Dissect the Common Molecular Origins of Polycystic Ovarian Syndrome and Alzheimer’s Disease: A Review Complemented with Protein-Protein Interaction Network Analysis

1. Introduction

There is a common cause-and-effect relation among modern-era diseases such as diabetes mellitus, obesity, cancer, neurodegenerative diseases, infertility, and polycystic ovarian syndrome (PCOS) [1]. PCOS is the most common cause of subfertility and infertility among women [2]. This modern syndrome was mainly known as a fertility disorder, however it affects the endocrine, metabolic, immune, and even nervous systems [3,4]. Therefore, PCOS is a complex disorder with multifactorial etiology and chronic consequences [5]. This syndrome does not involve only ovaries, and has various clinical manifestations [6,7,8,9,10]. Therefore, its initial name, Stein-Leventhal syndrome, may be a better option for PCOS as a disorder of ovary-organ axes. Accordingly, the treatment of PCOS are based on its clinical manifestations and must be investigated as a multi-organ disorder that leads to many neurological diseases.

In this continuum, Alzheimer’s disease (AD) is known as a complex, multifactorial, currently incurable, and eventually fatal disease [11,12,13,14,15,16]. AD is a neurodegenerative disease that includes multistage from preclinical, behavioral, mild cognitive impairment, and finally dementia [17,18,19]. There are 55 million individuals suffering from AD worldwide, and its prevalence is estimated to be doubled every five years [14,20]. It is the main cause of dementia, and 60-80% of dementia cases have AD [18,21,22,23,24], and its incidence is increased throughout aging [25]. It is reported that in 85 years old and over-65-year-old cases, there are 50% and 10% AD possibility, respectively [22]. The therapeutic, care, and social costs of AD will be high and noticeable regards to an estimated 152 million AD individuals and involving the patient’s family and health care systems; it is estimated to be about over $1 trillion by 2050 [14]. Therefore, AD is a major global health concern with a rising prevalence and substantial future burden.

Different hypotheses have been put forth about less-known pathogenesis of AD include tauopathy, amyloidogenic cascade, mitochondrial cascade, OS, neuroinflammation, cholinergic theory, glutamate dysfunction, germ theory, and vascular theory [12,13]. The whole gene set involved in human AD is exceled in the Supplementary file S1. In summary, neuronal cell bodies, axonal structure death and microglia-astrocyte activation (microgliosis and astrocytosis), altered brain neuronal apoptosis and autophagy, neuroinflammation, cerebrovascular changes, altered expression, transmission and function of acetylcholine, brain-derived neurotrophic factor (BDFN), and glutamate, as well as metabolic disorders, are among factors governing the pathogenesis of AD [18,21,25,26,27,28,29,30,31,32,33,34,35]. In this context, pathological aging causes neuroinflammation, synaptic damage, and eventually cognitive impairment; however, it should be kept in mind that AD is not a normal aging process [24,32].

The modulating role of metabolic disorders on neurons and organs was accepted, and the brain works with high metabolic rate, glucose and oxygen consumption, and lipid content [14,20,32]. Altered insulin signaling has been reported in AD, and its receptors are widely distributed in the hippocampus, the memory center [20]. The altered signaling in IR causes metabolic disorders of glucose, OS, a decline in neuronal energy production, amyloid-β production, tau protein phosphorylation, and leads to AD by impairing in mitochondrial function, cognition, and memory [19,36]. Therefore, managing metabolic status has a pivotal role in the AD monitoring strategy.

There are no effective treatments for AD and its therapy may be more effective through improving temporary memory than managing AD progression [15,24]. Nevertheless, there are several common and emerging novel therapies for AD. In general, higher education level, rich social network, regular physical activities, and more intellectually stimulating work have improving effects in AD treatment, as well [16]. The effectiveness of AD therapy relies upon the disease stage, genetic factors, general health, and age [13]. However, the novel therapeutic approaches are focused on delaying brain aging, CNS modulation, and slowing AD progression via neurotrophins [15,37]. An array of drugs that were prescribed for AD is found in the market (https://go.drugbank.com/categories/DBCAT001745; https://go.drugbank.com/indications/DBCOND0049114). Therefore, AD therapies currently aim to slow progression and support cognitive function rather than provide a strict cure.

