One of the major reproductive medicine achievements of the last three decades has been the development of assisted reproductive technologies (ART) for infertility management, especially in vitro fertilization (IVF), followed by embryo transfer into the uterine cavity. Despite the progress in ART, the pregnancy rate remains lower than previously anticipated and is only 33–35%. According to a number of experts, the low IVF success rate is related to embryonic and endometrial factors [1, 2].

The successful implantation and pregnancy maintenance require a number of obligatory conditions including the receptive endometrium, a functional viable blastocyst and a delicate “molecular dialogue” between the blastocyst and endometrium.

When transferring embryos with high-quality morphological and chromosomal characteristics, implantation rates remain low, comprising only 25% –35%. According to a retrospective cohort study conducted in the USA in 2011-2015, including 998 cycles with preimplantation genetic examination of the fetus, the implantation rate remained below 60% even after genetic screening [3].

Ovarian follicles are known to be the functional units of the ovary, they support the development and maturation of oocytes, which then undergo ovulation and fertilization. The key to a viable embryo is a morphologically and functionally normal oocyte. Nowadays, many women seek pregnancy at the age of 35–40 years, when the ovarian reserve starts to decline, and 1–12% of women in the population are at risk of premature ovarian insufficiency (POI) development, which is usually diagnosed late and leads to infertility [4]. IVF with oocyte donation is number one procedure for this category of patients, used in attempt to achieve pregnancy, but its effectiveness does not exceed 6% [5]. However, most patients with POI are willing to have a native child. Thus, the problem of POI is extremely urgent and important nowadays, so the scientific search for innovative methods of early diagnosis and treatment of this disease is continued.

The first surgery aimed at activating ovarian function in a patient with POI was done in Russia in 2019. It was headed by L.V.Adamyan. This operation includes the ovarian cortical layer sampling, fragmentation and re-implantation, all done within one surgical procedure. This manipulation leads to the primary follicles’ growth activation and increase in estradiol level, making it possible to perform superovulation in the IVF program and have genetically native offspring. This method also improves the quality of life in this category of patients [5].

It should be noted that various causes of POI have been described, but the molecular and genetic factors of etiology and pathogenesis remain unclear and most cases are idiopathic.

It is also believed that low pregnancy rates in IVF cycles are largely due to impaired endometrial receptivity. Maturation of the endometrium must be synchronous with the development of the embryo in order to achieve successful implantation and further maintenance of pregnancy. Increasing the endometrium receptivity can improve the results of in vitro fertilization and embryo transfer [1, 6, 7].

Endometrial receptivity is a physiological state in which blastocyst implantation is possible. This period of endometrial receptivity, also known as the window of implantation (WOI), lasts a limited time, during which the endometrial epithelium state is favorable for embryo adhesion, subsequent invasion of the endometrium and placentation. The suggested human WOI theoretically coincides with the 20th to 24th day of the 28-day menstrual cycle [1, 6] or 6-10 days after the luteinizing hormone (LH) surge in the blood [8].

The endometrium is a fairly dynamic tissue that undergoes cyclic proliferation, differentiation, cellular transport (especially migration of the immune cells), degeneration and regeneration under the influence of estrogens and progesterone in accordance with the cyclic changes in their concentration during the menstrual cycle. The main goal of this complex process is successful adhesion, invasion and placentation of the fetus, followed by postpartum regeneration [9]. Various intercellular, molecular and morphological interactions, the expression of adhesion molecules, matrix metalloproteinases, growth factors and cytokines that perform paracrine, autocrine, intracrine regulation, necessary for successful implantation, occur in the tissues of the glandular and integumentary epithelium of the endometrium, stromal cells and vessels, the extracellular matrix before implantation [9].

