madman
Super Moderator
Androgens and Non-Genomic vascular responses in hypertension (2022)
Mercedes Perusquía
ABSTRACT
Arterial hypertension is a global public health concern. In the last few years, the interest in androgen deficiency has been growing, and the association between androgens and high blood pressure (BP) is still controversial. One purpose of this review was to summarize the available findings in order to clarify whether male sex steroid hormones have beneficial or harmful effects on BP. The second purpose was to enhance the recognition of the acute non-genomic sex–independent vasorelaxing effect of androgens. Remarkably, BP variation is expected to be a consequence of the androgen-induced vasorelaxation which reduces systemic BP; hence the in vivo vasodepressor, hypotensive, and antihypertensive responses of androgens were also analyzed. This article reviews the current understanding of the physiological regulation of vascular smooth muscle contractility by androgens. Additionally, it summarizes older and more recent data on androgens, and some of the possible underlying mechanisms of relaxation, structural-functional differences in the androgen molecules, and their designing ability to induce vasorelaxation. The clinical relevance of these findings in terms of designing future therapeutics mainly the 5-reduced metabolite of testosterone, 5β-dihydrotestosterone, is also highlighted. Literature collected through a PubMed database search, as well as our experimental work, was used for the present review.
1. Introduction
1.1. General considerations of androgens
Steroids are important in biology, chemistry, and medicine. These compounds are divided into two different classes: natural or synthetic biologically active organic compounds, with a molecular structure of 17 carbon atoms arranged in four fused rings (A, B, C, and D) and called “cyclopentenoperhydrophenanthrene” They are found in plants, animals and fungi, and they regulate several biological functions. Sex steroid hormones can be grouped into three families based on their number of carbon atoms: pregnanes (21), androstanes (19), and estranes (18). Therefore, male sex steroids (androgens) are also classified as C19-steroids which are traditionally considered to maintain male sexual characteristics, although their relevant biological roles in the bone, muscle, prostate, and adipose tissue and the reproductive, cardiovascular, immune, neural, and hematopoietic systems have also been documented [1,2]. In this review, it has been highlighted the physiological effects of C19-steroids (androgens) on the cardiovascular system which have previously been scarcely considered.
It is important to take into account the fact that androgens are metabolized into different steroids each with its own variable effects and potency. Fig. 1 summarizes the classical gonadal pathway biosynthesis of androgens and the enzymes involved therein. In males, the testes synthesize testosterone (TES) from dehydroepiandrosterone (DHEA) and androstenedione, which is then mainly converted into 5αand 5β-dihydrotestosterone (5α-DHT and 5β-DHT, respectively) in target tissues, such as the prostate and seminal vesicles. In females, TES produced from the theca cells is predominantly converted into estrogens in the granulosa cells of the ovary (Fig. 1) (for details, see [3]). The major androgens produced by adult humans in decreasing order of magnitude are DHEA, androstenedione, TES, and 5α- and 5β-DHT. It is important to emphasize that androgens are not exclusive to males and estrogens are not exclusive to females; both sexes, male and female produce androgens and estrogens but their proportions are different. Thus, the plasma concentration of TES in males is 10–35 nM [4] and 0.7–2 nM in females [5].
However, little attention has been paid to the metabolism and changes in the chemical structure of androgens. TES is reduced to 5αDHT, which is considered the most potent androgen [6–9]; however, its 5β-reduced isomer (5β-DHT) has been mistakenly considered biologically inactive which will be discussed later.
1.2. Genomic vs non-genomic mechanisms of androgen action
Androgens may modify cellular function through both (i) their well-known classical genomic action and (ii) their more recently identified non-genomic mechanisms of action.
The genomic mechanism of action has been extensively studied. It regulates many of the most known biological functions such as secondary male sexual characteristics, sexual behavior, spermatogenesis, increment of muscle mass, loss of adipose tissue, prostate growth, prevention of osteoporosis, and roles in the hematopoietic, immune, and nervous systems, and in skin hair follicles, emotions, and libido [1,2]. Classically, androgens may mediate their biological responses by binding to the androgen receptor (AR). ARs are expressed in many tissues, with the highest concentrations occurring in male reproductive organs [10]. Briefly, the AR signaling pathway for genomic action signaling involves androgens crossing the plasma membrane, entering the cytoplasm, and binding to the AR (androgen-receptor complex), resulting in the dissociation of chaperone proteins and translocation of the complex to the nucleus where it dimerizes and binds to the androgen response element (ARE) for modulating gene transcription, and leads to protein synthesis creating a biological response, which has been extensively studied (for review see [11–13]). This classical androgen response is relatively slow, occurring over a period of hours, weeks, or years (Fig. 2).
In contrast to the genomic pathway, the non-genomic mechanism of action began to be recognized much later, roughly 40 years ago. Different steroids, including androgens, can also act via this mechanism, which is independent of the ligand-dependent transactivation function of nuclear receptors [14,15]. This is known as ‘non-genomic’ signaling, which typically occurs within a short time frame (seconds to minutes) (Fig. 2) [16,17]. Fig. 2 summarizes the two mechanisms of androgen action.
The effects mediated by the non-genomic type of action have been observed in most excitable tissues, such as the central nervous system and reproductive and non-reproductive smooth muscles, among others. Several studies have found that the majority of the physiological nongenomic androgen responses occur through membrane- or membrane-associated AR/binding proteins, which would decrease cell membrane fluidity, changes in intracellular calcium concentration ([Ca2+]i), activation of the second messenger pathway, or a combination of the above mechanisms (reviewed in [18–21]). Notably, non-genomic mechanisms can act independently or be associated with other genomic effects in order to elicit a physiological response, i.e., the genomic and non-genomic pathways of androgen action are likely interlinked.
1.3. Classical effects of androgens
1.3.1. Reproductive effects
As mentioned before, the major biological actions of TES include the development and maintenance of genetically mediated secondary sexual characteristics in males. These effects are mediated directly by TES and its two major active metabolites, 5α-DHT and 17β-estradiol. The genomic action, mediated by AR, of TES and 5α-DHT is responsible for sexual differentiation and virilization. Likewise, these androgens play an important role in the deepening of the voice, increase in sebum production, and an enlargement of the external genitalia, including penile length. The conversion of TES to 5α-DHT, but not to 5β-DHT, in the prostate is essential for its growth and development, and, importantly the concentration of this 5α-reduced metabolite is 10 times higher than that of TES. Similarly, erectile dysfunction, hypogonadism, and metabolic syndrome have all been associated with TES deficiency. In addition to the masculinizing effects of androgens, they also have anabolic properties. TES is known for producing muscle hypertrophy by increasing fractional muscle protein synthesis [22,23]. The correct term for these steroids is anabolic–androgenic steroids (AAS): “anabolic” refers to muscle building and “androgenic” refers to increased male sex characteristics. AAS stimulates growth in many types of tissues, especially bone and muscle, and their effects involve binding to the AR and therefore genomically mediated. AAS has been used since the 1950 s in pharmacological doses (administered orally and by several parenteral ways) in an attempt to maximize the well-known anabolic effects of TES, reduce the rate of its hepatic inactivation, and decrease its aromatization to 17β–estradiol [24]. AAS as TES derivates and analogs include the endogenous sex male steroids i.e., TES and 5α-DHT, as well as other agents and synthetics that behave like these androgens, some of which are nandrolone, decanoate, oxandrolone, oxymetholone, stanozolol, TES cypionate, and TES enanthate.
1.3.2. Effects of androgens on reproductive and non-reproductive smooth muscles
It has been demonstrated that androgens target reproductive and non-reproductive smooth muscles (SM) by inducing non-genomically mediated relaxation.
In urogenital SM, the uterine contractility is notably relaxed by androgens with a higher utero-relaxant potency than progestins [25–27]. Therefore, the uterus is a target organ for androgens that induce myometrial quiescence, which plays a key functional role in the maintenance of normal gestation. Androgens also lead to marked relaxation of barium-induced contraction of the rat epididymis and seminal vesicles [28]. The relaxing effects of TES were also confirmed in rabbit seminal vesicles [29]. The neck of the urinary bladder is also a target of androgen-induced relaxation; thus, TES and 5α-DHT have been found to elicit similar concentration-dependent relaxation in both male and female pigs [30,31]. In summary, the acute relaxing effect of androgens in other types of male urogenital SM has rarely been studied.