Despite an extensive history of AD detection, many facets of its etiology remain unclear; as with PCOS, its multifactorial nature involves interactions of genes, environment, and epigenetics [26]. This synaptic disorder occurs along with insulin resistance (IR), type-2 diabetes mellitus (T2DM), obesity, hypercholesterolemia, cardiovascular disease, thrombosis, cortisol and hormonal disturbances, dysphagia, malnutrition, pneumonia, Coronavirus disease 2019 (COVID-19), depression, anxiety, and higher suicide risk which increase the fatality in AD patients [19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49]. The early and accurate diagnosis can be a decisive step in its management. Its current diagnostic methods include positron emission tomography scan, cerebrospinal fluid analysis, AD confirmatory-prognostic biomarkers (p-tau-181), and blood neurofilament light chain. Clinically, the amyloid and tau protein deposition and neurodegeneration were proposed for AD detection [18], however, there is an urgent need for more accurate and novel diagnostic approaches [22,25].

Regarding different involvement of biological disorders, specifically metabolic and inflammatory ones, and a preview of proteins involved in the pathogenesis of PCOS and AD, it seems that common proteins expressed in both diseases. In this review, among 83 proteins that are involved in the pathogenesis of both PCOS and AD, we sought to highlight the shared biomolecular landscape in these two disorders to define a novel strategic approach for the simultaneous prevention, management, and treatment of AD and PCOS.

2. Results

2.1. Text-Mining of Polycystic Ovarian Syndrome

2.1.1. Current Status of Polycystic Ovarian Syndrome in Clinics

PCOS was reported with bilateral polycystic ovaries, amenorrhea or oligomenorrhea, obesity, acne, and hirsutism [4,43,44,45,46,47]. There are four major PCOS phenotypes, which include classic phenotypes [10,48,49,50]. The three criteria of phenotype A are hyperandrogenism, oligoanovulation, and polycystic ovarian morphology (PCOM), while in phenotype B, ovarian morphology is normal and the other two criteria are present. Phenotype C is characterized by hyperandrogenism, normal menstruation, and PCOM, and eventually phenotype D is manifested by normal androgen level besides the presence of PCOM and oligoanovulation [36,51,52]. The phenotype A and D are the most and least severe forms, respectively [50]. The most frequent PCOS phenotype is a classic one presented with hyperinsulinemia, menstrual dysfunction, IR, atherogenic dyslipidemia, increased anti-Müllerian hormone (AMH) level, risk of obesity, hepatosteatosis, and metabolic syndrome (MetS) [49,51]. The mild elevation of insulin, androgen, and lipid profile levels in phenotype C brings more metabolic consequences (e.g., MetS) among high socioeconomic individuals [49,51]. The minimum metabolic dysfunction, lower levels of thyroid hormones, and lower luteinizing hormone (LH) to follicular stimulating hormone (FSH) (${\displaystyle \frac{\mathrm{L}\mathrm{H}}{\mathrm{F}\mathrm{S}\mathrm{H}}})$ ratio, and higher sex hormone binding globulin (SHBG) was reported for phenotype D [49,51].

The definitive diagnosis of PCOS in the contemporary era is usually based on Rotterdam criteria, which were discussed (vide supra), and its diagnosis is confirmed if there are two out of the three mentioned criteria [2,46,47,49,53,54,55,56,57,58,59,60]. The volume of more than 10 ml and the presence of 12 or more follicles with 2-9 mm diameter can be mentioned as the pelvic ultrasonic diagnostic criteria of PCOS [61]. The differential diagnosis to other diseases with similar symptoms, such as adrenal congenital hyperplasia, Cushing’s syndrome, androgen-secreting tumors, raised prolactin, insufficient LH, thyroid gland dysfunction, and 21-hydroxylase deficiency is to be considered [62,63]. It should be noted that the early PCOS diagnosis is a tough task due to overlapping symptoms [49]. The diagnostic value is depending on the individual’s age. It can encompass various aspects of psycho-metabolo-gynecological health and social status [49]. PCOS can impose a heavy economic burden on human society. For instance, an estimation in the United States of America has revealed an 8-billion-dollar economic burden of PCOS in 2020 [9,64]. Therefore, early PCOS detection can be effective in estimating therapeutic costs and even their decrement.

Changes in gene expression and protein, metabolic pathways, hormonal imbalance, gut dysbiosis, and dysfunctions in ovarian granulosa cells are involved in PCOS etiology [56,65,66,67,68,69]. In essence, not only the ovaries, but all other organs related to sex hormones will be affected by PCOS.

