1 Introduction

Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders that affects approximately 15 to 20% of women of reproductive age [1], which is accountable for up to 80% of the causes of anovulation and decreased female fertility [2]. PCOS contemplates multiple pathological mechanisms, in which the interaction between genetic and environmental factors appears to be involved, although the exact aetiology remains elusive [3]. Despite the heterogeneity of manifestations, PCOS typical clinical features include hyperandrogenism and ovarian dysfunction. The hallmarks of ovarian dysfunction include follicular development arrest, ovaries with multiple microcysts and anovulation, which can lead to decreased fertility [4].

Due to the disorder's complexity and the uncertain aetiology, PCOS treatment is targeted towards mitigating patients’ symptoms, namely the clinical signs of hyperandrogenism and/or anovulation leading to infertility. According to the international evidence-based guideline for the Assessment and Management of PCOS, anovulatory infertility can be managed with ovulation induction therapies, including first- and second-line drugs (i.e., letrozole, clomiphene citrate, and gonadotropins, used alone or in combination), while ovarian stimulation for assisted reproductive technology (ART) with in vitro fertilization (IVF) can be considered as a third-line intervention in case of failure of the former [5]. However, despite women with PCOS undergoing ovarian stimulation for IVF usually being good responders, with a higher than average number of growing ovarian follicles, the reproductive outcomes of these women tend to be poorer when compared to those with other causes of infertility [6,7,8,9]. Therefore, additional factors related to oocyte quality [10], as well as endometrial competence [11] have been proposed as potentially implicated in impairing the reproductive outcomes of women with PCOS. Hence, PCOS has been hypothesized to endanger the follicular microenvironment compromising oocyte’s physiological functions [12, 13].

The follicular fluid (FF) is a biological fluid present in growing secondary follicles that consists of a plasma exudate enriched in secretory products from granulosa cells (GCs) and thecal cells [14]. There is a tight relationship between the oocyte and the surrounding GCs through the FF microenvironment, which is a via for nutrient exchange and biological signal transmission, hence the importance of the FF microenvironment for follicle development and maturation [15, 16]. Notwithstanding, as consequence of the intimate contact between the FF and the cumulus-oocyte complex, FF’s profiling harbours the potential to unveil molecular signatures and biomarkers that could pinpoint to the mechanisms of disease and predict oocyte quality and reproductive outcomes of women in general, and specifically those with PCOS. Indeed, FF of women with PCOS assessed after oocyte retrieval for IVF treatment, was recently found to depict a considerably different molecular composition when compared to the FF of unaffected women [17, 18]. Nevertheless, how each molecule can potentially contribute for creating an adverse environment to the developing oocyte and in which extent it affects the reproductive outcomes of women with PCOS, has not yet been fully elucidated.

Thus, the aim of this systematic review was to summarize the available evidence on how the FF molecular profile of women with PCOS undergoing IVF treatment differs from the FF of unaffected women. Our goal was also to identify which of the identified biomarkers were demonstrated to correlate with oocyte competence and, subsequently could be useful to predict reproductive outcomes.

2 Methods

2.1 Protocol and registration

The present systematic review was conducted in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 statement [19]. The protocol was registered at inception on February 16th, 2021 in the international database of prospectively registered systematic reviews (PROSPERO) [20] and can be accessed with the registration number CRD42021237734.

2.2 Information sources and search approach

An electronic search in PubMed and Web of Science databases was conducted for reports published up to December 2021, using a search strategy based on the following concepts: polycystic ovarian syndrome, follicular fluid, in vitro fertilization, and reproductive outcomes related to oocyte-embryo development. Medical Subject Headings in PubMed and synonyms were used to outspread the search. No restrictions regarding language and publication year or other filters were used.

