Lavoro comune di Federica Collino, Silvia De Francia, Antonina Germano, Simona Perga, Monica Pradotto.
INTRODUCTION
Focusing on female fertility problems, according to recent studies, many evidences show that women who suffer from polycystic ovary syndrome present, in most cases, thyroid disorders, as hyperthyroidism or autoimmune thyroiditis which is often associated with hypothyroidism or at risk of future hypothyroidism. In fact, hypothyroidism may lead to lower levels of sex hormone binding globulin (SHBG), which in turn leads to high concentrations of testosterone, one of the factors that contribute to the onset of some symptoms of PCOS such as infertility, polycystic ovaries, hirsutism and acne.
Hypothyroidism, in fact, if untreated, leads to a worsening of other aspects of the syndrome and in particular the metabolic and the potential cardiovascular one.
Moreover, women with PCOS may be predisposed to "estrogen dominance" and excessively high levels of estrogen may reduce the effectiveness of thyroid hormones.
For this reason, if a woman is suffering from polycystic ovary syndrome is advisable to check the proper functioning of the thyroid through the measurement of TSH, an hormone produced by the anterior lobe of the pituitary gland, which controls the secretory activity of thyroid hormones, and viceversa.
p<>. In case of mild hypothyroidism, the diagnosis may be more difficult and this could require more complex laboratory tests compared to only TSH tests, such as control of body temperature for a certain period and careful assessment of symptoms and the clinical history of the patient.
It is also important to consider the opposite situation, in fact patients with thyroid disorders are frequaently affectd by PCOS.
Why PCOS is frequently associated with thyroid disorders and vice versa?
There is no a definitive answer, although many connections have been highlighted. Both the thyroid and ovaries are part of the endocrine system and belong to a common hormonal axis consisting of hypothalamus - pituitary – thyroid - ovaries. But the real link between PCOS and thyroid function seems to be the insulin resistance (IR).
Insulin resistance (IR) constitutes a common and broadly prevalent metabolic disorder, which seems to govern the pathophysiology of diabetes mellitus, metabolic syndrome, and obesity. Furthermore, IR appears to be a clinically important manifestation of various endocrine diseases, including, as said above, polycystic ovary syndrome (PCOS), thyroid and adrenal diseases, as well as their complications.
Many independent studies support the close connection between ovary function, thyroid function and insulin-resistance. In particular, as will be discussed below, thyroid hormones show a fundamental role in glucose metabolisms and insulin sensitivity, in turn related with correct ovarian function. As a consequence, pathological conditions which determine an altered level of thyroid hormones can lead to a situation of insulin resistance that contribute to the onset of PCOS, a major cause of infertility in women.
From a pathophysiological point of view, IR appears to be the end result of a complex interaction between genetic predisposition and environmental factors.
In general, IR indicates the presence of an impaired peripheral tissue response to endogenously secreted insulin. It is typically manifested as both decreased insulin-mediated glucose uptake (IMGU) at the level of adipose and skeletal muscle (SM) tissue, and as an impaired suppression of hepatic glucose output. A significant body of evidence supports the critical role of SM for the development of IR, most commonly through an interactive cross-talk with adipose and liver tissue.
1. Thyroid Disorders and Insulin resistance.
Thyroid hormones constitute important mediators of body metabolism and affect various metabolic aspects involving glucose and insulin metabolism, through a variety of mechanisms.
Evidence for a relationship between T4 and T3 and glucose metabolism appeared over 100 years ago when the influence of thyroid hormone excess in the deterioration of glucose metabolism was first noticed. Since then, it has been known that hyperthyroidism is associated with insulin resistance. More recently, hypothyroidism has also been linked to decreased insulin sensitivity. The explanation to this apparent paradox may lie in the differential and opposite effects of thyroid hormones at the liver and peripheral tissues level, obtained by a direct effect of THs on the regulation of the expression of different genes that regulate glucose homeostasis at the liver and peripheral tissues (muscle, fat tissue, and fibroblasts) (Tabella 1).
