Tone

The acute vasodilatory effect of estradiol is largely mediated via generation of nitric oxide (NO) because the vasodilatory effect of estradiol is blocked by NO synthesis inhibitors (31, 35, 139). Estradiol also enhances NO synthesis in endothelium denuded rat aorta (20). Evidence for the role of ERs in mediating the effect of estradiol on NO synthesis comes from the findings that the vasodilatory effect of estradiol and the effect of estradiol on NO synthesis by endothelial cells are blocked by ICI-182780 (31, 35), an ER antagonist. With regard to the role of ER- and ER- in regulating NO synthesis, studies show that compared with wild-type mice, estradiol-induced NO synthesis and relaxation of the aorta are reduced in mice lacking ER- (232). Moreover, increased endothelial-derived nitric oxide synthase (eNOS) activation in response to estradiol is observed in endothelial cells overexpressing ER- (35). Additionally, a significant association between the number of ERs and basal release of NO is observed in ER--knockout mice (232); however, there was no correlation between NO release and estradiol levels. These findings suggest that the effects on NO synthesis are ER- mediated. The fact that these effects are triggered within 5 min suggests that the effects are nongenomic (262) and that NO plays a major role in mediating the acute vasodilatory effects of estradiol. This notion is further supported by the observation that estradiol-induced activation of eNOS is not abrogated by actinomycin D (35).

The mechanisms by which estradiol induces the rapid increase in NO release from endothelial cells are unclear and under debate. Recent findings suggest the involvement of calcium-dependent translocation of eNOS from cell membrane to the nucleus (81). NO release by estradiol may involve a plasma membrane rather than a nuclear ER (123). Plasmalemmal caveolae possess ER (123), eNOS, and caveolin; plasma membrane-impermeable BSA-conjugated estradiol stimulates NO release; and the effects of estradiol on NO release are blocked by agents that decrease intracellular calcium and by the ER antagonist ICI-182780 (123). Recent studies demonstrate ER- in endothelial cell membranes, and this membrane receptor induces acute effects on NO synthesis (235). Although, the above findings and vast majority of studies suggest a role of intracellular calcium in regulating the rapid release of NO in response to estradiol, this conclusion is not supported by one study (31). Other mechanisms that may be involved are activation of tyrosine kinase and mitogen-activated protein kinase pathways because inhibitors for these pathways block the effects of estradiol on NO production (35, 168). It is proposed that the nongenomic stimulation of NO synthesis may involve intracellular proteins, such as hsp90, which is known to bind to and activate eNOS and may interact with ERs (168). Estradiol upregulates the expression of hsp25 in endothelial cells (45). Alternatively, the antioxidant effects of estradiol may potentiate the activity of NO released under basal conditions. In this regard, estradiol increases rat aorta endothelium-dependent relaxing factor (EDRF) activity in the absence of changes in eNOS gene expression and this increase in EDRF is associated with decreased generation of O in response to estradiol and could account for the enhanced EDRF-NO bioactivity and decreased peroxynitrite release (13, 55)

The role of NO in mediating the acute, nongenomic vasodilatory effects of estradiol is well established. Even so, other mechanisms may also mediate estradiol-induced vasodilation. In this regard, membrane fluidity and ion-channel activity (e.g., Maxi-K and voltage-dependent L-type calcium channels) may play an important role. Application of estradiol stimulates rapid discharge of transient outward currents and induces an increase in the intracellular free calcium concentration in endothelial cells (234). Estradiol also causes coronary vasodilation by opening calcium-activated K⁺ channels (200) and relaxes endothelium-denuded porcine coronary arteries by opening large-conductance (BK_Ca), calcium-activated, and voltage-activated K⁺ channels via NO and cGMP (44). Estradiol inhibits voltage-dependent L-type calcium currents in SMCs (186) and induces potent stimulatory effects on large-conductance, calcium-activated, and voltage-activated K⁺ channels in coronary artery SMCs. Because the vasodilatory effects of estradiol in intact arteries is abrogated by BK_Ca channel blockers and inhibitors of cGMP-dependent protein kinase, it is feasible that estradiol induces its vasodilatory effects by opening BK_Ca channels via NO and cGMP-dependent pathways (295, 297).

