Acute and chronic effects of oestrogen on endothelial tissue-type plasminogen activator release in postmenopausal women
- Greta L Hoetzer*,
- Brian L Stauffer*†,
- Heather M Irmiger*,
- Marilyn Ng*,
- Derek T Smith* and
- Christopher A DeSouza*‡
- *Integrative Vascular Biology Laboratory, Department of Integrative Physiology, University of Colorado Boulder, CO 80309
- †Divisions of Cardiology, Department of Medicine, University of Colorado, Health Sciences Center Denver, CO 80262, USA
- †Divisions of Geriatrics, Department of Medicine, University of Colorado, Health Sciences Center Denver, CO 80262, USA
- Corresponding author C. DeSouza: Integrative Vascular Biology Laboratory, Department of Integrative Physiology, University of Colorado, 354 UCB Boulder, CO 80309, USA. Email: desouzac{at}colorado.edu
Abstract
The capacity of vascular endothelium to locally release tissue-type plasminogen activator (t-PA) represents an important endogenous defence mechanism against intravascular fibrin deposition and thrombosis. We determined the influence of chronic and acute oestrogen administration on endothelial t-PA release in postmenopausal women. Sixty-three healthy postmenopausal women were studied: 31 non-users (age 58 ± 1 years) and 32 users of hormone replacement therapy, including oestrogen alone (ORT: 62 ± 2 years; n = 15) and in combination with progesterone (HRT: 57 ± 1 years; n = 17). Net endothelial t-PA release was determined in vivo, in response to intrabrachial infusions of bradykinin and sodium nitroprusside. To examine the acute effects of oestrogen on endothelial t-PA release, bradykinin and sodium nitroprusside dose-response curves were repeated in the presence of 17 β-oestradiol in 20 of the 31 non-users. Net endothelial release of t-PA was ≈30 % higher (P < 0.01) in women taking ORT (from 2.0 ± 1.0 to 83.6 ± 9.2 ng (100 ml tissue)−1 min−1) compared with those taking HRT (from 1.4 ± 0.4 to 63.5 ± 5.6 ng (100 ml tissue)−1 min−1) and those not taking supplementation (1.0 ± 0.7 to 63.0 ± 4.7 ng (100 ml tissue)−1 min−1). Intra-arterial infusion of 17 β-oestradiol significantly potentiated bradykinin-induced t-PA release. Net endothelial release of t-PA was ≈45 % higher (P < 0.01) after (from 1.0 ± 0.8 to 87.4 ± 9.9 ng (100 ml tissue)−1 min−1) versus before (1.2 ± 0.6 to 60.8 ± 5.6 ng (100 ml tissue)−1 min−1) acute 17 β-oestradiol administration. Our results suggest that oestrogen has a direct modulatory effect on the capacity of the endothelium to release t-PA in healthy postmenopausal women. However, progesterone appears to oppose the favourable influence of oestrogen on endothelial fibrinolytic capacity.
Endogenous fibrinolysis is vital to vascular health as it represents an important defense mechanism against intravascular fibrin deposition and thrombosis (Kooistra et al. 1994). The vascular endothelium plays a central role in the regulation of thrombolysis through the synthesis and release of tissue-type plasminogen activator (t-PA), the key enzyme activating fibrinolysis (van den Eijnden-Schrauwen et al. 1995). t-PA is essential for dissolving embolized clots (Bugge et al. 1996) and managing intravascular fibrin deposition (Christie et al. 1999). The ability of the endothelium to locally and rapidly release t-PA is critical for efficient thrombolysis. Indeed, fibrin degradation is accelerated if t-PA is locally available before or during, rather than after, thrombus formation (Brommer, 1984; Fox et al. 1984). Moreover, the interaction between t-PA and its biological inhibitor, plasminogen activator inhibitor-1 (PAI-1), has a second-order rate constant of ≈107 M s−1 (Ouimet & Loscalzo, 1994). Thus local endothelial release of t-PA is essential for effective delivery of active (unbound) t-PA to the developing thrombus.