2.1.2. Epidemiology of Polycystic Ovarian Syndrome

PCOS has accrual incidence and prevalence with serious psychophysiological consequences, fertility issues, socioeconomic costs, and accompanying morbidity and even mortality [70,71]. The incidence of PCOS showed a 30.4% increase of incidence between 1990 and 2019 [9,10,64,72,73,74]. A 37.9% increase in PCOS incidence in North Africa and the Middle East was reported [75]. Furthermore, the number of PCOS cases in the Asian continent has risen during the last 30 years [10]. The prevalence of PCOS is dissimilar because of the demographic differences between population, study methodology, and employed diagnostic criteria [60,71,76,77,78]. This geographical variation are results of hormonal, metabolic, and neural makeup and environmental conditions, lifestyles, and individuals' genetics [79]. Therefore, this difference can dictate the existence of various PCOS phenotypes. In this context, the PCOS prevalence in Black, Caucasian, Spanish Caucasian, Chinese, and Indian was 4.0, 6.5, 5.6, and 9.13, respectively [49,51]. The prevalence of PCOS was higher in South Asia in comparison to the Caucasus (e.g., 40-50% for Pakistan) [49,51]. In contrast, another study indicated a higher PCOS prevalence in Caucasian and mixed ethnicities in comparison to Asians [80]. The higher hirsutism rate was reported in Middle Eastern, Mediterranean, Indian, and South Asian women in comparison to Eastern Asian and Caucasian women, while higher prevalence of hirsutism, acne, IR, T2DM, metabolic disorders, and elevated fasting insulin levels was reported in Hispanic women than in non-Hispanic white ones [61]. Another study indicated a higher PCOS prevalence in Black and Middle Eastern populations in comparison to the Chinese and White populations [70]. In an ethnical survey, Middle Eastern and Chinese women have the highest PCOS prevalence, and among Middle Eastern women, the PCOS prevalence is high in the Persian Gulf region [81]. Another interesting point is that the PCOS incidence in female Caucasians who live in America or Europe is lower than that in women resident in the Middle East; of course, the maximum PCOS incidence in these regions belongs to African American black or Afro-Brazilian [79]. There are different reports on PCOS symptoms and phenotypes between geographical regions, human races, and diagnostic methods, which one can mention to higher body mass index and waist circumference in African American and Hispanic women than in Northern European and Counterpart Asian [72]. The global incidence and prevalence of PCOS in different studies regarding diagnosis type have been reported as 6-20% and 2.2-26.7%, respectively [7,58]. Therefore, geography will be a determinant for the prevalence of PCOS.

PCOS is considered the main global cause of female infertility (30-60%), especially the anovulatory type (70-80%), while 90-95% of anovulatory infertile clients referred to infertility clinics have PCOS [49,51,69,82,83,84]. The high PCOS prevalence (55-60%) in first-degree relatives indicates its possibility of genetic inheritance [51,84,85,86]. The endocrine disruption is the dominant component of PCOS, in which there are increased LH pulse frequency, hyperandrogenism, and ${\displaystyle \frac{\mathrm{L}\mathrm{H}}{\mathrm{F}\mathrm{S}\mathrm{H}}}$ ratio in 70-75%, 60%, and over 90% of PCOS cases, respectively [87,88]. Indeed, another study reported high adrenal and ovarian androgens such as dehydroepiandrosterone (DHEA) and dehydroepiandrosterone sulfate (DHEAS) in 20-30% of PCOS cases; furthermore, 80% of females with hyperandrogenic symptoms suffered from PCOS [49,85]. Surprisingly, 71% of PCOS hyperandrogenic daughters are born to PCOS mothers [87]. Therefore, the hormonal milieu will be a determinant for PCOS, and future studies are acknowledged to study an integrated hormone network of PCOS.