Tabella1
In particular it has been extensively demonstrated that thyroid hormones, and specifically T3, have insulin antagonistic effects at the liver level that lead to an increased glucose hepatic output, via an enhanced rate of gluconeogenesis and glycogenolysis (Tabella 1).
With regards to lipid metabolism in the liver, both lipogenesis and lipolysis are stimulated by T3. However, in the context of insulin resistance, the conversion of glucose into fatty acids together with nonsuppressed gluconeogenesis is simply perpetuating the hyperinsulinemic state. Furthermore, nutritional influences, such as those of high-fat diets, should also be taken into consideration as modifiers of the effects of thyroid hormones on insulin sensitivity.
Opposite to what occurs at the liver level, at peripheral tissues, thyroid hormones have been shown to exert some of their actions synergically with insulin. The upregulation of the expression of genes such as GLUT-4 or phosphoglycerate kinase (PGK), involved in glucose transport and glycolysis respectively, is a good proof of concept.
In skeletal muscle (SM), the main site of insulin-mediated glucose disposal, glucose transporter GLUT4, is induced by T3. Moreover, recent data reports that in muscle and fat tissue, in parallel to its transcriptional action on the SLC2A4 gene (which codifies for the glucose transporter GULT 4), T(3) exerts a rapid post-transcriptional effect on GLUT4 mRNA polyadenylation, which might increase transcript stability and translation efficiency, leading to the increased GLUT4 content and availability to skeletal muscle, as well as on GLUT4 translocation to the PM, improving the insulin sensitivity. In this way THs can increase basal and insulin-stimulated glucose transport in these tissue.
Another T3 target in skeletal muscle is mitochondrial uncoupling protein 3 (UCP3). Unveiling this association may be important since progressive reduction of UCP3 levels results in insulin resistance accompanied by decreased fatty acid oxidation and a less intense Akt/PKB and 5_ adenosine monophosphate activated protein kinase (AMPK) signaling. Although discrepancies between the regulation by T3 of UCP3 expression in rats, humans, and mice have been observed, the rat model has shed some light into T3 actions in this tissue. T3 intravenous (i.v.) administration in hypothyroid rats showed a rise in serum fatty acid levels concomitant with a rapid increase in UCP3 expression in gastrocnemius muscle. These findings point to UCP as a possible molecular determinant of the action of T3 on energy metabolism.
_In human adipocytes , T3 increases the mRNA levels of the lipolytic β2-AR, favouring catecolamine-induced lipolysis and it also downregulates Sterol regulatory element binding protein (SREBP1c), involved in lipogenesis, which may constitute a link between hyperthyroidism and insulin resistance.
_Skin fibroblasts have been also used to study thyroid hormone-responsive genes involved in metabolism in human cells. In cultured human fibroblasts, Moeller et al. observed that, opposite to a posttranscriptional regulation as reported for other growth factors and hormones, the mRNA of the transcription factor HIF-1α (Hypoxia-inducible factor 1), a key mediator of glycolysis, increased in response to T3.
As the glucose transporter GLUT1, several enzymes of glycolysis, and the lactate exporter SLC16A3 were all also found induced by T3 and are target genes of the transcription factor HIF-1α, the authors postulated that the effect of thyroid hormones on the induction of these genes most probably was indirect and HIF-1α mediated. Furthermore, a new mechanism of thyroid action was unraveled by this group of researchers. It was shown that T3 bound to TRbeta, in lieu of initiating gene transcription in the nucleus, activates the phosphatidylinositol 3-kinase (PI3K) signaling pathway in the cytosol in order to activate HIF-1α gene expression.
At the cellular level, thyroid hormones can also increase mitochondrial biogenesis, fatty acid oxidation, and TCA cycle activity. These findings are quite relevant since the role of mitochondrial dysfunction, leading to cellular lipid excess and impaired oxidative metabolism, has been clearly demonstrated in the pathogenesis of type 2 diabetes. Furthermore, it has been described that in skeletal muscle, the lack of thyroid hormones might dysregulate mitochondrial gene expression.