Several studies have suggested that oestrogen-based hormone replacement therapy, either unopposed oestrogen (ORT) or oestrogen combined with progesterone (HRT), improves endogenous fibrinolysis in healthy post-menopausal women. This postulate is based primarily on the observation that oral regimes of ORT and HRT reduce circulating levels of t-PA antigen, PAI-1 antigen and PAI-1 activity (Kroon et al. 1994; Gilabert et al. 1995; Koh et al. 1997). However, oestrogen-induced reductions in circulating plasma concentrations of t-PA and PAI-1 may not necessarily reflect increased endothelial fibrinolytic capacity but rather changes in hepatic synthesis and/or clearance of these proteins (Koh et al. 1997; Giltay et al. 2000). For example, oestrogen has been reported to increase t-PA clearance through upregulation of the mannose receptor, a hepatic clearance receptor for t-PA (Lansink et al. 1999). In addition, the ineffectiveness of transdermal compared with oral oestrogen administration in lowering basal t-PA and PAI-1 levels supports the concept that the hepatic effects of oestrogen are important in the regulation of t-PA and PAI-1 clearance (Kroon et al. 1994; Koh et al. 1997; Giltay et al. 2000; Teede et al. 2000). Although the effects of postmenopausal oestrogen therapy on circulating fibrinolytic factors have been well studied, it is the local endothelial rate of release of t-PA and not the systemic plasma concentration that determines endogenous thrombolytic potential (Kooistra et al. 1994). Little information is available regarding the effects of oestrogen, either chronic use or acute administration, on endothelial fibrinolytic function in postmenopausal women.
Accordingly, the experimental aims of the present investigation were to determine: (1) the influence of chronic ORT and HRT use on the capacity of the vascular endothelium to release t-PA in healthy postmenopausal women; and (2) the effects of acute oestrogen administration on endothelial t-PA release in postmenopausal women not taking hormone therapy.
METHODS
Subjects
Sixty-three healthy sedentary postmenopausal women ranging in age from 50 to 80 years were studied. All of the women were at least 1 year postmenopausal before the start of the study. Among the 63 postmenopausal women, 31 either had never used or had not used hormone supplementation for at least 1 year prior to the study, 17 were users of oral HRT (oestrogen in combination with progestins: oestrogen 0.5–1.0 mg day−1; natural progesterone or medroxyprogesterone acetate, 0.09–400 mg), and 15 were users of oral ORT (unopposed oestrogen: 0.325–1.0 mg day−1). Six of the 17 women using HRT were taking cyclic progesterone and the remaining 11 women were taking a non-cyclic regimen. Women taking hormone supplementation had done so continuously for at least 1 year prior to the study (HRT: 7 ± 1; ORT 11 ± 3 years). All subjects were free of overt disease as assessed by medical history, physical examination, resting and exercise electrocardiograms and fasting blood chemistries. None of the subjects smoked, were taking medications (other than hormone therapy) or had participated in a regular exercise programme for at least 1 year before the start of the study. Prior to participation, all of the subjects had the research study and its potential risks and benefits explained fully before providing written informed consent according to the guidelines of the University of Colorado at Boulder. All experiments conformed to the Declaration of Helsinki.
Body composition
Body mass was measured to the nearest 0.1 kg using a medical beam balance. Body fat percentage was determined by dual energy X-ray absorptiometry (Lunar Corp., Madison, WI, USA). Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared. Minimal waist circumference was measured according to published guidelines (Lohman et al. 1988).
Maximal oxygen consumption (
)
To assess aerobic fitness subjects performed incremental treadmill exercise using a modified Balke protocol. Maximal oxygen
consumption (
) was measured using on-line computer-assisted open circuit spirometry as previously described (DeSouza et al. 1998).
Metabolic measurements
Fasting plasma lipid and lipoprotein, glucose and insulin concentrations were determined using conventional methods by the clinical laboratory affiliated with the General Clinical Research Center as previously described (DeSouza et al. 1998).
