Pharmacological sympathetic denervation prevents the development of polycystic ovarian syndrome in rats injected with estradiol valerate

Polycystic ovary syndrome (PCOS) is the most common endocrine disorder which affects between 6 and 10% of women in reproductive age [1, 2]. In 2006, the Androgen Excess & PCOS Society concluded that in the diagnosis of PCOS, the disorder predominantly consists of androgen excess and ovulatory dysfunction (i.e., chronic ovulatory dysfunction or polycystic ovarian morphology), with the exclusion of specific disorders, such as nonclassic adrenal 21-hydroxylase deficiency, Cushing’s syndrome, hyperprolactinemia, and androgen-producing tumors [3]. Frequently, PCOS is accompanied by obesity, hirsutism, and in the vast majority of cases, infertility [4]. In addition, some women with PCOS present metabolic alterations such as hyperinsulinemia and insulin resistance, the severity of both disorders increases in the presence of obesity [5] and the risk of cardiovascular diseases [6]. Some of the medical treatments for inducing ovulation in patients with the syndrome are: treatment with clomiphene citrate (an antiestrogen) [7], treatment with exogenous follicle stimulating hormone (FSH) [8], wedge resection of the ovaries [9] or treatment with low frequency electroacupuncture [10].

The etiology of PCOS is currently unknown. Evidence suggesting that the syndrome may originate in the hypothalamus, due to a primary neuroendocrine defect in gonadotropin-releasing hormone (GnRH) secretion that leads to increased frequency and amplitude in the pulses of the secretion of luteinizing hormone (LH) [11]. Recent studies suggests that ovarian innervation also plays a role in the physiopathology of the syndrome, since it has been observed that, in both, rodent with the induced pathology and women with the syndrome, there is an increase in sympathetic ovarian nervous activity [12‐15].

In the rat, the ovary receives sympathetic innervation from two neural pathways: the ovarian plexus nerve (OPN) and the superior ovarian nerve (SON). The SON-fibers innervate the thecal cells and provide the ovary with norepinephrine (NE), mainly; but also, with neuropeptide Y (NPY) and vasoactive intestinal peptide (VIP) [16, 17]. In prepubertal rats, the bilateral section of the SON decreases ovarian NE concentrations at the first 24 h post-denervation [16, 18].

In prepubertal and adult rats, injection with 2 mg of estradiol valerate (EV), a long-acting estrogen, induces PCOS [19‐22]. Rats with EV-induced PCOS are characterized by early puberty [21, 22], loss of estrous cycles [21, 23], high concentrations of androgens [22], anovulation, and around 60 days after EV injection, follicular cysts in the ovaries are observed [19]. The polycystic ovaries developed by the EV-injection show a reduction in their size, absence of corpora lutea, development of follicular cysts and poor follicular development [21, 24]. Because the structure of the ovaries of rats injected with EV is different from that of the ovary of women with PCOS, some authors have suggested having precautions when comparing these polycystic ovaries [25].

The development of PCOS has also been associated with changes in the catecholaminergic activity of the ovary. These changes include increase ovarian NE concentration and in the activity of tyrosine hydroxylase (TH), a limiting enzyme in NE biosynthesis, and a decrease in the number of beta-adrenergic receptors [12]. In this PCOS model, spontaneous ovulation occurs after the bilateral section of the SON and the androgens concentration in the ovary is normalized [20‐22]. Parra et al. [26] showed that in addition to ovarian NE concentration increases, EV-treated rats also show an increase in VIP concentration, which could stimulate the onset of PCOS. This is possible, since ovaries of rats with PCOS cultivated in the presence of VIP release higher amounts of androgens and oestradiol than ovaries of rats without the syndrome.

Since surgically sectioning the SON of rats with PCOS does not allow us to distinguish whether the reestablishment of ovulation and serum androgen concentrations is due to a decrease in ovarian NE or VIP levels, the aim of the present study was to analyze the participation of the noradrenergic system in the onset of PCOS. For this purpose, we used guanethidine monosulfate (GTD) as a pharmacological tool that destroys only peripheral noradrenergic fibers [27], as GTD does not cross the blood-brain barrier [28]. GTD does not accumulate in other cell types (i.e. sensory neurons) nor does it cause their destruction [28, 29]. Our working hypothesis was that eliminating of noradrenergic fibers before or during PCOS onset would result in spontaneous ovulation and normal testosterone levels, while the same treatment after PCOS onset would reverse the ovulation blockage caused by the syndrome. In the present study, the treatment consisted of chronically injecting GTD for three weeks, before, during or after EV treatment to induce PCOS. Our data suggest that the noradrenergic fibers play a stimulant role in the establishment of PCOS.

