Increasing evidence shows that sex hormones exert a protective effect on the vasculature, especially in the regulation of the active vasomotor responses. However, whether sex hormones affect vascular remodelling is currently unclear. In the present study, we tested the hypothesis that testosterone in males and β-oestradiol in females prevent inward remodelling, possibly through inhibition of cross-linking activity induced by enzymes of the TG (transglutaminase) family. Small mesenteric arteries were isolated from male and female Wistar rats. Dose-dependent relaxation to testosterone and β-oestradiol was inhibited by the NO synthase inhibitor L-NAME (NG-nitro-L-arginine methyl ester), confirming that these hormones induce NO release. When arteries were cannulated, pressurized and kept in organ culture with ET-1 (endothelin-1) for 3 days we observed strong vasoconstriction and inward remodelling. Remodelling was significantly inhibited by testosterone in males, and by β-oestradiol in females. This preventive effect of sex hormones was not observed in the presence of L-NAME. Inward remodelling was also reduced by the inhibitor of TG L682.777, both in males and females. In arteries from female rats, ET-1 increased TG activity, and this effect was prevented by β-oestradiol. L-NAME induced a significant increase in TG activity in the presence of sex hormones in arteries from both genders. We conclude that testosterone and β-oestradiol prevent constriction-induced inward remodelling. Inward remodelling, both in males and females, depends on NO and TG activity. In females, inhibition of inward remodelling could be mediated by NO-mediated inhibition of TG activity.
- nitric oxide
- sex hormone
- transglutaminase (TG)
- vascular remodelling
- vascular structure
• Increasing evidence shows that sex hormones exert a protective effect on the vasculature, although the effect of these hormones on vascular remodelling is currently unclear.
• In the present study, we showed that testosterone and β-oestradiol prevent constriction-induced inward remodelling.
• These observations indicate the possibility that hormone replacement therapy could prevent vascular remodelling.
The role of sex hormones has become an important theme in vascular biology during the last decades, as accumulating evidence suggests that both male and female sex hormones may play protective roles on the vasculature . Many studies have demonstrated a beneficial effect of endogenous sex hormones on vascular tone regulation, showing for example, that these hormones increase NO release and/or bioavailability, decrease oxidative stress, and reduce the synthesis of vasoconstrictor prostanoids [2–6]. Studies regarding the role of sex hormones in vascular remodelling have been mainly focused on growth factors and signal cascades related to smooth muscle cell proliferation [7–9]. However, little is known about the effect of sex hormones on small artery remodelling and distensibility.
Small arteries constitute the main source of resistance to blood flow in the vascular system . Regulation of vascular diameter involves adaptation of both vascular tone and structure . Changes in vascular structure include inward remodelling, which is characterized by a decrease in the outer and lumen diameter, and an increase in the media/lumen ratio . The underlying mechanisms of remodelling are still unclear, but it could be the result of a reorganization of the extracellular matrix, with either a repositioning of the cells or a balanced process of apoptosis and proliferation. Inward remodelling has been described in hypertension, in response to low blood flow, and after chronic vasoconstriction [13–15]. It is one of the mechanisms responsible for the increased vascular resistance in hypertension and has a prognostic value for cardiovascular disease [16,17].
In previous work from this group, it was reported that TG (transglutaminases), in particular type 2 TG, play a role in vascular remodelling associated with chronic vasoconstriction in vitro , hypertension  and reduced blood flow . The expression of TG occurs in several cell types, including the endothelium and smooth muscle cells . TG is located both in and outside of the cell  and it has been suggested that its release depends on its activation . An important function of TG is the cross-linking of a glutamine residue of a protein/peptide substrate to the primary amino group of a lysine residue . Thus, a large variety of proteins can provide a substrate for the enzyme. The physiological role of TG is still being discussed, although its involvement in cell adhesion, wound healing, apoptosis and matrix reorganization is currently accepted . The activity of TG is dependent on calcium and can be inhibited by NO and GTP [25,26].
The aim of the present study was to investigate the effect of testosterone and β-oestradiol on constriction-induced remodelling in mesenteric arteries from male and female rats. In addition, we studied the role of TG activity and NO in this process.
