Current and Future Approaches to Slow Human Ovarian Aging Based on Molecular Insights

As in rodents, the HPO axis has also been shown to be altered in perimenopausal women. As compared to healthy young women, most aged women approaching menopause failed to generate an LH surge in response to estradiol challenge [16]. Moreover, a gradual increase in gonadotropin secretion has been observed to start as early as 27–28 years old and to accelerate dramatically after 49, concerning mainly FSH rather than LH [17]. Changes in gonadotropin secretion precede any detectable menstruation or ovulation failure and appear to result from an altered secretion pattern of the gonadotropin-releasing hormone GnRH [18,19].

Despite the absence of knowledge of the exact mechanism of the above neuroendocrine changes, circumstantial evidence suggests that they may somehow be related to abnormal neurotransmitter secretion in the central neural system and points to glutamate as being potentially responsible. In fact, the vesicular glutamate transporter 2 (VGLUT2), a protein that packages the excitatory neurotransmitter glutamate into synaptic vesicles to be released at synapses, has been localized by immunocytochemistry to the GnRH-secreting neurons in the preoptic region of the rat brain [20,21]. A decline in glutamate release from the preoptic area in aged female rats correlates with reduced GnRH secretion, postpones LH surge [15], and is accompanied by an attenuation of the expression of glutamate receptors in GnRH-secreting neurons [22].

In addition to glutamate, gamma-aminobutyric acid (GABA) and its GABAA receptor also play an important role in the regulation of GnRH neural secretory activity in rats [23], entering an intricate interplay with the glutamate receptors [24].

Both the glutamate- and GABA-driven signaling in GnRH neurons have been considered as potential targets of therapeutic action. In the rat model, researchers succeeded in restoring normal LH surge via combined treatment with a GABA antagonist and a glutamate agonist, suggesting that normal GnRH neural output can be reached by balancing the effects of excitatory and inhibitory neurotransmitters in the GnRH-secreting neurons [25].

Other neurotransmitters supposed to be involved in controlling the secretory activity of GnRH neurons include the peptides kisspeptin (K) [26], neurokinin B (N) [27], and dynorphin (Dy) [28]. These three peptides colocalize in the so-called KNDy cells—a group of neurons of the arcuate nucleus of the hypothalamus that co-express K, N, and Dy—and are conserved across a range of species from rodents to humans, and their interaction is believed to regulate both hypothalamic GnRH secretion and OA [28,29]. There is sufficient evidence indicating that the gene expression of these three peptides is altered in postmenopausal women and female monkeys, which is likely to cause an imbalance of excitatory and inhibitory input to GnRH neurons and thus to cause the neuroendocrine disorders involved in OA [11,30,31,32,33,34]. Whereas HPO axis (Figure 1) alterations may be the primary trigger of OA, it is also possible that it is a mere consequence of a decline in follicle numbers, leading to reduced ovarian hormone (mainly estradiol and inhibin) output and thus decreasing the negative feedback in pituitary FSH secretion (Figure 1), resulting in elevated serum FSH levels and exacerbating ovarian follicle depletion [11,34].

Compared to the relative scarcity of data on the HPO axis, HPO-independent ovarian decay has received much more attention across the latest decades, resulting in a variety of treatment strategies becoming clinically available. The depletion of ovarian follicles, resulting in hypoestrogenism, with all its negative health consequences, and the decrease in the quality of the remaining oocytes are the two principal consequences of OA [2,6,11,12]. However, in order to search for clinically effective prevention and treatment strategies, we have to ask how the different types of cells present in the ovary contribute to this process. The ovary is composed of multiple cell types, including oocytes, granulosa cells (GC), theca cells, stromal cells, immune cells, and smooth muscle cells. The relative contribution of each of these cell types to the overall picture of OA has been greatly facilitated since 2009, when single-cell RNA sequencing (scRNA-seq) was first used to analyze the transcriptome of mouse oocytes and embryos [35]. Since then, scRNA-seq has increasingly been used to analyze individual cell types in normal and aging ovaries (reviewed in Liang et al. [36]). By using this approach, a cascade of events—in which immune cell infiltration with stromal cell fibrosis worsens the ovarian microenvironment and GC apoptosis accelerates fertility loss, leading to reduced oocyte quality—has been mapped, along with the single-cell transcriptomic level [37,38,39].

It has long been known that oxidative stress, caused by the accumulation of reactive oxygen species (ROS), is intimately involved in the process of OA. ROS are by-products of normal cell (mitochondrial) metabolism, required for normal cell function, but their excessive accumulation within cells can have serious negative consequences, namely, follicle atresia and diminished oocyte quality [13]. This situation can result from an age-related decline in ROS scavenging, enhanced ROS production promoted by different lifestyle and environmental factors (see Section 4), or a combination of both. As to the former, it has been demonstrated that older women exhibit lower superoxide dismutase (SOD) levels in GC, causing decreased antioxidant efficacy with OA [13]. In addition to ROS, advanced glycosylation end-products (AGEs), formed by reactive carbonyl species and free amino groups gained from a series of nonenzymatic reactions [40,41], also contribute to OA [11].