The "molecular dialogue" that occurs between the embryo and the endometrium before and during implantation leads to the synthesis and release of various molecules into the endometrium. The expression of these molecules is dependent on the developmental phase of the endometrium. During the secretory phase of the menstrual cycle, endometrial secretion is rich in carbohydrates, glycoproteins, lipids, binding and nutritional transport proteins, ions, glucose, cytokines, enzymes, hormones, growth factors, proteases and their inhibitors and other substances [10]. The endometrial secretion composition varies with the menstrual cycle as a result of changes in the concentration of ovarian steroids in serum. All these substrates are important sources of nutrients necessary for metabolic processes in the fetoplacental system, regulate the development of the placenta, and also modulate maternal immunological reactions to placental tissues [9].

However, the role of the endometrium is not limited to implantation of embryos. It also creates an adequate environment for the further development of the embryo providing signaling molecules and being a continuous source of nutrients for the growing embryo during the post-implantation period. Thus, scientists came to the conclusion that the endometrial microenvironment is important for successful implantation, adequate nutrition of the embryo and its development [11].

An important problem remains the identification of the period of maximum correspondence between embryo development and endometrial responsivity. Many studies searching for appropriate endometrial receptor biomarkers have been conducted in the last decade. The endometrial biopsy and subendothelial blood flow study are among the most widely used markers, but, as it turned out, they both have low prognostic significance [12].

Nowadays, both invasive and non-invasive methods are used to assess endometrial receptivity. Endometrial biopsy is the major invasive study. Non-invasive methods include examination of cervical mucus and endometrial secretions obtained by aspiration and/or lavage. The following methods are used to study the materials: immuno-histochemistry with estrogen, progesterone receptor and pinopodia expression analysis [13], thin-layer chromatography of uterine secretions [14], DNA microarray [15], microRNA studies [16] and proteomic analysis of endometrial or cervical mucus secretions using mass spectrometry [17], the study of endometrial secretions [16].

The following molecular markers were determined using these research methods: integrin ανβ3 [18, 19]; leukemia inhibiting factor (LIF) [19]; vascular endothelial growth factor (VEGF) [3]; heparin-binding epidermal growth factor (HB-EGF) [13, 20]; interleukins (IL) IL-1, IL-6, IL-11 [18]; matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs) [21]; glycodelin A [22]; cadherins, MUC1 [7] and others.

Endometrial Receptive Array (ERA) is also used nowadays, which allows to study the expression of genes available in the microchip. Every gene has diagnostic significance for determining endometrial receptivity [23]. This test has several limitations, including the need for an endometrial biopsy and the potential variability of gene expression caused by the collection of a small sample of endometrial tissue from a specific local area, which probably may not characterize the general state of endometrial susceptibility, as well as the lack of expediency of embryo transfer in the same cycle. With the technology development, the use of proteomic methods for studying endometrial secretions or cervical mucus has become an alternative to the biopsy of endometrium for the endometrial receptivity assessment [12].

Thus, high-tech achievements enabled scientists to conduct research in the fields of proteomics, epigenetics, and DNA microarrays for mRNA and miRNAs, as well as to obtain data based on immunological factors to identify potential markers for endometrial receptivity evaluation. However, the scientists dealing with this problem themselves note that the increase in the number of biomarkers has further complicated the understanding of the state of endometrial receptivity, since there are no highly specific biomarkers available and current studies don’t provide clear data on the factors responsible for homeostasis in endometrial cells and blastocyst, and their interaction [1].

It must be noted that high concentrations of hormones during ovarian superovulation in IVF cycles affect the composition and expression of these factors, leading to the disruption of homeostasis in the cells of the endometrium and embryos. There is evidence that the implantation rate is much higher in natural cycles with frozen embryo transfer than in directly stimulated cycles during IVF [1].

It is known that invasion of the embryo into the basement membrane of the endometrium is mediated through apoptosis, which brings the trophoectoderm in close contact with the endometrial epithelium. Spontaneous periodic apoptosis in the normal endometrium within the menstrual cycle is a key factor in maintaining the normal structure and function of the endometrium. Cellular autophagy plays an important role in apoptosis of endometrial cells in various phases of the endometrial cycle [24].