Intra-arterial fibrinolytic protocol
All measurements were performed between 07.00 and 10.00 h after a 12 h overnight fast in a temperature-controlled room. Twenty minutes before testing an intravenous catheter was placed in a deep antecubital vein and a 5 cm, 20 gauge catheter was introduced into the brachial artery of the non-dominant arm. Forearm blood flow (FBF) was measured using strain-gauge venous occlusion plethysmography (D. E. Hokanson, Bellevue, WA, USA). Drug infusion rates were normalized per 100 ml forearm tissue and infused at 4 ml min−1 by a syringe pump. Following the measurement of resting blood flow for 5 min, bradykinin was infused intra-arterially at 12.5, 25 and 50 ng (100 ml tissue)−1 min−1 and sodium nitroprusside at 1.0, 2.0 or 4.0 μg (100 ml tissue)−1 min−1 for 5 min at each dose. Both bradykinin and sodium nitroprusside were dissolved in preservative-free, sterile 0.9 % saline and diluted to required concentrations. Forearm volume was determined by water displacement. To avoid an order effect the sequence of drug administration was randomized.
Net endothelial release of t-PA antigen and PAI-1 antigen in response to bradykinin and sodium nitroprusside was calculated according to Jern et al. (1997) Briefly, arteriovenous concentration gradients were determined by subtracting the measured values in simultaneously collected venous and arterial blood. For both t-PA and PAI-1, a positive difference indicated a net release and a negative difference, net uptake. Net release or uptake rates were calculated as follows:
Net release = (CV–CA)(FBF[101 – haematocrit/100]),
where CV and CA represent the concentration in the vein and artery, respectively. Haematocrit was measured in triplicate using the standard microhaematocrit technique and corrected for trapped plasma volume within the trapped erythrocytes (Chaplin & Mollison, 1952). The total amount of t-PA antigen released across the forearm in response to bradykinin was calculated as the total area under each curve above baseline using a trapezoidal model.
A questionnaire designed to detect and document recent infection/inflammation (within less than 2 weeks) was administered prior to the phlebotomies. Subjects with a history of recent infection/inflammation did not undergo the fibrinolytic protocol, and were rescheduled, in order to avoid confounding effects from potential infection/inflammation-associated endothelial and fibrinolytic changes (Macko et al. 1996).
Acute oestradiol administration
In 20 of the 31 women not using oestrogen therapy, we determined the acute effects of 17 β-oestradiol (Clinalfa AG, Switzerland) on the capacity of endothelium to release t-PA. After allowing sufficient time (≈20 min) for FBF and plasma fibrinolytic concentrations to return to baseline following the initial infusions of bradykinin and sodium nitroprusside described above, 17 β-oestradiol was dissolved and diluted in sterile 0.9 % saline (5 ng ml−1) and infused at a constant rate (4 ml min−1) for 20 min. The dose selected has been shown to increase forearm oestradiol concentrations to levels typical of reproductive-age women at midcycle (≈150 pg ml−1) (Abraham et al. 1995). After 20 min, the 17 β-oestradiol infusion was maintained at the same rate whilst the bradykinin and sodium nitroprusside dose-response curves were repeated in the same order as performed earlier. Net endothelial release rates of t-PA antigen and PAI-1 antigen were determined at time 0 and after 20 min of 17 β-oestradiol infusion, and after each dose of bradykinin and sodium nitroprusside infused thereafter.
Blood sampling and fibrinolytic assays
Arterial and venous blood samples were collected simultaneously at the beginning and the end of each drug dose to determine t-PA and PAI-1 antigen concentrations. All samples were collected into tubes containing 0.45 M sodium citrate buffer, pH 4.3 (Stabilyte, Biopool AB, Sweden), aliquoted and stored for analysis as previously described by our laboratory (DeSouza et al. 1998). Plasma concentrations of t-PA antigen and PAI-1 antigen were determined by enzyme immunoassay (DeSouza et al. 1998).
Statistical analysis
Subject baseline characteristics were analysed by between-groups analysis of variance (ANOVA). Group differences in FBF and net endothelial t-PA and PAI-1 release in response to bradykinin and sodium nitroprusside were determined by repeated measures ANOVA. When indicated by a significant F value, a post hoc test using the Newman-Keuls method was performed to identify differences amongst the groups. Changes in basal and agonist-stimulated endothelial t-PA and PAI-1 release in response to concomitant 17 β-oestradiol administration were determined by repeated measures ANOVA. All data are expressed as means ± s.e.m. Statistical significance was set a priori at P < 0.05.