In gastrointestinal SM, progesterone, 17β-estradiol, and TES are also capable of inhibiting the contractibility of intestinal SMs such as those in the ileum [32,33]. In comparison with female steroids, the high potency of androgens in inducing non-genomic gastrointestinal relaxation was recognized [33,34]. TES and 5α-DHT have also been found to be capable of inducing similar levels of relaxation in gallbladder SMs in male guinea pigs [34]. Similarly, several androgens (DHEA, TES, and its 5-reduced metabolites) are capable of inducing non-genomic bronchorelaxation by acting on the airway SMs as well as inducing genomic anti-inflammatory responses, suggesting a potential therapeutic use in the treatment of severe asthma (reviewed in [35]). A substantial point of interest in this review is that vascular SM (VSM) is also a target of TES and its metabolites in acute non-genomic relaxation. In particular, in the last few years, there has been a growing interest in androgen-induced vasorelaxation, which is discussed in detail in Section 2.
*Therefore, the available data indicate that multiple types of SM are targets for androgen-induced relaxation and that androgens may be important in the regulation of SM excitability in general.
2. Vascular effects of androgens
2.1. Vasorelaxation/vasodilation of isolated vessel studies
It has been widely accepted that estrogens exert protective effects on female vasculature [36,37]. Recent clinical and experimental animal studies have revealed that TES, the primary and most well-studied androgen, directly protects against risk factors for cardiovascular diseases (CVD), such as atherosclerosis and hypertension, and reduces the incidence of CVD, such as cardiac failure and myocardial ischemia [38–44]. Likewise, several clinical trials have pointed out that androgen deficiency seems to exacerbate many risk factors and pathologies associated with cardiovascular and metabolic diseases [41,45–50]. Based on these findings, the protective effect of TES on the cardiovascular system is gaining recognition. In this context, the main purpose of the present review article is to show the direct and rapid non-genomic effects at the vascular level, particularly in the regulation of blood pressure (BP).
Sex steroid hormones, including androgens, mediate vasorelaxation. Particularly regarding androgens, in the last 25 years, there has been a plethora of evidence that TES is capable of inducing vasorelaxation. The acute vasorelaxing effect of TES on contractions induced by depolarization, or several agonists in a number of isolated vascular beds of various species of mammals including humans (revised in [18,20,21,51–54]).
Notably, the acute concentration-dependent vasorelaxing effects of androgens have been reported in isolated human blood vessels, including the internal mammary artery [55], umbilical arteries [51,56], and pulmonary arteries and veins [57]. TES and 5α-DHT have also been reported to induce vasodilating effects on the penile and cavernosal arteries [58–60], and internal spermatic veins [61], the lack of vasodilation may cause non-psychogenic erectile dysfunction associated with hypoandrogenism (androgen deficiency). Moreover, androgens have been shown to regulate prostatic blood flow [62]. As reviewed by Traish et al. [63], TES–propionate has been used for the last six decades to treat sexual issues to augment penile vasodilation.
TES has direct effects on vascular tone causing vasodilatation, and an increase in coronary blood flow [42,43]; thus, it has been observed that the coronary circulation is more sensitive to the vasorelaxant effect of TES than large vessels with a benefit on time to cardiac ischemia. Notably, clinical studies have highlighted the beneficial effects of TES therapy on exercise-induced cardiac ischemia, an effect mediated by coronary vasodilation [64], which persists for 3 months and indeed 12 months [65].
*Together, these findings reveal that TES exerts beneficial effects on the cardiovascular system by reducing vascular tone, which involves direct vasodilatory action on the peripheral vasculature and reduction of BP. For their acute and marked vasorelaxing effects, the male sex steroids have been named “vasoactive androgens”.
2.2. Systemic hypotensive actions
The vasorelaxing effect induced by androgens implies an increase in blood flow, and therefore, a reduction in BP.
High blood pressure (hypertension), sometimes called “the silent killer,” is a serious medical condition and a global public health problem. Hypertension is very common in older people and is a major health concern. The pathogenesis of essential hypertension, the most common type of hypertension, is multifactorial and highly complex. Factors that play an important role in this pathology include genetics, activation of the neurohormonal systems, such as the sympathetic nervous system and renin-angiotensin-aldosterone system, obesity, and increased dietary salt intake. However, the etiology of hypertension remains unknown.
Sex steroid hormones have also been associated with hypertension, however, there is a polemic regarding whether female sex steroids are beneficial and male sex steroids are detrimental to BP. Male sex is considered a major contributing risk for CVD. A direct effect of androgens on promoting hypertension has been hypothesized [66] and was attributed to the myths that TES is capable of inducing adverse effects on the cardiovascular system. One of these effects is an increase in BP.
The theory of increased hypertension risk in response to TES remains largely unsubstantiated. In view of the non-genomic vasorelaxing properties of androgens, it is difficult to consider that TES can be detrimental to the cardiovascular system, since TES, the most important male sex endogenous hormone, plays a vital role in the regulation of biological processes in males. This review aims to clarify the reasons for this uncertainty.
Several lines of evidence, contradict the traditional view of the adverse effects of androgens on BP. The first evidence that androgens are associated with the regulation of BP through BP reduction was collaterally observed in classic studies by Selye, when the anesthetic effect of steroids was reported in the early 1940 s [67,68]. Later, a body of information associated low circulating levels of TES with cardiovascular dysfunction, particularly with hypertension [43,69–72]. This finding was confirmed in several reports indicating that low concentrations of TES in men are associated with higher BP [73–76]. Other studies have suggested that a low TES concentration in men may be a novel risk factor for hypertension or CVD. An important clinical study documented that acute administration of TES in humans with cardiac failure decreases peripheral vascular resistance and improves cardiac output [77]. An observational study in men showed that long-term TES therapy, for up to 8 years, produced marked and sustained gradual decreases in systolic and diastolic BP [78]. One recent study showed that hypertensive men had lower levels of free TES, and bioavailable TES, compared to normotensive men [79].
More recently, we reported important in vivo experimental research on animals. It is important to highlight that the consequence of androgen-induced vasorelaxation, widely discussed in the present review, is a decrease in BP. In this regard, numerous studies have suggested that low BP is linked to androgens (reviewed in [54]). We first reported that some TES metabolites are capable of blocking vasopressor responses to noradrenaline or a Ca2+ channel agonist (Bay K 8644) in anesthetized vagosympathectomized, pithed, Wistar male rats [80]. Our recent in vivo experimental research has also demonstrated that BP regulation is an expected consequence of androgen-induced vasorelaxation [81–84]. We documented that TES and its 5β–reduced metabolite (5β-DHT) induced a marked reduction in BP in both conscious, normotensive, Sprague-Dawley rats and testicular feminized male rats (Tfm; androgen receptor-deficient) [81]. Afterward, we reported that androgens can also elicit a significant antihypertensive response in conscious, spontaneously hypertensive, male rats (SHR) and in normotensive WKY rats who underwent TES deprivation by orchiectomy for inducing hypertension, which was then prevented by TES replacement therapy [82]. Similarly, it was demonstrated that in an in vivo rat model of preeclampsia, the elevated mean arterial BP (MAP) was reduced significantly by an intravenous bolus injection of DHEA, TES, or its 5-reduced metabolites (5α- and 5β-DHT) [83]; notably, this study indicates that androgens are also antihypertensive in females. More recently, a model for studying androgen deficiency in conscious hypertensive rats caused by orchiectomy helped us to understand the progression of hypertension in aging men since our experiments clearly demonstrated that hypertension is a disease associated with aging. We also observed that the intravenous bolus administration of DHEA, TES, or 5β-DHT is capable of inducing antihypertensive responses in hypertensive aging rats [84]. The above information is summarized in Table 1.
Notably, TES is an acute vasorelaxant in human subcutaneous resistance arteries [85]. In addition, functional studies have also shown that the vasorelaxing potency of 5β–DHT is higher in the mesenteric artery than in the thoracic aorta of hypertensive and normotensive rats [86]. These findings suggested that TES and particularly 5β-DHT regulate vascular resistance in the mesenteric arterial bed, explaining its outstanding antihypertensive response.
It seems that physiological androgen levels regulate BP by placing a “brake” for developing hypertension, and indeed hypotestosteronemia is a risk factor for hypertension. Several studies have reported that men with CVD, type 2 diabetes mellitus, obesity, metabolic syndrome, and dyslipidemia have low levels of TES (reviewed in [46,53,71,87,88]). In this respect, the incidence of CVD increases in aging men when TES production decreases. In line with these findings, it has been suggested that hypertension, comorbidity highly susceptible to increased COVID19 severity, may be due to lower androgen levels in older men patients [89]. These findings indicate that TES is beneficial for the cardiovascular system.
*In parallel, these findings have also shown that DHEA and 5β-DHT are significantly more potent than TES in evoking an antihypertensive response, indicating that androgens may provide a therapeutic potential for the regulation of BP.