Obesity is another PCOS-related issue both as an etiological factor and as a consequence, with a prevalence of 50-70% among individuals with PCOS [4,10,89]. Accordingly, 25-70% of PCOS patients are obese, and 40-88% of PCOS cases have obesity or overweight [54,74,84,86]. Moreover, diverse mechanisms involved in PCOS pathogenesis which are glucose regulation disturbances (31.1%) [10], higher fasting glucose (10%) [90], glucose intolerance (40%) [54], IR (65-70%, 75%, 50-70%) [54,85,91,92], T2DM (7.5%, 8-10%, 15.4%, 82%) [10,54,71,85], TD1M (40.5%) [85], arterial hypertension (40%, 19.8%) [44,71], MetS (43.9%, 30%, 28.6%) [71,93], autoimmune thyroid disease [94], hirsutism (70%) [85], diminished ovarian reserve (16.9%) [95], acne (15-30%) [85], early pregnancy loss (30-50%) [96], abortion (42-73%) [85], depression (16-55%, 27-50%, 18.7%) [54,58,71], anxiety (38.6%, 17.6%) [54,71], and more attempt for suicide (40%) [97]. There are variations in the menstrual cycle of PCOS patients, in which 85-90% and 30-40% of oligomenorrhea and amenorrhea cases suffer from PCOS, respectively [85]. In this line, obesity and adipose tissue dysfunction can alter PCOS development [98]. Therefore, PCOS can be explained as a multisystem disorder with various epidemiological patterns, and more comprehensive treatments are needed to be managed.

2.1.3. Pathobiology and Possible Therapy of Polycystic Ovarian Syndrome

In essence, different organs and functions are affected in PCOS, however the dysfunction of ovarian granulosa cells is the crystal-clear sign of PCOS [3,56,66]. PCOS has a polygenic inheritance pattern that encompasses individual genes, gene interaction, and altered gene-environment interplay [2,10,49]. Accordingly, the whole gene set involved in human PCOS is included in Supplementary file S1 to give a big picture for genes governing this phenotype (vide infra).

In this context, H19 is overexpressed in PCOS and manifests fluctuation in gene expression of connective tissue growth factor and microRNAs (miRNA) miR-19b that affect germ cells and kartogenin proliferation besides apoptosis, ovarian fibrosis and PCOS [3]. Another PCOS-gene related finding is altered expressions of neurokinin B, kisspeptin, and neurokinin 3 receptor, which are expressed in the central nervous system (CNS) and ovary and alter gonadotropin secretion, regulation of hypothalamic-pituitary-gonadal axis and follicle development, and ovulation timing [53,54,66,67,71,99,100,101]. PCOS-gonadotropin hormone-related genes include gonadotropin-releasing hormone (GnRH), LH, LH receptor (LHR), LH/chorionic gonadotropin receptor (LHCGR), FSH receptor (FSHR), FSH subunit beta (FSHβ), AMH, androgen receptor (AR), SHBG, and prolactin receptor [6,7,10,49,53,55,57,61,82,102,103,104,105,106,107]. The changes in expression of cytochromes P450 (CYP) genes (CYP11a, CYP17, CYP19, and CYP21) participated in adrenal-ovarian steroidogenesis, are another part of PCOS pathogenesis and are active in the production of estrogen, hydroxyprogesterone, 11-deoxycortisol, and progesterone, respectively [6,49,51,91,102,106].

The insulin gene expression and glycoxidative pathway are altered in PCOS. These genes include insulin (INS), insulin receptor substrate (IRS)-2, IRS-1, calpain 10 (CAPN10), insulin receptor (INSR), CAPN2, gastric inhibitory polypeptide receptor (GIPR), PPARG, fat mass obesity (FTO), high-mobility group (HMGA) 1-2, thyroid adenoma-associated (THADA), RAS oncogene family member 5B (RAB5B), and adenylyl cyclases (ACY5) [7,10,49,51,56,77,91,102,103]. In this context, INS was a hub gene in the PPI network of PCOS (Table 1).

The inflammatory adipokines and immune pathway related genes are inseparable compartments of PCOS including IL1β, IL8, leukemia inhibitory factor (LIF), nitric oxide synthase (NOS2), prostaglandin-endoperoxide synthase 2 (PTGS2), metastasis-associated lung adenocarcinoma transcript 1 (MALT1), matrix metallopeptidase 2 (MMP2), adenosine deaminase RNA specific (ADAR), transforming growth factor beta-1 (TGF-β1), tumor necrosis factor alpha (TNFα), and IL6 [47,102,107,108,109,110,111]. The sepia translocation (SET) is another gene involved in PCOS, inhibiting protein phosphatase 2A (PP2A), however the remarkable point about this gene is that the site of SET overexpression can bring different outcomes; in other words, its overexpression in the brain and ovary causes tau hyperphosphorylation in AD, hyperandrogenism in PCOS, and ovarian carcinoma [111,112]. The changes in controlling genes of serotonin (5HT) formation, appetite, mood, and circadian rhythm, such as basic helix-loop-helix ARNT-like 1 (BMAL1) are associated with eating, sleep, and metabolic disorders [64,113,114].