RESULTS
Table 1 presents selected subject characteristics. There were no differences in age, body composition, and maximal oxygen consumption amongst the groups of postmenopausal women. Although well within clinically normal levels, the women taking hormone supplementation (both ORT and HRT) demonstrated significantly higher resting systolic blood pressure than the women not taking supplementation. By design none of the subjects were dyslipidaemic, but plasma concentrations of HDL cholesterol were higher and insulin levels lower (P < 0.05) in the women using ORT compared with non-users. As expected, plasma oestradiol concentrations were lowest in the women not taking supplementation.
Selected subject characteristics
Figure 1 shows the FBF responses to bradykinin and sodium nitroprusside. Vasodilator responses to bradykinin were significantly higher in the women taking ORT compared with women not taking supplementation. However, FBF responses were not different between the ORT and HRT groups. There were no differences in the FBF responses to sodium nitroprusside amongst the groups.
Values are means ± s.e.m. *P < 0.05 vs. non-users of hormone therapy.
Influence of postmenopausal hormone replacement on endothelial t-PA release
Figure 2 shows the net release rates and total amounts of t-PA released in response to bradykinin amongst the groups. Endothelial release of t-PA was ≈30 % higher (P < 0.01) in women taking ORT (from 2.0 ± 1.0 to 83.6 ± 9.2 ng (100 ml tissue)−1 min−1) compared with those taking HRT (from 1.4 ± 0.4 to 63.5 ± 5.6 ng (100 ml tissue)−1 min−1) and those not taking supplementation (1.0 ± 0.7 to 63.0 ± 4.7 ng (100 ml tissue)−1 min−1). The total amount of t-PA released (the area under the curve) was also greater (P < 0.01) in the ORT group (427 ± 38 ng (100 ml tissue)−1) than either the HRT (309 ± 31 ng (100 ml tissue)−1) or the non-user (298 ± 24 ng (100 ml tissue)−1) groups. There were no significant differences in either net release rates or total amount of t-PA released across the forearm between the women taking HRT and those not taking hormone therapy. Infusion of sodium nitroprusside resulted in no significant changes in t-PA release in any of the groups (data not shown). In addition, neither bradykinin nor sodium nitroprusside produced consistent or significant changes in the net release of PAI-1 amongst the groups. For example, at the highest dose of bradykinin the women taking ORT demonstrated marginal release (2.2 ± 1.9 ng (100 ml tissue)−1 min−1) compared with minimal uptake in HRT users (−3.2 ± 4.2 ng (100 ml tissue)−1 min−1) and non-users (−1.3 ± 2.7 ng (100 ml tissue)−1 min−1) of hormone therapy.
Values are means ± s.e.m. *P < 0.05 vs. HRT and non-users of hormone therapy.
Effects of oestradiol on endothelial t-PA release
Infusion of 17 β-oestradiol significantly increased forearm venous oestradiol levels from 8.7 ± 0.4 to 169.0 ± 13.6 pg ml−1. Forearm blood flow responses to bradykinin and sodium nitroprusside were not significantly different with 17 β-oestradiol (Fig. 3). Additionally, there was no change in baseline net release rates of either t-PA or PAI-1 to 17 β-oestradiol. However, co-infusion of 17 β-oestradiol significantly increased net endothelial t-PA release in response to bradykinin. Net release of t-PA was ≈45 % higher (P < 0.01) after (from 1.0 ± 0.8 to 87.4 ± 9.9 ng (100 ml tissue)−1 min−1) versus before (1.2 ± 0.6 to 60.8 ± 5.6 ng (100 ml tissue)−1 min−1) the administration of 17 β-oestradiol (Fig. 4). As a result, the total amount of t-PA released in response to bradykinin increased 65 % (from 263 ± 26 to 431 ± 54 ng (100 ml tissue)−1 with 17 β-oestradiol treatment (Fig. 4). There was no significant effect of 17 β-oestradiol on the t-PA response to sodium nitroprusside or on the PAI-1 response to either bradykinin or sodium nitroprusside (data not shown).
Values are means ± s.e.m. *P < 0.05 vs. saline.
Values are means ± s.e.m.