Apoptosis is a programmed cell death. However, autophagy is one of the alternative mechanisms contributing to the degradation and death of cells under the influence of many unfavorable factors. The term "autophagy" means "self-eating". Although autophagy has been studied for 50 years, progress in understanding this process at the molecular level has accelerated significantly over the past two decades [25].

Autophagy is an evolutionary process aimed to maintain cell homeostasis by degradation of large protein aggregates, amino acid recycling, reduction of damaged proteins or foreign organelles, processing of nutrients and regulation of protein levels in response to extracellular signals, which contributes to cell survival under stressful conditions [26]. Starvation, hypoxia, oxidative stress, infections, hormonal stimulation are among the well-known autophagy inducers [27].

Three main types of autophagy are known: macroautophagy, microautophagy, and chaperone-mediated autophagy [19]. Macroautophagy is a pathway resulting in mass degradation of cytoplasmic components, such as proteins and organelles, in the lysosome. Induction of autophagy leads to the formation of a two-layered structure, known as an isolating membrane (phagophore), which continues growing and surrounding cytoplasmic components, forming a structure called autophagosome. The outer membrane of the autophagosome then fuses with the lysosome, forming an autolysosome, allowing the lysosomal enzymes to destroy the cytoplasmic components contained in the autophagosome. Autophagosome life is short compared to other organelles. Amino acids produced by the degradation of cytosolic components can be reused by the cell. Autophagosomes are involved not only in the lysosomal degradation, but also in many other processes, including phagocytosis, exocytosis, secretion, antigen presentation, induction and modulation of the inflammatory response [28]. In microautophagy, the lysosome absorbs damaged macromolecules and organelles by invagination of the lysosome membrane [29]. Chaperone-mediated autophagy promotes the transport of cytosolic proteins to lysosomes. The intensity of chaperone-mediated autophagy and macroautophagic activity decrease with age [30].

Thus, autophagy, like apoptosis, is a mechanism of cell death. When the disruption of apoptosis arises, induction of autophagy occurs. However, the mechanisms of interaction between these two processes are not well understood. During cell starvation, its homeostasis is initially regulated via autophagy, but with the exhaustion of proteins and organelles, irreversible cell death is triggered by apoptosis [17].

Autophagy involves phosphatidylinositol-3-kinase (PI3K)/Akt pathway, AMP-dependent protein kinase (AMPK), mitogen-activated protein kinases, inositol triphosphates and calcium [27].

The most studied molecular pathway regulating autophagy is the mTOR (Target of Rapamycin), which is located in the mTOR 1 complex (mTORC1) and inhibits autophagy under conditions of increased nutrition to maintain homeostasis. The mTOR pathway forms the nucleus of two different signal complexes mTORC1 and mTORC2 [31]. mTORC1 can promote lipid biogenesis and energy metabolism by inducing transcription factors and suppressing autophagy [31]. However, under fasting conditions, mTOR activity decreases, leading to UNC-51-like kinase 1 (ULK1) activation, being the initial stage of autophagy. Becklin 1, consisting of a complex of beclin 1, a family of Bcl-2, class III PI3K (vesicular protein 34, VPS34) and ATG14L, also plays a role in the regulation of autophagy [32]. Stimulation of this complex promotes phosphatidylinositol-3-phosphate production, which regulates the formation of autophagosomes. Meanwhile, the PI3K/Akt signaling pathway downregulates the beclin 1 complex and stimulates mTOR, thereby inhibiting autophagy. Autophagosome elongation occurs through two ubiquitin-like conjugation systems, the Atg5-Atg12 conjugation system, and the LC3 conjugation system. LC3-II, which is formed by conjugation of phosphatidylethanolamine with LC3-I, is incorporated into autophagosomes and thus is considered to be a marker of autophagosome formation [33].

