The Impact of Estrogen and Estrogen-Like Molecules in Neurogenesis and Neurodegeneration: Beneficial or Harmful?

FB-B and MM-R: drafting original manuscript. PL-C, LM, CF, and LB: reviewing and editing. LM, CF, and LB: validation. AO, FR, CF, and LB: supervision and funding acquisition. All authors contributed to the article and approved the submitted version.

Stem cells are able to undergo proliferative divisions that produce additional stem cells with the same properties, and divisions that produce daughter cells that differentiate into multiple cell types ( Homem et al., 2015 ). Stem cells can be “pluripotent” precursor cells that give rise to all cell types within an organism, or “multipotent” precursor cells that have more restricted potential, and differentiate into a subset of cell types. Neural stem cells (NSCs) are multipotent stem cells that can also be referred to as “progenitor” cells ( Martinez-Cerdeno and Noctor, 2018 ). Thus, in this article we will refer to these cells as neural stem/progenitor cells (NSPCs), leaving the term of neural progenitor cells (NPCs) to those cells that are descendants of NSCs and have a restricted cell linage potential: neuronal progenitor cells, astrocyte progenitor cells, oligodendrocyte progenitor cells, etc. Postnatal neurogenesis has been well demonstrated in two specific regions of the mammalian brain: the subventricular zone (SVZ) of the lateral ventricles and the subgranular zone (SGZ) of the dentate gyrus (DG) in the hippocampus ( Gross, 2000 ; Kaplan, 2001 ). In both regions, NSPCs are immersed in a particular microenvironment composed of diverse cellular constituents, such as astrocytes, pericytes, ependymocytes, and vascular and microglial cells ( Batiz et al., 2015 ; Bjornsson et al., 2015 ). Besides, it has been described other neurogenic niches able to generate new neurons and glial cells in response to brain injury or some other inductive stimuli. However, their magnitude and relevance remain still under discussion ( Kokoeva et al., 2005 ; Cameron and Dayer, 2008 ; Lin and Iacovitti, 2015 ). In the SVZ, type B cells—adult multipotent NSPCs— are located below ependymal and ependymal-like cells, which line the walls of the lateral ventricles ( Mirzadeh et al., 2008 ). Ultrastructural and immunolabeling studies have revealed that type B cells display only a small fraction of apical surface and a single basal body. These cells contact the cerebrospinal fluid by short apical processes, and blood vessels by their long basal processes. Thus, the organization of these cells, i.e., ependymal, ependymal-like, and type B cells, confers a pinwheel architecture to the ventricular surface ( Mirzadeh et al., 2008 ). Also, type B cells express diverse stem and glial markers ( Batiz et al., 2015 ). As mentioned earlier, type B cells can divide and differentiate into type C cells, which represent a population of intermediate/transit-amplifying progenitor cells that, in turn, give rise to neuroblasts or type A cells ( Doetsch et al., 1997 ; Parras et al., 2004 ; Merkle and Alvarez-Buylla, 2006 ). Finally, type A cells are guided to the olfactory bulb, where they gradually differentiate into two types of postmitotic GABAergic interneurons: the granular neurons and the periglomerular neurons ( Gage, 2000 ; Kriegstein and Alvarez-Buylla, 2009 ; Ming and Song, 2011 ). Namely, the displacement of type A cells from the SVZ of the lateral ventricles to the olfactory bulb occurs through a tunnel-like structure composed of vascular cells and an astrocyte network driving the rostral flow of neuroblasts: the rostral migratory stream ( Nguyen-Ba-Charvet et al., 2004 ). In the SGZ, radial glia-like or type 1 cells—adult multipotent NSPCs— also exhibit a multistep differentiation process as described for the SVZ niche, beginning with the expansion of type 2a cells (transient amplifying progenitor cells) ( Seri et al., 2001 ; Batiz et al., 2015 ). Then, type 2a cells divide into type 2b cells, which are more restricted intermediate progenitors, giving rise to type 3 cells (neuroblasts). Finally, type 3 cells divide into glutamatergic granule cells that are guided by astrocyte processes that function as instructive railways for their migration and subsequent integration into existing neuronal circuits in the