OUP user menu

Hyperthyroidism enhances endothelium-dependent relaxation in the rat renal artery

E. Büssemaker, R. Popp, B. Fisslthaler, C.M. Larson, I. Fleming, R. Busse, R.P. Brandes
DOI: http://dx.doi.org/10.1016/S0008-6363(03)00326-2 181-188 First published online: 1 July 2003


Objective: Hyperthyroidism has pronounced effects on vascular function and endothelium-dependent relaxation. The aim of the present study was to identify mechanisms underlying hyperthyroidism-induced alterations in endothelial function in rats. Methods: Animals were subjected to either a single injection (36 h) or 8 weeks treatment with the thyroid hormone triiodothyronine (T3, i.p.). Vascular reactivity and agonist-induced hyperpolarization were studied in isolated renal arteries. Endothelial nitric oxide (NO) synthase expression and cyclic AMP accumulation were determined in aortic segments. Results: Endothelium-dependent relaxations to acetylcholine (ACh) were enhanced by T3 36 h after injection and after treatment for 8 weeks. Thirty-six hours after T3 application, relaxation mediated by the endothelium-derived hyperpolarizing factor (EDHF) and by endothelium-derived NO were significantly enhanced. After 8 weeks treatment with T3, however, EDHF-mediated relaxation was impaired, whereas NO-mediated relaxation remained enhanced. KCl- and ACh-induced hyperpolarizations were more pronounced in arteries from rats treated with T3 for 36 h compared to control, whereas in arteries from rats treated with T3 for 8 weeks both responses were attenuated. In rats treated for 36 h, vascular cyclic AMP levels were enhanced in the aorta and inhibition of protein kinase A attenuated EDHF-mediated relaxations of the renal artery without affecting responses in arteries from the control group. In the aorta from rats treated with T3 for 8 weeks, the expression of the endothelial NO synthase was markedly up-regulated (463±68%). Conclusions: These data indicate that short-term treatment with T3 increases endothelium-dependent relaxation, most probably by increasing vascular cyclic AMP content. Following treatment with T3 for 8 weeks, expression of the endothelial NO synthase was enhanced. During this phase, NO appears to be the predominant endothelium-derived vasodilator.

  • Endothelial function
  • Nitric oxide
  • Acetylcholine
  • Hormones
  • Na/K-pump

Time for primary review 30 days.

1 Introduction

Thyroid hormones (THs) exert multiple effects on the cardiovascular system (for review see [1]). By increasing cellular oxygen consumption THs increase the demand for blood in the periphery. Moreover, THs directly affect cardiac function and vascular tone, leading to an increase in cardiac output and a decrease in total peripheral resistance [1]. Although THs also exert non-genomic effects including a partially endothelium-dependent vasodilatation of resistance arteries [2], many of their actions can be attributed to changes in gene expression. In the heart for example, THs modulate the expression of the Na–K-ATPase, the Ca2+-ATPase, several potassium channels and isoforms IV and V of adenylyl cyclase [3]. In addition, as a consequence of the increases in cardiac output, THs enhance fluid shear stress at the endothelial cell surface and thus indirectly affect gene expression within the vasculature, particularly in the endothelium.

In humans, hyperthyroidism is associated with an enhanced endothelium-dependent vasodilatation of the forearm microcirculation and the normalization of thyroid function restores endothelium-dependent responses [4]. The precise mechanism by which THs enhance endothelium-dependent relaxation is, however, unknown.

The vascular endothelium generates three vasodilator principles, which elicit the relaxation of vascular smooth muscle cells by distinct pathways that may be differentially affected by thyroid hormones. Nitric oxide (NO), generated by the endothelial NO synthase (eNOS), induces relaxation by activating the soluble guanylyl cyclase in smooth muscle cells and increasing the formation of cyclic GMP, while prostacyclin (PGI2), generated from arachidonic acid by the prostacyclin synthase elevates smooth muscle cyclic AMP levels [5]. The endothelium-derived hyperpolarizing factor (EDHF) elicits relaxation by hyperpolarizing smooth muscle cells, which results in the closure of voltage-dependent calcium channels and a decrease in smooth muscle calcium [6]. There has been considerable debate regarding the identity as well as the effector mechanisms involved in EDHF-mediated responses [7]. Depending on the vascular bed studied, three main mechanisms have been proposed to account for the EDHF phenomenon: the generation of hyperpolarizing epoxyeicosatrienoic acids by cytochrome P450 enzymes; the release of K+ ions from endothelial cells and the subsequent activation of inwardly rectifying K+ channels; and/or the Na–K-ATPase, or a mechanism involving cyclic AMP and myo-endothelial gap junctional communication [7].