Autophagy plays an important role in the menstrual cycle, embryogenesis, implantation and the development of pregnancy [34, 35].

Recent studies have revealed active apoptosis and autophagy in endometrial epithelial cells during the late secretory phase and during menstruation, while low activity of these processes was detected during the proliferative phase and at the beginning of the secretory phase. The expression level of MAP1LC3A-II (Microtubule associated protein-1 light chain-3), which plays an important role during the autophagosome elongation stage, was evaluated as a biomarker of the endometrial cell autophagy in the study. The expression of MAP1LC3A-II increased and reached a maximum level in the glandular cells of the endometrium during the late secretory phase of the menstrual cycle. In addition, intense immunoreactivity of cleaved caspase-3, a marker of apoptosis, was observed in MAP1LC3A-positive glandular cells of the secretory endometrium [36]. According to available data, apoptosis of the endometrial glands and stromal cells during the menstrual cycle is controlled by ovarian steroids — estrogen and progesterone. During the proliferative phase, estrogen-induced activation of anti-apoptotic and inhibition of proapoptotic genes occur. On the other hand, progesterone inhibits estrogen-induced anti-apoptotic effects in endometrial epithelium [22]. If pregnancy does not develop, menstruation occurs, implying the activation of proapoptotic factors and the induction of apoptosis. Thus, autophagy of endometrial cells can be directly involved in the regulation of the endometrial cycle and is closely related to the induction of apoptosis [37].

During the first trimester of pregnancy, the placenta develops under conditions of low glucose levels and physiological hypoxia with oxygen tension approximately 2% [38]. While hypoxia and nutrient deficiency are usually harmful to other cells, they are preferable for trophoblasts in early pregnancy (up to 11 weeks of gestation), leading to extravillary trophoblast penetration of the endometrium to 1/3 of the uterine myometrium and their migration along the lumen of the spiral arterioles [35]. Thus, trophoblasts evolutionarily receive stress adaptation mechanisms that are based on autophagy. Trophoblastic invasion and vascular remodeling are necessary for proper perfusion of the placenta and maintenance of fetal growth [35]. This is confirmed in a study using a mouse model in which the important autophagy gene Atg7 was deleted in trophoblast cells. Atg7-knocked out placentas were characterized by a smaller size, degree of trophoblast invasion, and impaired vascular remodeling compared to the control group. This indicates that a placentation defect occurs in the absence of autophagy induction, which is also seen in pregnancy complicated with preeclampsia. Thus, the authors showed that physiologic hypoxia promotes autophagy, regulating the invasion of primary trophoblasts and thereby ensuring adequate placentation [39].

There is a hypothesis regarding the etiology of preeclampsia, which links its development to an insufficient trophoblast invasion and impaired vascular remodeling [20]. Some studies, evaluating autophagy activity association with trophoblast invasion, showed somewhat conflicting results. According to some published data, soluble endoglin, which is present in high concentrations in the serum of pregnant women with preeclampsia, suppresses the activity of autophagy, causing insufficient invasion in extravillous trophoblasts. The expression of the p62 autophagy inhibition marker in extravillous trophoblasts of decidua basalis is significantly increased in pregnant women with preeclampsia compared with pregnant women with normotension [40].

Autophagy is activated during preimplantation development of the embryo, leading to mass degradation of maternal proteins, and degradation products become nutrients and structural materials for subsequent embryonic development. There are two systems of degradation: the ubiquitin-proteasome pathway and the autophagy-lysosome system. It is believed that these systems support early embryonic development [41].

Proteasomal degradation selectively recognizes short-living ubiquitin-bound proteins, whereas lysosomal-mediated degradation is aimed at long-lived proteins. Cytosolic components, damaged organelles, as well as ubiquitin-positive proteins are delivered to lysosomes through autophagosomes, while extracellular materials are transported through endocytosis [42].