All experiments were carried out in strict accordance with the Mexican Law of Animal Treatment and Protection Guidelines, and the specifications of the Mexican Official Standard, NOM-062-ZOO-1999. The Committee of the Facultad de Estudios Superiores Zaragoza approved the experimental protocols. All efforts were taken to minimize the number of animals used, and all procedures were undertaken in a humane manner.

The study was performed using pre-pubertal female rats of the CIIZ-V strain from our own breeding stock. Animals were maintained under controlled lighting conditions (lights on from 05:00 to 19:00 h); with free access to food (Purina S.A., México) and tap water.

Taken together, the results of the present study shows that PCOS onset in pre-pubertal EV-treated rats depends on the integrity of the ovary’s peripheral noradrenergic system.

The ovary is innervated by sympathetic fibers through the SON and OPN. In addition to containing catecholamines synthetizing fibers, the SON are formed by fibers that synthesize other peptides; such as VIP and NPY; the supply of substance P (SP) and calcitonin gene-related peptide (CGRP) to the ovaries by the OPN [34]. The surgical section of these nerves has been used to analyze the role of innervation in ovarian functions. A research model to target specific types of neurons consist of treatments with neurotoxins, such as GTD; which when administered chronically and in high doses produce sympathectomy at the peripheral level [31, 35‐37].

Studies with male rats by Heath and Burnstock, [38] show that chronic high-doses GTD treatment has no apparent effect on parasympathetic cholinergic neurons or sensory neurons. According to Angeletti et al. [39], injecting three doses of GTD decreases the number of catecholaminergic neurons in the superior cervical ganglion. Lara et al. [31] showed that in the pubertal rat, the chronic injection of GTD destroys the ovary’s catecholaminergic neurons. Tyrosine hydroxylase is the limiting enzyme in catecholamines biosynthesis, and is widely used as a norepinephrine biosynthesis marker [31]. An increase in TH activity and expression in the ovary has been documented in rats with EV-induced PCOS [12, 24, 40]. Our results support these findings, since the ovaries of EV-treated rats showed as increase in TH expression. Tyrosine hydroxylase expression was almost completely lost in GTD-treated animals before, during, or after EV-treatment. This result reinforces the accounts of a GTD-induced reduction of catecholaminergic neurons and agrees with the observations presented by Angeletti et al. [39] and Lara et al. [31].

To date, despite the exhaustive research on the etiology of PCOS, the causes of the syndrome are still unknown. One of the hypotheses proposed to explain the origin of PCOS suggests that increases in ovarian sympathetic nerve activity results in abnormal ovarian function [12, 13, 20, 24]. In women with PCOS, the ovaries show greater density of nerve fibers [14] and higher sympathetic nerve activity [41]. In EV-treated rats, and in rats exposed to chronic cold stress, the onset of PCOS is accompanied by increases in sympathetic nerve activity [13, 42]. In the EV-induced PCOS model, the surgical sectioning of the SON restores estral cyclicity, ovulatory capacity, and the presence of corpora lutea [20, 21]. This response has been explained by decreases in adrenergic tone; however, as previously described, in addition to providing adrenergic fibers to the ovary the SON is also a source of other peptides such as VIP.

Hence the importance of using a more specific tool, such as GTD, that only destroys catecholaminergic neurons [37]. In the present study, the results obtained from estimating TH expression in control and EV-treated animals with GTD denervation shows that regardless of the age of treatment, the number of pixels with the immunoreactive mark was significantly lower in GTD treated animals. These results confirm the success of catecholaminergic fibers denervation.

According to the criteria established in 2006 by the Society of Excess of Androgens and Polycystic Ovarian Syndrome, hyperandrogenism is the most important parameter in PCOS diagnosis. In EV-induced PCOS rats, hyperandrogenism is explained by increases in steroidogenic activity in theca cells of cystic and pre-cystic structures in the ovaries [43] and increases in NE concentrations [13].

This hypothesis is supported by present study results, by the time the rats reached adulthood EV-treated animals with chronic GTD treatment stopped PCOS onset (anovulation and high of testosterone concentrations).

According to Lara et al. [24], an increase in ovarian neural growth factor (NGF) concentration is observed 15 days after EV treatment, resulting in higher TH activity in the superior mesenteric celiac ganglion (CSMG) and the subsequent release of NE to the ovary that precedes development of ovarian cysts.

In models 1 and 2 of the present study, denervation with GTD began before the ovarian NGF increases, suggesting that TH activity increases was not present. We can’t rule out that PCOS onset had taken place and was resolved by the nerves inactivity. New experiments are necessary to evaluate such possibility.