Although the role of ERs in mediating the effects of estradiol on ion channels has not been intensively investigated, findings from some studies provide evidence that they may not participate in the process. The inhibitory effects of supraphysiological concentrations (>100 nM) of estradiol on L-type calcium channels are mimicked by the ER antagonist ICI-182780 (233). Moreover, in endothelium-denuded vessels, attenuation of high-K⁺-induced force development and myosin light chain phosphorylation are not blocked by an ER antagonist and are mimicked by estradiol analogs with negligible affinity for ERs (126). Interestingly, recent studies demonstrate that at concentrations of >100 nM (pharmacological concentration) estradiol binds to the -subunit of Maxi-K channels in vascular SMCs and induces Maxi-K channel activation. This finding provides the first evidence that the direct interaction of estradiol with a voltage-gated channel subunit may be responsible for the direct and acute relaxing effects of estradiol on SMCs (280).

Estradiol rapidly activates adenylyl cyclase activity and increases cAMP production (52, 61). Moreover, via the cAMP-adenosine pathway, estradiol induces the production of adenosine in vascular SMCs (52). The fact that the effects of estradiol on cAMP and adenosine synthesis are blocked by an ER antagonist, but not by cyclohexamide, suggests that these effects are mediated via nongenomic, but ER-linked, mechanisms. Because adenylyl cyclase is a membrane-bound enzyme, it is feasible that estradiol may directly interact with adenylyl cyclase. Alternatively, activation of ion channels (as discussed above) may be responsible for the stimulatory effects of estradiol on adenylyl cyclase. Additional studies are required to elucidate the mechanisms by which estradiol induces cAMP and adenosine synthesis.

Although the above findings provide evidence that estradiol induces cAMP production, whether these effects are induced by physiological concentrations of estradiol remains unresolved. In this regard, some studies (37) show that physiological concentrations of estradiol are unable to induce significant increases in cAMP production. However, studies from our laboratory provide evidence that the effects of physiological concentrations of estradiol are significantly blocked by the adenylyl cyclase inhibitor dideoxyadenosine, suggesting that physiological concentrations of estradiol are capable of stimulating cAMP production (52).

Finally, in addition to the above pathways, other mechanisms may also be responsible for the acute nongenomic effects of estradiol on the vasculature. For instance, because estradiol is a phenol, it may induce antioxidant effects and protect the vasculature from free radical-induced deleterious effects.

The rapid-onset, nongenomic effects of estradiol may play an important role in regulating vascular tone; however, the long-term genomic effects of estradiol also importantly contribute to the vascular protective effects of estradiol. Long-term treatment with estradiol induces eNOS (294) and abrogates vasoconstrictor effects on vascular tissues (39, 122, 232). Compared with premenopausal women, NO synthesis is decreased in postmenopausal women and is normalized by estradiol replacement therapy (110, 228). This effect of estradiol replacement therapy, however, is diminished by coadministration of synthetic progestin, medroxyprogesterone, and cyproterone acetate (110). Vascular NO synthesis is decreased in mice lacking ER- (232), and long-term administration of estradiol increases acetylcholine-mediated coronary vasodilation in nonhuman primates (298, 299), male-to-female transsexuals (197), postmenopausal women (95), and in postmenopausal women with angina and normal coronary arteries (224). The role of genomic effects of estradiol on vascular tone is further supported by the recent observation that impaired endothelium-dependent vasorelaxation (267) and early coronary calcification (266) are observed in a subject lacking functional ER-. However, the vasodilatory response to estradiol was adequate, suggesting a nongenomic response and that a lack of ER- may not be as deleterious as previously hypothesized (266, 267). Indeed, estradiol prevents neointima formation in ER--knockout mice (107). Estradiol also increases the synthesis of prostacyclin by inducing the expression of prostacyclin synthase (33, 169) and cyclooxygenase (115).