granular cell layer ( Seri et al., 2001 ; Aimone et al., 2014 ; Batiz et al., 2015 ). Of note, the processes of migration and integration of neurons in the hippocampus are remarkably shorter than through the rostral migratory stream ( Shapiro et al., 2005 ). In the two postnatal neurogenic niches described previously, the SVZ of the lateral ventricles and in the SGZ of the DG, multipotent NSPCs -type B and type 1 cells, respectively- express estrogen receptors. Interestingly, different studies have shown the expression of estrogen receptors in embryonic and adult NSPCs. Brannvall et al. (2002) demonstrated that ERα and ERβ are both expressed by cortical embryonic NSPCs obtained at early developmental stages. Interestingly, ERα levels are relatively high at embryonic day (E) 15 and decrease in later developmental stages. On the other hand, ERβ levels remain largely constant during development and are higher than ERα levels in cortical adult NSPCs ( Brannvall et al., 2002 ). Interestingly, Waldron et al. (2010) have shown that both ERa and ERb increase their levels in cortical adult NSPCs isolated from aged rats. Regarding the hippocampus, both ERα and ERβ have been observed in hippocampal NSPCs of human post-mortem samples and a human hippocampal progenitor cell line ( Smeeth et al., 2020 ). Recent studies have shown that NSPCs obtained from Sprague Dawley rat embryos not only express ERα and ERβ, but also GPER-1 ( Zhang et al., 2019 ). Besides multipotent NSPCs, Haumann et al. (2020) recently observed the expression of GPER1 in transient amplifying progenitor cells and neuroblasts from the SVZ of recently born and adult female rats. Experiments performed by Wang et al. (2008) using human-derived NSPCs revealed a significant increase in their proliferative activity after estrogen stimulation. In fact, 17β-estradiol (E2) induces time- and dose-dependent effects associated with an increase in DNA replication and up-regulation of cell cycle proteins. Remarkably, ERβ-, but not ERα-specific ligands, lead to increased phosphorylation of ERK 1/2, which initiates cell cycle entry (demonstrated by the up-regulation of mitotic markers such as CDK1/cdc2 and PCNA), DNA replication, and duplication of the centrosome -the major microtubule-organizing center in the cell- in dividing cells ( Wang et al., 2008 ). Thus, in these cells, E2 effects appear to be predominantly mediated by ERβ. Similarly, primary cultures of embryonic rat-derived NSPCs respond to E2 increasing their proliferation ( Brannvall et al., 2002 ). GPER-1 has been found to be widely expressed in NSPCs ( Zhang et al., 2019 ; Zhong et al., 2019 ). Functionally, NSPCs isolated from Sprague-Dawley rats, and treated with 10 nM E2 for 2 days, significantly increase their proliferative rate, while doses of 50 nM notably inhibit this process. Additionally, treatment with 10 nM E2, for 7 days, stimulated differentiation to the neuronal lineage, and inhibited differentiation to astrocytes ( Zhang et al., 2019 ). However, the participation of GPER-1 in these cellular responses was not specifically verified. Recently, Zhong et al. (2019) stimulated mouse-derived NSPCs with different concentrations of G1, a selective GPER-1 agonist, showed a dose-dependent decreased proliferation, which was blocked by selectively inhibiting GPER-1. It has also been demonstrated that ERs are pivotal in the maintenance and functioning of oligodendrocyte progenitor cells (OPCs) and post-mitotic oligodendrocytes (OLs), respectively. Crawford et al. (2010) showed that diarylpropionitrile (DPN), a synthetic and non-steroidal ERβ agonist, improves the generation of OLs and remyelination of the corpus callosum and spinal cord of mice with experimental autoimmune encephalomyelitis (EAE), a model for multiple sclerosis, even restoring nerve conduction and refractoriness of callosal axons. DPN induces a marked improvement in the number of mature OLs, myelin density and axonal conduction, leading to functional recovery in EAE mice ( Khalaj et al., 2016 ). Also, the spinal cord of EAE mice treated with DPN display increased activation of Akt/mammalian target of rapamycin (mTOR) axis, which is not observed in a cohort of EAE mice with conditional deletion of ERβ ( Khalaj et al., 2016 ). This finding correlates with those