Pregnant women receiving donor oocytes are at a greater risk of preeclampsia and gestational hypertension than pregnant women with their own oocytes. The p62 accumulation in extravillous trophoblast is significantly higher in pregnant women with donor oocytes, meaning that inhibition of autophagy correlates with preeclampsia [40]. But there are conflicting data reporting high activity of autophagy in preeclamptic placentas. Electron microscopy revealed the presence of autophagic vacuoles both in the syncytial layers and in the endothelium of the placentas of pregnant women with preeclampsia [43]. An increase in MAP1LC3-II and p62 decrease were recorded in the placentas of women with hypertension compared to those who have normotensive pregnancy, which indicates activation of autophagy [44].

Autophagy disorders lead to the accumulation of amyloid protein aggregates, which in addition contribute to the development of neurodegenerative diseases [45]. It is assumed that amyloid proteins may play an important role in the development of preeclampsia. Amyloid protein aggregates are present at higher levels in the urine of women with preeclampsia than in healthy pregnant women [46]. Thus, autophagy prevents protein aggregation in trophoblasts. These proteins can interfere with placental development through apoptosis and cell aging induction. Cell aging is known to be caused by apoptosis or inhibition of autophagy in trophoblasts and leads to shortening or dysfunction of telomeres. This process is observed in preeclampsia and is associated with aging of the placenta. Aging cells also change their microenvironment, secreting proinflammatory cytokines, chemokines, growth factors and proteases, all together known as the secretory phenotype associated with aging [47].

Tsukamoto et al. (2008) conducted a study and reported that autophagy does not play an important role in folliculogenesis or oogenesis, since oocytes lacking Atg5 gene are fertilized in vivo [48]. After fertilization, the mother proteins in oocytes are degraded and new proteins, encoded by the zygotic genome, are synthesized. Autophagy is rapidly activated during the first 3-4 hours after fertilization and lasts for a short period of time. Then its activity decreases from the unicellular to the early bicellular stage. A possible explanation for the autophagy reduction at this stage is the need to avoid accidental degradation of nuclear factors, crucial for embryonic development. Autophagy is reactivated in the middle of the bicellular stage and maintains a high level of activity from the four— to eight-cell stage, before its activity declines again in subsequent stages. The major qualitative changes in the protein synthesis occur at the four to eight-cell stage in mice. Proteins received from the mother are stored in the oocyte and can be readily degraded and be replaced by proteins originating from the embryonic genome [48].

Latz et al. (2013) reported that autophagy-deficient embryos obtained from Atg5-deleted oocytes stop at four— to eight-cell stages when they are fertilized by spermatozoa without Atg5, whereas after fertilization with Atg5+ sperm, they pass to the blastocyst phase. This study indicates that a complete autophagy deficiency can be fatal for embryos. In addition, the autophagy-deficient embryos show maternal proteins’ accumulation inside the oocytes due to decreased protein synthesis rate. Thus, autophagy degrades maternal factors necessary for preimplantation development [49].

In another study, the authors showed that deficiency of Atg7 (autophagy induction gene), specific for germ cells, led to a 79% decrease in the number of primary ovarian follicles compared with the control group, causing the development of premature ovarian depletion. It was also demonstrated that Atg7 deficiency led to excessive loss of follicles in 49% of cases and decrease in oocytes in 52% of cases in the neonatal period. The authors concluded that autophagy can protect follicles and oocytes from excessive atresia through apoptosis in the ovaries of newborns under conditions of starvation that occurs after birth, suggesting that autophagy is necessary for the survival of germ cells during the neonatal period. The authors also showed that active induction of autophagy occurred immediately after birth and within the first 3–6 hours postpartum, which was confirmed by the high expression of the LC3 autophagy marker during immune staining. It should be noted that inhibition of autophagy in the ovaries of newborn mice, cultured in vitro, led to a loss of germ cells under fasting conditions. This means that germ cells are very sensitive to fasting, and during the neonatal period there is a massive loss of germ cells. During the fetus-to-newborn transition the cells undergo severe starvation until the nutrition is restored with the milk nutrients. Most organs adapt to this postpartum fasting through autophagy, which produces amino acids to maintain energy homeostasis through autophagic degradation of “native” proteins [4].