2.3. Mechanism of vasorelaxing action of androgens
2.3.1. Non-genomic action
The action of androgens in inducing vasorelaxation has been defined as follows: androgen-induced vasorelaxation is rapid (~2 min). Likewise, several lines of pharmacological evidence have established that this immediate effect induced by androgens is non-genomically mediated since the vasorelaxing effect of androgens is observed in the presence of AR antagonists, protein synthesis, and transcription inhibitors [51,52,90,91]. The response is observed even when the androgen is conjugated to molecules, such as bovine serum albumin (BSA), which prevents it from entering the cytoplasm [92], and the AR deficient testicular-feminized rat inhibitors [90]. Indeed, these data taken together indicate that the mechanism of action of androgens is unlikely to involve nuclear receptor activation.
2.3.2. Endothelium- and nitric oxide (NO)-independent mechanism
Numerous studies have reported that the vasorelaxing effect caused by 17β-estradiol is mediated in part by via generation of endotheliumderived NO and is attenuated by NO synthesis inhibitors, suggesting an endothelium-dependent mechanism [37], revised in [93] but this seems not to be the case for androgens. Androgen-induced vasorelaxation is observed in endothelium-denuded vascular preparations, as well as in the presence of NO synthase inhibitors such as L-NAME [43,51,52,94–97], as revised by [20]. These findings indicate that androgen-induced vasorelaxation is endothelium–independent. In this respect, it is relevant to consider the following convincing facts: (i) androgens are powerful vasorelaxants in the human umbilical artery [51], notably, this blood vessel has an atypical endothelium due to the fact that biosynthesis of endothelial NO is absent; and (ii) androgen-induced antihypertensive responses in preeclamptic rats [83]. Clinical and experimental studies have suggested that endothelial dysfunction is present in preeclamptic subjects [98–102]. In summary, the vasorelaxing effect of androgens strongly suggests an NO–independent mechanism [51].
In marked contrast, some contradictory data have reported that androgen-induced vasorelaxation is partially, but not completely, inhibited through pretreatment with L-NAME [103–106], suggesting that a partial endothelium-dependent mechanism might be involved. A study also documented that vasodilatation to TES in human pulmonary arteries is dependent upon the presence of an intact endothelium [57]. However, this contradiction may be related to differences in the vessel type or experimental animal species and/or, as previously suggested by Jones and Kelly [88], can also be explained by the fact that TES is converted to 17β-estradiol (see Fig. 1); an estrogen that mediates its vasorelaxing effect through stimulation of NO production, it is currently not known whether the reported vasorelaxing effect of TES is a direct endothelium-dependent effect or an indirect effect via estrogens.
*Although these data suggest that the endothelium is not involved in androgen-induced vasorelaxation, further investigations are needed to clarify this controversy.
2.3.3. Ion channels
We provided the first experimental evidence that the mechanism of vasorelaxation of androgens was due to the blockage of voltage- and receptor-dependent Ca2+ channels, a mechanism that restricts the availability of extracellular Ca2+ in the contractile machinery [94,95]. Vasorelaxation of 17β-estradiol, progesterone, and TES, inhibited the entry of Ca2+ but not its release [107]. More specifically, in single vascular myocytes, it has been shown that the mechanism of action of androgens in inducing vasorelaxation depends on the blockade of L–voltage-operated Ca2+ channels (L-VOCCs) [91,97,108–110], revised by [88]. A compelling fact is that TES acts on the same site as dihydropyridines (nifedipine), the α1C subunit of the L-VOCCs, in order to block the extracellular Ca2+ entry into intracellular space [110]. We also found that 5β-DHT acts exclusively as an antagonist to blocking Ca2+ currents via L-VOCCs (from nM to µM concentrations), whereas TES possesses different mechanisms: (i) a dual effect on L-VOCCs (antagonist at nM concentrations/agonist at µM concentrations), (ii) an increase in [Ca2+]i and (iii) an increase in cAMP production. This is the first evidence that can help us understand the mechanism underlying the acute vasorelaxing action of 5β-DHT and the controversial response of TES upon vascular reactivity, i.e., the vasorelaxing and/or antivasorelaxing effects [91].
Several functional studies in VSM and other SMs demonstrated the TES may also relax contractions induced by KCl and several contractile agonists, such as acetylcholine, carbachol, noradrenaline, oxytocin, ovalbumin, phenylephrine, prostaglandins, and serotonin, among others, it cannot be ruled out that androgens may also inactive non-voltage-gated pathways, i.e. the store-operated Ca2+ entry (SOCE) and receptor-operated Ca2+ channels (ROCCs), suggesting that TES has an antagonistic effect on L-VOCCs, ROCCs, and/or SOCE, showing that the vasorelaxation is induced by blockade of extracellular Ca2+ entry through these channels. In this respect, in addition, there are also the experimental findings that in the bronchodilating effect of androgens in airway SM [35] SOCE and ROCC are partly involved.
In addition to the above, a human study examined the effect of TES therapy on vascular reactivity in isolated human subcutaneous arteries it was observed that after TES therapy, the vasoconstrictor effect of noradrenaline increased and reduced the vasodilator response to acetylcholine and sodium nitroprusside (which breaks down to NO in the circulation). This effect physiologically would divert blood flow toward the vital organs, an important survival effect in chronic heart failure [85], revised in [88] which is evidence for a different mode of action of test on vascular reactivity.
.
In addition to Ca2+ channels, it has been repeatedly documented that in functional studies in some vascular beds, TES-induced vasorelaxation may also activate several types of K+ channels, which are: voltage-sensitive K+ channels (KV), large-conductance Ca2+-activated K+ channels (BKCa), ATP-sensitive K+ channels (KATP), and inward-rectifier K+ channels (KIR) [56,92,96,103,111]. One study through patch-clamp experiments reported that TES produced activation of BKCa channels [96]. In conclusion, the main K+ channels involved in this process are BKCa and Kv.
With these findings, there is controversy. The most intriguing conclusions, which can be drawn from the numerous observations of nongenomic vasorelaxing mechanisms are as follows: either (i) the mechanism differs depending on blood vessels [52]; or (ii) in agreement with [21], androgen-induced vasorelaxation is due to the fact that Ca2+ channels are closely related to K+ channels, where membrane hyperpolarization by K+ channel activation closes the Ca2+ channels and leads to vasorelaxation.
However, little attention has been paid to the in vivo mechanism of action. The mechanism of androgen action in eliciting antihypertensive responses was studied for the first time in conscious (in vivo) rats. 5βDHT abolished the pressor response to Bay K 8644 (an L-VOCC antagonist) in vivo, suggesting a blockade of L-VOCC [82]. Our data in conscious orchidectomized hypertensive rats revealed that: (i) the blockade of NO synthase by L-NAME did not alter the antihypertensive response to 5β-DHT; and (ii) endothelin-1, by binding to ETA receptors, interferes with NO synthesis and acts to counter NO-induced vasodilation; under these circumstances, dose-dependent antihypertension induced by 5β-DHT was not abolished in the presence of L-NAME or endothelin-1 [84].
Fig. 3 illustrates the mechanisms of action of androgens described here to produce vasorelaxation/hypotension.
3. Differences between male and female steroid hormones in the cardiovascular system
It is relevant to note that when the efficacies and potencies of androgen-induced vasorelaxation were compared to female sex steroids, it was observed that TES and its two immediate dihydrometabolites, 5αand 5β-DHT, but not TES, have turned out to be more potent vasorelaxants than the female steroids, estrogens, and progestins in inducing vasorelaxation (see section 5). Our early observations were that 5β-DHT was the most potent steroid, even when compared to the vasorelaxing response of progestins in the rat aorta [94]. Remarkably, the concentration-response curve to 5β-DHT exhibited the highest potency for vasorelaxation of the human umbilical artery, even exceeding the potency of several progestins [51]. This is probably the only published report comparing the efficacy and potency of TES and its 5-reduced metabolites with progesterone and its 5-reduced progestins.
Regarding estrogens, the concentration-response curves showed that 5β-DHT was significantly more potent than 17β-estradiol in inducing vasorelaxation in rat aorta [95]. In the pulmonary vasculature, the vasorelaxing efficacy of 17β-estradiol was significantly lower than that of TES [47]. In patch-clamp electrophysiological studies, the concentration-response curves to TES, 5β-DHT, and 17β-estradiol on Ca2+ currents in single aortic myocytes from rats indicated that androgens were more potent in blocking Ca2+ currents than 17β-estradiol [91].
