An inverse association between serum soluble receptor of advanced glycation end products and hyperandrogenism and potential implication in polycystic ovary syndrome patients

Polycystic ovary syndrome (PCOS) is the most common endocrinological disorder in women of reproductive age [1], which is characterized by hyperandrogenism, chronic anovulation and/or PCO morphology, with the exclusion of the adrenal, ovary and pituitary disorders [2]. PCOS is also characterized by metabolic aberrations, such as hyperinsulinemia, insulin resistance (IR), impaired glucose tolerance (IGT), diabetes mellitus (DM) and increased several cardiovascular risk factors [3, 4]. However, the etiology of PCOS remains largely unknown. Advanced glycation end products (AGEs) are the end products of a chemical procedure called Maillard reaction [5], which are formed by nonenzymatic modification of proteins, lipids, and nucleic acids [6]. Serum AGEs have been shown to be elevated in women with PCOS, compared with controls. In recent years, AGEs have been shown to be implicated in the pathogenesis of PCOS [7‐9]. In addition, they also have been proposed to be among the main intermediaries of several diseases, such as type 2 diabetes mellitus (T2DM), obesity, metabolic syndrome, aging and inflammation [9, 10].

The action of AGEs is classified as receptor independent (by binding to the extracellular matrix) and receptor dependent (by binding to a cellular receptor called RAGE) manners [6]. The AGE-RAGE interaction leads to activation of signaling pathways that in turn activate nuclear factor-kappa B(NF-κB), promoting the development of a proinflammatory state, cellular toxicity and damage [11]. An extracellular form of RAGE circulates in the blood and follicular fluid is called soluble receptor of AGEs (sRAGE). In contrast to RAGE, sRAGE acts as a decoy receptor, as it binds circulating AGEs and prevents the adverse intracellular events of the AGE-RAGE binding [11, 12]. Therefore, sRAGE is often considered as a ‘good’ receptor since its levels have shown to be down-regulated in hyperglycemia and obesity [13, 14]. However, there were very few studies of the variant levels of serum AGEs and sRAGE in PCOS patients with different body mass index (BMI), waist circumference (WC), homeostatic model assessment-insulin resistance (HOMA-IR), total testosterone (TT) and free androgen index (FAI).

The aim of this study is to investigate the variations of AGEs/sRAGE levels in Chinese Han population with PCOS. The relations between their levels and clinical marks, BMI, WC, HOMA-IR, TT and FAI, have also been studied. Finally, the cut-off values of sRAGE at insulin resistance risk were determined, as well.

With an estimated prevalence of up to 10% [17], PCOS is widely accepted as the most common cause of infertility with clinical and/or biochemical signs of excess androgen secretion in the reproductive-aged women, and most of the patients have metabolic abnormalities such as hyperinsulinemia, obesity et al, implicating long-term sequelae which may affect women’s long-term health [4, 18, 19]. Therefore, searching for a new therapeutic target for PCOS is one of the highlights today. In recent years, some studies, including our research group’s preliminary work, have found that there was an elevated level of serum AGEs in PCOS women [7‐9]. AGE-RAGE system is closely related to insulin resistance and hyperandrogenism, which are two core pathophysiological processes in PCOS. In addition to normal receptors, one form of AGE receptor, lacking both the cytosolic and the transmembrane domains, are called soluble RAGEs (sRAGE). They are secreted extracellularly, and can be detected in circulating blood [12, 20]. sRAGE can bind with their ligands (AGEs) in the circulation, thus preventing the adverse effects of the AGE-RAGE axis. sRAGE is then often considered a ‘good’ receptor. sRAGE levels have been shown to be down-regulated in hyperglycemia and obesity, which has been explained by the beneficial role of sRAGE and its function as decoy to capture the circulating AGEs preventing activation of RAGE signaling pathway as it still possesses the V-ligand essential for ligand binding. [13, 21, 22] Moreover, the decoy function theory is supported by an in vivo study which demonstrated that the treatment with sRAGE normalized the increases in most of the inflammatory markers in diabetic rats to those seen in age matched, non-diabetic controls. [23] But the concept of sRAGE as a decoy receptor is not unanimously supported, or at least it seems not to be complete. Because the role of sRAGE as a protective factor for vascular damage has been questioned. High levels of sRAGE have been reported associated with cardiovascular mortality in diabetic patients. Furthermore, sRAGE has been proposed as a biomarker of acute coronary syndrome, however during a very narrow time period, indicating that the time of sampling needs attention. [24] So the other soluble RAGE, esRAGE, a product of alternative splicing of the RAGE gene, also should be considered [25].