The aim of the present study was to determine the effect of chronic and acute hyperthyroidism in vivo on endothelium-dependent relaxation in an isolated rat vessel. In particular, we focused on the differential contribution of EDHF and NO to endothelium-dependent relaxation and on the mechanism(s) underlying the TH-induced changes in endothelium-dependent responses. Therefore, studies were performed in the renal artery. The acetylcholine (ACh)-induced relaxation of this conduit vessel exhibits pronounced NO- and EDHF-mediated components [8]. In the rat aorta in contrast, relaxations are mediated predominantly by NO, while in the microcirculation, EDHF is the predominant vasodilator [9].

2 Methods

2.1 Materials

Sp-5,6-DCI-cBIMPS and Rp-8-Br-cAMPS were from BioLog (Bremen, Germany). The polyclonal platelet endothelial cell adhesion molecule-1 (PECAM-1) antibody and the monoclonal eNOS antibody were from BD Transduction (Heidelberg). The soluble guanylyl cyclase polyclonal antibodies were generated from the keyhole limpet hemocyanin conjugated peptide sequence CKDVEDGNANFLGKAS and RNYGPEVWEDIKKEC (position 547–561 and position 15–28 in the human soluble guanylyl cyclase α1 and β1 subunits, respectively) by Eurogentec (Seraing, Belgium). All other compounds were obtained from Sigma (Deisenhofen, Germany).

2.2 Animals and study protocol

The study was approved by the local district government (Approval number II25.3-19c20/15-F28/01) and conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH Publication No. 85-23, revised 1996). Male Wistar–Kyoto rats (WKY, 8 weeks) were obtained from Charles-River (Sulzfeld, Germany). Chronic hyperthyroidism was induced by injection of triiodothyronine (300 μg/kg, i.p. dissolved in Na2CO3 solution 1%, 1000 μg/ml) every other day, over a period of 8 weeks. Acute hyperthyroidism was induced by a single injection of triiodothyronine (300 μg/kg, i.p.) 36 h prior to killing. Control rats received an equivalent volume of solvent. The doses of triiodothyronine chosen have previously been shown by others to reliably induce hyperthyroidism at both time points [10–13]. Experiments were performed sequentially in a paired design so that one treated and one control rat were used every day in order to allow paired comparison of the observations made.

2.3 Organ chamber experiments

Animals were anesthetized with isofuren and killed by decapitation. The renal arteries and the abdominal aorta were isolated, cleaned of connective tissue and cut into rings. Renal artery rings were mounted on stainless steel wires connected to force transducers and placed in individual organ chambers containing modified Tyrode's solution of the following composition (in mmol/l): NaCl 132, KCl 4, CaCl2 1.6, MgCl2 1.2, NaH2PO4 0.36, NaHCO3 23.8, Ca2+-EDTA 0.05, glucose 10 (administered with 20% O2/5% CO2/75% N2, pH 7.4). Diclofenac (10 μmol/l) was included in all experiments to inhibit prostaglandin synthesis. Passive tension was gradually increased to 1 g and each ring was challenged three times with KCl-rich (80 mmol/l) Tyrode solution. Precontraction was elicited using phenylephrine at a concentration adjusted to obtain a similar level of precontraction in each ring (approx. 80% of initial KCl-induced contraction). When a stable contraction was obtained, cumulative concentration–relaxation curves to either acetylcholine (ACh,1 nmol/l to 10 μmol/l) or KCl (4–10 mmol/l) were obtained. EDHF-mediated responses were defined as that portion of the endothelium-dependent hyperpolarization and relaxation that was insensitive to the combined presence of Nω-nitro-l-arginine (l-NA, 300 μmol/l) and diclofenac.