Despite these findings, studies by Khalil and colleagues demonstrated that progesterone and TES are less potent than 17β-estradiol in inducing vasorelaxation [107,112,113]. In this respect, it is important to note that TES is a less potent vasorelaxant than DHEA and its 5β-reduced metabolite; thus, the roles of TES have been inappropriately generalized as common to all androgens (see section 5)
*However, in a general context, androgens, particularly 5β-DHT, have turned out to be more potent vasorelaxants than female steroids, estrogens, and progestins.
4. Sex differences in vasorelaxing effects of androgens
Historically, female sex steroid hormones (estrogens and progestins) have been studied extensively, which has led to a substantial understanding of human female physiology and the diverse effects of these hormones on physiological, as well as reproductive functions.
A huge body of evidence has created the wrong belief that estrogens are cardioprotectors in females and that androgens are detrimental to the male cardiovascular system. Recently, the interest in the role of sex regulation of metabolic homeostasis, BP, and CVD has been growing. Men typically develop CVD earlier than women [66]; however, it is important to consider that very few studies have examined androgen signaling and hypertension in females.
Some studies have suggested that androgens are as important for BP regulation in females as they are in males. The available data suggest that in vitro relaxation induced by TES is independent of sex. No significant differences between the vasorelaxing effect of TES in the coronary arteries and aortas of male vs female rabbits were observed [90] or in the aortas of male vs female rats [114]. TES has also been reported to induce sex-independent vasodilation in vivo in canine coronary conductance and resistance arteries [104]. TES elicits vasorelaxation in both males and females, however, rat pulmonary and coronary arteries in males were more sensitive to the effects of TES than those in females [47].
In addition, androgens also possess the ability to relax the female human vasculature, such as the internal mammary artery [55], pulmonary arteries and veins [57], and umbilical arteries [51]. This effect of TES on the human umbilical artery has also been confirmed by others [56]. Moreover, recently it has been reported that androgens produce antihypertensive responses in an in vivo preeclampsia rat model [83].
*Therefore, these studies have concluded that androgen-induced relaxation of VSM appears to be sex-independent, which indicates that male sex hormones may also regulate physiological processes in both women and men.
5. Molecular changes in androgen structure associated with vascular reactivity
Relatively few studies have documented the effects of other androgens, metabolites, and analogs of TES on VSM, which is of considerable interest. The 5-reduced non-aromatized metabolites of TES (5α-, 5βDHT, and a few terahydo metabolites) also induce vasorelaxation, demonstrating that these steroid metabolites are also vasoactive steroids.
Our research group has systematically studied the vascular effects of androgen metabolites and reported that the vasorelaxing effects of 5βDHT are notably more efficacious and potent than those of its precursor, TES, in the rat aorta [20,86,91,94,95], dog coronary and femoral arteries [52], pig coronary and prostatic small arteries [96,115], and in the human umbilical artery [51]. Taken together, these findings provide clear evidence that 5β-DHT exhibits the highest efficacy and potency in inducing vasorelaxation in isolated blood vessels, even when compared to the vasorelaxing effects of progestins and estrogens [47,91,94,95].
In marked contrast, the 5α isomer (5α-DHT) of 5β-DHT, was found to be less potent or equipontent to TES in relaxing the rat aorta [92], dog coronary and femoral arteries [52], pig coronary and prostatic small arteries [96,115], and human umbilical arteries [51,51]. Similar studies have shown that TES and 5α-DHT may also relax human penile arteries [58,59], cavernosal arteries [60], and internal spermatic veins [61] without any differences in their relaxing efficacy or potency. Moreover, as a nonaromatizable dihydro-androgen metabolite of TES, 5α–DHT, has been frequently used to verify that aromatization of TES into estrogen is not required for this androgen to produce vasorelaxation [92,96,103,104].
The subsequent 3α or 3β hydroxylation of the 5α- and 5β-reduced dihydro metabolites are generally less potent than TES in inducing vasorelaxation [61], although androsterone was found to be equipotent to TES [51,92].
On the other hand, the esters of TES, which, by virtue of their greater lipid solubility, exhibit longer-acting androgenic effects in vivo, are less efficacious and potent than TES in inducing vasorelaxation in vitro. Thus, TES-enanthate and TES-hemisuccinate were less efficacious and potent than TES in the rat aorta [92] and in rabbit coronary arteries and aortas [90]. In isolated human vessels, TES-propionate was found to relax the internal spermatic vein [61] and to be more efficacious than TESphenylpropionate, TES-decanoate, and TES-isocaproate in relaxing the radial and internal mammary arteries and spermatic veins [55,111]. However, the weakness of these studies was their failure to include TES and/or its metabolites as a control group.
Consistent with the in vitro effects of TES metabolites on vasorelaxation, studies have shown that the 5-reduced metabolites of TES are also capable of blocking vasopressor responses of noradrenaline or Bay K 8644 in vagosympathectomized, pithed rats in vivo. Again, 5β-DHT was more potent than 5α-DHT or the 3α,5β-reduced metabolite in inducing a rapid vasodepressor response [80]. Similarly, TES and 5β-DHT both produced dose-dependent reductions in the mean arterial BP of conscious male normotensive and testicular-feminized male rats (Tfm), and 5β–DHT was found to produce greater reductions in BP than TES [81]. Likewise, the order of potency of producing an acute antihypertensive response in conscious male spontaneously hypertensive rats (SHR) was as follows: 5β-DHT > TES > 5α-DHT [82].
Conformational analysis of different androgens showed that there is a difference between the 5α- and 5β-reduced metabolites. Ring A in the steroid molecule of 5β-steroids is folded under or below the plane of the remainder of the molecule, similarly to Δ4-3keto configuration observed in TES, Δ5,3β hydroxy structure in DHEA, and the 5α-reduction (5αDHT) with an A/B trans ring–junction (see molecular structures in Fig. 1). Thus, the A/B cis conformation in 5β-DHT is responsible for a major non-genomic efficacy and potency to induce vasorelaxation and subsequently reduction of BP than that induced by TES and 5α-DHT.
Remarkably, 5β-reduction does not bind to the AR and is completely devoid of androgenic responses. This androgen is also non-aromatizable and is deprived of estrogenic properties; nonetheless, possesses acute vasorelaxing and antihypertensive responses non–genomically mediated. Therefore, 5β-DHT is not a biologically inactive metabolite. 5β–Reduced steroids are increasingly being recognized as active compounds with important regulatory functions. The present review highlights the role that this 5β-reduced androgen plays in the regulation of BP and that it is an excellent therapeutic option. The cited findings regarding the vasorelaxing, hypotensive, and antihypertensive responses indicate that the beneficial actions of androgens might be very relevant for understanding and treating hypertension, which is one of the most important health problems worldwide. In support of this view, there is direct evidence indicating that the activity of 5β-reductase, the enzyme that catalyzes the conversion of TES into 5β-DHT, is significantly lower in patients with essential hypertension than in normotensive controls [116].
Furthermore, it should be noted that although castration-induced prostate regression is a well-known phenomenon, this is not a good alternative for treating prostatic cancer. Our findings suggest that the reduction in prostatic blood flow may be treated by using 5β-DHT, which is devoid of androgenic and anabolic properties due to its lack of genomic activity; however, this androgen is capable of regulating prostatic function, similarly to TES, through control of blood flow accompanied by its vasorelaxing and antihypertensive non-genomic responses.
Finally, of considerable interest is to note a frequent mistake in the scientific literature: the 5–reduced dihydroandrogens (5α- and 5β-DHT) are only identified as “DHT,” and relatively few studies distinguish between the 5α or 5β configurations; however, it has become clear that the biological effects of these molecular configurations are fundamentally different.
Admittedly, the type and dose of androgen, as well as the future route of administration of androgen replacement therapies, will require substantially more investigation.
6. Perspectives
In an optimistic manner, the present lines of evidence can conclude that C19-steroids (androgens) in general, and 5β-DHT in particular, are vasoactive male sex steroids, which may play an important role in cardiovascular regulation. A deficiency in endogenous TES may be a risk factor for hypertension and CVD, and exogenous androgens may be used therapeutically for the prevention or treatment of hypertension. Although a deficiency in either TES or 5α-DHT may be involved in the development of hypertension, treatment with 5β-DHT would be preferred, since it not only exhibits a higher efficacy and potency than TES but also lacks genomic effects on reproductive target tissues. In summary, the properties of 5β-DHT i.e., no bioconversion to estrogens lacking estrogenic effects and androgenic action, and outstanding vasorelaxing action, point to this androgen as an excellent candidate for hypertension control. Taking into account, the rapid non-genomic effect (1–2 min) of androgens in inducing vasorelaxation and consequently reduction of BP, particularly 5β-DHT might be administered intravenously as a bolus and could be considered as first-line therapy for an acute hypertensive emergency. Admittedly, further clinical studies are urgently needed.