Briefly, the genes involved in PCOS were reported in the main literature [7,10,57,61,66,78,91,92,102,103,110,115]. In summary, tachykinin precursor 3 (TAC3), tachykinin receptor 3 (TACR3), steroid 5β-reductase (SRD5B), SRD5A, DENN domain containing 1A (DENND1A), paraoxonase 1 (PON1), extracellular signal-regulated kinases (ERK)1/2, FK506 binding protein 52 (FKBP52), AR, patatin-like phospholipase domain containing 3 (PNPLA3), maestro , fatty acid desaturase 2 (FADS2), triggering receptor expressed on myeloid cells 1 (TREM1), TOX3, small ubiquitin-like modifier (SUMO) 1P1, melanocortin 4 receptor (MC4R), prospero homeobox 1 (PROX1), oxytocin receptor gene (OXTR), GATA binding protein 4 (GATA4), RAD50, Erb-B2 receptor tyrosine kinase 4 (ERBB4)/HER4, Nei like deoxyribonucleic acid (DNA) glycosylase 2 (NEIL2), vitamin D receptor (VDR), and leptin (LEP) are major players in PCOS (see Supplementary file S1).

The hypothalamic-pituitary-ovarian axis is one of the major player in PCOS [77,116]. The GnRH cells regulate pituitary gonadotropins, the synthesis of gonadal sex hormones, and the release of adrenocorticotropic hormone and adrenal hormones [53]. The outcomes of increase GnRH neurons stimulation, GnRH secretion and neuronal pathway hyperactivation by different bioactive molecules such as glucagon-like peptide-1 (GLP-1), kisspeptin, AMH, and gamma-aminobutyric acid (GABA) innervation amplified LH amplitude and pulse frequency, decreased or normal FSH level, increased ${\displaystyle \frac{LH}{FSH}}$ ratio [2,53,60,61,67,87,88,95,99,105,108,117]. The increased LH level leads to increased AMH, ovarian theca cell hyperplasia, excess estrogen and androgen, ovarian dysfunction, metabolic disorders mapped as increased leptin, adiponectin, and insulin, and decreased white fat browning, and IR, which decreases SHBG [2,53,61,117,118,119]. FSH fluctuations also cause follicular maturation disorder and cyst formation, terminate to anovulation [2]. The LH level is higher than FSH in PCOS, which disturbs follicular growth and leads to hirsutism through higher conversion of androstenedione to testosterone [108]. Androgens (DHEA and DHEAS) are produced in the ovaries and adrenal glands [49,60], and their increased levels give rise to hirsutism, increased LH levels, hyperinsulinemia, and insulin receptor desensitization, which cause a decrease in hepatic SHBG and, in turn, a higher free testosterone level [49,59]. Notably, there is adrenal hyperandrogenism in 20-30% of PCOS cases, which causes ovarian premature atresia, ovarian cyst, anovulation, and persistent estrogen that are risk factors for endometrium carcinoma [44,49]. The lower hepatic SHBG causes higher free estradiol, testosterone, and insulin levels. The insulin increment stimulates androgen production from the ovary and subsequently PCOS [5,51].

The impaired follicular growth, hyperandrogenism, and increased hepatic free fatty acid (FFA) flux can upturn insulin levels [90,108,116,119]. This increment boosts LH and androgen level in a vicious cycle, IR, cessation of follicular development, and higher cyst formation [49,108,116,120]. Following IR, the hepatic SHBG and insulin-like growth factor (IGF1) binding protein (IGFBP)-1 levels decrease, while androgens increase, which stimulates visceral adipose tissue and causes FFA production, thereby aggravating IR [103]. Insulin alters steroidogenesis by PPAR-γ activation, suppressing hepatic SHBG, increasing blood free steroid hormone, and being a gonadotrophic factor that is a direct stimulator of steroidogenic enzymes, synergy with LH and FSH, and increased pituitary responsiveness to GnRH [116]. One of the dermatological manifestations of PCOS is acanthosis nigricans, skin fold hyperpigmentation, which is considered an IR manifestation [92,116]. Due to the pivotal roles of insulin in metabolism and energy regulation of the brain, any disturbances in this case can lead to neurological disorders as well [121,122]. Therefore, regulation of insulin biosynthesis can be an effective strategy for managing this multisystem disorder.