Estradiol reduces blood pressure in various animal models (26, 43, 142, 249), and the modulatory effects of estradiol on vascular tone may be responsible for estradiol-induced effects on blood pressure. Multiple clinical studies show that both long-term and short-term administration of estradiol replacement therapy is associated with either the lowering of blood pressure (92) or blood pressure-neutral effects (275) in most postmenopausal women. However, estradiol replacement therapy infrequently increases blood pressure in postmenopausal women (152). The blood pressure-lowering effect of estradiol may be mediated by direct effects of estradiol on ion channel activity or increased synthesis of vasodilatory substances such as NO, cGMP, cAMP, adenosine, and prostacyclin. Alternatively, estradiol may lower blood pressure by reducing the synthesis of ANG II and endothelin-1 (ET-1) or by interfering with the synthesis of and decreasing the plasma levels of catecholamines (150, 212, 244).

Effects on Vascular Growth

Interactions with SMCs. Abnormal activity of SMCs is one of the key processes responsible for vascular pathology. In this regard, estradiol inhibits SMC proliferation, migration and extracellular matrix (collagen) synthesis induced by serum, PDGF, ANG II, ET-1, fibronectin, free radicals, IGF-1, bFGF, oxidized-LDL, and mechanical pulsatile stretch (122, 132, 156, 203, 205, 229).

Interactions with endothelial cell, macrophages, and monocytes. Interactions of estradiol with endothelial cells, macrophages, and monocytes may contribute to the vasoprotective effects of estradiol. Under normal circumstances, the endothelium is thought to induce a net inhibitory effect on SMC growth (55), and damage or dysfunction of the endothelium by balloon catheters, vascular cuffs, and immune reactions results in abnormal proliferation of SMCs and neointima formation, effects that are abrogated by estradiol (23, 34, 36, 65, 66, 68, 88, 107, 144, 153, 179, 204, 238, 270).

Modulation of circulating growth factors. Estradiol also induces antivasoocclusive effects by modulating the synthesis of circulating factors that are mitogenic for SMCs and damaging to ECs. In this regard, estradiol downregulates the expression of angiotensin-converting enzyme (ACE) in serum as well as in the aorta and reduces ANG II formation (71). ACE expression is increased at sites of balloon injury and angioplasty (63), ANG II is a potent mitogen for SMCs (49, 54), and ANG II is known to induce vascular remodeling processes associated with cardiovascular disease. The finding that estradiol downregulates ACE and reduces ANG II synthesis suggests that estradiol may abrogate SMC growth in part by decreasing ANG II biosynthesis. Estradiol replacement therapy decreases ANG II levels and ACE activity in postmenopausal women (218) and suppresses renin levels (244). Estradiol also downregulates the expression of AT₁ receptors in SMCs (198). Because these receptors mediate the mitogenic effects of ANG II, estradiol may abrogate the effects of ANG II on growth. Indeed, our studies show that estradiol inhibits ANG II-induced growth of human SMCs in vitro (229). Additionally, estradiol induces the synthesis of ANG 1-7, a vasodilator and SMC growth inhibitor (55).

Antioxidant effects. The antioxidant effects of estradiol and its metabolites may play a critical role in the antivasoocclusive effects of estradiol. Indeed, physiological and supraphysiological concentrations of estradiol induce antioxidant activity in vitro and ex vivo/in vivo, respectively (245, 252), and estradiol inhibits oxidized-LDL-induced endothelial damage (191) and superoxide anion (30)- and cholesterol-induced SMC growth (99). We recently demonstrated that physiological concentrations of 2-hydroxyestradiol, a prominent estradiol metabolite, is a potent antioxidant that prevents peroxyl radical-induced peroxidation of vascular SMC membrane phospholipids and inhibits peroxyl radical-induced proliferation and migration of SMCs (56). 2-Hydroxyestradiol prevents the oxidation of acidic membrane phospholipids (phosphatidylinositol and phosphatidylserine), which are known to activate PKC activity and play a critical role in regulating SMC growth (56). Makides et al. (162) demonstrate that methoxy metabolites of estradiol are also potent antioxidants. Because SMCs rapidly convert 2-hydroxyestradiol to 2-methoxyestradiols (53a), it is feasible that the antioxidant effects of 2-hydroxyestradiol on membrane lipids are mediated by methoxyestradiol. Estradiol also reduces glycooxidative damage in the artery wall (288), suggesting that the antioxidant effects of estradiol and its metabolites at the cell membrane level may play a critical role in mediating the growth regulatory and antimitogenic effects of estradiol. Finally however, even though multiple in vitro and ex vivo studies support the notion that estrogen increases the resistance of LDL to oxidation, ample studies in postmenopausal women receiving estrogen show no difference (see review in Ref. 245). Taken together, the evidence for in vivo antioxidant effects of physiological concentrations of estrogen remains controversial.

