BACKGROUND Functional ovarian hyperandrogenism, including the polycystic ovary syndrome (PCOS), is very prevalent in women with type 1 diabetes (T1D). The pathogenic mechanisms of this association remain unclear.

HYPOTHESIS Individual factors expose or protect women with T1D to/from the development of androgen excess and PCOS. Such androgen excess in women with T1D may increase their cardiometabolic risk.

MAIN OBJECTIVE Unveiling the pathogenic mechanisms behind functional hyperandrogenism in women with T1D from a sex/gender-medicine and sexual dimorphism perspective.

MATERIAL AND METHOS We have designed a cross-sectional comparative clinical study, including 5 groups of study subjects with 12 participans per group:

i) Women with T1D & PCOS. ii) Women with T1D without PCOS. iii) Men with T1D and normal gonadal function. iv) Women with PCOS without diabetes mellitus v) Non-hyperandrogenic control women without T1D. All groups will show similar age and body mass index. T1D groups will be matched for duration of disease.

OUTCOMES 1.1 Insulin sensitivity (hyperinsulinaemic euglycaemic clamping). 1.2 Body composition (dual-energy x-ray absorptiometry, bioelectrical impedance analysis & sonographic studies).

1.3 Ovarian and adrenal steroidogenesis. 2.1 Differential pattern in genetic variants related with insulin signalling and response, inflammation, adiposity, gonadal function, steroidogenesis, and PCOS itself by whole exome sequencing.

2.2 Microbiopsy studies in deep subcutaneous adipose tissue and skeletal muscle tissue: 2.2.1 Differential DNA methylation patterns in genes associated with PCOS. 2.2.2 Differential transcriptomic pattern in genes associated with PCOS.

2.2.3 Differential proteomic patterns in adipose and muscle tissues. 3. Interaction between T1D and PCOS on parameters of metabolic control (intersticial blood glucose monitoring) and morbidities associated with T1D itself.

Functional ovarian hyperandrogenism, and its most frequent phenotypic expression, polycystic ovary syndrome (PCOS), has a worldwide prevalence similar to that of other pandemic metabolic entities such as type 2 diabetes, with figures ranging from 6.5% of women in our environment using strict classical criteria to 17-21% of premenopausal women from Europe and the United States, according to the most recent and inclusive diagnostic criteria. Almost all classic and non-classic cardiovascular risk factors cluster in women with this condition from their early lifespan. Conditions such as obesity, type 2 diabetes, hypertension, or dyslipidaemia, place this prevalent population at a higher risk of cardiovascular events compared to non-hyperandrogenic women. PCOS is a complex syndrome with familial aggregation, in which protective and facilitating environmental factors trigger the onset of the hyperandrogenic phenotype and its metabolic repercussions on a predisposing genotype.

POLYCYSTIC OVARY SYNDROME AND TYPE 1 DIABETES MELLITUS One of the metabolic events inherent to PCOS is a deficient organ-dependent insulin action primarily at the liver. Acting upon this central defect, obesity is the major contributor to peripheral insulin resistance in these women. Since insulin acts as a co-gonadotrophin on theca cells by stimulating various enzymes involved in ovarian and adrenal steroidogenesis, any condition associating endogenous hyperinsulinism, such as obesity or type 2 diabetes, may be associated with PCOS. However, our research group first described the association between PCOS and T1D more than 20 years ago. In that pivotal publication these adolescent and young adult women, who suffered from complete impairment of insulin secretion as the primary mechanism of disease instead of insulin resistance and compensatory hyperinsulinism, had a 3-fold increase in the prevalence of classic PCOS compared with unselected non-hyperandrogenic women from the general population.