There is also a bidirectional relationship between obesity-induced inflammation and PCOS, i.e., one is the other’s cause. On one hand, IR, increased insulin, androgens, and dyslipidemic lipid profile presented with an increment of triglycerides (TGs), low-density lipoprotein (LDL), very-low-density lipoprotein (VLDL), and a decrement of high-density lipoproteins (HDL) are all behind reasons for obesity; on the other hand, obesity aggravates this scenario and induces PCOS [120,123]. Although obesity might be necessary in PCOS development, it is not sufficient as there are PCOS cases among many anorexic and thin cases [43,124]. Leptin, a biomolecule involved in obesity, affects the hypothalamic-pituitary-ovarian axis and is a key regulator of ovulation; therefore, its possible involvement in PCOS is evident [43,113,120]. Consequently, not only does obesity cause inflammatory cytokine release such as TNFα, adipokinin activation, mitochondrial dysfunction, IR, and metabolic and hormonal disorders, but also IR ends in obesity, compensatory increased insulin release, and inflammation, and the result of both pathways is a vicious cycle [5,50,68]. Vitamin D deficiency, oxidative stress (OS), immunological disturbances, and chronic inflammation are occurred in PCOS, which can lead to autoimmune diseases and higher susceptibility to infections [47,68,94,95,123,125,126,127,128]. The more detailed immunological research will be acknowledged in elucidating PCOS pathogenesis, deeply. In conclusion, PCOS has genetic, epigenetic, neuroendocrine, metabolic, hormonal, and inflammatory clinical correlates that affect not only the reproductive system but also general health status through various pathobiological pathways. Consequently, there is an urgent prerequisite for a more comprehensive diagnostic and management approach for this syndrome.

Due to complexity, various unknown issues of PCOS, and absence of a comprehensive understanding of molecular mechanisms involved in this syndrome, there are not a determined therapy up to now, and this issue can be extended to the prevention, control, and early diagnosis [129]. Therefore, the pharmacology of PCOS, including metformin, hormone therapy, 3 hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors, and lifestyle interventions focused on management of the symptoms regard to patient's referral because of one of the signs, not solving the main problem [2,9].

2.2. Text-Mining of Alzheimer’s Disease

Causes of Alzheimer’s Disease

Oxidative stress in the AD pathogenesis gives rise to overproduction of reactive oxygen species (ROS) and advanced glycoxidation end products (AGEs) which are neurotoxic and brings the inflammatory responses by releasing IL1 and TNFα. The outcomes of these inflammatory pathways are Aβ deposition, lipid oxidation, cellular apoptosis, and finally AD manifestation [14,24]. High lipid peroxidation and increased protein carbonyl content, indicate an OS condition in AD [24]. The point is a close relationship between OS and mitochondrial disorders. Following the mitochondrial dysfunctions, the consequences include decreased cytochrome C oxidase activity in the hippocampus and increased ROS production, which causes neuronal apoptosis, cortical atrophy, and AD [14,25]. This is a confirmation of mitochondrial hypometabolism that is reported in AD [20]. Therefore, OS can also alter metabolic pathways, which may cause overproduction of deteriorating biomolecules in the brain.

The common outcomes of all pathways occurred in AD are extracellular Aβ deposition and intra-neuronal tau tangles in the cerebral cortex and subcortical gray matter [13,17,18,32,36] which lead to neuronal damages and destruction [32], neuronal hypo-connectivity [22,25], neuronal apoptosis and brain atrophy [22,25], impaired synaptic plasticity [25], cognitive decline, and neuroinflammation [22]. Aβ, the hydrolytic product of amyloid-beta precursor protein, leads to synaptic dysfunction, cognitive impairment, microglial immune responses, cytokine release, cell death, and neurodegeneration [13,14,15,25,33,36,37,39,50]. In conclusion, Aβ is one of the most reliable targets for AD prevention and therapy.