DISCUSSION
The primary new findings of the present study are as follows. Firstly, in healthy postmenopausal women chronic ORT, but not HRT, use is associated with increased endothelial fibrinolytic capacity. Postmenopasual women taking ORT demonstrated significantly higher rates of t-PA release compared with women taking HRT and those not taking hormone therapy. Notably, endothelial t-PA release was not different between users of HRT and non-users. Secondly, acute intra-arterial infusion of 17 β-oestradiol potentiates endothelial t-PA release in response to bradykinin in healthy postmenopausal women. Taken together these results suggest that, in healthy postmenopausal women, oestrogen, either chronic supplementation or acute administration, favourably affects endothelial fibrinolytic potential. However, the modulatory influence of oestrogen on the capacity of the endothelium to release t-PA appears to be opposed by progesterone.
It is well recognized that endothelial cells play an important role in maintaining vascular health through the synthesis and release of a variety of factors that regulate vascular tone, smooth muscle cell proliferation, inflammatory responses, and thrombolysis (Gokce et al. 1998). Clinical and animal studies have demonstrated that oestrogen has many potential beneficial effects on endothelial function. For example, oestrogen increases both basal and stimulated endothelial nitric oxide production, improving vasodilator function (Guetta et al. 1997). In addition, oestrogen inhibits cytokine-induced expression of cell adhesion molecules, promotes re-endothelialization of damaged intima and stimulates angiogenesis (Rubanyi et al. 2002; White, 2002). The results of the present study demonstrate, for the first time, that oestrogen also influences endothelial fibrinolytic function. Indeed, the capacity of the endothelium to release t-PA was significantly greater in users of ORT compared with non-users of hormone therapy. Moreover, acute administration of 17 β-oestradiol, producing plasma oestradiol concentrations typical of premenopausal women at mid-cycle, significantly increased bradykinin-induced t-PA release in oestrogen-deficient postmenopausal women. Although the clinical relevance of oestrogen's influence on the capacity of endothelium to release t-PA remains to be determined, it may reflect an important antithrombotic effect of the hormone, counterbalancing its procoagulant (Caine et al. 1992) and plaque-destabilizing properties (Haynes et al. 2000). Indeed, animal studies have shown that rapid local release of t-PA in response to a thrombogenic stimulus results in pronounced fibrin degradation and preserved vascular patency (Giles et al. 1990; Kruithof et al. 1997).
The mechanisms by which oestrogen may influence endothelial release of t-PA are unclear. Endothelial cells synthesize and store t-PA in secretory granules, enabling both constitutive and stimulated release. In vitro studies have demonstrated that increases in cytoplasmic calcium concentrations and diacylglycerol activation of protein kinase C are important triggers for constitutive and stimulated release of t-PA (Kooistra et al. 1994). In fact, rapid increases in intracellular calcium concentrations have been shown to be sufficient to induce acute t-PA release (Tranquille & Emeis, 1990). Bradykinin is thought to stimulate t-PA release through receptor/G-protein activation of the phospholipase C-phosphatidylinositol signalling pathway, resulting in an increase in cytoplasmic calcium concentrations, diacylgycerol and protein kinase C activity (Emeis & Tranquille, 1992; Kooistra et al. 1994). Stefano et al. (2000) recently reported that physiological doses of 17 β-oestradiol rapidly increase intracellular calcium concentrations through activation of an endothelial cell-surface oestrogen receptor pathway. Thus, it is reasonable to suggest that the rapid 17 β-oestradiol-induced potentiation in t-PA release in response to bradykinin was due, at least in part, to augmented intracellular calcium concentrations. Moreover there is evidence that oestrogen may also activate protein kinase C (Haynes et al. 2002). In addition to its integral signalling role in stimulating t-PA release, protein kinase C activation is associated with an up-regulation of t-PA expression and synthesis (Levin et al. 1998). Increased production and storage of t-PA resulting from chronic oestrogen-induced activation of protein kinase C may underlie the greater amount of t-PA released in response to stimulation in our postmenopausal women taking unopposed oestrogen therapy. It should be noted that the lack of effect of sodium nitroprusside on net endothelial t-PA release amongst the groups and in the presence of 17 β-oestradiol indicates that the observed differences in bradykinin-induced t-PA release were not due to increases in limb blood flow (Stein et al. 1998). Whatever the mechanisms involved, the observed rapid and apparently chronic influence of oestrogen on endothelial t-PA release suggests that oestrogen has both non-genomic and genomic effects on the capacity of the endothelium to release t-PA (Mendelsohn, 2002).