PPAR gamma coactivator- 1 alpha (PGC-1 alpha), a key transcriptional regulator of mitochondrial content and function, fatty acid oxidation, and gluconeogenesis, has been involved in the process whereby thyroid hormones regulate mitochondrial function. It has been shown that PGC-1 alpha gene expression is increased by T3, as much as 13-fold 6 hours after T3 treatment. The regulation pattern of T3 on PGC-1 alpha is complex and may occur through nongenomic activation of kinases to induce the expression of PGC-1 alpha or through transcriptional upregulation via the presence of a thyroid responsive element (TRE) in the PGC-1 alpha promoter or by genomic upregulation of a transcription factor (via a TRE), which then binds to the PGC-1 alpha promoter and increases PGC-1 alpha transcription. It is hypothesized that PGC-1 alpha can be dysregulated by reduced T3 levels, thus contributing to insulin resistance. Not only low circulating but also, intracellular T3 levels, could count for this matter.
In addition, bile acids are potent stimulators of the enzyme PGC-1 and may play an important role in the relationship between thyroid action and glucose metabolism. On the other hand, the natural occurrence of polymorphisms of deiodinase type 2 such as Thr92Ala, with a lower activity, has also been implicated with increased risk for diabetes type 2.
1.1 Insulin Resistance as a Consequence of Hyperthyroidism
Thyrotoxic subjects frequently show impaired glucose tolerance. This is a result of increased glucose turnover with increased glucose absorption through the gastrointestinal tract, postabsorptive hyperglycemia, elevated hepatic glucose output, with elevated fasting and/or postprandial insulin and proinsulin levels, elevated free fatty acid concentrations and elevated peripheral glucose transport and utilization. The literature about this topic is vast and has been previously comprehensively reviewed by Dimitriadis and Raptis.
Thyrotoxic diabetic patients are more prone to ketosis. Although ketoacidosis may result per se from the insulin resistance present in thyrotoxicosis, the stimulatory action of thyroid hormones in excess on lipolysis and free fatty acids availability can also contribute to increased ketogenesis.
1.2. Hypothyroidism Can Lead to Insulin Resistance
Although seldom happening, hypothyroid patients can experience hypoglycaemia. This phenomenon can be interpreted in the light of reduced levels of gluconeogenesis leading to decreased liver glucose output. On the other hand, insulin resistance has been shown to be present in peripheral tissues in hypothyroid animal models. Hence, a poor utilization of glucose in hypothyroidism may be offset by a reduced release to circulation maintaining a balance of the glucose metabolism.
Studies performed in adipocytes and skeletal muscle of rats of mature rats rendered hypothyroid have shown that these tissues are less responsive to insulin with regards to glucose metabolism. In particular, it was observed that glucose conversion to glycogen was partially inhibited while the glycolytic flux stimulation by insulin was totally frustrated. This decrease in insulin sensitivity occurred without an impaired membrane insulin effector system. The authors postulated that they rather occurred through a post-receptor mechanism that may include abnormal phosphorylation of insulin signaling proteins.
According other studies on animals, another important factor of decreased insulin responsiveness in hypothyroidism includes the dysregulation of leptin action at the hypothalamus.
Also adipocyte-myocyte crosstalk by adipokines has been reported to play a significant role in skeletal muscle insulin resistance and may partially explain insulin resistance present in hypothyroidism. However, other factors associated with insulin resistance in hypothyroidism, such as altered blood flow, impaired GLUT4 translocation, decreased glycogen synthesis, and decreased muscle oxidative capacity have to be also considered.
Besides animal studies, in literature are present also different studies in humans demonstrating insulin resistance in hypothyroidism, even if, compared to the number of reports about insulin resistance in hyperthyroid patients, there are relatively fewer studies in humans dealing with the effects of hypothyroidism on glucose metabolism.
Rochon et al. measured whole-body sensitivity of glucose disposal to insulin in hypothyroid patients using the euglycemic-hyperinsulinemic clamp technique. They demonstrated that hypothyroidism induced a decrease in the insulin-mediated glucose disposal that reverted upon treatment. Handisurya et al. confirmed these findings and added the knowledge that glucose-induced insulin secretion is diminished by thyroid replacement corresponding well with the observed improvement of insulin sensitivity.