2.3 Study design and selection criteria

After removing duplicates, the study selection process consisted in a stepwise approach by eliminating non-eligible studies based on the information provided in the abstract, followed by full text analysis of studies considered potentially eligible in the first phase as described in further detail in Fig. 1. The eligibility inclusion criteria were based on PICOS (Population, Intervention, Control intervention, Outcome, Study design) and included observational studies (cohort and case-control studies); studies conducted in humans, irrespective of race and with no restrictions on population size; studies with at least one group including women with normal ovarian function and another group comprising women with PCOS diagnosis. The exclusion criteria included papers written in languages other than English or Portuguese; studies whose full text could not be accessed; grey literature; abstracts; reviews or position statement papers; studies in which PCOS diagnosis was not established based on Rotterdam criteria; studies that did not provide information on participants’ diagnosis and inclusion and exclusion criteria; studies with pooled data pertaining to women with PCOS and other conditions associated with impaired ovarian function (e.g., endometriosis); studies that included women with conditions that could potentially impair ovarian function as control group; studies that analysed serum and/or plasma or embryo culture medium or follicular cells but not FF and; studies that did not assess the relationship between FF composition and IVF outcomes.

Fig. 1
figure 1

Flowchart of search and selection process for studies included in the review, adapted from PRISMA 2020 flow diagram. FF = Follicular Fluid, IVF-ET = In Vitro Fertilization – Embryo Transfer, PCOS = Polycystic Ovary Syndrome

Full size image

2.4 Quality assessment and data retrieval

The Newcastle-Ottawa Scale was used for assessing the quality and risk of bias of the selected studies based on inclusion and exclusion criteria. Only papers that passed the quality and risk of bias assessment were included. Data from each original study included was independently extracted and checked in a cross-over manner by two independent researchers. For each study the information retrieved included: study design, population characteristics (sample size of study groups, age, and body mass index), intervention - ovarian stimulation protocol used, and IVF outcomes. The information was then summarized in tables and complemented by a narrative description whenever required due to data heterogeneity.

3 Results

3.1 Study selection

The initial search retrieved 664 papers: 402 at Web of Science and 262 at PubMed. After eliminating duplicates, 463 papers were identified for analysis. Among those, 231 papers were excluded based on the information provided in the abstract and 189 additional papers were excluded when full text analysis was considered. Then the remaining 43 papers that met the eligibility criteria underwent risk of bias assessment, which resulted in the exclusion of an additional paper. This yielded a final selection of 42 papers to be included in the systematic review. The flowchart of the search and selection process for studies included in the review is outlined in Fig. 1.

Data retrieved from papers included in this systematic review, revealed that the FF composition of women with PCOS compared to those without PCOS predominantly diverged at the following molecular classes: oxidative stress, inflammatory biomarkers, growth factors and hormones.

3.2 Markers of oxidative stress

Oxidative stress markers assessment in the FF of women with PCOS have been addressed by several studies (Table 1). A lower total antioxidant capacity (TAC) [21,22,23], along with higher levels of reactive oxygen species (ROS) [22] and 8-Isoprostane (8-IP) [24] has been described in the FF pertaining to women with PCOS comparing to ovulatory non-PCOS women. ROS present chemical properties that often confer reactivity to different biological targets, such as the lipids in membranes, amino acids in proteins, and nucleic acid bases in DNA. Levels of malondialdehyde (MDA), an end-product of lipid peroxidation chain reaction, were also found to be higher in the FF of women with PCOS [21,22,23].

The activities of several antioxidant enzymes and other molecules involved in redox reaction were also reported to be different in the FF of women with PCOS as compared to controls. Namely, a lower superoxide dismutase (SOD) activity has been described in the FF of PCOS women [23, 25]. In addition, Naigaonkar et al. found that glutathione status was compromised in those patients, as evidenced by decreased levels of reduced glutathione and increased levels of oxidized glutathione [23]. Furthermore, in the same study, the expression of genes involved in antioxidant-related machinery in GCs revealed lower transcript mRNA levels of SOD, glutathione peroxidase (GPx) and glutathione reductase in women with PCOS, in agreement with the activities of the respective enzymes found in the FF [23].

During oxidative stress conditions, oxidizing molecules tend to react with thiol groups (-SH) to form reversible disulphide bonds. Therefore, thiol/disulphide homeostasis can be used as a surrogate marker of oxidative stress. Several authors demonstrated the presence of decreased thiol groups [21, 23, 26], and increased disulphide as well as disulphide/total thiol groups ratio in the FF of women with PCOS [26]. Interestingly, Tola and colleagues also found a positive correlation between FF thiol levels and fertilization rate among women with PCOS [26].