This association has been confirmed in subsequent studies conducted by different groups worldwide. In a recent meta-analysis and systematic review by our group, we reported an increased prevalence of PCOS; in its upper range, such prevalence might reach 34% of patients with T1D, tripling the figures observed in the general population for the classic PCOS phenotype, which is the most severe in terms of cardiometabolic consequences. From a pathophysiological view, and given the obvious absence of endogenous hyperinsulinism in T1D, this relationship must be necessarily supported by exogenous subcutaneous insulin delivery. In healthy subjects, insulin directly reaches the liver through the portal circulation after its pancreatic secretion. After exerting its actions at this level, with its subsequent hepatic clearance, insulin passes into the systemic circulation at much lower concentrations than those found in the portal circulation. In PCOS and other insulin-resistant states, insulin resistance is compensated by increased pancreatic insulin secretion, resulting in portal and systemic hyperinsulinism. In contrast, in subjects with T1D, insulin is administered subcutaneously (non-physiologically) and exerts its actions on different organs and tissues, including the gonads and adrenal cortex, before reaching liver tissue at concentrations sufficient to suppress gluconeogenesis; hence, systemic insulin concentrations are necessarily supraphysiological for this reason. In contrast, hepatic insulin levels are not excessive, since this would precipitate hypoglycaemia, and this explains the constant finding of normal levels of sex hormone binding globulin (SHBG) in women with T1D and PCOS. In contrast, in women with PCOS without T1D, SHBG synthesis and secretion is decreased as a consequence of portal hyperinsulinism (certain adipokines secreted by visceral adipose tissue also contribute to this inhibition), a situation that does not occur in women with T1D in the absence of hepatic hyperinsulinism.

A few studies have reported a negative interaction between T1D and PCOS in terms of micro- or macrovascular complications. Nevertheless, the possibility that androgen excess negatively affects adipose tissue distribution in women with T1D and PCOS is plausible, placing them at risk for increased abdominal adiposity and insulin resistance. On the other hand, similarly to those women at high risk of PCOS in the general population - for example, those with obesity, even extreme obesity, or first-degree relatives of women with PCOS - who do not develop characteristics of the syndrome, not all patients with T1D, universally treated with subcutaneous insulin, will develop PCOS, indicating that a predisposition to PCOS is a conditio sine qua non for its occurrence.

DEFECTS IN OVARIAN AND ADRENAL STEROIDOGENESIS Both the fact that PCOS is not universal in women with insulin resistance and hyperinsulinism, and that insulin resistance is not universal in all women with PCOS, suggests that there is a primary defect that favours androgen excess in affected women, and that this is essential for the development of the syndrome in response to insulin or other triggers. The elegant studies conducted by the McAllister's group 20 years ago at the University of Pennsylvania demonstrated that, after several passes in primary culture, theca cells from patients with PCOS produced an excess androgen secretion compared to those of control women without the syndrome. This occurred because an intrinsically increased expression and activity of the enzyme 17α-hydroxylase/17,20-desmolase (P450c17, CYP17), the qualitative regulatory enzyme of sex steroid synthesis that catalyses both the 17α-hydroxylation of pregnenolone and progesterone, and the conversion of 17α-hydroxypregnenolone to dehydroepiandrosterone. The 17,20-desmolase activity of the enzyme is regulated by post-transcriptional mechanisms, including phosphorylation of the serine/threonine residues of the enzyme itself. The same group also demonstrated increased expression in theca cells of women with PCOS of the P450scc enzyme, which is the quantitative regulatory enzyme of ovarian and adrenal steroidogenesis that catalyses the conversion of cholesterol to pregnenolone after cholesterol transfer from the outer to the inner mitochondrial membrane by the steroidogenesis acute regulatory protein (StAR). Since any factor that could influence these theca cells in vivo were, obviously, not present after several culture passes, these alterations place excessive androgen production as a primary ovarian defect in PCOS, corroborating clinical findings such as hyper-responsiveness of 17-hydroxyprogesterone to gonadotrophin stimulation or persistent hyperandrogenemia after suppression of adrenal steroidogenesis.