2.3. The Protein-Protein Interaction Network Analysis

According to the PPI network results, there are 83 overlapping proteins interacted with 956 edges in the common PCOS-AD network, while the number of nodes (proteins) and edges (interactions) in the PCOS and AD networks are 370-5812 and 734-7817, respectively (Figure 1 and Table 1). Therefore, AD has a bigger network than PCOS and the common PCOS-AD network. The average node degree, an indicator of the number of connections per protein, is highest in PCOS (31.4) compared with AD (21.3) and their common network (23.0). These finding highlight their connectivity and the importance of shared proteins in functionality. There is a modular and interconnected structure, especially in related biological processes regarding the local clustering coefficient (0.674), the average node degree (23.0), and the presence of 956 edges in the common PCOS-AD network. The expected random interactions are lower than the observed ones (p < 1.0e^-16), which indicate the possible biological relevance of these networks and not by chance. The highly interactive hub genes include insulin (INS), interleukin 6 (IL6), tumor necrosis factor (TNF), tumor protein p53 (TP53), apolipoprotein E (APOE), peroxisome proliferator-activated receptor gamma (PPARG), estrogen receptor 1 (ESR1), and interleukin 1 beta (IL1B). In conclusion, the presence of common hub genes in AD and PCOS specifies possible common molecular elements and pathophysiological overlap of these two disorders.

3. Discussion

PCOS and AD have many common features including their complexity and unknowns, being multifactorial entity, absence of effective therapy, catalytic role of modern lifestyle on their prevalence because of dietary habits, sedentary life style, stress, chronic chemical exposure, altered working hours (i.e., night shifts), inflammation, OS, metabolic makeup and disorders, gut-brain axis contribution, brain architectural changes, altered NTs, sleep disorders, common comorbidities including IR, T2DM, obesity, vascular injuries, CVDs, thyroid diseases, immunity disorders, and higher susceptibility to COVID-19, depression, anxiety, and suicide [62,68,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156] which outlined in Figure 2.

With the expansion of molecular and cellular knowledge, the relationship between many diseases in terms of the existence of common genetic and molecular factors has received more attention, because it may be possible to manage diseases more effectively. It seems that PCOS and AD are also included in this topic, because they have common biological pathways, including metabolic processes (INS, PPRG), lipid metabolism (APOE), cell cycle control-programmed cell death (TP53), and inflammation (IL6, IL1B, and TNF; vide supra).

The complications of disruption in any of the aforementioned pathways, namely IR, metabolic dysfunction, and chronic inflammation, can result from the coordinated effect of these shared proteins, which is also indicated by the high clustering coefficient in the common network of AD and PCOS. APOE, INS, and IL6 are not only involved in metabolic disorders but also play critical roles in neurodegenerative diseases, especially AD. The systemic inflammation and impaired neuronal insulin signaling are two examples of pathologic consequences of altered levels of TNF, IL6, and insulin.

The identification of INS, IL6, TNF, APOE, IL1B, TP53, PPARG, CCL2, and ESR1 as common hub proteins in the PPI networks of PCOS and AD suggests that these disorders share interconnected molecular mechanisms despite their distinct clinical manifestations. The clustering of these proteins highlights four major biological processes: IR, chronic inflammation, hormonal dysregulation, and cellular stress responses. Collectively, these findings support the growing evidence that PCOS is not only a reproductive disorder but also a systemic metabolic condition that may predispose affected women to neurodegenerative diseases later in life [108]. To the best of our knowledge, there is currently no large prospective cohort study that conclusively demonstrates PCOS as an independent predictor of AD, although a proteomic study confirmed the interconnected nature of these disorders [128]. Therefore, the design of epidemiological investigations may uncover common aspects of these disorder.

Among the identified hub proteins, INS occupies a central role, emphasizing IR as a critical molecular link between PCOS and AD. IR is a hallmark feature of PCOS and contributes to hyperinsulinemia, obesity, and MetS. In CNS, impaired insulin signaling has been associated with reduced neuronal glucose utilization, increased OS, amyloid-β accumulation, and tau hyperphosphorylation, leading some researchers to characterize AD as a form of "type 3 diabetes" [108]. Thus, disturbances in insulin signaling may provide a mechanistic bridge between metabolic dysfunction in PCOS and neurodegeneration in AD.

The inflammatory mediators including IL6, TNF, IL1B, and CCL2 identified as shared hubs further highlight the significance of chronic low-grade inflammation in both disorders. Women with PCOS exhibit elevated circulating levels of pro-inflammatory cytokines, which contribute to IR and ovarian dysfunction. Similarly, in AD, activated microglia release IL6, TNF-α, and IL-1β, promoting neuroinflammation, neuronal injury, and progression of amyloid pathology. CCL2 regulates immune cell recruitment and sustains inflammatory signaling within both peripheral tissues and the brain [157]. Persistent systemic inflammation may therefore represent a common pathogenic mechanism linking PCOS to increased susceptibility to AD.