3.3 Inflammation markers

Numerous studies have shown the presence of increased levels of inflammatory markers in the FF of women with PCOS (Table 2). Higher levels of C-reactive protein (CRP) [27, 28] and pro-inflammatory cytokines, including interleukin 6 (IL-6) [21, 27], interleukin 8 (IL-8) [21], interleukin 18 (IL-18) [29], and tumour necrosis factor (TNF-α) [21, 27, 30, 31] were found in the FF of women with PCOS when compared to ovulatory non-PCOS.

Leukaemia inhibitory factor (LIF) is a pleiotropic cytokine and acts as a local regulatory factor in ovary [32]. LIF concentrations in serum and FF of PCOS women were revealed to be lower when compared to controls without PCOS, and negatively correlated with oestradiol concentrations [33]. Progranulin (PGRN) levels, an adipokine recently shown to be associated with low-grade chronic inflammatory state and with a functional role in obesity-related insulin resistance, were reported to be higher in the FF of women with PCOS. PGRN mRNA expression in GCs was found to be higher in PCOS women than in ovulatory non-PCOS, in addition to being higher in overweight when compared to normal-weight women in both PCOS and control patients. Furthermore, PGRN expression levels in the overweight PCOS group were significantly higher than in the overweight control groups [31].

The soluble receptor for advanced glycation endproducts (sRAGE) has been proposed as biomarker of severity in inflammatory and metabolic conditions [34]. Several studies reported the presence of lower sRAGE levels in the FF of women with PCOS when compared to non-PCOS [27, 35, 36]. Furthermore, a multi-adjusted regression analysis showed that higher sRAGE levels in FF predicted the need for lower gonadotrophin doses, higher number of oocytes retrieved and better IVF outcomes in the non-PCOS group [36]. Granulocyte colony-stimulating factor (G-CSF) is a cytokine that stimulates granulocyte cells proliferation and differentiation. Kahyaoglu et al. found that women with PCOS had increased G-CSF levels in both serum and FF compared with ovularory non-PCOS women. The authors hypothesized that as women with PCOS presented more follicles, G-CSF levels could be derived from granulosa cells [37].

Oxidative stress is considered to play an important role in the inflammatory response. Indeed, Artimani and colleagues found a positive correlation between pro-inflammatory cytokines and MDA as well as total oxidant status levels, further corroborating the association between ROS and inflammatory response [21].

3.4 Growth factors

The FF of women with PCOS was found to depict a distinctive growth factor profile, which could potentially contribute to some of the clinical manifestations observed in this condition (Table 3). Namely, elevated levels of vascular endothelial growth factor (VEGF) have been described in the FF of women with PCOS [27, 36, 38]. Further, Savchev and colleagues investigated two isoforms of VEGF present in FF, more specifically VEGF121 and VEGF165, and found an association between higher VEGF165 levels and the diagnosis of PCOS in women with BMI≥ 30 kg/m2, and age ≥40 years [39]. The levels of placental growth factor (PlGF), a VEGF family member, were also found to be increased in the FF of women with PCOS when compared to controls. The last study also found that FF of women with PCOS presented lower levels of the soluble form of VEGF receptor 1 (sFlt-1), which is known to bind VEGF and PlGF and block its angiogenic effects on VEGFR [40]. Fang et al. disclosed that the levels of growth differentiation factor-8 (GDF-8) in FF at oocyte retrieval were higher in women with PCOS as compared to the FF of women without PCOS, this study also provided evidence that aberrant expression of GDF-8 in the FF of women with PCOS was associated with subnormal progesterone secretion and poor pregnancy outcomes [41]. Recent animal studies suggested that hepatocyte growth factor (HGF)- and c-Met-mediated epithelial mesenchymal mechanisms are crucial for follicle development [42, 43]. A study conducted by Sahin and co-workers found that the levels of HGF in serum and FF and the mRNA expression of c-Met in GCs were slightly higher in patients with PCOS than in control patients, however these differences were not statistically significant [44]. The fibroblast growth factor (FGF) family is known for being involved in the regulation of ovarian function and follicular development [45]. Liu et al. found that the prevalence of elevated FF testosterone levels was significantly higher in the PCOS patients with elevated FF-FGF13 levels than in those without. Furthermore FF testosterone and increased ovarian volume (> 10 mL for one or both ovaries) were positively correlated with FF-FGF13 in PCOS patients [46].