Obesity is an additional common feature of PCOS women, present in 30–75% of cases, and an aggravating factor in the cluster of clinical entities of the metabolic syndrome [26]. In previous work, it was demonstrated that serum AGEs levels are directly involved in adipogenesis [27] and production of inflammatory mediators in adipocytes, leading to complications related to obesity [28]. All these findings suggested a potential role of AGEs in obesity-related comorbid conditions. In present study, we found that serum levels of AGEs were increased, while serum levels of sRAGE were decreased along with increased BMI. Correlation analysis showed that, in line with other data [14, 29, 30], serum sRAGE levels were inversely correlated with BMI, while AGEs had a positive correlation with BMI. Multiple regression analysis found that BMI was the independent predictor factor of sRAGE, which further confirmed their findings.

Insulin resistance/hyperinsulinemia was identified as a significant contributor and possible mediator of the underlying pathophysiology in PCOS, proximately 50–70% of women with PCOS have some degree of insulin resistance [20]. In vitro studies had shown that AGEs are implicated in the pathogenesis of insulin resistance [31]. In this study, there was no correlation between AGEs and increased HOMA-IR in reproductive women with PCOS, which was consistent with Diamanti-Kandarakis E et al. research. They found that serum AGEs elevated in PCOS independently of the presence of IR, although they use the quantitative insulin sensitivity check index (QUICKI) instead of HOMA-IR represents the IR. On the other hand, earlier studies showed that serum AGEs are positively correlated with insulin and HOMA-IR in women with PCOS, as long as without hyperglycemia [12]. The hyperglycemia might account for their observation results. We found that there was an inverse relationship between sRAGE and HOMA-IR, which was accordance with the findings from Basta G et al. study [13]. The study showed that serum sRAGE concentrations were downregulated in insulin resistance status.

Hyperandrogenism is another significant contributor in the pathogenesis of PCOS. In the present study, there was no difference in AGEs with increased total testosterone levels and FAI in reproductive-aged women with PCOS. In contrast, there was an inverse relationship between sRAGE and FAI, which might be contributed to the high level of insulin inhibited liver SHBG production, and then increased FAI. sRAGE acts as a decoy receptor, which had been proved by numerous studies, and prevents the development of vascular complications [32, 33]. The decreased sRAGE levels in obese PCOS patients with more severe insulin resistance states and higher FAI could be explained by reasons below: 1) sRAGE secretion was decreased; 2) sRAGE was used more by its binding to AGEs, which were higher in these subgroup patients.

AGEs can be produced both endogenously and exogenously. The endogenous AGEs are usually formed slowly under physiological conditions, which are often relate to elevated blood glucose levels. Exogenous sources of AGEs are diet and smoking. Foods rich in protein and fat with extensive cooking, baking, grilling and frying dramatically increased AGE generation [34, 35]. Tantalaki E et al. [36] found that serum levels of AGEs, testosterone, and HOMA-IR index in high AGEs diet group were significantly higher compared to low AGEs diet. These results indicated that overexposure to exogenous AGEs may exacerbate the metabolic and hormonal disorders. Animal experiment also showed that high AGEs feed rats demonstrated higher levels of fasting glucose, insulin, testosterone, and serum AGEs than control and low AGEs-fed rats. Furthermore, the high AGEs diet group showed increased AGEs localization in the theca cells, and increased RAGE expression in granulosa cells compared to control and low AGEs-fed rats. The metabolic and hormonal alterations in conjunction with the increased dietary glycotoxins deposition in ovarian tissues in high AGEs-fed animals compared to the controls suggest an impact of environmental factors on ovarian tissue [37]. Recently, Vazzana N et al. [38] found that sRAGE significantly increases following diet-induced weight loss in obese women, and Brix JM et al. [39] also found a significant increase of sRAGE levels after a dramatic weight loss induced by bariatric surgery in adults with morbid obesity. These finding indicated that low AGEs intake, lifestyle changes along with weight loss to increase the level of sRAGE, which may effectively reduce AGE-RAGE axis’ adverse effect in PCOS patients.

Fujii EY et al. [40] found sRAGE concentrations in blood were positively correlated with the concentrations in follicular fluid (FF). Thus, the concentrations of sRAGE in blood might be able to serve as a useful biologic marker of the follicular environment. In our study, ROC curve analysis determined an optimal sRAGE cut-off point, 704.097 mmol/L in this study, for the presence of insulin resistance, reminding the requirements of intervention AGEs/sRAGE system. Another important question should be considered is whether ovary sRAGE could be biologic markers of the oocyte environment, egg quality, and outcome of fertilization, which may be more crucial in reproductive system.

Taken together, our study suggested that in reproductive-aged PCOS patients, BMI was an independent risk predictive factor of AGEs and sRAGE levels. Compared with AGEs, the variation of serum sRAGE levels was more closely related to insulin resistance and hyperandrogenism (FAI), two major pathological manifestations of PCOS. AGEs/sRAGE system may be a promising intervention target in the women with PCOS, especially in obese patients, although the role of sRAGE needs be further clarified.