2.4 Membrane potential recordings

Renal artery rings were opened longitudinally, pinned to the sylgard base of a heated bath with the intimal side upward and superfused (5 ml/min, 37°C) with modified Tyrode's solution. All experiments were performed in the presence of l-NA (300 μmol/l), diclofenac (10 μmol/l) and phenylephrine (30 nmol/l) to mimic conditions in the organ chamber experiments. The membrane potential was recorded with glass capillary microelectrodes (tip resistance of 80–120 MΩ) filled with KCl (3 mol/l) and connected to a high impedance amplifier (intra 767, WPI, Berlin). Successful impalements were characterized by a sudden negative drop in potential from the baseline (zero potential reference) followed by a stable negative potential for at least 3 min. ACh (1 μmol/l) or KCl (5 mmol/l) were applied as bolus injections into the bath.

2.5 Cyclic AMP radio-immunoassay

Segments of the abdominal aorta were cut into rings and incubated in HEPES-modified Tyrode solution (37°C) of the following composition (in mmol/l): NaCl 140, KCl 4.7, CaCl2 1.3, MgCl2 1, HEPES 10, glucose 5 (pH 7.4) for 30 min in the presence or absence of the phosphodiesterase inhibitor, isobutyl-methylxanthine (IBMX, 500 μmol/l) and presence of diclofenac (10 μmol/l) and l-NA (300 μmol/l). Subsequently, rings were frozen in liquid nitrogen and homogenized in ice-cold trichloracetic acid (10%). Cyclic AMP was extracted with water-saturated diethylether, acetylated and quantified by radio-immunoassay using a commercially available kit (Perkin Elmer Life Science, Boston, MA, USA)

2.6 Western blotting

Aortic segments were incubated with Triton X-100 (1%) lysis buffer (pH 7.5, containing in mmol/l: Tris–HCl 50, NaCl 150, sodium pyrophosphate 10, sodium fluoride 25, EDTA 2, EGTA 2, orthovanadate 2, and the protease inhibitors; trypsin inhibitor 10 μg/ml, leupeptin 2 μg/ml, pepstatin A 2 μg/ml, antipain 2 μg/ml and phenylmethysulfonyl fluoride 40 μg/ml; (4°C, final volume 50 μl) and agitated continuously for 15 min. The Triton X-100 soluble supernatant (20 μg) was subjected to SDS PAGE (8%) and blotted onto nitrocellulose membranes. Proteins were detected using their respective antibodies and were visualized by enhanced chemiluminescence using a commercially available kit (Amersham, Germany). Expression of eNOS was normalized to that of the endothelial cell marker PECAM-1, to exclude any potential influence of changes in the intima/media ratio on the results of the densitometry.

2.7 Statistical analysis

Values presented are the mean±S.E.M and were compared by paired- and unpaired t-test or ANOVA for repeated measurements followed by the Newman–Keuls test, as appropriate. A P value less than 0.05 was considered to be significant.

3 Results

Consistent with the development of hyperthyroidism, treatment with T3 for 8 weeks induced a pronounced increase in the heart to body weight ratio, whereas a single injection of T3 had no significant effect (Table 1).

View this table:
Table 1

Effect of a single dose of thyroid hormone (36 h) and of 8 weeks of thyroid hormone treatment (8 weeks) on heart and body weight

Heart weightBody weightRation
T3 36 h997±20321±2.63.2±0.47
T3 8 weeks1176±30*309±8*3.8±0.01*7
  • *P<0.05 vs. the respective control group.

Both treatment protocols enhanced the endothelium-dependent relaxation of isolated renal artery rings to ACh (Fig. 1A,B). Endothelium-independent relaxation to cromakalim and sodium nitroprusside were not affected by T3 treatment (data not shown). To identify which components of the global endothelium-dependent response could be attributed to NO and EDHF, concentration–relaxation curves to ACh were obtained in the presence of diclofenac (10 μmol/l) and KCl (40 mmol/l) or diclofenac and l-NA (300 μmol/l), respectively.