3.5 Hormones

PCOS is predominantly associated with gonadotrophic axis dysfunction, besides several other hormone imbalances. Elevated levels of LH [49,50,51,52], testosterone [53] and Anti-Müllerian hormone (AMH), are the hallmark of gonadotrophic axis dysregulation associated with PCOS [54, 55] (Table 4). AMH is expressed by GCs of pre-antral and small antral follicles. Women with PCOS tend to depict raised AMH circulating levels in parallel to the number of antral follicles [56]. Indeed, several independent studies documented the higher AMH levels in the FF of women with PCOS when compared with controls [50, 52, 57, 58]. Although, the same finding was not observed by Yilmaz et al. in women with PCOS but without obesity or hyperandrogenism [59].

Gonadotrophin and ovulation are regulated by the KISS1/KISS1R system that thus, is essential for female reproductive system [60]. Interestingly, Hu and colleagues found that KISS1 expression levels were significantly upregulated in human granulosa lutein cells obtained from women with PCOS and were highly correlated with AMH serum levels [50].

Women with PCOS often present glucose and energy homeostasis disruptions, along with obesity and metabolic syndrome, which arise associated with raised circulating levels of leptin and insulin [61]. Since leptin is an adipocyte derived hormone, leptin levels are usually proportional to body fat mass [62]. Li and colleagues reported that leptin levels in the FF of infertile women with PCOS was higher when compared to those of women without PCOS. However, this study included normal weight and overweight women (range 19.81–30.62 Kg/m2) in both experimental groups [51]. Contrarily, Garruti et al. found that serum and FF leptin levels of non-overweight women with PCOS were lower when compared to those of non-overweight control women [49]. Regarding other adipokines, adiponectin levels were reported to be lower in the FF of lean women with PCOS in comparison with lean non-PCOS women [63]. Furthermore, another study demonstrated that overweight and PCOS women, exhibit increased serum and FF irisin levels compared to normal weight women. Irisin levels were positively correlated with BMI, dyslipidemia and the number of oocytes retrieved and fertilized [64].

PCOS has been shown to be correlated with the endogenous opioid system activity, which could contribute to the pathophysiology of the disorder. Particularly, β-endorphins are opioids involved in female reproductive functions, playing a role in regulating the normal menstrual cycle and possibly in the beginning of puberty [65, 66]. A study conducted by Zhang et al. found higher levels of β-endorphins in the FF of women with PCOS [52], while other authors reported no differences in β-endorphin levels between in FF of women with or without PCOS [67].

Previous studies have reported the association between PCOS and specific thyroid diseases [68]. Gao et al. found higher levels of thyroid-stimulating hormone (TSH) in FF of PCOS women than those in the non-PCOS patients. Furthermore, the expression of TSH receptor in ovarian GCs was significantly upregulated at both mRNA and protein levels in the PCOS group compared with non-PCOS patients [69]. In PCOS patients, TSH levels, both in serum and FF, were negatively correlated with IVF oocyte maturation rate and fertilization rate [69].

3.6 Other molecules

A large amount of data reveals that women with PCOS have an altered FF profile with variations in several biomolecules (Table 5). MicroRNA are small noncoding single-stranded RNA molecules that play important roles in regulating gene expression at the post-transcriptional level. The study of the expression of small non-coding microRNA in FF of women with and without PCOS who underwent the same IVF protocol for subfertility, found 29 miRNAs that differed significantly between the two groups, being the top 7 of these correlated with age, FAI, inflammation and AMH in women with PCOS [73].

Circulating nutrients and exogenous substances can flow through the basal lamina and enter the follicular antrum. A study conducted by Sun and colleagues showed that FF trace elements profile was different in women with PCOS (namely, Cu, Mg, Ca, Ti, As), being the Copper (Cu) concentrations significantly higher. Also, FF Cu levels were positively correlated with FF progesterone and testosterone levels in women with PCOS [74].