MATERIALS AND METHODS
Rats and vessel isolation
Small mesenteric arteries from 4-month-old male and female Wistar rats were used. Rats were anaesthetized by isoflurane inhalation and killed by decapitation. Small mesenteric artery segments were carefully dissected out and immediately placed in ice-cold Mops buffer [145 mmol/l NaCl, 4.7 mmol/l KCl, 1.2 mmol/l NaH2PO4, 1.2 mmol/l MgSO4, 2 mmol/l CaCl2, 3 mmol/l 3-(N-morpholino) propanesulfonic acid, 5 mmol/l glucose and 2 mmol/l pyruvate, pH 7.4]. Arteries were cut into equal-sized pieces where one segment was randomly assigned as control and other segments subjected to various interventions. All experiments were approved by the Committee for Animal Experiments of the Academic Medical Centre Amsterdam (permission 102242).
Dose–response relationships to sex hormones were recorded using a Mulvany wire myograph (Danish Myo Technology). After dissection, vessels were placed in the organ bath containing PSS (physiological saline solution; 119 mmol/l NaCl, 25 mmol/l NaHCO3, 4.7 mmol/l KCl, 1.18 mmol/l KH2PO4, 1.17 mmol/l MgSO4, 2.5 mmol/l CaCl2, 0.027 mmol/l EDTA and 5.5 mmol/l glucose, pH 7.4) continuously gassed with 5% CO2/95% air at pH 7.4 and 37°C. Arteries were mounted on two 40-μm wires and stretched up to its optimal position (normal distension, 0.9 L100), which was estimated according to the Laplace relationship (pressure=tension/radius) that gives an estimated diameter at 100 mmHg . Segments were exposed to PSS containing 120 mmol/l KCl and 10 μmol/l noradrenaline (norepinephrine) to assess its maximum contractility. After a washout period, the integrity of the endothelium was assessed by adding 10 μmol/l methacholine to segments precontracted with 10 μmol/l noradrenaline. Then, concentration–response curves to β-oestradiol (0.1 nmol/l–10 μmol/l) in females or testosterone (0.1 nmol–10 μmol/l) in males were recorded on segments precontracted with 10 μmol/l noradrenaline. To determine the contribution of NO in this response the competitive NO synthase inhibitor L-NAME (NG-nitro-L-arginine methyl ester, 100 μmol/l) was added to the bath 30 min before construction of the concentration–response curves.
Pressure myography and remodelling
Vessels were also studied using a pressure myograph, where segments were tied to glass cannulas on both ends and pressurized using an electro-pneumatic converter (model T5200, Fairchild). The organ bath was mounted on top of a microscope equipped with a digital camera connected to a computer. Inner and outer vessel diameters were measured using MatLab software (Math Works) and recorded continuously. The temperature of the setup was kept at 37°C. Vessels were kept in Leibovitz medium containing 1% of a mix of antibiotic–antimycotic solution. The perfusate but not the superfusate was supplemented with 10% HI-FBS [heat-inactivated FBS (fetal bovine serum]. FBS may contain approx. 20 pg/ml oestrogen , which corresponds to a final concentration of approx.7 pmol/l. After checking for leaks, a passive pressure–diameter curve (10–120 mmHg) was constructed in the presence of papaverine (0.1 mmol/l) to rule out the influence of vasomotor tone. After washing out papaverine, the superfusate was supplemented with different compounds depending on the experimental group. Three different groups were used in each gender. For males, the groups consisted of ET-1 (endothelin-1), ET-1+testosterone, ET-1+L682.777 (an inhibitor of TG activity) and ET-1+testosterone+L-NAME. For females, the groups consisted of ET-1, ET-1+β-oestradiol, ET-1+L682.777 and ET-1+β-oestradiol+L-NAME. Concentrations used were: 10 nmol/l ET-1, 10 μmol/l L682.777 and 0.1 mM L-NAME. The concentration chosen for both sex hormones was 10 nmol/l in order to mimic the physiological serum levels of these hormones in rats [1,29,30]. To ensure full access of sex hormones to the vessel wall, β-oestradiol or testosterone were also added in the same concentration to the perfusate in corresponding groups. Experiments that involved L-NAME were done using DMEM (Dulbecco's modified Eagle's medium) containing 1% of a mix of antibiotic–antimycotic solution, as the competitive inhibitor L-NAME is an L-arginine analogue and Leibovitz medium contains a relatively high concentration of L-arginine. The perfusate, but not the superfusate, was supplemented with 10% HI-FBS. In these experiments, incubation was done in a CO2 incubator. Arteries were maintained for 3 days at 80 mmHg, with the medium refreshed daily. At the end of the culture period, a second passive pressure–diameter curve (10–120 mmHg) was recorded for each artery after full dilation with 0.1 mmol/l papaverine.