The identification of APOE as a shared hub protein is particularly noteworthy because APOE represents the strongest genetic risk factor for sporadic AD. APOE influences cholesterol transport, neuronal repair, amyloid-β aggregation, and neuroinflammation. The APOE ε4 allele significantly increases AD risk and accelerates disease onset. In parallel, dyslipidemia is a common metabolic abnormality in PCOS, characterized by elevated TGs and LDL-C levels [158]. Therefore, APOE-mediated disturbances in lipid metabolism may constitute an important molecular intersection between PCOS-associated metabolic dysfunction and AD pathogenesis.

Another key hub identified in the network is PPARG, a transcription factor involved in adipogenesis, insulin sensitivity, glucose homeostasis, and inflammatory regulation. PPARG polymorphisms have been associated with PCOS susceptibility, whereas activation of PPARG signaling has demonstrated neuroprotective effects in experimental models of AD through modulation of inflammatory pathways and improvement of metabolic function [159]. The presence of PPARG among the common hub proteins suggests that metabolic dysregulation and inflammation converge through shared regulatory mechanisms in both diseases.

The identification of ESR1 highlights the contribution of endocrine signaling to the molecular overlap between PCOS and AD. Estrogen signaling is essential for glucose metabolism, mitochondrial function, synaptic plasticity, and neuronal survival. Women with PCOS often exhibit hormonal imbalances involving altered estrogen-androgen signaling, while declining estrogen levels during aging and menopause have been associated with increased AD risk. Consequently, ESR1 may represent an important molecular mediator linking reproductive endocrine dysfunction to neurodegenerative processes [108].

The shared presence of TP53 further suggests that OS and cellular senescence contribute to the pathophysiology of both disorders. TP53 regulates DNA repair, apoptosis, and responses to oxidative damage [160]. Increased OS has been extensively reported in PCOS and is also a major driver of neuronal dysfunction and cell death in AD [24,59]. Therefore, TP53-mediated stress responses may represent another common pathway underlying disease progression.

Overall, the shared hub proteins identified through PPI network analysis indicate that IR (INS), chronic inflammation (IL6, TNF, IL1B, and CCL2), lipid dysregulation (APOE), endocrine imbalance (ESR1), metabolic regulation (PPARG), and cellular stress responses (TP53) constitute a highly interconnected molecular network linking PCOS and AD. These findings support the hypothesis that long-term metabolic and inflammatory disturbances in PCOS may contribute to an increased risk of neurodegeneration and cognitive decline. Furthermore, they suggest that therapeutic interventions targeting IR, inflammation, and hormonal dysregulation may have potential benefits for preventing or delaying AD-related pathology in women with PCOS.

4. Materials and Methods

Firstly, literature search using related keywords including AD, PCOS, their association, and related bioinformatics analyses was conducted across PubMed, ScienceDirect, and Google Scholar databases without any filtrations during the search process to distillate the common pathogenesis of two disorders through text mining. Secondly, the entire gene set involved in human PCOS and AD was curated (from Phenopedia, a knowledge-based component of the Public Health Genomics and Precision Health Knowledge Base (version 10.0) launched at https://phgkb.cdc.gov/PHGKB/startPagePhenoPedia.action. The retrieved gene lists were converted to their corresponding proteins reported in the UniProt launched at https://www.uniprot.org/ using the SynGO gene set analysis tool (https://www.syngoportal.org/) and submitted to the STRING (version 12) launched at https://string-db.org/ for constructing their protein-protein interaction (PPI) networks. To improve the reliability of the network, interactions were filtered using a STRING confidence score threshold of 0.4, and only interactions meeting or exceeding this threshold were retained for subsequent analysis. Hub genes were identified from the resulting PPI networks based on node degree of proteins in network. The STRING’s built-in functional enrichment analysis included without exhausted statistical analyses.

5. Conclusions

This study highlights a significant molecular overlap between AD and PCOS ontologically involved in metabolism, cell cycle, and inflammation. These common biological pathways highlight a potential intertwining between AD and PCOS. This study provides a rationale for investigating the likelihood of PCOS patients developing AD, and offers therapeutic approaches to target the shared molecules and mediators specifically in inflammation and insulin signaling, with dual advantages. To achieve this goal, more integrated and accurate laboratory and clinical studies will be appreciated at future.