Fatty acids can be significant biological markers of aberrant lipid metabolism and have a strong impact on gene expression, which in turn can lead to an altered metabolism, cell growth and differentiation [75]. Niu et al. demonstrated that the metabolic profiles of both plasma and FF indicated higher levels of two long-chain fatty acids, palmitic acid and oleic acid, in women with PCOS compared to controls [76]. Interestingly, the embryo fragmentation score was significantly positively correlated with the oleic acid concentration in all women with PCOS, which may contribute to the mechanisms driving to poor pregnancy outcomes in those patients [76].

FF is known for providing a unique environment for oocyte development and this biological fluid is full of metabolites built up during this process. Therefore, FF metabolites can potentially reflect the oocyte maturation process as well as affect the quality of the mature oocyte. Zhang et al. investigated the FF metabolomics profile from women with PCOS and healthy controls using proton nuclear magnetic resonance (1H-NMR). This study revealed that the FF from women with PCOS presented higher levels of glycoproteins, acetate, and cholesterol and lower levels of lactic acid, glutamine, pyruvate, and alanine. The authors hypothesised that differential metabolite profile suggested the presence of altered pyruvate, amino acid and lipid metabolism [77].

Secreted frizzle-related protein-5 (Sfrp-5), a member of the SFRP family, is a novel antagonistic of the Wnt signaling pathway. Recent studies found that Sfrp-5 regulated lipid metabolism, negatively regulated adipogenesis and reduced metabolic dysfunction. Thus, this protein seems to be involved in the pathogenesis of a variety of metabolic diseases [78,79,80]. Inal and colleagues investigate the effect of serum and FF Sfrp-5 levels on IVF–ICSI outcomes in patients with PCOS and revealed higher FF Sfrp-5 levels in a selected population of nonobese, nonhyperandrogenic PCOS patients. The authors suggest that the Wnt signaling pathway and Sfrp-5 may play an important role in the etiopathogenesis of PCOS [81].

It has been reported that proteoglycans degradation by a disintegrin-like and metalloproteinase with thrombospondin type motifs-1 (ADAMTS-1) is important during folliculogenesis, ovulation, and fertilization via extracellular matrix remodelling [82]. A study conducted by Tola et al. investigated FF aggrecan (proteoglycan) and ADAMTS-1 levels in developing and preovulatory follicles obtained from PCOS and normal ovulatory infertile patients undergoing IVF procedures. The authors found elevated aggrecan and ADAMTS-1 levels in the FF of PCOS women compared with control groups. Also, it was found a positive predictor effect of ADAMTS-1 on implantation, which may indicate that follicular ADAMTS-1 levels could be a potential marker of high-quality embryos for transfer in PCOS patients [83].

Serum amyloid-associated protein has recently been proposed as an inflammatory marker that may be more specific to inflammation related to PCOS. Timur et al. investigated serum and FF amyloid A protein levels in non-obese non-hyperandrogenic patients with PCOS undergoing in IVF [84]. However, no significant difference was found between two groups regarding the serum and FF amyloid A protein levels on the day of oocyte retrieval.

4 Discussion

PCOS is a multifactorial disorder characterized by hyperandrogenemia, anovulation, and metabolic dysfunction. Several neural, immune and endocrine axis components seem to be involved in the mechanisms leading to PCOS. The increased pulse amplitude of gonadotropin-releasing hormone release in the hypothalamus promotes LH secretion over FSH, which consequently leads to abnormal sex steroids production and ovarian dysfunction [89]. In the ovary, the hyperandrogenic environment thwarts normal follicular growth, maturation, and ovulation. Besides androgens, circulating AMH levels also tend to be higher in women with PCOS [57, 58]. AMH is a glycoprotein hormone and a member of the transforming growth factor β family of growth and differentiation factors [90]. AMH play a crucial role in male sexual differentiation since is responsible for regression of the female Müllerian ducts. However, AMH has been shown to be involved in postnatal ovarian function whereas this hormone is expressed by GCs of growing follicles from the primary up to the small antral stage [91]. The AMH levels reflect the number of growing follicles recruited from the primordial follicle pool and therefore has been often suggested as a marker for ovarian reserve. Accordingly, Chen and colleagues found a positive correlation between FF AMH levels and antral follicle count in women with PCOS, suggesting that AMH could also be used as a reliable biomarker of ovarian reserve in these patients [58]. Yilmaz and colleagues investigate the FF AMH levels of non-obese non-hyperandrogenemic PCOS, excluding the possible effect of hyperandrogenemia on AMH levels. They observed that FF AMH levels of PCOS patients were higher than the control group although the difference was not significant. Those findings suggest a possible association between FF AMH and androgens [59].