Fig. 1

Effect of in vivo treatment with triiodothyronine (T3) on endothelium-dependent relaxation in the rat renal artery. Rats were treated with solvent (CTL), T3 (300 μg/kg) for 36 h or T3 (300 μg/kg) every other day for 8 weeks. Cumulative concentration–relaxation curves to acetylcholine (ACh) were obtained. (A) Effect of short-term T3 treatment (▵) and (B) effect of long-term T3 treatment (♢) compared to time-matched controls (▪). Experiments were performed in the continuous presence of diclofenac (10 μmol/l). Results are the mean±S.E.M., n = 7 in each group, P<0.05 versus control.

In the presence of KCl, which depolarizes vascular cells and prevents EDHF-mediated responses, both a single T3 injection and 8 weeks treatment enhanced the ACh-induced relaxation of renal artery rings compared to control (Fig. 2A,B). In the presence of l-NA, to inhibit NO synthases, a single injection of T3 was associated with an enhanced ACh-induced, EDHF-mediated relaxation after 36 h, whereas in vessels from rats treated for 8 weeks, EDHF-mediated responses were attenuated (Fig. 2C,D). Accordingly, the ACh-induced hyperpolarization of renal artery smooth muscle cells was increased following single T3 injection, but was attenuated after 8 weeks of treatment. The smooth muscle resting membrane potential was similar in arteries from the solvent and the 36 h T3 groups but was significantly hyperpolarized after T3 treatment for 8 weeks (Fig. 3A,B).

Fig. 3

Effect of in vivo T3 treatment on the membrane potential and EDHF-mediated hyperpolarization of isolated renal artery rings. Resting membrane potential (RMP, mV) and acetylcholine- (ACh, 1 μmol/l) induced changes in membrane potential (ΔMP, mV) in renal artery rings from solvent-treated (CTL, closed bars, A,B), short-term (300 μg/kg, 36 h) T3-treated (A, open bars) and long-term (8 weeks) T3-treated animals (B, open bars). Experiments were performed in the continuous presence of Nω-nitro-l-arginine (300 μmol/l) diclofenac (10 μmol/l) and phenylephrine (30 nmol/l) to mimic conditions in the organ chamber studies. Results are the mean±S.E.M., n = 5 in each group, *P<0.05 versus control.

Fig. 2

Contribution of nitric oxide (NO) and the endothelium-derived hyperpolarizing factor (EDHF) to the endothelium-dependent relaxation of renal arteries from hyperthyroid rats. Concentration–relaxation curves to acetylcholine (ACh) in renal artery rings from control (CTL, ▪), short-term T3-treated (300 μg/kg, 36 h ▵, A,C) and long-term T3-treated (8 weeks, ♢, B,D) animals. (A,B) Experiments were performed in the presence of KCl (+KCl, 40 mmol/l) and diclofenac (10 μmol/l) to reveal NO-mediated responses. (C,D) Experiments were performed in the presence of Nω-nitro-l-arginine (+l-NA, 300 μmol/l) and diclofenac (10 μmol/l) to reveal the contribution of EDHF to endothelium-derived relaxation. Results are the mean±S.E.M., n = 7 in each group, P<0.05 versus control.

Relaxation and hyperpolarization to KCl, which can mimic EDHF-mediated responses in some vascular beds [14], was increased 36 h after a single T3 application (Fig. 4A,B) but attenuated in arteries from the animals receiving T3 for 8 weeks (Fig. 4C,D). These results indicate that different mechanisms underlie the enhanced endothelium-dependent relaxation of renal artery rings from rats treated with T3 for 36 h versus 8 weeks.