Interaction with lipids. Estradiol influences the vascular effects of LDL cholesterol. Estradiol, which is a phenol with antioxidant properties, prevents the oxidation of LDL and very-low-density lipoprotein (VLDL) to oxidized LDL and oxidized VLDL, both in vivo and in vitro (163, 223, 236) and protects the vasculature against the deleterious effects of oxidized lipids. Estradiol also prevents compromise of the endothelial barrier mediated by LDL and attenuates the accumulation of LDL and oxidized LDL in the artery wall (73). Also, TNF--mediated oxidation and accumulation of LDL in the artery wall are prevented by estradiol (288). Moreover, estradiol increases the catabolism of LDL, LDL apolipoprotein (B), and -VLDL via LDL receptor-dependent and -independent mechanisms (41, 46). In this regard, estradiol has been shown to increase the expression of LDL receptors (46). Pharmacological levels of estradiol increase hepatic mRNA for the LDL receptor (18, 96, 259) and increase the synthesis of LDL receptor protein out of proportion to the increase in hepatic LDL receptor mRNA, indicating both transcriptional and posttranscriptional regulation of the LDL receptor by estradiol (18). Estradiol also improves the clearance of VLDL, decreases LDL production (302), reduces LDL particle size, increases the clearance of both the light and dense LDL (27), increases expression of VLDL and LDL receptors in the left ventricles of the heart (48, 165), and induces sterol-27-hydroxylase activity, which decreases LDL production (136). Estradiol-induced removal of VLDL is associated with increased activities of hepatic lipase, lipoprotein lipase, and expression of LDL receptors (46). There is also evidence for crosstalk between ERs and LDL receptors and for upregulation of ER associated with an increase in LDL receptor expression, and these effects can be blocked by the ER antagonist tamoxifen (210).

Evidence for non-receptor-mediated antimitogenic effects of estradiol. Vascular ECs and SMCs express the functional ERs, ER- and ER- (98, 262). However, whether the cardiovascular effects of estradiol are ER mediated is still unresolved and under debate. Recent studies by Isfrati et al. (107) show that estradiol induces vasoprotective effects in mice lacking ER-, suggesting that the effects may be mediated via ER-. The notion that ER- mediates antivasoocclusive effects of estradiol is further supported by the finding that, similar to estradiol, low concentrations (25 µg · kg¹ · day¹) of genistein, a phytoestrogen that binds exclusively to ER- and has negligible affinity for ER- binding (135), inhibits balloon injury-induced neointima formation (157). However, this study does not provide any evidence regarding whether the effects of genistein are blocked by ER antagonists such as ICI-182780; moreover, inhibitory effects of genistein on SMC growth are observed at concentrations of >2.5 µM. Hence it is feasible that the effects of genistein are mediated via some alternative pathway not involving ERs. Indeed, at micromolar concentrations genistein is a tyrosine kinase inhibitor and inhibits growth of cancer cells containing or lacking ERs (292). Moreover, our previous study demonstrates that the inhibitory effects of genistein on human aortic SMCs are not reversed by ICI-182780 (50). Also, Karas et al. (118) provide evidence that estradiol protects against vascular injury in mice in which ER- is genetically disrupted. Together, these finding provide evidence that estradiol can induce vasoprotective effects via mechanisms independent of ERs. However, the possibility that the ER- may compensate for ER- in ER--knockout mice, and vice versa, remains an unresolved issue. In a recent study, Bakir et al. (11) demonstrated that the inhibitory effects of estradiol on neointima formation are blocked by ICI-1827980, suggesting that the effects are ER mediated. However, the authors also observed that concentrations of ICI-182780 that blocked the effects of estradiol also induced circulating levels of estradiol (11). Because estradiol metabolites are more potent than estradiol in inhibiting SMC growth and ICI-182780 inhibits estradiol metabolism (53a), it is feasible that the effects of ICI-182780 may be due to inhibition of metabolism of estradiol and not to the antagonistic effects on ERs.