Some of the typical clinical features associated with PCOS include insulin resistance, dyslipidaemia, and abnormal glucose metabolism. Furthermore, the prevalence of overweight and obesity is also higher in women with PCOS, as compared to unaffected women, which tends to exacerbate the symptoms, namely hyperandrogenism and anovulation [92]. Adipokines are adipocyte secreted molecules that have been hypothesized to play a role in PCOS. Adipokine levels including leptin, adiponectin and chemerin were identified to be altered in women with PCOS [28, 49, 51, 63]. Leptin is a peptide hormone, usually released in proportion to adipose tissue mass, that is involved in the control of body weight and energy homeostasis. High leptin levels are frequently found in women with PCOS and suggested to be involved in the pathogenesis of PCOS [51], although are still controversial since other studies present conflicting results [49]. Li and colleagues aimed to investigate whether serum and FF leptin levels, as well as the expression of leptin receptors (Ob-R) and b-R signaling in GCs were altered in women with PCOS. The authors reported higher leptin levels in serum and FF and decreased expression of p-STAT3 in GCs of women with PCOS compared to non-PCOS women. Further analysis showed that those findings were related to poor IVF outcome, suggesting that abnormalities of leptin secretion and signaling pathway could be involved in PCOS-associated infertility [51]. On the other hand, Garruti and colleagues reported lower serum and FF leptin levels in non-overweight women with PCOS (22.55±2.6 Kg/m2) when compared to BMI-matched (22.55±2.6 Kg/m2) non-overweight non-PCOS women. The disparities between studies could be attributed to demographic differences of study cohorts (ethnicity, age, BMI) and/or heterogeneity of the syndrome itself.

Oxidative stress and chronic low-grade inflammatory status have been anticipated as key contributors to PCOS pathogenesis since several reports have disclosed higher oxidative stress and inflammatory mediators in those women. ROS can induce inflammatory response prompting the release of cytokines, and both phenomena are directly associated with PCOS comorbidities including obesity, insulin resistance, dyslipidemia and hyperandrogenism. As a matter of fact, a study conducted by Artimani and colleagues demonstrated that in women with PCOS there was a positive correlation between inflammatory cytokines and pro-oxidant agents and a negative correlation with antioxidant levels [6]. This is very relevant since the physiological balance between pro-oxidants and antioxidants is deemed to ensure several reproductive processes including ovarian steroid genesis, oocyte maturation and ovulation [93]. However, an altered redox homeostasis is often observed in PCOS. Indeed, several studies included in this review demonstrated a tendency for women with PCOS to have higher levels of pro-oxidant markers (e.g. ROS/TOS and MDA) [6, 22, 23] and lower antioxidant capacity (e.g. TAC, free thiols groups and SOD activity)[6, 22, 23, 25]. Lipid peroxidation (LPO) is a well characterized consequence of oxidative stress on cell membrane structure and function. LPO is a process characterized by the molecular modification of lipids containing carbon-carbon double bond(s), especially polyunsaturated fatty acids, when exposed to free radicals [94]. During LPO different aldehydes can be formed as secondary products, among which malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) have been widely used as OS surrogate markers. Women with PCOS have been reported to present higher MDA levels than non-PCOS patients [6, 22, 23]. Naigaonkar et al. also found that women with PCOS and insulin resistance had higher levels of LPO and hypothesized that those levels could be a consequence of free fatty acid levels abundancy and accumulation in the mitochondria, which is a prominent source of ROS [23]. The human body holds a complex antioxidant defense system that relies on endogenous enzymatic and non-enzymatic antioxidants, which are responsible to scavenge oxidants and prevent oxidative stress-related damages. The three major antioxidant enzymes include SOD, catalase (CAT) and glutathione peroxidase (GPx) [95]. The metalloenzyme SOD promotes the dismutation of superoxide (O2•−) into molecular oxygen (O2) and hydrogen peroxide (H2O2) and is considered the first line of defense against ROS. There is also evidence that the FF of women with PCOS has lower SOD activity and GCs have lower SOD transcript expression [23, 25]. These findings demonstrated that redox homeostasis in follicular microenvironment of women with PCOS is altered towards a pro-oxidant environment that can compromise oocyte quality. Furthermore, GCs seem to play a critical role in maintaining redox balance within the follicle, as transcript expression of antioxidant enzymes in GCs was in line with its activity at the FF [23]. Thiol redox reaction is another mechanism of protection against OS. Thiol is a compound that contains a sulfhydryl group (-SH) functional group. The –SH group in thiols reacts with oxidants to form disulphide bonds, exerting a protective effect against OS. Consistent with the aforementioned, disulphide levels were found to be higher [26] while thiol levels were reported as decreased [23, 26] in the FF of women with PCOS than without PCOS. Further, a positive correlation between fertilization rate and thiol levels among women with PCOS was also reported [26].