Fig. 4

Effect of in vivo T3 treatment on the KCl-induced hyperpolarization and relaxation of renal artery rings (A,C) Concentration–relaxation curves to KCl (4–8 mmol/l) and (B,D) effect of KCl (4–8 mmol/l) on the membrane potential (MP) of isolated endothelium-intact renal artery rings in the presence of phenylephrine (concentration adjusted to 80% of max. KCl-induced contraction), Nω-nitro-l-arginine (300 μmol/l) and diclofenac (10 μmol/l). (A,B) Comparison of responses obtained in rings from solvent (CTL, closed symbols and bars)- and short-term T3-treated (open symbols and bars) animals. (C,D) Comparison of responses obtained in rings from solvent (CTL, closed symbols and bars)- and long-term T3-treated (open symbols and bars) animals. Results are the mean±S.E.M., n = 6 in each group, *P<0.05 versus control.

Eight weeks of T3 treatment led to a marked increase in eNOS expression in the aorta (eNOS/PECAM-1 protein levels were increased by 463±68% relative to control, n = 4, P<0.01), without affecting the expression of the soluble guanylyl cyclase (Fig. 5, n = 4, P = ns). A single application of T3 did not increase eNOS expression, as determined by Western blot analysis (data not shown).

Fig. 5

Expression of eNOS and the α and β subunits of the soluble guanylyl cyclase in aortae from solvent and long-term T3-treated rats. Representative Western blots showing the expression of eNOS and the soluble guanylyl cyclase sub-units α1 (sGC α1) and β1 (sGC β1) in the Triton-X100 soluble fraction of rat aortic rings from animals subjected to long-term T3 treatment (T3, 300 μg/kg every other day) and from the control group (CTL).

As cyclic AMP is reported to enhance EDHF-mediated effects [15], we assessed the role of cyclic AMP in the ACh-induced relaxation of arteries from rats treated with T3 for 36 h. Cyclic AMP levels were increased by approximately twofold in abdominal aortic rings obtained from T3-treated compared to solvent-treated animals. Similar increases in cyclic AMP levels were recorded in the absence and in the presence of a phosphodiesterase inhibitor (Fig. 6A). Consistent with a potentiating role of cyclic AMP/protein kinase A activation on EDHF-mediated responses, the pre-treatment of renal artery rings from solvent-treated rats with the protein kinase A activator, Sp-5,6-DCI-cBIMPS (30 μmol/l), significantly increased EDHF-mediated relaxation (Fig. 6B). Concentration–relaxation curves to Sp-5,6-DCI-cBIMPS were significantly shifted to the right in vessels from T3-treated animals (Fig. 6C). Moreover, the protein kinase A inhibitor, Rp-8-Br-cAMPS (10 μmol/l), significantly impaired EDHF-mediated relaxations in renal artery rings from T3-treated animals. In the presence of Rp-8-Br-cAMPS relaxations of renal arteries from T3- and sham-treated animals were no longer different (Fig. 6D).

Fig. 6

Evidence for the involvement of cyclic AMP and protein kinase A in the altered endothelium-dependent relaxation of renal arteries from short-term T3-treated rats. (A) Cyclic AMP levels in aortic rings from rats of the control group (CTL, open bars) and the short-term T3-treated group (T3, 300 μg/kg, 36 h) with or without pre-treatment with isobuthyl methylxanthine (IBMX; 500 μmol/l, 30 min); n = 6, *P<0.05. (B) Endothelium-dependent relaxation to acetylcholine (ACh) of rat renal artery rings from control animals (CTL) in the presence of diclofenac (10 μmol/l) and Nω-nitro-l-arginine (300 μmol/l) and presence (♦) or absence (○) of the protein kinase A activator Sp-5,6-DCI-cBIMPS (30 μM). The concentration of phenylephrine was adjusted to obtain identical levels of precontraction; n = 7, P<0.05. (C) Relaxation–response curves to Sp-5,6-DCI-cBIMPS in renal artery rings from the control group (CTL, ▪) and short-term T3 treated group (T3 36 h, ▵), n = 8, P<0.05. (D) Concentration–relaxation curves to acetylcholine (ACh) in the presence of diclofenac (10 μmol/l) and Nω-nitro-l-arginine (300 μmol/l) in renal artery rings from the control group (CTL, ▪) and short-term T3-treated group (T3 36 h, ▵) in the presence of the protein kinase A inhibitor Rp-8-Br-cAMPS (10 μmol/l), n = 7, P = ns, all data are mean±S.E.M.