Mercedes Perusquía
ABSTRACT
Arterial hypertension is a global public health concern. In the last few years, the interest in androgen deficiency has been growing, and the association between androgens and high blood pressure (BP) is still controversial. One purpose of this review was to summarize the available findings in order to clarify whether male sex steroid hormones have beneficial or harmful effects on BP. The second purpose was to enhance the recognition of the acute non-genomic sex–independent vasorelaxing effect of androgens. Remarkably, BP variation is expected to be a consequence of the androgen-induced vasorelaxation which reduces systemic BP; hence the in vivo vasodepressor, hypotensive, and antihypertensive responses of androgens were also analyzed. This article reviews the current understanding of the physiological regulation of vascular smooth muscle contractility by androgens. Additionally, it summarizes older and more recent data on androgens, and some of the possible underlying mechanisms of relaxation, structural-functional differences in the androgen molecules, and their designing ability to induce vasorelaxation. The clinical relevance of these findings in terms of designing future therapeutics mainly the 5-reduced metabolite of testosterone, 5β-dihydrotestosterone, is also highlighted. Literature collected through a PubMed database search, as well as our experimental work, was used for the present review.
1. Introduction
1.1. General considerations of androgens
Steroids are important in biology, chemistry, and medicine. These compounds are divided into two different classes: natural or synthetic biologically active organic compounds, with a molecular structure of 17 carbon atoms arranged in four fused rings (A, B, C, and D) and called “cyclopentenoperhydrophenanthrene” They are found in plants, animals and fungi, and they regulate several biological functions. Sex steroid hormones can be grouped into three families based on their number of carbon atoms: pregnanes (21), androstanes (19), and estranes (18). Therefore, male sex steroids (androgens) are also classified as C19-steroids which are traditionally considered to maintain male sexual characteristics, although their relevant biological roles in the bone, muscle, prostate, and adipose tissue and the reproductive, cardiovascular, immune, neural, and hematopoietic systems have also been documented [1,2]. In this review, it has been highlighted the physiological effects of C19-steroids (androgens) on the cardiovascular system which have previously been scarcely considered.
It is important to take into account the fact that androgens are metabolized into different steroids each with its own variable effects and potency. Fig. 1 summarizes the classical gonadal pathway biosynthesis of androgens and the enzymes involved therein. In males, the testes synthesize testosterone (TES) from dehydroepiandrosterone (DHEA) and androstenedione, which is then mainly converted into 5αand 5β-dihydrotestosterone (5α-DHT and 5β-DHT, respectively) in target tissues, such as the prostate and seminal vesicles. In females, TES produced from the theca cells is predominantly converted into estrogens in the granulosa cells of the ovary (Fig. 1) (for details, see [3]). The major androgens produced by adult humans in decreasing order of magnitude are DHEA, androstenedione, TES, and 5α- and 5β-DHT. It is important to emphasize that androgens are not exclusive to males and estrogens are not exclusive to females; both sexes, male and female produce androgens and estrogens but their proportions are different. Thus, the plasma concentration of TES in males is 10–35 nM [4] and 0.7–2 nM in females [5].
However, little attention has been paid to the metabolism and changes in the chemical structure of androgens. TES is reduced to 5αDHT, which is considered the most potent androgen [6–9]; however, its 5β-reduced isomer (5β-DHT) has been mistakenly considered biologically inactive which will be discussed later.
1.2. Genomic vs non-genomic mechanisms of androgen action
Androgens may modify cellular function through both (i) their well-known classical genomic action and (ii) their more recently identified non-genomic mechanisms of action.
The genomic mechanism of action has been extensively studied. It regulates many of the most known biological functions such as secondary male sexual characteristics, sexual behavior, spermatogenesis, increment of muscle mass, loss of adipose tissue, prostate growth, prevention of osteoporosis, and roles in the hematopoietic, immune, and nervous systems, and in skin hair follicles, emotions, and libido [1,2]. Classically, androgens may mediate their biological responses by binding to the androgen receptor (AR). ARs are expressed in many tissues, with the highest concentrations occurring in male reproductive organs [10]. Briefly, the AR signaling pathway for genomic action signaling involves androgens crossing the plasma membrane, entering the cytoplasm, and binding to the AR (androgen-receptor complex), resulting in the dissociation of chaperone proteins and translocation of the complex to the nucleus where it dimerizes and binds to the androgen response element (ARE) for modulating gene transcription, and leads to protein synthesis creating a biological response, which has been extensively studied (for review see [11–13]). This classical androgen response is relatively slow, occurring over a period of hours, weeks, or years (Fig. 2).
In contrast to the genomic pathway, the non-genomic mechanism of action began to be recognized much later, roughly 40 years ago. Different steroids, including androgens, can also act via this mechanism, which is independent of the ligand-dependent transactivation function of nuclear receptors [14,15]. This is known as ‘non-genomic’ signaling, which typically occurs within a short time frame (seconds to minutes) (Fig. 2) [16,17]. Fig. 2 summarizes the two mechanisms of androgen action.
The effects mediated by the non-genomic type of action have been observed in most excitable tissues, such as the central nervous system and reproductive and non-reproductive smooth muscles, among others. Several studies have found that the majority of the physiological nongenomic androgen responses occur through membrane- or membrane-associated AR/binding proteins, which would decrease cell membrane fluidity, changes in intracellular calcium concentration ([Ca2+]i), activation of the second messenger pathway, or a combination of the above mechanisms (reviewed in [18–21]). Notably, non-genomic mechanisms can act independently or be associated with other genomic effects in order to elicit a physiological response, i.e., the genomic and non-genomic pathways of androgen action are likely interlinked.
1.3. Classical effects of androgens
1.3.1. Reproductive effects
As mentioned before, the major biological actions of TES include the development and maintenance of genetically mediated secondary sexual characteristics in males. These effects are mediated directly by TES and its two major active metabolites, 5α-DHT and 17β-estradiol. The genomic action, mediated by AR, of TES and 5α-DHT is responsible for sexual differentiation and virilization. Likewise, these androgens play an important role in the deepening of the voice, increase in sebum production, and an enlargement of the external genitalia, including penile length. The conversion of TES to 5α-DHT, but not to 5β-DHT, in the prostate is essential for its growth and development, and, importantly the concentration of this 5α-reduced metabolite is 10 times higher than that of TES. Similarly, erectile dysfunction, hypogonadism, and metabolic syndrome have all been associated with TES deficiency. In addition to the masculinizing effects of androgens, they also have anabolic properties. TES is known for producing muscle hypertrophy by increasing fractional muscle protein synthesis [22,23]. The correct term for these steroids is anabolic–androgenic steroids (AAS): “anabolic” refers to muscle building and “androgenic” refers to increased male sex characteristics. AAS stimulates growth in many types of tissues, especially bone and muscle, and their effects involve binding to the AR and therefore genomically mediated. AAS has been used since the 1950 s in pharmacological doses (administered orally and by several parenteral ways) in an attempt to maximize the well-known anabolic effects of TES, reduce the rate of its hepatic inactivation, and decrease its aromatization to 17β–estradiol [24]. AAS as TES derivates and analogs include the endogenous sex male steroids i.e., TES and 5α-DHT, as well as other agents and synthetics that behave like these androgens, some of which are nandrolone, decanoate, oxandrolone, oxymetholone, stanozolol, TES cypionate, and TES enanthate.
1.3.2. Effects of androgens on reproductive and non-reproductive smooth muscles
It has been demonstrated that androgens target reproductive and non-reproductive smooth muscles (SM) by inducing non-genomically mediated relaxation.
In urogenital SM, the uterine contractility is notably relaxed by androgens with a higher utero-relaxant potency than progestins [25–27]. Therefore, the uterus is a target organ for androgens that induce myometrial quiescence, which plays a key functional role in the maintenance of normal gestation. Androgens also lead to marked relaxation of barium-induced contraction of the rat epididymis and seminal vesicles [28]. The relaxing effects of TES were also confirmed in rabbit seminal vesicles [29]. The neck of the urinary bladder is also a target of androgen-induced relaxation; thus, TES and 5α-DHT have been found to elicit similar concentration-dependent relaxation in both male and female pigs [30,31]. In summary, the acute relaxing effect of androgens in other types of male urogenital SM has rarely been studied.