Dimitriadis et al. explored glucose uptake in muscle and adipose tissue of hypothyroid and control subjects by means of the arteriovenous difference technique in the anterior abdominal subcutaneous adipose tissue and the forearm muscles after the consumption of a mixed meal.
A decreased net extraction of glucose and blood flow after the meal in hypothyroid muscle and adipose tissue was reported. This impairment in the ability of insulin to increase blood flow rate to the hypothyroid tissues is an alternative explanation to the mechanism whereby hypothyroidism can induce lower glucose disposal.
In conclusions, thyroid hormones have a large impact on glucose metabolism. A direct regulation on thyroid responsive genes at the target organ has been described and more recently an indirect effect involving hypothalamic pathways that regulate glucose metabolism via control of the sympathetic nervous system has been reported. Furthermore, thyroid hormone effects can be insulin agonistic, such as demonstrated in muscle or antagonistic such as observed in the liver. In hyperthyroidism, dysregulation of this balance may end in glucose intolerance mainly due to hepatic insulin resistance.
In hypothyroidism the results are less evident. However, the available data suggest that insulin resistance is present mainly at the peripheral tissues. Possible explanations hypothesized to explain this phenomenon span from the dysregulation of mitochondrial oxidative metabolism to the reduction of blood flow in muscle and adipose tissue under hypothyroid conditions.
2. PCOS and Insulin Resistance (IR).
PCOS is the most common endocrine disorder among young women. Despite its steadily increasing prevalence, the fundamental underlying defect remains still speculative and seems to be multifactorial in origin. Beyond the reproductive abnormalities (chronic anovulatory dysfunction, infertility), women with PCOS display several metabolic abnormalities as well, including disorders of glucose metabolism and* insulin action*, which underlie the increased risk of developing impaired glucose tolerance and type 2 diabetes.
In fact, many studies have identified that insulin resistance appears to be responsible for many of these long term health consequences. Women with PCOS exhibit basal hyperinsulinemia, decreased glucose-stimulated insulin release and IMGU, due to reduced hepatic insulin clearance and pancreatic β-cell dysfunction.
In addition, they exhibit a generalized insulin resesitance, which mainly involves an impaired insulin responsiveness of adipose tissue and SM.
Obesity contributes to the risks, but not all women with PCOS are obese.
IR in PCOS is associated with a unique postbinding defect in insulin receptor signaling due to a complex interaction between intrinsic (genetically determined) and environmental factors. It is supported that the disrupted insulin receptor tyrosine kinase activity in adipocytes and IRS-1- associated PI3K activity in SM are the key elements in the IR pathogenetic process. In addition, extrinsic factors including inflammatory mediators, adipokines, androgens, free fatty acids (FFAs), amino acids, and increased glucose levels have been all implicated in the pathogenesis of IR in PCOS.
As far as leptin is concerned, Pehlivanov and Mitkov reported higher serum leptin in PCOS demonstrating a positive correlation with IR, independently of markers of adiposity.
The existing data indicate that adipose tissue, which is also an insulin resistant site in PCOS, and especially when it is centrally accumulated (visceral fat), secretes increased levels of adipokines, FFA, and inflammatory mediators (TNF-α, IL-6), which in turn promote IR at the level of SM via a vicious cycle. This was well described by Ek et al. who reported an increased rate of visceral fat lipolysis in PCOS, suggesting a genetically determined upregulation in visceral fat lipolysis, associated with a selective increase in the function of both protein kinase A (PKA) and hormone-sensitive lipase (HSL). It is obvious that the increased FFA influx has detrimental effects for insulin metabolic signaling in skeletal muscle (SM).
In parallel, multiple studies suggest the association between androgen excess and IR in women with PCOS, but their cross-sectional nature does not allow safe conclusions about causality.