Determination of TG activity
TG activity was visualized using the pseudo-substrate cadaverine linked to FITC. Arteries were mounted in a pressure myograph, using DMEM medium containing 1% of antibiotic–antimycotic solution. The perfusate, but not the superfusate, was supplemented with 10% HI-FBS. In all groups, FITC-cadaverine (100 μmol/l) was added to the superfusate, which was also supplemented with different compounds depending on the group. Four different groups were used in each gender: control, ET-1, ET-1+testosterone and ET-1+testosterone+L-NAME in males; and control, ET-1, ET-1+β-oestradiol and ET-1+β-oestradiol+L-NAME in females. Last groups with L-NAME were studied in a different set of experiments in which they were compared with new vessels incubated with ET-1+corresponding hormone. Concentrations used were: 10 nmol/l ET-1, 10 nmol/l testosterone, 10 nmol/l β-oestradiol and 0.1 mM L-NAME. The pressure was set at 80 mmHg and the myographs were placed in a CO2 incubator at 37°C for 24 h. After washing with PBS for 5 min and fixation with formaldehyde for 10 min, vessels were mounted on glass slides using Vectashield/DAPI (4′,6-diamidino-2-phenylindole) and imaged on a confocal microscope (Leica TCS SP2). Laser and photomultiplier settings were unchanged during the acquisition of images from the different groups in each rat. TG activity was quantified using ImageJ software. Data were corrected for vessel size and depicted in arbitrary units.
Calculations and statistics
From each rat, three to four equal-sized segments were obtained and studied in simultaneous experiments, being assigned to different experimental groups. Wall-to-lumen ratio was calculated by dividing wall thickness (the difference between outer and inner diameters) by the lumen diameter. Results are expressed as means±S.E.M. for the number of animals indicated. Student's t tests were used to compare data, followed by Bonferroni corrections as appropriate. To compare pressure–diameter relationships on day 0 against day 3, two-way repeated-measures ANOVAs were also performed. Data were considered significant when P<0.05.
Chemicals and solutions
Leibovitz medium, DMEM medium and the antibiotic–antimycotic solution were obtained from GIBCO. FBS was obtained from BioWhitaker. Papaverine, noradrenaline, methacholine, L-NAME and β-oestradiol were obtained from Sigma, and testosterone was obtained from Fluka. ET-1 was obtained from Bachem, and L682.777 was obtained from Zedira. Salts were obtained from Merck. FITC–cadaverine was obtained from AnaSpec. Vectachield/DAPI was obtained from Vector Laboratories.
The acute effect of testosterone and β-oestradiol on vascular function was tested in a wire myograph setup. Mesenteric arteries were pre-contracted with noradrenaline. Then, male vessels were exposed to increasing doses of testosterone. This induced a dose-dependent relaxation (Figure 1). Segments from female rats were exposed to increasing doses of β-oestradiol, which also induced dose-dependent relaxation. The relaxation induced by both sex hormones was attenuated in the presence of the competitive NO synthase inhibitor, L-NAME (Figure 1). These experiments showed that both hormones induce relaxation through an increase in the NO bioavailability, and confirmed the advisability of the doses of sex hormones that we would later use in the vascular remodelling experiments.
To study the effect of physiological concentrations of sex hormones on inward remodelling associated with chronic vasoconstriction, we used an organ culture approach. Small mesenteric arteries were cannulated and exposed to ET-1 for 3 days. Arteries from both genders showed inward remodelling after the culture period (Figures 2 and 3). Remodelling was defined as the difference in the passive diameter between the first and the second measurement at a given pressure in each segment. When testosterone was also added to the culture medium of male arteries (Figure 2A) or β-oestradiol to the culture medium of female arteries (Figure 2B), remodelling was reduced. This showed that both sex hormones partially prevent constriction-induced vascular remodelling. In the presence of the NO synthesis inhibitor L-NAME, strong inward remodelling was observed, despite the presence of the sex hormones (Figures 2C and 2D).
Remodelling was also partially prevented in both genders by incubation with the TG inhibitor L682.777 (Figure 3), indicating that TG activity is involved in the remodelling process.
During the 3 days of incubation, the constriction induced by ET-1 was maintained, and this was not modified by co-incubation with testosterone, β-oestradiol or L682.777 (Figure 4).
Wall cross-sectional area, measured at 80 mmHg, was not altered by culture period in any group (Table 1), indicating that the observed remodelling is eutrophic. The wall to lumen ratio was increased after culture period with ET-1 in both groups, confirming that the vessels were inwardly remodelled (Table 1). Differences in this parameter were not statistically significant between ET-1 and ET-1+hormone groups. L-NAME also increased the wall to lumen ratio in vessels from both genders. In males, this increase was also significant when compared with vessels with ET-1+testosterone.