In gastrointestinal SM, progesterone, 17β-estradiol, and TES are also capable of inhibiting the contractibility of intestinal SMs such as those in the ileum [32,33]. In comparison with female steroids, the high potency of androgens in inducing non-genomic gastrointestinal relaxation was recognized [33,34]. TES and 5α-DHT have also been found to be capable of inducing similar levels of relaxation in gallbladder SMs in male guinea pigs [34]. Similarly, several androgens (DHEA, TES, and its 5-reduced metabolites) are capable of inducing non-genomic bronchorelaxation by acting on the airway SMs as well as inducing genomic anti-inflammatory responses, suggesting a potential therapeutic use in the treatment of severe asthma (reviewed in [35]). A substantial point of interest in this review is that vascular SM (VSM) is also a target of TES and its metabolites in acute non-genomic relaxation. In particular, in the last few years, there has been a growing interest in androgen-induced vasorelaxation, which is discussed in detail in Section 2.
*Therefore, the available data indicate that multiple types of SM are targets for androgen-induced relaxation and that androgens may be important in the regulation of SM excitability in general.
2. Vascular effects of androgens
2.1. Vasorelaxation/vasodilation of isolated vessel studies
It has been widely accepted that estrogens exert protective effects on female vasculature [36,37]. Recent clinical and experimental animal studies have revealed that TES, the primary and most well-studied androgen, directly protects against risk factors for cardiovascular diseases (CVD), such as atherosclerosis and hypertension, and reduces the incidence of CVD, such as cardiac failure and myocardial ischemia [38–44]. Likewise, several clinical trials have pointed out that androgen deficiency seems to exacerbate many risk factors and pathologies associated with cardiovascular and metabolic diseases [41,45–50]. Based on these findings, the protective effect of TES on the cardiovascular system is gaining recognition. In this context, the main purpose of the present review article is to show the direct and rapid non-genomic effects at the vascular level, particularly in the regulation of blood pressure (BP).
Sex steroid hormones, including androgens, mediate vasorelaxation. Particularly regarding androgens, in the last 25 years, there has been a plethora of evidence that TES is capable of inducing vasorelaxation. The acute vasorelaxing effect of TES on contractions induced by depolarization, or several agonists in a number of isolated vascular beds of various species of mammals including humans (revised in [18,20,21,51–54]).
Notably, the acute concentration-dependent vasorelaxing effects of androgens have been reported in isolated human blood vessels, including the internal mammary artery [55], umbilical arteries [51,56], and pulmonary arteries and veins [57]. TES and 5α-DHT have also been reported to induce vasodilating effects on the penile and cavernosal arteries [58–60], and internal spermatic veins [61], the lack of vasodilation may cause non-psychogenic erectile dysfunction associated with hypoandrogenism (androgen deficiency). Moreover, androgens have been shown to regulate prostatic blood flow [62]. As reviewed by Traish et al. [63], TES–propionate has been used for the last six decades to treat sexual issues to augment penile vasodilation.
TES has direct effects on vascular tone causing vasodilatation, and an increase in coronary blood flow [42,43]; thus, it has been observed that the coronary circulation is more sensitive to the vasorelaxant effect of TES than large vessels with a benefit on time to cardiac ischemia. Notably, clinical studies have highlighted the beneficial effects of TES therapy on exercise-induced cardiac ischemia, an effect mediated by coronary vasodilation [64], which persists for 3 months and indeed 12 months [65].
*Together, these findings reveal that TES exerts beneficial effects on the cardiovascular system by reducing vascular tone, which involves direct vasodilatory action on the peripheral vasculature and reduction of BP. For their acute and marked vasorelaxing effects, the male sex steroids have been named “vasoactive androgens”.
2.2. Systemic hypotensive actions
The vasorelaxing effect induced by androgens implies an increase in blood flow, and therefore, a reduction in BP.
High blood pressure (hypertension), sometimes called “the silent killer,” is a serious medical condition and a global public health problem. Hypertension is very common in older people and is a major health concern. The pathogenesis of essential hypertension, the most common type of hypertension, is multifactorial and highly complex. Factors that play an important role in this pathology include genetics, activation of the neurohormonal systems, such as the sympathetic nervous system and renin-angiotensin-aldosterone system, obesity, and increased dietary salt intake. However, the etiology of hypertension remains unknown.
Sex steroid hormones have also been associated with hypertension, however, there is a polemic regarding whether female sex steroids are beneficial and male sex steroids are detrimental to BP. Male sex is considered a major contributing risk for CVD. A direct effect of androgens on promoting hypertension has been hypothesized [66] and was attributed to the myths that TES is capable of inducing adverse effects on the cardiovascular system. One of these effects is an increase in BP.
The theory of increased hypertension risk in response to TES remains largely unsubstantiated. In view of the non-genomic vasorelaxing properties of androgens, it is difficult to consider that TES can be detrimental to the cardiovascular system, since TES, the most important male sex endogenous hormone, plays a vital role in the regulation of biological processes in males. This review aims to clarify the reasons for this uncertainty.
Several lines of evidence, contradict the traditional view of the adverse effects of androgens on BP. The first evidence that androgens are associated with the regulation of BP through BP reduction was collaterally observed in classic studies by Selye, when the anesthetic effect of steroids was reported in the early 1940 s [67,68]. Later, a body of information associated low circulating levels of TES with cardiovascular dysfunction, particularly with hypertension [43,69–72]. This finding was confirmed in several reports indicating that low concentrations of TES in men are associated with higher BP [73–76]. Other studies have suggested that a low TES concentration in men may be a novel risk factor for hypertension or CVD. An important clinical study documented that acute administration of TES in humans with cardiac failure decreases peripheral vascular resistance and improves cardiac output [77]. An observational study in men showed that long-term TES therapy, for up to 8 years, produced marked and sustained gradual decreases in systolic and diastolic BP [78]. One recent study showed that hypertensive men had lower levels of free TES, and bioavailable TES, compared to normotensive men [79].
More recently, we reported important in vivo experimental research on animals. It is important to highlight that the consequence of androgen-induced vasorelaxation, widely discussed in the present review, is a decrease in BP. In this regard, numerous studies have suggested that low BP is linked to androgens (reviewed in [54]). We first reported that some TES metabolites are capable of blocking vasopressor responses to noradrenaline or a Ca2+ channel agonist (Bay K 8644) in anesthetized vagosympathectomized, pithed, Wistar male rats [80]. Our recent in vivo experimental research has also demonstrated that BP regulation is an expected consequence of androgen-induced vasorelaxation [81–84]. We documented that TES and its 5β–reduced metabolite (5β-DHT) induced a marked reduction in BP in both conscious, normotensive, Sprague-Dawley rats and testicular feminized male rats (Tfm; androgen receptor-deficient) [81]. Afterward, we reported that androgens can also elicit a significant antihypertensive response in conscious, spontaneously hypertensive, male rats (SHR) and in normotensive WKY rats who underwent TES deprivation by orchiectomy for inducing hypertension, which was then prevented by TES replacement therapy [82]. Similarly, it was demonstrated that in an in vivo rat model of preeclampsia, the elevated mean arterial BP (MAP) was reduced significantly by an intravenous bolus injection of DHEA, TES, or its 5-reduced metabolites (5α- and 5β-DHT) [83]; notably, this study indicates that androgens are also antihypertensive in females. More recently, a model for studying androgen deficiency in conscious hypertensive rats caused by orchiectomy helped us to understand the progression of hypertension in aging men since our experiments clearly demonstrated that hypertension is a disease associated with aging. We also observed that the intravenous bolus administration of DHEA, TES, or 5β-DHT is capable of inducing antihypertensive responses in hypertensive aging rats [84]. The above information is summarized in Table 1.
Notably, TES is an acute vasorelaxant in human subcutaneous resistance arteries [85]. In addition, functional studies have also shown that the vasorelaxing potency of 5β–DHT is higher in the mesenteric artery than in the thoracic aorta of hypertensive and normotensive rats [86]. These findings suggested that TES and particularly 5β-DHT regulate vascular resistance in the mesenteric arterial bed, explaining its outstanding antihypertensive response.
It seems that physiological androgen levels regulate BP by placing a “brake” for developing hypertension, and indeed hypotestosteronemia is a risk factor for hypertension. Several studies have reported that men with CVD, type 2 diabetes mellitus, obesity, metabolic syndrome, and dyslipidemia have low levels of TES (reviewed in [46,53,71,87,88]). In this respect, the incidence of CVD increases in aging men when TES production decreases. In line with these findings, it has been suggested that hypertension, comorbidity highly susceptible to increased COVID19 severity, may be due to lower androgen levels in older men patients [89]. These findings indicate that TES is beneficial for the cardiovascular system.
*In parallel, these findings have also shown that DHEA and 5β-DHT are significantly more potent than TES in evoking an antihypertensive response, indicating that androgens may provide a therapeutic potential for the regulation of BP.