Oxidative stress and inflammation are tightly linked pathophysiological processes. The role of inflammation in oocyte quality and reproductive function has been broadly investigated. TNF-α and IL-6 are proinflammatory cytokines involved in modulation of inflammatory response. These cytokines have been associated with obesity, insulin resistance and attendant comorbidities, such as type 2 diabetes [96]. Indeed, FF levels of TNF-α and IL-6 were found to be higher in infertile women with PCOS compared to control women [6, 27, 31]. Some authors even postulated that it is the imbalance between pro- and anti-inflammatory cytokines that leads to altered steroidogenesis, delayed follicular maturation and ovarian dysfunction, which are some of PCOS key clinical features [97]. RAGE is a transmembrane protein that belongs to the immunoglobulin superfamily. This receptor binds a variety of stress associated molecules, including advanced glycation end products (AGEs), a family of compounds formed in the presence of glucose by non-enzymatic alteration of proteins, lipids, and nucleic acids [98]. The sRAGE is an extracellular form of RAGE that bind its ligands (AGEs), thus interrupting the adverse intracellular signaling triggered by the AGE-RAGE axis [99]. There is growing evidence supporting the AGE-RAGE system contribution to increased oxidative stress and inflammation [100]. This system has been implicated in the pathogenesis of multiple metabolic diseases and more recently in PCOS and infertility [99]. It has been demonstrated that women with PCOS have lower FF sRAGE levels when compared to non-PCOS, however the effects of reduced sRAGE levels in PCOS remains unclear [27, 35, 36]. AGE–RAGE interactions can also stimulate VEGF production and consequently modify the follicular environment [99]. A study conducted by Wang and colleagues demonstrated that treatment of GCs, isolated from women with PCOS, with recombinant sRAGE decreased VEGF mRNA expression [36]. VEGF is a potent angiogenic factor that was first described as an essential growth factor for vascular endothelial cells. Macrophages and GCs are considered the most significant sources of VEGF in female reproductive tract. VEGF expression was found to be associated to important events in the course of the ovarian cycle, including follicular growth, ovulation, corpus luteum development, and ovarian steroidogenesis [101]. VEGF has been suggested to play a role in the selection of the dominant follicle by increasing vascularization and, VEGF seems to be involved in the ovulation process through the activation of plasminogen activator system and plasminogen activator inhibitor 1 [102]. Numerous studies found higher VEGF levels in the FF of women with PCOS compared to control women [27, 36, 38, 39]. It was proposed that the elevated circulating VEGF levels in women with PCOS may partially explain the ovarian stroma hypervascularization, which characterizes PCOS [103]. In fact, the profuse vascularization has been hypothesized to facilitate abnormal theca interna cell growth, responsible for androgen steroidogenesis. Kudsy et al. also hypothesized that granulosa lutein cells VEGF secretion could be stimulated by insulin, since GCs of women with PCOS are more sensitive to insulin [38]. Since insulin resistance tends to be more frequent in women with PCOS than in the general population, it is reasonable to hypothesize that the consequent hyperinsulinism could contribute for VEGF hypersecretion [104]. Given the aforementioned, VEGF appears to play a critical role in the pathophysiology of PCOS.