Since we found that ET-1-induced remodelling depends on TG activity and is influenced by sex hormones, we studied the relationship between sex hormones and TG activity. The activity of TG in the vessel wall is reflected by the incorporation of fluorescent cadaverine (FITC–cadaverine), a substrate for TG (Figure 5). TG activity was detected in the adventitia, smooth muscle layer and endothelium. Differences were analysed in the smooth muscle layer where it was located in the cell membrane. In males, neither ET-1 nor testosterone modified TG activity. In females, ET-1 induced a significant increase in TG activity. This increase was completely abolished in vessels incubated with β-oestradiol. In an additional set of experiments, the effect of L-NAME was tested in vessels incubated with both ET-1 and the corresponding hormone. L-NAME increased TG activity, with respect to the ET-1+hormone group, in arteries from both genders (Figure 6).
In the present study, we investigated the effect of sex hormones on vascular tone and vascular structure. Clinical and epidemiological studies have shown an inverse correlation between serum sex hormones levels and cardiovascular disease in both genders [31–33]. Serum levels of sex hormones decrease with age, and evidence suggests that they play an important role in the regulation of vascular function [1,34–36]. Thus, low serum levels are independently related to cardiovascular disease in both genders [37–39]. The protective effect of sex hormones can be attributed to several of their actions, such as modulation of the lipid profile, vascular tone and vascular structure [2,34,36]. Even so, studies concerning vascular structure and sex hormones have focused mainly on hypertrophic remodelling, and there are few data concerning the effect of sex hormones on eutrophic inward remodelling [40–42]. In the present study, we tested whether testosterone and β-oestradiol (the main male and female sex hormones respectively) could prevent eutrophic inward remodelling induced by 3 days of maintained constriction with ET-1 under pressurized conditions. We observed that in both genders, corresponding hormones are able to inhibit vascular remodelling.
The observed remodelling can be defined as eutrophic, since there is an increase in the wall to lumen ratio with no change in the wall cross-sectional area. As regards vessel stiffness, the remodelling can be interpreted as a reduction in distensibility. As can be seen in the pressure–diameter relationships, most of the reduction in diameter is observed at higher pressure levels.
Previous work showed that an active, persistent constriction leads to eutrophic inward remodelling. Consequently, remodelling could be prevented by vasodilators such as papaverine and verapamil . Considering these data, the reported effect of the hormones could be related to its vasodilator properties. The wire myograph experiments showed that both sex hormones indeed induced a dose-dependent relaxation. This relaxation was inhibited by L-NAME, indicating that it is mediated by NO. Direct measurement of NO levels during the various interventions could further substantiate the current findings. However, the analysis of the average tone of the pressurized arteries during the 3 days of organoid culture revealed that constriction was similar during the culture period in all groups. Thus, although the inhibitory effect of sex hormones on inward remodelling may be related to NO, it appears not to be due to its vasoactive effect. We further investigated the participation of NO in the remodelling process by performing 3-day incubation experiments in the presence of the NO synthesis inhibitor L-NAME. These data showed that NO inhibition counteracted the effect of sex hormones, suggesting that the effect of the sex hormones is associated with NO production. These findings point towards a functional endothelium during culture, although endothelial function and acute effects of sex hormones at the end of the culture have not been specifically tested. In fact, previous studies have reported that the presence of HI-FBS in culture medium, which is also present in our experiments, has been shown to preserve endothelial function [43,44]. Nevertheless, although vessels maintain normal overall function in the pressurized culture system, changes in acute responsiveness to sex hormones cannot be ruled out.
Vascular remodelling of small arteries induced by reduced blood flow, hypertension and exposure to vasoconstrictors depends on TG activity [18–20]. Thus, inhibition of TG activity prevents vascular remodelling [18,20,45], an observation that was confirmed in the present study. Although the way by which TG exerts its effect on remodelling is not totally clear, TG could play a role in providing mechanical strength to the vessel wall by cross-linking structural proteins within the extracellular matrix. It has been suggested that acting this way, TG could fixate the constricted artery in a more narrowed state [14,46]. Some proteins have been found to be substrates for TG in the extracellular matrix of remodelling-activated smooth muscle cells, such as fibronectin, collagen α1 chain, fibulin-2 and nidogen-1 precursors . Further investigation showed that TG activity associated with remodelling is located to the cell membrane, suggesting that translocation of the enzyme is necessary to induce remodelling. Once activated and located in the extracellular cell surface, TG could cross-link its extracellular substrates.