In about one third of women with PCOS, hyperandrogenism also has an adrenal component. An adrenal hyper-responsiveness to corticotrophin stimulation is common in women with PCOS and hyperandrogenism of adrenal origin. A decrease in peripheral 11β-hydroxysteroid dehydrogenase-dependent cortisol regeneration can induce a compensatory hyperactivation of the corticotrophic axis, and this defect can be found at the hepatic level in PCOS, induced by ovarian hyperandrogenism itself. Another factor related to adrenal activation would be the increased expression and activity of the 17α-hydroxylase/17,20-desmolase enzyme previously mentioned. Increased phosphorylation of the serine/threonine residues of this enzyme together with that of the serine residues of the insulin receptor substrate-1 (IRS-1) could constitute a common event linking ovarian and adrenal hyperandrogenism and insulin resistance. Additionally, a partial functional defect in the action of 3β-hydroxysteroid dehydrogenase could lead to the increased Δ5/Δ4 adrenal precursor ratio observed in some women with PCOS, either caused by the hyperinsulinism characteristic of the syndrome, or by common alterations in the regulation of 17α-hydroxylase/17,20-desmolase and 3β-hydroxysteroid dehydrogenase by the MEK/ERK signalling pathway.

ADIPOSITY, VISCERAL ADIPOSITY AND BODY COMPOSITION IN PCOS The contribution of obesity to gonadal dysfunction in women and men is well-known. Excess weight increases the prevalence of classical PCOS up to 4-fold in unselected women in our setting, and 30%-50% of women with extreme obesity have features consistent with a diagnosis of PCOS. Furthermore, the estimated prevalence of obesity in women with PCOS is close to 50%, although even more remarkable is the prevalence of central adiposity, derived from a 'masculinised' distribution of adipose tissue, consisting of visceral deposition. This visceral deposition may begin in childhood and during puberty, reinforcing the hyperandrogenic phenotype in predisposed girls. During adulthood, obesity and central adiposity worsen the cardiovascular profile of women with PCOS.

Androgen excess itself influences gene expression and the proteome of visceral adipose tissue, and thus a male-like pattern of adipokine expression in visceral adipose tissue has been described. Androgens are also potent inhibitors of adipogenic differentiation, limiting the number of adipocytes and their storage capacity in subcutaneous adipose tissue, so that androgen excess may limit the adaptive capacity of adipose tissue in women with PCOS by creating a state of lipotoxicity and dysfunctional adipocytokine secretion. Oxidative stress and subclinical chronic inflammation inherent to PCOS also play a role in the inability to physiological expansion of adipose tissue and its metabolic dysregulation. Ultimately, the interrelationship between PCOS and abdominal obesity is the result of a vicious circle in which androgenic excess favours the deposition of abdominal visceral fat, which directly facilitates androgenic excess of ovarian and adrenal origin through the secretion and effect of autocrine, paracrine and endocrine mediators [down-regulation of adiponectin and up-regulation of tumour necrosis factor-alpha (TNF-α), IL-6 and leptin], or indirectly through the induction of insulin resistance and compensatory hyperinsulinism.

INSULIN RESISTANCE, INFLAMMATION AND OXIDATIVE STRESS IN PCOS PATHOPHYSIOLOGY Euglycaemic clamp studies are consistent with a post-ligand binding defect in the insulin receptor affecting the intracellular metabolic signalling pathway in adipose tissue; an alteration supported by findings from metabolomic studies showing an increase in plasma long-chain fatty acids and glycerol, suggesting increased lipolysis, likely related to decreased sensitivity to insulin action in adipose tissue. Consistent with this, a substantial number of women with PCOS have increased phosphorylation of serine residues of the insulin receptor, whose function is inhibited by reducing the intrinsic kinase activity of its tyrosine residues. Such phosphorylation is mediated by TNF-α, especially in those patients with abdominal obesity, a scenario derived from the hyperandrogenic stimulus to the release of this cytokine by mononuclear cells, both after fasting and in response to glucose intake. In other women, phosphorylation of IRS-1 or activation of phosphatidylinositol-3-kinase aggravates insulin resistance. However, insulin resistance at the adipocyte level may also be mediated by another complementary mechanism, namely, the infiltration of this tissue by macrophages. Inflammatory macrophages characterised by CD11c expression cluster around adipocytes forming crown-like structures, the density of which directly correlates with the degree of insulin resistance. These macrophages will secrete cytokines such as TNF-α, promoting a local inflammatory environment. Women with PCOS show increased CD11c expression in macrophages and crown-like structures in subcutaneous adipose tissue in direct proportion to circulating testosterone levels. In muscle tissue of women with PCOS, the mitogenic pathway of the insulin receptor, MAPK-ERK 1/2, is also constitutively activated enhancing increased phosphorylation of IRS-1 serine residues. The presence of mitochondrial dysfunction in women with functional hyperandrogenism may contribute to the presence of insulin resistance. Oxidative stress induced by this mitochondrial dysfunction is involved in the pathogenesis of PCOS and metabolic complications related to hyperandrogenism. Although some studies have failed to confirm a primary impairment of mitochondrial function in PCOS, biomarkers of oxidative stress are increased in these women regardless of the presence of weight excess. Moreover, the alterations in mitochondrial function of hyperandrogenic women with insulin resistance resemble those of patients with type 2 diabetes.