4 Discussion

The results of the present study demonstrate that in vivo treatment with the thyroid hormone, triiodothyronine (T3), enhances the endothelium-dependent relaxation of the isolated rat renal artery via two potentially different mechanisms. Within 36 h after T3 application, aortic cyclic AMP content was increased, and was associated with enhanced EDHF- and NO-mediated relaxation in the renal artery. Following chronic treatment with T3, most probably as a consequence of an elevated fluid shear stress at the endothelial cell surface, eNOS expression was markedly enhanced in the aorta and was associated with an increase in NO-mediated relaxation in the renal artery. In the same tissues, EDHF-mediated relaxation was attenuated.

The most prominent effect of prolonged hyperthyroidism on the vascular system is a marked reduction in peripheral resistance, which is associated with attenuated responsiveness to vasoconstrictor catecholamines [1,16]. Although the direct application of thyroid hormone (TH) to isolated arteries is reported to elicit relaxation by inhibiting the myosin light chain kinase [17], this effect requires supra-physiological concentrations of the hormone and is not observed in vivo [2,18]. A number of experimental observations indicate that the TH-induced decrease in vascular resistance can be attributed to the elevated production of NO. Indeed, inhibition of NOS restores normal constrictor responsiveness in arteries from hyperthyroid rats [10,16,19], and the conversion of l-arginine to l-citrulline is increased in the heart, kidney and vessels from hyperthyroid rats [20]. In the latter study however, pharmacologically-induced hypothyroidism also led to an increase in NOS activity, and the authors speculated that, depending on the TH state, different NOS isoforms may be involved. In the present study we observed that the expression of eNOS in the rat aorta was significantly enhanced after 8 weeks treatment with T3, a phenomenon which was associated with an increased ACh-induced NO-mediated relaxation. Although it is only possible to speculate about the mechanisms responsible for the up-regulation of eNOS in an in vivo model, the experimental data available to date indicate that an increase in fluid shear stress may underlie this effect. Indeed, it is generally accepted that shear stress regulates eNOS expression [21,22] and increased eNOS levels have been reported in in vivo models of high vascular shear stress induced by arterio-venous fistulae [23] or chronic exercise [24]. Although it is not possible to exclude a direct effect of circulating T3 on eNOS expression, eNOS levels in cultured endothelial cells tend to decrease rather than increase in response to incubation with T3 for up to 72 h (authors unpublished observations, 2002). As the complete renal artery was required for the organ bath and electrophysiological studies, eNOS protein expression was determined in the aorta. Although this is a potential flaw in the experimental design, the marked increase in endothelium-dependent NO-mediated relaxation of the renal artery after prolonged T3 treatment makes it tempting to suggest that eNOS level would also be upregulated in the renal artery.

The smooth muscle cells in renal arteries from animals treated with T3 for 8 weeks were significantly hyperpolarized relative to the control group, and this phenomenon may also be the consequence of a chronic elevation in shear stress, as this stimulus is known to enhance the expression of potassium channels on endothelial cells [25]. Indeed THs are reported to induce many proteins involved in the control of the membrane potential. In cardiac myocytes for example, TH increases the expression of the voltage-dependent K+-channel, Kv1.5 [26] and the hyperpolarization-activated cyclic nucleotide-gated channel, HCN2 [12]. Furthermore, TH enhances the expression of the Na–K-ATPase expression in cardiac tissue [27–29], as well as in glia [30] and mesangial cells [16].