In animals of Model 3, in which EV-induced PCOS is fully established, denervation of the noradrenergic fibers reaching the ovary is efficient at restoring ovulation, but the testosterone concentration decrease was not significant. This can be explained by the persistence of some pre-cystic structures in the ovaries of these rats. The persistence of pre-cystic structures in the ovaries in animals from Model 3 could be explained by the short time between denervation and autopsy. Unlike the animals in Model 1 and 2, where autopsy was performed at least 50 days after denervation treatment, animals in Model 3 were sacrificed immediately after denervation treatment, and the time elapsed between denervation treatment and autopsy was not enough to allow the remodeling of the ovarian structure.

In addition to preventing hyperandrogenism and anovulation, peripheral noradrenergic denervation in EV-treated animals prevented morphological alterations of the ovaries, a characteristic of animals with PCOS. The ovaries of these animals showed corpora lutea and follicles in different growth stages, which allow us to suggest that the animals would probably be able to ovulate in later cycles.

In mammals, the sympathetic innervation of the ovary stimulates ovulation and secretion of steroid hormones [18, 44, 45]. According to several authors, the role of the ovarian sympathetic innervation varies according to the animal’s endocrine stage [16, 17, 31, 46, 47].

In the present study, the results obtained from Vh-treated and GTD-denervated animals support this idea. Uchida and Kagitani [48] reported that the electrical stimulation of the SON results in lower testosterone concentrations via the activation of alpha-1 adrenergic receptors, and in lower oestradiol levels by the activation of alpha-2 adrenergic receptors. The discrepancy between Uchida and Kagitani’s findings and the present study results on testosterone and oestradiol secretion regulation in the Vh-treated and GTD denervated animals may result from the destruction of noradrenergic neurons by GTD treatment, evidenced by the TH (activity / expression) reduction.

According to Kuncová et al. [49], GTD-treatment to pre-pubertal rats does not modify immunoreactivity to VIP in cardiac tissue. According to Benarroch et al. [50], adults rats treated with GTD for 5 weeks showed increased immunoreactivity to VIP neurons in the CSMG and not to VIP neurons in the superior cervical ganglion. The authors indicate their results were due to differences in afferent inputs to prevertebral and paravertebral ganglia. Working with adult rats, we’ve previously shown [51] that the effects on progesterone, testosterone, and oestradiol secretion resulting from unilaterally injecting VIP into the ovary’s intra-bursal depends on the injected ovary, the day of the estrous cycle treatment was performed, and the time elapsed between treatment and autopsy. These results suggest that the asymmetric response of the ovaries to VIP treatment is modulated, at least in part, by the innervation received by each ovary [51]. The effects of VIP on steroid hormones release by the ovaries are modulated by neural signals arriving through the SON (NE and/or NPY) ([52], in press). According to Lara et al. [31] there is no information on whether VIP-containing nerves in the SON or VIP-producing ovarian cells remain functional in GTD-treated rats. Therefore, in the present study we cannot rule out the possibility that the GTD-treatment had an impact on the VIP effects on the ovaries.

Studies have shown that VIP is a potent stimulant of steroid hormones secretion, and its effects have been explained through the regulation of enzymes involved in steroidogenesis [53]. The effect that VIP has on regulating steroid hormone secretion has been shown in experimental models, where the ovaries of adult EV-treated rats were removed and placed in a culture medium with VIP can release a greater quantity of androgens and estrogens than the ovaries of EV-treated rats cultured without the VIP-stimulus [26].

Taken together, the results of the present study support the notion that the increased peripheral noradrenergic tone seen in animals with EV-induced PCOS is an important factor in the onset of the pathophysiology.

There is evidence showing that even though the concentration of NE is high in rats with EV-induced PCOS, the reestablishment of ovarian function is not only due to a decrease in noradrenergic activity since ovulation restoration occurs in the innervated ovary and not in the denervated ovary of animals with EV-induced PCOS and unilateral sectioning of the SON [22].

Furthermore, in rats with EV-induced PCOS ovulation reestablishment occurs after unilateral or bilateral sectioning of the vagus nerve [30]. Morales-Ledesma et al. [54] showed that development the PCOS induced by subcutaneous injecting testosterone propionate was not modified by the unilateral or bilateral sectioning of the SON, suggesting that the etiology of PCOS depends not only on the hyperactivity of NE fibers and that other neurotransmitters are participating in the onset and maintenance of PCOS.

The present study contributes to the set of evidence that support the participation of the sympathetic innervation in PCOS onset and could lay the foundations to propose therapeutic alternatives that improve PCOS alterations. There is evidence that some women with PCOS can ovulate after treatment with low frequency electroacupuncture applied at the ovarian innervation origin level [55]. In clinical studies, low GTD doses have been used to reduce hypertension [56]. Since some patients with PCOS present metabolic alterations, including high blood pressure [57, 58], GTD treatment could reduce sympathetic activity and improve some of the alterations related to the PCOS. More studies are necessary to support such possibility.