GENETIC PREDISPOSITION PCOS is a complex multigenic condition that arises from the interaction between protective and predisposing genetic variants, which may have been selected over the ages by evolution due to an ancestral survival advantage, with environmental factors playing a determining role in the expression of the hyperandrogenic phenotype. Among others, genomic variants in genes regulating androgen biosynthesis, insulin action and inflammation may be associated with predisposition to PCOS. A meta-analysis of large-scale genome-wide association studies, including more than 10,000 women with PCOS and 100,000 controls, all of European origin, identified 14 loci associated with hyperandrogenism, gonadotrophin regulation and testosterone concentrations in affected women, and correlations with obesity, insulinemia, type 2 diabetes, lipid levels and coronary atherosclerotic disease, indicating a shared genetic architecture between these metabolic alterations and PCOS.

These pathophysiological mechanisms described in women with PCOS from the general population have not been studied in women with T1D. Therefore, the description of facilitating and protective factors associated with the development or absence of functional hyperandrogenism and PCOS in women with T1D is the main objective of the present research project.

HYPOTHESIS There are individual factors that expose or protect women with T1D to/from the development of androgen excess and PCOS, despite the iatrogenic systemic hyperinsulinism they all experience as a result of the subcutaneous administration of insulin they require for survival. Androgen excess in women with T1D may increase their cardiometabolic risk.

STUDY OBJECTIVES MAIN OBJECTIVE: Unveiling the pathogenic mechanisms behind functional hyperandrogenism in women with T1D from a sex medicine perspective.

1. Primary aims: To identify predisposing/protective factors associated with the development of functional hyperandrogenism in women with T1D by phenotyping these women using state-of-the-art techniques for the assessment of:

   1.1. Insulin sensitivity. 1.2. Body composition. 1.3. Ovarian and adrenal steroidogenesis.

2. Secondary aims: Identification of relevant molecular-genetic factors associated with the development of functional hyperandrogenism in women with T1D 2.1. Identify genomic variants in genes/proteins related to insulin signalling and action, adiposity, inflammation, steroidogenesis, and PCOS itself (Genomics).

   2.2. Description of differential DNA methylation patterns in deep subcutaneous adipose and skeletal muscle tissues (Epigenomics).

   2.3. Description of differential gene expression profiles in adipose and muscle tissues (Transcriptomics).

   2.2. Description of differential proteomic profiles in adipose and muscle tissue (Proteomics).

3. Exploratory aim: To evaluate the interaction between T1D and PCOS on parameters of metabolic control and morbidities associated with T1D itself.

1. Non--hyperandrogenic women with type 1 diabetes INCLUSION CRITERIA

   • Premenopausal women between 18 and 45 years old.
   • Diagnosis of type 1a diabetes at least 12 months before inclusion in the study, confirmed by positive autoimmunity and complete insulin deficiency.
   • Treatment with subcutaneous insulin therapy (multiple doses or continuous infusion).
   • Availability of metabolic control data (continuous interstitial blood glucose monitoring) at least in the month prior to study entry.
   • Menarche at least three years prior to study entry. EXCLUSION CRITERIA
   • Honeymoon period of T1D.
   • Pregnancy or lactation.
   • Thyroid hormone dysfunction or hyperprolactinaemia.
   • Diagnosis of non-classical congenital adrenal hyperplasia or other secondary causes of hyperandrogenism.
   • Diagnosis of other serious chronic disease.
   • Treatment with oral contraceptives or glucocorticoid therapy in the 3 months prior to inclusion in the study.