2.3. Mechanism of vasorelaxing action of androgens
2.3.1. Non-genomic action
The action of androgens in inducing vasorelaxation has been defined as follows: androgen-induced vasorelaxation is rapid (~2 min). Likewise, several lines of pharmacological evidence have established that this immediate effect induced by androgens is non-genomically mediated since the vasorelaxing effect of androgens is observed in the presence of AR antagonists, protein synthesis, and transcription inhibitors [51,52,90,91]. The response is observed even when the androgen is conjugated to molecules, such as bovine serum albumin (BSA), which prevents it from entering the cytoplasm [92], and the AR deficient testicular-feminized rat inhibitors [90]. Indeed, these data taken together indicate that the mechanism of action of androgens is unlikely to involve nuclear receptor activation.
2.3.2. Endothelium- and nitric oxide (NO)-independent mechanism
Numerous studies have reported that the vasorelaxing effect caused by 17β-estradiol is mediated in part by via generation of endotheliumderived NO and is attenuated by NO synthesis inhibitors, suggesting an endothelium-dependent mechanism [37], revised in [93] but this seems not to be the case for androgens. Androgen-induced vasorelaxation is observed in endothelium-denuded vascular preparations, as well as in the presence of NO synthase inhibitors such as L-NAME [43,51,52,94–97], as revised by [20]. These findings indicate that androgen-induced vasorelaxation is endothelium–independent. In this respect, it is relevant to consider the following convincing facts: (i) androgens are powerful vasorelaxants in the human umbilical artery [51], notably, this blood vessel has an atypical endothelium due to the fact that biosynthesis of endothelial NO is absent; and (ii) androgen-induced antihypertensive responses in preeclamptic rats [83]. Clinical and experimental studies have suggested that endothelial dysfunction is present in preeclamptic subjects [98–102]. In summary, the vasorelaxing effect of androgens strongly suggests an NO–independent mechanism [51].
In marked contrast, some contradictory data have reported that androgen-induced vasorelaxation is partially, but not completely, inhibited through pretreatment with L-NAME [103–106], suggesting that a partial endothelium-dependent mechanism might be involved. A study also documented that vasodilatation to TES in human pulmonary arteries is dependent upon the presence of an intact endothelium [57]. However, this contradiction may be related to differences in the vessel type or experimental animal species and/or, as previously suggested by Jones and Kelly [88], can also be explained by the fact that TES is converted to 17β-estradiol (see Fig. 1); an estrogen that mediates its vasorelaxing effect through stimulation of NO production, it is currently not known whether the reported vasorelaxing effect of TES is a direct endothelium-dependent effect or an indirect effect via estrogens.
*Although these data suggest that the endothelium is not involved in androgen-induced vasorelaxation, further investigations are needed to clarify this controversy.
2.3.3. Ion channels
We provided the first experimental evidence that the mechanism of vasorelaxation of androgens was due to the blockage of voltage- and receptor-dependent Ca2+ channels, a mechanism that restricts the availability of extracellular Ca2+ in the contractile machinery [94,95]. Vasorelaxation of 17β-estradiol, progesterone, and TES, inhibited the entry of Ca2+ but not its release [107]. More specifically, in single vascular myocytes, it has been shown that the mechanism of action of androgens in inducing vasorelaxation depends on the blockade of L–voltage-operated Ca2+ channels (L-VOCCs) [91,97,108–110], revised by [88]. A compelling fact is that TES acts on the same site as dihydropyridines (nifedipine), the α1C subunit of the L-VOCCs, in order to block the extracellular Ca2+ entry into intracellular space [110]. We also found that 5β-DHT acts exclusively as an antagonist to blocking Ca2+ currents via L-VOCCs (from nM to µM concentrations), whereas TES possesses different mechanisms: (i) a dual effect on L-VOCCs (antagonist at nM concentrations/agonist at µM concentrations), (ii) an increase in [Ca2+]i and (iii) an increase in cAMP production. This is the first evidence that can help us understand the mechanism underlying the acute vasorelaxing action of 5β-DHT and the controversial response of TES upon vascular reactivity, i.e., the vasorelaxing and/or antivasorelaxing effects [91].
Several functional studies in VSM and other SMs demonstrated the TES may also relax contractions induced by KCl and several contractile agonists, such as acetylcholine, carbachol, noradrenaline, oxytocin, ovalbumin, phenylephrine, prostaglandins, and serotonin, among others, it cannot be ruled out that androgens may also inactive non-voltage-gated pathways, i.e. the store-operated Ca2+ entry (SOCE) and receptor-operated Ca2+ channels (ROCCs), suggesting that TES has an antagonistic effect on L-VOCCs, ROCCs, and/or SOCE, showing that the vasorelaxation is induced by blockade of extracellular Ca2+ entry through these channels. In this respect, in addition, there are also the experimental findings that in the bronchodilating effect of androgens in airway SM [35] SOCE and ROCC are partly involved.
In addition to the above, a human study examined the effect of TES therapy on vascular reactivity in isolated human subcutaneous arteries it was observed that after TES therapy, the vasoconstrictor effect of noradrenaline increased and reduced the vasodilator response to acetylcholine and sodium nitroprusside (which breaks down to NO in the circulation). This effect physiologically would divert blood flow toward the vital organs, an important survival effect in chronic heart failure [85], revised in [88] which is evidence for a different mode of action of test on vascular reactivity.
.
In addition to Ca2+ channels, it has been repeatedly documented that in functional studies in some vascular beds, TES-induced vasorelaxation may also activate several types of K+ channels, which are: voltage-sensitive K+ channels (KV), large-conductance Ca2+-activated K+ channels (BKCa), ATP-sensitive K+ channels (KATP), and inward-rectifier K+ channels (KIR) [56,92,96,103,111]. One study through patch-clamp experiments reported that TES produced activation of BKCa channels [96]. In conclusion, the main K+ channels involved in this process are BKCa and Kv.
With these findings, there is controversy. The most intriguing conclusions, which can be drawn from the numerous observations of nongenomic vasorelaxing mechanisms are as follows: either (i) the mechanism differs depending on blood vessels [52]; or (ii) in agreement with [21], androgen-induced vasorelaxation is due to the fact that Ca2+ channels are closely related to K+ channels, where membrane hyperpolarization by K+ channel activation closes the Ca2+ channels and leads to vasorelaxation.
However, little attention has been paid to the in vivo mechanism of action. The mechanism of androgen action in eliciting antihypertensive responses was studied for the first time in conscious (in vivo) rats. 5βDHT abolished the pressor response to Bay K 8644 (an L-VOCC antagonist) in vivo, suggesting a blockade of L-VOCC [82]. Our data in conscious orchidectomized hypertensive rats revealed that: (i) the blockade of NO synthase by L-NAME did not alter the antihypertensive response to 5β-DHT; and (ii) endothelin-1, by binding to ETA receptors, interferes with NO synthesis and acts to counter NO-induced vasodilation; under these circumstances, dose-dependent antihypertension induced by 5β-DHT was not abolished in the presence of L-NAME or endothelin-1 [84].
Fig. 3 illustrates the mechanisms of action of androgens described here to produce vasorelaxation/hypotension.
3. Differences between male and female steroid hormones in the cardiovascular system
It is relevant to note that when the efficacies and potencies of androgen-induced vasorelaxation were compared to female sex steroids, it was observed that TES and its two immediate dihydrometabolites, 5αand 5β-DHT, but not TES, have turned out to be more potent vasorelaxants than the female steroids, estrogens, and progestins in inducing vasorelaxation (see section 5). Our early observations were that 5β-DHT was the most potent steroid, even when compared to the vasorelaxing response of progestins in the rat aorta [94]. Remarkably, the concentration-response curve to 5β-DHT exhibited the highest potency for vasorelaxation of the human umbilical artery, even exceeding the potency of several progestins [51]. This is probably the only published report comparing the efficacy and potency of TES and its 5-reduced metabolites with progesterone and its 5-reduced progestins.
Regarding estrogens, the concentration-response curves showed that 5β-DHT was significantly more potent than 17β-estradiol in inducing vasorelaxation in rat aorta [95]. In the pulmonary vasculature, the vasorelaxing efficacy of 17β-estradiol was significantly lower than that of TES [47]. In patch-clamp electrophysiological studies, the concentration-response curves to TES, 5β-DHT, and 17β-estradiol on Ca2+ currents in single aortic myocytes from rats indicated that androgens were more potent in blocking Ca2+ currents than 17β-estradiol [91].