described by Kumar et al., who demonstrated that ERβ is crucial to enhance OLs production, myelin protein synthesis, remyelination, and axon conduction in EAE mice through the activation of the PI3K/Akt/mTOR signaling pathway, without disturbing inflammation in the corpus callosum or spinal cord ( Kumar et al., 2013 ; Figure 2 ). A growing body of evidence suggests estrogens are relevant in oligodendrogenesis by promoting oligodendrocyte differentiation in a GPER-1-dependent manner. In this regard, it has been shown that the pharmacological treatment with the classical estrogen receptor antagonist Fulvestrant (ICI-182,780) modifies the proliferative activity of NSPCs without affecting the differentiation toward the oligodendroglial lineage ( Okada et al., 2010 ). Notably, the use of a cell-impermeable E2, a bovine serum albumin-conjugated E2 (E2-BSA), rapidly activates ERK1/2-dependent pathways, which is not inhibited by ICI-182,780 ( Okada et al., 2010 ). Furthermore, E2-BSA does not promote proliferation of NSPCs, but it does mimic the increase in the generation of oligodendrocytes, strongly suggesting that the oligodendroglial generation from NSPCs is likely to be stimulated via membrane-associated ERs ( Okada et al., 2010 ). In this regard, it has been shown that tamoxifen, a type of selective estrogen receptor modulator (SERM) that acts as a competitive antagonist or as a partial agonist depending on the cell type, favors the repair of demyelinated lesions in the CNS by increasing the mRNA levels of myelin basic protein (MBP) and 2′, 3′-Cyclic nucleotide 3′-phosphodiesterase (CNP), two genes highly expressed during oligodendroglial differentiation ( Gonzalez et al., 2016 ). Furthermore, the effects of tamoxifen were blocked when a pan-antagonist was added to the media, thus supporting that tamoxifen acts as an estrogen receptor agonist in OPC differentiation. Interestingly, this tamoxifen-induced differentiation occurs in a PKCα signaling independent-manner, and not only by activation of classical estrogen receptors but also by activation of GPER-1 ( Gonzalez et al., 2016 ). These observations correlate with those reported by Barrat et al., showing that tamoxifen-derived metabolites (i.e., 4-hydroxy-tamoxifen and endoxifen) also promote oligodendrogenesis in an estrogen receptor-dependent manner, including non-classical ERs ( Barratt et al., 2016 ; Table 1 ). Phytoestrogens have been previously described by their ability to promote the differentiation of diverse types of stem cells, including NSPCs ( Heim et al., 2004 ; Oberbauer et al., 2013 ). In vitro experiments performed on embryonic forebrain-derived NSPCs revealed that hops ( Humulus lupulus L.)-derived prenylflavonoids—the most potent type of phytoestrogens reported so far—can induce neurite outgrowth of dorsal root ganglion neurons and neuronal differentiation of NSPCs, measured by the activation of the promoter of neuronal fate-specific doublecortin (DCX) gene ( Oberbauer et al., 2013 ). Likewise, resveratrol (3,5,4′-trihydroxy-trans-stilbene), a polyphenolic compound found in grapes and wine that acts as a phytoestrogen ( Gehm et al., 1997 ), can regulate the proliferation and neuronal differentiation of embryonic and adult NSPCs and human bone marrow-derived mesenchymal stem cells through the activation of protein kinase A (PKA)/glycogen synthase kinase 3 β (GSK3β)/β-Catenin and PKA/ERK1/2 signaling pathways, as well as p38 kinases, sirtuin-1 (SIRT1) and AMP-activated protein kinase (AMPK) ( Figure 3 ; Taupin and Gage, 2002 ; Park et al., 2012 ; Kumar et al., 2016 ; Jahan et al., 2018 ). Administration of resveratrol (20 mg/kg body weight) to adult rats significantly increases the number of proliferating and newly generated cells in the hippocampus, with upregulation of p-CREB and SIRT1 proteins ( Kumar et al., 2016 ). In contrast, experiments performed in mice showed that resveratrol (10 mg/Kg body weight) reduced the proliferation of NSPCs in the hippocampus ( Park et al., 2012 ). In vitro , resveratrol exerts biphasic effects on rat-derived NPCs; low concentrations (10 μM) stimulate cell proliferation through increased phosphorylation of ERK1/2 and p38 kinases, whereas high concentrations (>20 μM) exhibit inhibitory effects ( Kumar et al., 2016 ). Xenoestrogens are