For a review: Effects of androgens on insulin action in women: is androgen excess a component of female metabolic syndrome? 2008
On one hand, hyperandrogenemia in women with PCOS appears to be an effect of the augmented steroidogenesis by hyperinsulinemia secondary to IR. On the other hand, hyperandrogenemia induces generalized and muscle IR, through either a direct effect of androgens on insulin action in AT and SM, or indirectly by affecting lipid metabolism and body fat distribution.
Hyperandrogenemia-induced IR is selective, affecting mainly the metabolic but not the mitogenic actions of insulin, since insulin-stimulated ovarian steroidogenesis is perfectly maintained. Androgen excess has been associated with some of the typical insulin-signaling defects in PCOS. Most of these effects have been mostly studied in female rats, while there is a relative paucity of similar clinical studies. Testosterone administration to female adult rats for 8–12 weeks caused hyperinsulinemia in both intact and ovariectomized animals, while in the latter it induced a 50% reduction in IMGU into SM. Impaired SM insulin action was combined with fewer type 1 muscle fibers (slow twitch, oxidative) and increased type 2 fibers (fast twitch, insulin resistant). In the same study, a decreased SM capillary density and an impaired muscle glycogen synthase activity were also reported, contributing to the observed IR of SM. It seems that post-insulin receptor signaling events are involved in testosterone-induced IR in SM in this rat model. In support of this, experimental data in primary differentiated rat myotubes have demonstrated a synergistic interaction between testosterone and insulin in phosphorylation of intracellular signaling proteins (phosphorylation of IRS-1 at serine residues), resulting in a dissociation of insulin receptor from the PI3K signaling cascade and an impaired insulin metabolic signaling.
Enhanced mitogenic signaling in skeletal muscle of women with polycystic ovary syndrome, 2006
Effects of testosterone on muscle insulin sensitivity and morphology in female rats, 1990
Effect of testosterone on insulin stimulated IRS1 Ser phosphorylation in primary rat myotubes--a potential model for PCOS-related insulin resistance, 2009
Human Studies. In vivo studies of PCOS patients, where serial SM biopsies were performed during hyperinsulinemiceuglycemic clamps, have revealed a significant impairment in IMGU, an increased expression of IRS-2, and normal expression of insulin receptor, IRS-1, and PI3K. These data suggest that the increased expression of IRS-2 might represent a compensatory adaptation for the decreased insulin-mediated IRS-1-associated PI3K activity, which is not, however, completely effective, since IMGU was not restored to normal. Most recently, Hojlund et al. reported small reductions in insulin-stimulated phosphorylation of Akt (PKB) and AS160 (Akt substrate of 160 kDa) in intact muscle of women with PCOS, which was partially reversed by treatment with insulin-sensitizing agents such as pioglitazone.
Recently, another candidate pathogenetic mechanism for muscle IR in women with PCOS has been proposed, consisting in a defective insulin regulation of ERK 1/2 (extracellular signal-regulated kinases 1/2).
Defects in insulin receptor signaling in vivo in the polycystic ovary syndrome (PCOS), 2001
Impaired insulin-stimulated phosphorylation of Akt and AS160 in skeletal muscle of women with polycystic ovary syndrome is reversed by pioglitazone treatment, 2008
Insulin resistance in polycystic ovary syndrome is associated with defective regulation of ERK1/2 by insulin in skeletal muscle in vivo, 2009
Summarizing the existing data, androgens promote IR at the tissue level of SM by reducing capillary network formation for adequate delivery of insulin to SM, switching muscle fiber isoforms, reducing glycogen synthase activity and impairing insulin-mediated GLUT4 plasma membrane translocation.
However, the finding of hyperinsulinemia in PCOS patients raises the questions what the cause is and what the effect is. Most of the existing clinical data suggest, without providing definitive confirmation, that hyperinsulinemia causes hyper androgenism, more than the other way around.
Based on the few existing human in vivo studies, it seems that in PCOS there is a severe functional defect in the insulin-signaling cascade within SM, consisting in an abnormal phosphorylation pattern of the insulin receptor or downstream key signaling proteins.