	ENDOGENOUS ESTROGEN: BIOLOGICAL POTENCY AND PATHOPHYSIOLOGICAL IMPORTANCE

TOP ABSTRACT INTRODUCTION CELLULAR AND BIOCHEMICAL... VASOPROTECTIVE EFFECTS OF... ENDOGENOUS ESTROGEN: BIOLOGICAL... VASCULAR PROTECTIVE EFFECTS OF... PROTECTIVE EFFECTS OF ESTRADIOL... CONCLUSION REFERENCES

In addition to estradiol, numerous other estrogens are present in vivo and are known to possess biological activities that are similar to estradiol as well as activity that is opposite to estradiol. Hence, the effects of estradiol may not reflect the effects of all estrogens. Conversely, the effects of other estrogens may not reflect the biological effects of estradiol. Our recent studies show that the inhibitory effects of various estrogens on mitogen-induced SMC growth and MAP kinase activity vary considerably. Estrone and estriol, which are present in large amounts endogenously, are not effective in inhibiting SMC growth and MAP kinase activity. In fact, they significantly abrogate the inhibitory effects of estradiol (49), suggesting that inactive estrogens may block the biological effects of estradiol. Indeed, this may explain in part why the use of conjugated estrogens apparently do not reduce cardiovascular risk in postmenopausal women (104). In contrast to estrone and estriol, metabolites of estradiol such as 2-hydroxyestradiol, 2-methoxyestradiol, and 4-methoxyestradiol, which show minimal binding to ER, are more potent inhibitors of mitogen-induced SMC growth compared with estradiol.

Apart from the differential potency of various estrogens on ERs, they also have significant differences in their chemical properties. In this regard, the antioxidant potency of estrone and estriol is much less than estradiol, whereas the catechol estradiols are more potent antioxidants than estradiol. Together, these findings suggest that the biological effects of estrogens may vary considerably.

The above findings may have clinical implications as several different estrogens are used for hormone replacement therapy. Indeed, recent in vitro studies from our laboratory provide evidence that clinically used estrogens differentially inhibit SMC growth and MAP kinase activity (49). Because direct interactions of estrogens with SMCs may be responsible for the protective effects of estrogens on the vasculature, it is feasible that estrogens unable to inhibit SMC growth would be inactive against neointima formation or vasoocclusive disorders.

Another important issue is what concentrations of estradiol are relevant to physiology. Although most studies use circulating levels of estradiol (10⁹ mM) as a physiological concentration, under in vivo conditions estradiol may exist in several biologically active forms. Thus the levels of estradiol in the plasma may not reflect the true physiological effects of estradiol. For example, the concentrations of 2-hydroxyestradiol range between 0.12 and 0.3 µM in peripheral blood, and the rate of urinary excretion of 2-hydroxyestradiol is 20-180 µg/24 h in urine (6, 78). Finally, the recent finding that vascular SMCs express aromatase activity and synthesize estradiol (96) suggests that the local levels of estradiol may be higher.