2. Women with type 1 diabetes and polycystic ovary syndrome INCLUSION CRITERIA

   • Women between 18 and 45 years old.
   • Diagnosis of type 1a diabetes at least 12 months before inclusion in the study, confirmed by positive autoimmunity and complete insulin deficiency.
   • Treatment with subcutaneous insulin therapy (multiple doses or continuous infusion).
   • Availability of metabolic control data (continuous interstitial blood glucose monitoring) at least in the month prior to study entry.
   • Menarche at least three years prior to study entry.
   • PCOS diagnosis based on the 2012 American NIH consensus criteria, including the Rotterdam and AE-PCOS.

   EXCLUSION CRITERIA

   • Honeymoon period of T1D.
   • Pregnancy/lactation.
   • Thyroid hormone dysfunction or hyperprolactinaemia.
   • Diagnosis of non-classical congenital adrenal hyperplasia or other secondary causes of hyperandrogenism.
   • Diagnosis of other serious chronic disease. reatment with oral contraceptives or glucocorticoid therapy in the 3 months prior to inclusion in the study.

3. Men with T1D and normal gonadal function of similar age, BMI, and duration of diabetes.

   INCLUSION CRITERIA

   • Age between 18 and 45 years old.
   • Diagnosis of type 1a diabetes at least 12 months before inclusion in the study, confirmed by positive autoimmunity and complete insulin deficiency.
   • Treatment with subcutaneous insulin therapy (multiple doses or continuous infusion).
   • Availability of metabolic control data (continuous interstitial blood glucose monitoring) at least in the month prior to study entry.

   EXCLUSION CRITERIA

   • Honeymoon period of T1D.
   • Thyroid hormone dysfunction or hyperprolactinaemia.
   • Diagnosis of non-classical congenital adrenal hyperplasia.
   • Diagnosis of male hypogonadism.

4. Women with PCOS of similar age and BMI. INCLUSION CRITERIA

   • Women between 18 and 45 years old.
   • Menarche at least three years prior to study entry.
   • PCOS diagnosis based on the 2012 American NIH consensus criteria, including the Rotterdam and AE-PCOS.

   EXCLUSION CRITERIA

   • Pregnancy/lactation.
   • Previously known carbohydrate metabolism abnormalities (prediabetes or type 2 diabetes).
   • Thyroid hormone dysfunction or hyperprolactinaemia.
   • Diagnosis of non-classical congenital adrenal hyperplasia or other secondary causes of hyperandrogenism.
   • Diagnosis of other serious chronic disease.
   • Treatment with oral contraceptives or glucocorticoid therapy in the 3 months prior to inclusion in the study.

5. Non-hyperandrogenic control women with regular menses of similar age and BMI. INCLUSION CRITERIA

   • Women between 18 and 45 years old.
   • Menarche at least three years prior to study entry.
   • Presence of regular menses.
   • Lack of signs or symptoms of functional hyperandrogenism. EXCLUSION CRITERIA
   • Pregnancy/lactation.
   • Previously known carbohydrate metabolism disturbances.
   • Thyroid hormone dysfunction or hyperprolactinaemia.
   • Diagnosis of non-classical congenital adrenal hyperplasia or other secondary causes of hyperandrogenism.
   • Diagnosis of other serious chronic disease.
   • Treatment with oral contraceptives or glucocorticoid therapy in the 3 months prior to inclusion in the study.

Premenopausal adult women with type 1 diabetes mellitus presenting with or without polycystic ovary syndrome.

All data sets generated and/or analyzed during the current study will not be publicly available but will be available from the Principal Investigator on reasonable request.