Despite these findings, studies by Khalil and colleagues demonstrated that progesterone and TES are less potent than 17β-estradiol in inducing vasorelaxation [107,112,113]. In this respect, it is important to note that TES is a less potent vasorelaxant than DHEA and its 5β-reduced metabolite; thus, the roles of TES have been inappropriately generalized as common to all androgens (see section 5)
*However, in a general context, androgens, particularly 5β-DHT, have turned out to be more potent vasorelaxants than female steroids, estrogens, and progestins.
4. Sex differences in vasorelaxing effects of androgens
Historically, female sex steroid hormones (estrogens and progestins) have been studied extensively, which has led to a substantial understanding of human female physiology and the diverse effects of these hormones on physiological, as well as reproductive functions.
A huge body of evidence has created the wrong belief that estrogens are cardioprotectors in females and that androgens are detrimental to the male cardiovascular system. Recently, the interest in the role of sex regulation of metabolic homeostasis, BP, and CVD has been growing. Men typically develop CVD earlier than women [66]; however, it is important to consider that very few studies have examined androgen signaling and hypertension in females.
Some studies have suggested that androgens are as important for BP regulation in females as they are in males. The available data suggest that in vitro relaxation induced by TES is independent of sex. No significant differences between the vasorelaxing effect of TES in the coronary arteries and aortas of male vs female rabbits were observed [90] or in the aortas of male vs female rats [114]. TES has also been reported to induce sex-independent vasodilation in vivo in canine coronary conductance and resistance arteries [104]. TES elicits vasorelaxation in both males and females, however, rat pulmonary and coronary arteries in males were more sensitive to the effects of TES than those in females [47].
In addition, androgens also possess the ability to relax the female human vasculature, such as the internal mammary artery [55], pulmonary arteries and veins [57], and umbilical arteries [51]. This effect of TES on the human umbilical artery has also been confirmed by others [56]. Moreover, recently it has been reported that androgens produce antihypertensive responses in an in vivo preeclampsia rat model [83].
*Therefore, these studies have concluded that androgen-induced relaxation of VSM appears to be sex-independent, which indicates that male sex hormones may also regulate physiological processes in both women and men.
5. Molecular changes in androgen structure associated with vascular reactivity
Relatively few studies have documented the effects of other androgens, metabolites, and analogs of TES on VSM, which is of considerable interest. The 5-reduced non-aromatized metabolites of TES (5α-, 5βDHT, and a few terahydo metabolites) also induce vasorelaxation, demonstrating that these steroid metabolites are also vasoactive steroids.
Our research group has systematically studied the vascular effects of androgen metabolites and reported that the vasorelaxing effects of 5βDHT are notably more efficacious and potent than those of its precursor, TES, in the rat aorta [20,86,91,94,95], dog coronary and femoral arteries [52], pig coronary and prostatic small arteries [96,115], and in the human umbilical artery [51]. Taken together, these findings provide clear evidence that 5β-DHT exhibits the highest efficacy and potency in inducing vasorelaxation in isolated blood vessels, even when compared to the vasorelaxing effects of progestins and estrogens [47,91,94,95].
In marked contrast, the 5α isomer (5α-DHT) of 5β-DHT, was found to be less potent or equipontent to TES in relaxing the rat aorta [92], dog coronary and femoral arteries [52], pig coronary and prostatic small arteries [96,115], and human umbilical arteries [51,51]. Similar studies have shown that TES and 5α-DHT may also relax human penile arteries [58,59], cavernosal arteries [60], and internal spermatic veins [61] without any differences in their relaxing efficacy or potency. Moreover, as a nonaromatizable dihydro-androgen metabolite of TES, 5α–DHT, has been frequently used to verify that aromatization of TES into estrogen is not required for this androgen to produce vasorelaxation [92,96,103,104].
The subsequent 3α or 3β hydroxylation of the 5α- and 5β-reduced dihydro metabolites are generally less potent than TES in inducing vasorelaxation [61], although androsterone was found to be equipotent to TES [51,92].
On the other hand, the esters of TES, which, by virtue of their greater lipid solubility, exhibit longer-acting androgenic effects in vivo, are less efficacious and potent than TES in inducing vasorelaxation in vitro. Thus, TES-enanthate and TES-hemisuccinate were less efficacious and potent than TES in the rat aorta [92] and in rabbit coronary arteries and aortas [90]. In isolated human vessels, TES-propionate was found to relax the internal spermatic vein [61] and to be more efficacious than TESphenylpropionate, TES-decanoate, and TES-isocaproate in relaxing the radial and internal mammary arteries and spermatic veins [55,111]. However, the weakness of these studies was their failure to include TES and/or its metabolites as a control group.
Consistent with the in vitro effects of TES metabolites on vasorelaxation, studies have shown that the 5-reduced metabolites of TES are also capable of blocking vasopressor responses of noradrenaline or Bay K 8644 in vagosympathectomized, pithed rats in vivo. Again, 5β-DHT was more potent than 5α-DHT or the 3α,5β-reduced metabolite in inducing a rapid vasodepressor response [80]. Similarly, TES and 5β-DHT both produced dose-dependent reductions in the mean arterial BP of conscious male normotensive and testicular-feminized male rats (Tfm), and 5β–DHT was found to produce greater reductions in BP than TES [81]. Likewise, the order of potency of producing an acute antihypertensive response in conscious male spontaneously hypertensive rats (SHR) was as follows: 5β-DHT > TES > 5α-DHT [82].
Conformational analysis of different androgens showed that there is a difference between the 5α- and 5β-reduced metabolites. Ring A in the steroid molecule of 5β-steroids is folded under or below the plane of the remainder of the molecule, similarly to Δ4-3keto configuration observed in TES, Δ5,3β hydroxy structure in DHEA, and the 5α-reduction (5αDHT) with an A/B trans ring–junction (see molecular structures in Fig. 1). Thus, the A/B cis conformation in 5β-DHT is responsible for a major non-genomic efficacy and potency to induce vasorelaxation and subsequently reduction of BP than that induced by TES and 5α-DHT.
Remarkably, 5β-reduction does not bind to the AR and is completely devoid of androgenic responses. This androgen is also non-aromatizable and is deprived of estrogenic properties; nonetheless, possesses acute vasorelaxing and antihypertensive responses non–genomically mediated. Therefore, 5β-DHT is not a biologically inactive metabolite. 5β–Reduced steroids are increasingly being recognized as active compounds with important regulatory functions. The present review highlights the role that this 5β-reduced androgen plays in the regulation of BP and that it is an excellent therapeutic option. The cited findings regarding the vasorelaxing, hypotensive, and antihypertensive responses indicate that the beneficial actions of androgens might be very relevant for understanding and treating hypertension, which is one of the most important health problems worldwide. In support of this view, there is direct evidence indicating that the activity of 5β-reductase, the enzyme that catalyzes the conversion of TES into 5β-DHT, is significantly lower in patients with essential hypertension than in normotensive controls [116].
Furthermore, it should be noted that although castration-induced prostate regression is a well-known phenomenon, this is not a good alternative for treating prostatic cancer. Our findings suggest that the reduction in prostatic blood flow may be treated by using 5β-DHT, which is devoid of androgenic and anabolic properties due to its lack of genomic activity; however, this androgen is capable of regulating prostatic function, similarly to TES, through control of blood flow accompanied by its vasorelaxing and antihypertensive non-genomic responses.
Finally, of considerable interest is to note a frequent mistake in the scientific literature: the 5–reduced dihydroandrogens (5α- and 5β-DHT) are only identified as “DHT,” and relatively few studies distinguish between the 5α or 5β configurations; however, it has become clear that the biological effects of these molecular configurations are fundamentally different.
Admittedly, the type and dose of androgen, as well as the future route of administration of androgen replacement therapies, will require substantially more investigation.
6. Perspectives
In an optimistic manner, the present lines of evidence can conclude that C19-steroids (androgens) in general, and 5β-DHT in particular, are vasoactive male sex steroids, which may play an important role in cardiovascular regulation. A deficiency in endogenous TES may be a risk factor for hypertension and CVD, and exogenous androgens may be used therapeutically for the prevention or treatment of hypertension. Although a deficiency in either TES or 5α-DHT may be involved in the development of hypertension, treatment with 5β-DHT would be preferred, since it not only exhibits a higher efficacy and potency than TES but also lacks genomic effects on reproductive target tissues. In summary, the properties of 5β-DHT i.e., no bioconversion to estrogens lacking estrogenic effects and androgenic action, and outstanding vasorelaxing action, point to this androgen as an excellent candidate for hypertension control. Taking into account, the rapid non-genomic effect (1–2 min) of androgens in inducing vasorelaxation and consequently reduction of BP, particularly 5β-DHT might be administered intravenously as a bolus and could be considered as first-line therapy for an acute hypertensive emergency. Admittedly, further clinical studies are urgently needed.