	VASCULAR PROTECTIVE EFFECTS OF ESTRADIOL: EPIDEMIOLOGICAL EVIDENCE

TOP ABSTRACT INTRODUCTION CELLULAR AND BIOCHEMICAL... VASOPROTECTIVE EFFECTS OF... ENDOGENOUS ESTROGEN: BIOLOGICAL... VASCULAR PROTECTIVE EFFECTS OF... PROTECTIVE EFFECTS OF ESTRADIOL... CONCLUSION REFERENCES

The evidence for a role of estradiol in regulating vascular structure and function is that ovarian dysfunction and estradiol deficiency are linked to an increased risk of cardiovascular disease in postmenopausal women. Thus compared with men, women within the reproductive age group are protected against cardiovascular disease (143, 203, 296), and these differences decrease with the onset of menopause (143). Also, premenopausal women who undergo premature surgical menopause (i.e., bilateral oophorectomy) have twice the risk of cardiovascular disease than do age-matched premenopausal controls (103, 225).

Because the ovaries are the main source of estradiol production, their dysfunction with age results in a decreased production of estradiol. In this regard, it is important to note that the levels of estradiol, which is the most potent natural estrogen and is largely produced in the ovary, decrease to undetectable levels in postmenopausal women. Compared with estradiol, the levels of other estrogens such as estrone, which can be synthesized in the peripheral compartment, do not fall as dramatically. This suggests that the beneficial effects of ovarian steroids on the cardiovascular system may largely be mediated by estradiol.

Several epidemiological studies demonstrate a reduction in risk of coronary heart disease in postmenopausal women treated with oral estrogen compared with untreated women (25, 83, 203). Three meta-analyses estimate an overall reduction in risk of ~50% for women who had at any time received estrogen therapy. Four studies examine the association between estrogen use and angiographically defined coronary artery disease in postmenopausal women and demonstrate that the calculated odds ratios for severe coronary stenosis or 70% occlusion are 0.50 or lower for estrogen users compared with nonusers. A reduction in mortality among women with diagnosed coronary atherosclerosis is reported in relation to estrogen use (101, 203, 261, 269). Although, estrogen replacement therapy may reduce vasoocclusive disorders in postmenopausal women, hormone replacement therapy using premarin (conjugated estrogen) plus medroxyprogesterone (the HERS study) does not reduce the overall incidence of coronary artery disease in postmenopausal women with established coronary heart disease (104). Several lines of evidence suggest that unlike progesterone, synthetic progestin such as medroxyprogesterone can abrogate the protective effects estradiol on vascular cells. In this regard medroxyprogesterone attenuates the inhibitory effects of estradiol on neointima formation (2, 144) and on NO synthesis (110). Hence it is possible that the lack of effects of estrogen replacement therapy in the HERS study is due to the negative effects of medroxyprogesterone. Alternatively, the lack of effects could also be due to the type of estrogen used. In this regard, our studies show that estrone, estriol, and estrone sulfate, which constitute a major portion of the conjugated estrogen, have little or no inhibitory effects on mitogen-induced SMC growth (53). Because direct interaction of estrogen with SMC in part contributes to the cardioprotective effects of estrogen, it is feasible that the lack of effect of conjugated estrogen is due to the lack of inhibitory estrogen in that specific preparation. Indeed, several studies show that estradiol replacement reduces neointimal thickening in the carotid arteries of postmenopausal women with established coronary artery disease and undergoing percutaneous transdermal coronary angioplasty (203). Taken together, the above findings provide evidence that overall estradiol induces protective antivasoocclusive effects on the cardiovascular system.

Whether estradiol induces antivasoocclusive effects in subjects with established coronary artery disease or has only acute protective effects in patients with minimal occlusive disorders is under intense debate. In this regard, the protective effects of estradiol on the vasculature may depend on the stage or condition of the vascular disease. Indeed, the antiatherogenic effects of estradiol are abolished by balloon catheter injury in cholesterol-clamped rabbits; moreover, the direct antiatherogenic effects of estradiol is present, absent, or reversed depending on the state of the arterial endothelial cells (99). Animal as well as human studies provide convincing evidence that vascular remodeling after balloon injury and percutaneous transdermal coronary angioplasty is significantly inhibited by estradiol. However, estradiol does not prevent neointima formation in a nonhuman primate model with preexisting atherosclerosis (1, 76). Estradiol is able to inhibit the further progression of atherosclero