Selective autophagy was discovered quite recently [19]. Under certain conditions, autophagosomes can selectively isolate and degrade mitochondria, endoplasmic reticulum, endosomes, lipid droplets, secretory granules, cytoplasmic aggregates, ribosomes, pathogens and viruses. For example, autophagy, including xenophagia (selective autophagic degradation of microbes), is known to be a mechanism for the destruction of pathogenic microorganisms that have entered the cell, suggesting that autophagy inducers may be a new class of antibacterial therapy [50].

After fertilization in most eukaryotic species, p62 autophagy receptors and the protein, associated with the gamma-aminobutyric acid receptor (GABARAP) remove sperm mitochondria via mitophagy. Mitochondrial sperm proteins are decomposed mainly through the ubiquitin – proteasome system, but selective autophagy is also involved in this process [51].

Experimental delayed blastocyst implantation in a mouse model using ovariectomy showed that autophagy in blastocysts is activated by 17β-estradiol (E2), which promotes the long-term survival of “sleeping blastocysts” [52].

Another study showed significant increase in the expression of autophagy-associated protein LC3-II in the endometrium of patients with recurrent implantation failure in combination with chronic endometritis compared to the control group. An increase in the amount of LC3 indicates the active formation of autophagosomes. Autophagy was mainly present in the stroma of the endometrium, where lymphocytes are located, but not in the endometrial glands, in patients with chronic endometritis. These data may indicate that autophagy has an important effect on the induction of the inflammatory response in chronic endometritis and subsequent changes in the local cytokine environment [53].

Autophagy markers are present in cytotrophoblast, syncytiotrophoblast, extravillous trophoblast and decidual stromal cells and are physiologically involved during early pregnancy. A significant decrease in mTORC1 expression was found in patients with chronic endometritis. In addition, women with recurrent implantation failure combined with chronic endometritis have abnormal expression of IL-10, TGF-β (transforming growth factor-β) and IL-17. MTOR has been reported to modify inflammatory responses by modulating immunoproteasomal degradation [54]. Therefore, a decrease in mTORC1 in chronic endometritis causes an increase in autophagy and local proinflammatory immune responses mediated by cytokines [49].

In conclusion, it should be noted that the abnormal expression of IL-10, TGF-β, and IL-17 in women with recurrent implantation failure and chronic endometritis in combination with decreased autophagy demonstrates the predominance of pro-inflammatory immune reactions of the endometrium, which is associated with a decrease in endometrial receptivity and pregnancy rate [49].

The research of the role of autophagy in the development of endometriosis has been started recently [55, 56]. Adamyan et al. reported that there is a violation of antioxidant defense mechanisms due to oxidative stress in endometriosis, leading to cell damage [57, 58]. These data are confirmed by other studies. For example, Liu et al. found that expression of hypoxia inducible factor-1α (HIF-1α) and the marker of autophagy LC3 was high in ectopic endometrium [5632].

Assessing these results, the researchers concluded that autophagy activity affects the formation of endometrioid heterotopia, and the suppression of autophagy mechanisms may become a new therapeutic target in the treatment of endometriosis [56].

Conclusion

Thus, the data presented suggest that autophagy plays an important role in the reproductive system homeostasis and the body as a whole. Current data on the etiology of many reproductive diseases are currently not completely clear, whereas treatment should be etiopathogenic in order to prevent relapse. To date, the biochemical and morphological markers used are not specific enough and are indicators of the reproductive system disturbances. Further study of autophagy, a deeper understanding of the etiopathogenic mechanisms of disorders will provide us with new opportunities for the development of diagnostic methods, treatment options and measures to prevent many diseases of the human reproductive system.

The authors declare no conflict of interest.