Recently, approaches in “omics” technologies have been explored in an attempt to further understand this heterogeneous and complex disorder. Amongst these, metabolomics, which examines the fingerprints of all metabolites in a biological system, allows the investigation of metabolite pathways that may provide new insights into the underlying biology of PCOS. Zhang and colleagues using 1H NMR metabolomics approach explored the metabolic variance at the FF from women with PCOS and healthy controls [77]. A total of 9 metabolite regions were shown to differentiate the FF from women with PCOS and healthy women. The decreased levels of pyruvate and lactate and increased levels of acetate suggested the occurrence of shifts in pyruvate metabolism and glycolysis. Aberrant insulin level may underlie these changes, as it contributes to decreased glucose uptake and thus glycolytic flux [77]. Also, it is possible that the decreased levels of some amino acids, namely alanine and glutamine, could be attributed to a compensatory increase in amino acids consumption as cellular substrate to produce energy. The increased glycerol and lipid levels, namely cholesterol is not unexpected since these are typical signs of dyslipidemia, which is a common comorbidity associated with PCOS. It is suggested that impaired lipase expression and altered lipolysis caused by insulin resistance, may be responsible for those events [105]. Glycoproteins are closely related to several inflammatory disorders and the high glycoprotein level found in the FF of women with PCOS suggest that the oocytes’ environment might also be characterized by low-grade inflammation [106]. Interestingly, the authors reported negative correlations between embryos quality and several amino acids, including creatine, leucine, isoleucine, as well as tyrosine. Most particularly, a negative correlation between glucose and both 2PN fertilization rate and cleavage rate and positive correlations between acetoacetate and 3-hydroxy-butyrate and 2PN fertilization rate have been reported in women with PCOS. The latest observations suggest that the metabolic shifts occurring in GCs and detected in the FF of women with PCOS might affect the ART process.

Taken all together, the data retrieved from papers included in this review revealed that the FF composition of women with PCOS compared to those without PCOS predominantly differ on the levels of oxidative stress and inflammatory biomarkers, growth factors, hormones and metabolites (Fig. 2). The identification of the metabolic pathways and molecular mechanisms involved in PCOS pathophysiology might allow the identification of specific biomarkers, which in turn could potentially be useful for early diagnosis of this endocrinopathy and be the outset for future tailored pharmacological therapies.

Fig. 2
figure 2

Main changes on the composition of follicular fluid in women with polycystic ovary syndrome. Abbreviations: AMH - anti-mullerian hormone; CRP – c-reactive protein; IL-6 – interleukin 6; LPO – lipid peroxidation; ROS – reactive oxygen species; SOD – superoxide dismutase; sRAGE – soluble receptor for advanced glycation end products; TAC – total antioxidant capacity; TNF-α – tumor necrosis factor; VEGF - vascular endothelial growth factor

Full size image

5 Conclusion

The characterization of FF and its molecular profile opens a window of opportunity to further understand PCOS. The FF is a biological matrix that is in intimate contact with the oocyte and contains a variety of bioactive molecules, which contribute for follicle development and maturation. The identification of molecular shifts within the FF harbors the potential of understanding the pathways involved in the ovarian dysfunction observed in women with PCOS. In conclusion FF can be a promising tool to identify novel biomarkers for the PCOS diagnosis as well to the development of targeted therapeutics for the treatment of PCOS.

Table 1 Summary of studies comparing oxidative stress markers in follicular fluid of PCOS and non-PCOS women
Full size table
Table 2 Summary of studies comparing inflammatory markers in follicular fluid of PCOS and non-PCOS women
Full size table
Table 3 Summary of studies comparing growth factors in follicular fluid of PCOS and non-PCOS women
Full size table
Table 4 Summary of studies comparing hormones in follicular fluid of PCOS and non-PCOS women
Full size table
Table 5 Summary of studies comparing other molecules in follicular fluid of PCOS and non-PCOS women
Full size table