The activity of the Na–K-ATPase plays a central role as an effector for EDHF-mediated hyperpolarization and relaxation in several vascular beds and low concentrations of potassium, which activate the Na–K-ATPase on vascular smooth muscle cells mimic EDHF-mediated responses [31]. Preventing the activation of the Na–K-ATPase, using low concentrations of ouabain, inhibits EDHF-mediated hyperpolarization and relaxation in the rat [14,32]. In the present investigation, we observed that hyperthyroidism for 8 weeks attenuated rather than enhanced EDHF-mediated relaxations, a result that was unexpected given that THs are generally accepted to enhance Na–K-ATPase expression. Preliminary experiments, however, indicate that T3 treatment for 8 weeks had no substantial influence on the expression of the mRNA encoding the α1, α2, α3 or γ subunits of the vascular Na–K-ATPase (Fisslthaler, 2002, unpublished observation). Nevertheless, vessels from these animals were markedly hyperpolarized under resting conditions. Consequently, the agonist-induced activation of the Na–K-ATPase will probably result in a smaller change in the membrane potential of vascular smooth muscle cells. Alternatively, it has been observed previously that NO has an inhibitory effect on EDHF-mediated responses [33,34]. In vessels from the rat, this might be a consequence of a NO-mediated inhibition of the Na–K-ATPase [35,36].

In contrast to the changes observed in renal arteries from T3-treated rats for 8 weeks, eNOS expression and resting membrane potential were not altered 36 h after a single injection of T3. Nevertheless, a marked increase in both NO- and EDHF-mediated relaxations in the renal artery was observed, and was associated with increased aortic levels of cyclic AMP. Although we have concentrated on the effects of this nucleotide on EDHF-mediated relaxation, cyclic AMP [37], and the subsequent activation of protein kinase A and phosphorylation of eNOS [38,39] has been shown to increase the activity of eNOS in cultured endothelial cells, which might be a potential explanation of the improvement of NO-mediated relaxation in the present study. With regard to EDHF, we and others have shown that cyclic AMP, via a pathway involving protein kinase A, has an important permissive effect on responses evoked by the vasodilator [15,40,41]. Although we have not addressed the mechanisms leading to the T3-induced increase in cyclic AMP levels in vessels from animals studied 36 h after injection of T3 and have measured cyclic AMP in the aorta only, several lines of evidence suggest that cyclic AMP and the protein kinase A underlie the enhanced EDHF-mediated relaxation observed in the renal artery. First of all, T3 treatment (36 h) induced a rightward shift in the concentration–relaxation curve to a cyclic AMP analogue indicating that the increase in cyclic AMP following T3 exposure was sufficient to activate protein kinase A. More importantly, inhibition of protein kinase A blocked the T3-induced increase in EDHF-mediated relaxation. In order to demonstrate that activation of protein kinase A increases EDHF-mediated responses, vessels were incubated with a cyclic AMP analogue and subsequently contracted to the same levels as control vessels. This treatment induced a marked increase in EDHF-mediated relaxation. Taken together these data suggest that 36 h of T3 increases vascular cyclic AMP levels, which lead to enhanced endothelium-dependent relaxation. Nevertheless, other so far unidentified factors may also be involved in the enhancement of EDHF-mediated relaxation following T3 treatment as a cyclic AMP analogue was less effective in enhancing EDHF-mediated relaxation than TH.

There is controversy regarding the effects of THs on cyclic AMP and although THs have been reported to increase the isoproterenol-induced accumulation of cyclic AMP in cardiomyocytes and small mesenteric arteries of the rat [42,43], THs decrease the expression of the forskolin—but not the isoproterenol-sensitive isoforms of adenylyl cyclase in the rat heart [44]. In cultured rat aortic smooth muscle cells, T3 stimulates prostacyclin production, which would be expected to stimulate the generation of cyclic AMP formation [45]. This mechanism, however, cannot account for the observations made during the present investigation, as all experiments were performed in the presence of the cyclooxygenase inhibitor diclofenac.

In conclusion, we have demonstrated that hyperthyroidism enhances endothelium-dependent relaxation in the renal artery most likely by two distinct mechanisms; initially via a cyclic AMP-mediated increase in EDHF—as well as NO-mediated relaxation and subsequently via the up-regulation of eNOS.


We are indebted to Sina Bätz, Ingrid Kempter and Isabel Winter for excellent technical assistance. This study was supported by the Deutsche Gesellschaft für Kardiologie-Herz- und Kreislaufforschung and the Else Kröner-Fresenius-Stiftung (stipend to E.B.) and a research grant from the Institut de Recherches Internationales Servier.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
View Abstract