As to the oxidative and mitochondrial stress caused by excessive ROS, an accumulation of the effects of negative lifestyle factors (see Section 3) is to be blamed. In order to compensate for this age-related condition, antioxidant and mitochondria-protective treatments, such as vitamins C, D, and E, melatonin, and others, have been used for decades to slow ovarian aging [42] (see Section 3 for the latest updates). Oocytes, as some of the most long-living cells in the organism, are particularly vulnerable to oxidative stress. The deficiencies of the mitochondrial function brought about by oxidative stress lead to damage to the proper mitochondrial DNA (mtDNA), further increasing the problem, and subsequently to damage to other cell components, mainly those of the microtubules and chromosome cohesive elements, leading to improper chromosome and chromatid separation during the first and the second meiotic division and, consequently, oocyte aneuploidy [43].

In general, antioxidant and mitochondrial therapies (Figure 2), aimed at slowing down OA, have proven to be both the most effective and best-tolerated therapies [44]. They have been shown to reduce oxidative stress caused by lifestyle factors such as smoking, unhealthy diet, drug abuse, or high alcohol consumption [45]. Among the variety of antioxidants available [11,42,43,44,46], a special position is reserved to melatonin. In fact, a synthesis of the recent data from both animal experimental and human clinical studies (reviewed in Tesarik and Mendoza Tesarik [47]) indicates that not only is melatonin a powerful antioxidant, acting both as a hormone and as a direct ROS scavenger; it also performs additional actions that benefit ovarian health, namely, anti-inflammatory and immunomodulatory activities (Figure 3). In fact, disequilibrium between individual elements of the immune system, mainly lymphocytes and macrophages, in addition to being involved in other gynecological pathologies, accelerate OA through a chronic, low-grade inflammatory state. At doses of 4–6 mg per day, melatonin is advisable both for delaying the onset of OA and for the treatment of women who already suffer from it [46,47].

Other non-invasive, recently validated treatment approaches include pharmacological inhibition of follicle recruitment, hormonal and growth factor modulation, and the use of advanced mitochondrial-targeted antioxidants. Briefly, GnRH agonists and antagonists were used to inhibit follicular recruitment during anticancer chemotherapy, aiming at the protection of the ovarian follicular pool against the adverse effects of anticancer drugs. After early encouraging data obtained both in male patients [48] and in an experimental system using female monkeys [49], this strategy was applied to female cancer patients. However, several systematic reviews and meta-analyses have suggested that this approach may be controversial, and the outcomes may differ among the different cancer types and anticancer drugs used [50,51,52,53]. Further research into this subject is warranted, focusing on the individual clinical context, dosage optimization, and timing.

Hormonal and growth factor modulation has also been reported to give encouraging results in some patients facing OA. In fact, dehydroepiandrosterone (DHEA) supplementation led to improved serum ovarian reserve markers, such as inhibin B and antimullerian hormone (AMH), antral follicle count, as evaluated by a vaginal ultrasound scan, and in vitro fertilization outcomes [54]. Other recent (2023–2025) small randomized clinical trials have also highlighted the possibilities of using insulin-like growth factor 1 (IGF-1) and vascular endothelial growth factor (VEGF) to improve antral follicle counts (reviewed in Hirano et al. [13]), and further extensive clinical validation has been suggested to develop combination-based therapies (DHEA + IGF-1 + VEGF) by creating personalized formulas adapted to each individual biomarker profile, making further use of modern precision medicine. As early as 2005, growth hormone (GH), supposedly acting—at least partially—through IGF-1, was shown to improve IVF outcomes in women aged > 40 years when administered during ovarian stimulation [55], and subsequent studies basically confirmed these findings and suggested the underlying mechanisms of GH action [56,57]. New formulations of GH delivery systems, based on novel biomaterials (see Section 4.1), have been shown to improve the bioavailability of GH and may lead to innovative therapeutic strategies for preventing and treating ovarian dysfunction [58]. Nonetheless, these treatments still have limitations. In particular, while the short-term beneficial effects on oocyte quality were demonstrated, further studies are needed to confirm their long-term effects in humans, including the absence of negative consequences for offspring health.