An interesting finding of our cross-sectional study was that, in stark contrast to ORT, HRT use was not associated with enhanced t-PA release from the vascular endothelium. Women taking HRT demonstrated net endothelial t-PA release rates almost identical to those of women not taking hormone therapy. This novel finding indicates that progesterone may antagonize the action of oestrogen on endothelial t-PA release. One potential site for this antagonistic effect is protein kinase C activation. Progesterone has been shown to blunt basal and stimulated protein kinase C activity (Lachowicz et al. 2000). From a clinical perspective, it has been suggested, based primarily on plasma fibrinolytic markers, that the procoagulant effects of oestrogen may be offset by increased fibrinolytic capacity in postmenopausal women taking HRT (Gilabert et al. 1995; Koh et al. 1997). The results of the present study do not support this postulate. The lack of effect of HRT use on endothelial t-PA release suggests that any procoagulant effects of oestrogen may not be neutralized by increased fibrinolytic activity in healthy postmenopausal women taking HRT. Thus, an imbalance in the coagulation-fibrinolytic axis, favouring thrombogenesis, may indeed contribute to the increased risk of vascular events associated with HRT use (Bracamonte & Miller, 2001; Writing Group For The Women's Health Initiative Investigators, 2002).
There are a number of important experimental considerations regarding the present study. Firstly, with respect to our cross-sectional study, we cannot rule out the inherent possibility that genetic and/or other lifestyle behaviours may have influenced the results of our group comparisons. In fact, women who use postmenopausal hormone supplementation have been reported to lead healthier lifestyles (Mathews et al. 1996). Therefore, we attempted to minimize potential lifestyle differences by studying women across the postmenopausal age-range who were non-smokers and who did not differ in body composition or habitual physical activity. Although we observed similar net endothelial t-PA release rates in women who received an acute administration of 17 β-oestradiol compared with those taking ORT, prospective trials are needed to confirm the possible beneficial effects of chronic oestrogen supplementation on endothelial fibrinolytic capacity. Secondly, potential regional differences in both the capacity of the endothelium to acutely release t-PA and in oestrogen receptor number and distribution may limit extrapolation of our findings from the forearm to other vascular beds. However, there is evidence to suggest that acute t-PA release in the forearm provides an excellent surrogate measure of t-PA release in the coronary circulation (Newby et al. 2001). Moreover, both acute and chronic oestradiol administration have been shown to elicit similar vascular responses in the peripheral and coronary circulation (White, 2002). Thirdly, differential effects of oestrogen in combination with natural progesterone or medroxyprogesterone acetate on endothelial function have been reported (Ganz, 2002). Although we observed no differences in bradykinin-induced endothelial t-PA release amongst women taking oestrogen combined with either progesterone or medroxyprogesterone acetate in the present study, the relatively small sample size of the HRT group limit our interpretation of this data. In addition, we were unable to account for any potential temporal effect of cyclic versus non-cyclic regimens of progesterone use on endothelial t-PA release.
In conclusion, the results of the present study indicate that oestrogen favourably effects the capacity of the endothelium to release t-PA in healthy postmenopausal women. This may represent an important cardioprotective effect of oestrogen. However, the influence of oestrogen on endothelial t-PA release appears to be negated when combined with progesterone. Given the clinical importance of understanding the effects of different regimens of postmenopausal hormone replacement therapy on cardiovascular function, future studies are needed to address the potentially negative interaction between oestrogen and progesterone on endothelial fibrinolytic function.
Acknowledgments
We would like to thank all of the subjects who participated in the study as well as Yoli Casas and Jared Greiner for their technical assistance. This study was supported by National Institutes of Health awards HL03840 and DK62061, American Diabetes Association Clinical Research Award and American Heart Association awards 0060430Z, 0255921Z (Dr DeSouza). Dr Stauffer was supported by American Heart Association award 0120679Z. Dr Smith was supported by American Heart Association award 0110221Z. Greta Hoetzer was supported by a predoctoral fellowship on NIH AG00279.
Footnotes
-
- Received March 31, 2003.
- Accepted June 16, 2003.
- © The Physiological Society 2003