We found that in females, ET-1 increased TG activity, which was located in the cell membrane. The observed increase in TG activity was completely inhibited by addition of β-oestradiol. This suggests that the inhibitory effect of β-oestradiol on remodelling is mediated by modulation of TG activity. An attractive hypothesis is that the effect of β-oestradiol on TG is mediated by NO. NO release induced by sex hormones induces S-nitrosylation of different proteins [48–50]; and TG can be inactivated by S-nitrosylation [25,51]. In fact, TG activation and associated remodelling can be inhibited by NO donors [18,47]. Future work should also address the role of oxidative stress reduction in prevention of inward remodelling. However, we speculate that β-oestradiol inhibits TG activity through release of NO. Our results show that inhibition of NO formation prevents the inhibitory effect of β-oestradiol on TG activity, support-ing the idea that the preventive effect of β-oestradiol on inward remodelling may be due to inactivation of TG through NO release.
In males, ET-1 did not significantly increase TG activity and testosterone did not modify this. Nevertheless, L-NAME induced an increase in TG activity, suggesting that basal NO production is down-regulating the activity of TG. Although testosterone prevented inward remodelling, it does not seem to act through modulation of TG activity. Further work is therefore necessary to clarify the inhibitory effects of testosterone on remodelling in males. Conversion of testosterone into β-oestradiol through aromatase, which is present in both smooth muscle and endothelial cells  could be speculated upon. However, since neither ET-1 nor testosterone affected TG activity, a similar pathway in males as in females via the conversion of testosterone to β-oestradiol appears unlikely. Taken together, these data show that the mechanism of action of sex hormones is different in each gender. In fact, the incidence of cardiovascular disease is lower in women compared with age-matched men but it increases notably in women after menopause [39,53]. This effect led to the idea that female sex hormones are vascular protective, whereas male sex hormones are deleterious, although more recent evidence indicates that male sex hormones also play a preventive role in the vascular function . Nevertheless, the way by which sex hormones from both genders exert this protection is different, and often female sex hormones seem to be more efficient in this respect . Interestingly, regarding the stiffening of the vessel wall, women exhibit a greater age-related rise in arterial stiffness after menopause compared to aged matched men , and this difference seems to be due to the intrinsic gender differences of the arteries, that become evident when the sex steroid secretion is low. Nevertheless, sex hormones also modulate this stiffness . Thus, apart from the differences in the action of the hormones, specific gender properties of the tissue may determine the way of action of sex hormones.
We conclude that physiological concentrations of testosterone in males and β-oestradiol in females prevent constriction-induced remodelling. We found that NO plays an important role in inward remodelling in arteries from both males and females. In addition, we observed that NO modulates TG activity in the vessel wall of both males and females. In females, the beneficial effect of β-oestradiol may be due to the inhibition of TG activity through NO release. In males, the inhibitory effect of testosterone on inward remodelling appears not to act via NO-mediated inhibition of TG activity. These observations could be of relevance for the treatment of human cardiovascular disease with hormone replacement; since the process of small artery inward remodelling reported in hypertension  may be reversible at its early stages . Studies in humans on large arteries have also shown an inverse relation between serum hormone levels and arterial stiffness, which can be reversed when serum levels are restored . Further studies should therefore address the possibility that the hormone replacement therapy could reverse or at least prevent vascular remodelling.
This study was supported a fellowship (to L.d.C) from the Spanish Ministry of Education [AP2007-00936] and Fondo Investigaciones Sanitarias [grant number PI11406 (to M.F.)].
Lara del Campo contributed to the conception and design of the study, collected, analysed and interpreted the data, drafted the paper and contributed to the critical revision of the study. Bilge Guvenc Tuna collected, analysed and interpreted the data. Mercedes Ferrer contributed to the conception and critical revision of the study. Ed van Bavel contributed to the conception and critical revision of the study. Erik Bakker contributed to the conception and design of the study, collected, analysed and interpreted the data, drafted the paper and contributed to the critical revision of the study.
Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; DMEM, Dulbecco's modified Eagle's medium; ET-1, endothelin-1; FBS, fetal bovine serum; HI-FBS, heat-inactivated FBS; L-NAME, NG-nitro-L-arginine methyl ester; PSS, physiological saline solution; TG, transglutaminase
- © The Authors Journal compilation © 2013 Biochemical Society