In addition to the above non-invasive treatments, other more invasive treatments are also currently in use. The intra-ovarian injection of platelet-rich plasma (PRP), prepared from the patient's own blood, has shown promising results in promoting the activation of dormant primordial follicles, which is believed to be due to the action of platelet-derived growth factors [59]. Fertility preservation, either through oocyte recovery and freezing or ovarian tissue cryopreservation in view of subsequent autotransplantation, is another available option [60]. In addition to being used for restoring fertility, ovarian tissue cryopreservation and autotransplantation may also be considered for endocrine function restoration [61]. Mitochondrial donation, carried out via pronuclear transfer, has recently been shown to be effective (8 live births out of 22 attempts) in women with pathogenic mtDNA variants [62]. Other advanced strategies, tested in experimental animal models, are described below (Section 4).

Research into new biomaterials in regenerative medicine, including those recreating the three-dimensional environment for in vitro cultured cells, has been aimed at creating a microenvironment favoring cell interactions, promoting good passive and active targeting, and allowing for high drug loading capacity and controlled drug release, while still displaying good stability and biodegradability. Two types of biomaterials have been investigated in animal models of OA—natural ones and synthetic ones, each showing its own advantages and disadvantages [64]. Briefly, natural biomaterials, derived from biological sources like humans, plants, animals, and microorganisms, are valued for being biocompatible and biodegradable and can be categorized based on their composition, i.e., as protein-based and polysaccharide-based.

In the context of OA prevention and therapy, recent research highlights several types of natural biomaterials, including stem cell-derived extracellular vesicles, human amniotic cell-derived, follicular fluid-derived and menstrual blood stromal cell-derived exosomes, the decellularized extracellular matrix, collagen, hyaluronic acid, fibrin, and alginate. All of these have been evaluated and have shown promising results in animal models. Likewise, synthetic (polyethyleneglycol) supramolecular hydrogels were tested in animal models of OA, with promising results (reviewed in Wu et al. [64]).

Nanoparticles (NPs) based on cerium dioxide and silver have attracted widespread attention due to their antioxidant and anti-inflammatory action, likewise silver and magnesium oxide NPs, which, in addition, were shown to alleviate the hormone imbalances occurring during OA. In turn, zinc oxide NPs have not only exhibited antioxidant and anti-inflammatory effects but have also been shown to be particularly useful in the context of OA associated with diabetes, enhancing insulin sensitivity and regulating glucose metabolism (reviewed in Wu et al. [64]).

Knowledge derived from ancestral natural medicine formulas has recently been re-evaluated in animal models of OA. For example, celastrol, a compound extracted from the Chinese plant Triprerygium wilfordii, has been shown to promote ovarian follicle development in mice and pigs by regulating granulosa cell proliferation and apoptosis [65]. A 2-month treatment of mice with low-molecular-weight chitosan, a natural fibrous polysaccharide derived from chitin, led to delayed OA, and this effect was apparently mediated by enhanced macrophage phagocytosis and improved ovarian tissue homeostasis [66]. Another study has demonstrated that collagen extracted from the sturgeon swim bladder can counteract OA in mice exposed to cyclophosphamide through a variety of mechanisms, including antioxidant effects, the activation of the PI3K/Akt and Bcl-2/Bax pathways, and the inhibition of the mitogen-activated protein kinase pathway, thereby reducing apoptosis [67].

The analysis of the mechanisms causing ovarian cellular senescence, including the accumulation of advanced glycation end products, oxidative stress, mitochondrial dysfunction, DNA damage, telomere shortening, and exposure to chemotherapy in six distinct cell types (oocytes, granulosa cells, theca cells, immune cells, ovarian surface epithelium, and endothelial cells), has suggested the potential of ovarian senotherapeutics in the treatment of OA [68]. They include the use of personalized (i.e., adapted to the current condition of each individual) mixtures of specific agents targeting the key factors involved in the senescence of different ovary cell types (see above), including desatinib, quercetin, rapamycin, metformin, resveratrol, melatonin, and coenzyme Q10 [68]. New vehicles of GH administration, facilitating its bioavailability, are also under investigation (see Section 3). However, it has to be stressed that all these therapies are currently in the pre-clinical stage, and their application in humans will require more studies concerning their short- and long-term safety.

The employment of new biomaterials and biomedicines, though showing promising results in animal experiments, awaits further validation before clinical application with regard to the efficacy and safety of these materials in humans. Above all, taking into account the basic principal of "first do not harm," a fundamental ethical principle in medicine, the eventual clinical use of these therapies should follow a thorough review of their potential effects in terms of both short-term and long-term offspring health.

Finally, the current trend towards repeating in vitro fertilization/ovarian stimulation cycles has aroused concern about possible long-term adverse effects on the ovarian reserve (i.e., the stock of oocytes in the ovaries). This issue was addressed by a recent review [69]. The data presented suggest that repeated ovarian stimulation can induce changes in the immune response and increase oxidative stress in the ovarian microenvironment, leading to an accelerated loss of ovarian reserve. This risk further explains the current interest in the use of melatonin (see Section 3.1) before, during, and after ovarian stimulation (see Section 3.1).