considered endocrine disruptors commonly associated with harmful effects in the CNS, affecting several cellular functions. Growing evidence indicates that bisphenol-A can modify the neurogenic process, affecting the proliferation/differentiation balance of NSPCs and neuronal maturation. Gestational exposure to low doses of bisphenol-A (20 ng/kg) change neuron distribution across cortical layers. A single injection of 5-bromo-2′-deoxyuridine (BrdU) at E14.5 showed a decrease in the number of proliferative (BrdU-positive) cells of the IV cortical layer, whereas layers V and VI showed more BrdU-labeled cells in bisphenol-A-treated than control mice ( Nakamura et al., 2007 ). Experiments performed by Tiwari et al. (2015b) showed that bisphenol-A has a dual role on NSPCs proliferation. Treatment of neurospheres derived from rat hippocampus with bisphenol-A at doses lower than 100 μM showed a significant increase in proliferation and cell viability. However, doses of bisphenol-A higher than 100 μM induce a substantial reduction the viability and the number of proliferating NSPCs in neurospheres, as well as the number of neurons and astroglial cells derived from them ( Tiwari et al., 2015b ). Similar results have been found in vivo , doses of 4 μg/kg body weight of bisphenol-A enhanced the number of BrdU + cells in the rat hippocampus, whereas doses of 40 and 400 μg/kg significantly reduce them ( Tiwari et al., 2015b ). Interestingly, most of the bisphenol-A-dependent detrimental effects on hippocampal neurogenesis are ameliorated or prevented by curcumin treatment ( Tiwari et al., 2016 ). On the other hand, bisphenol-A decreases the number and size of oligodendroglial-committed neurospheres. Indeed, both in vitro and in vivo findings reveal that bisphenol-A alters the proliferation and differentiation of OPCs, and reduces the expression of myelination-related genes ( Tiwari et al., 2015a ). High concentrations of bisphenol-A also reduce the number of neurons and significantly increases the number of glial cells located in the rat DG, suggesting a pro-glial differentiation effect on proliferating NSPCs ( Tiwari et al., 2015b ). Bisphenol-A treatment also reduces the number of differentiating neurons from an immortalized human NSPC line known as ReNcell VM ( Fujiwara et al., 2018 ). Recent investigations have demonstrated that bisphenol-A affects mitochondrial biology in hippocampal NSPCs, thus leading to increased oxidative stress and apoptosis ( Agarwal et al., 2016 ; Figure 3 ). The dynamin-related protein (Drp-1) and its mitochondrial adaptor fission protein 1 (Fis-1) are the main responsible for mitochondrial fission, a process positively related to mitochondrial fragmentation and apoptosis ( Frank et al., 2001 ; Lee et al., 2004 ). On the other hand, the mitochondrial dynamin-like GTPase (Opa-1) and mitofusin 1 and 2 (Mfn-1/2) are the pro-fusion proteins responsible for maintaining mitochondrial circuitry functions. Bisphenol-A treatments upregulate the levels of Drp-1 in NSPCs in vitro and in the rat hippocampus, leading to decreased elongated mitochondria, abnormal cristae depletion, apoptosis, and consequently loss of NSPCs proliferation/differentiation balance. In parallel, bisphenol-A reduces mitochondrial membrane potential and ATP ( Agarwal et al., 2016 ). Bisphenol-A also decreases the levels of superoxide dismutase (SOD) and catalase, enhancing the cellular levels of ROS. In that sense, genetic and pharmacological inhibition of Drp-1 reduces all these bisphenol-A-mediated inhibitory effects, indicating that impaired mitochondrial dynamics and cytotoxicity are Drp-1 dependent ( Agarwal et al., 2016 ). Additionally, bisphenol-A’s inhibitory effects are also diminished by pharmacological treatment with the antioxidant N-acetyl-cysteine (NAC), which similarly reduces Drp-1 levels in treated NSPCs cultures ( Agarwal et al., 2016 ). Similarly, bisphenol-A provoked an accumulation of ROS levels in HT-22 cells, an immortalized mouse hippocampal cell line ( Pang et al., 2019 ). Recent investigations have shown that some of the effects of bisphenol-A in neurogenesis are mediated by the suppression of the Wnt/β-catenin signaling pathway ( Tiwari et al., 2015b , 2016 ). In that sense, bisphenol-A promotes the expression of extrinsic Wnt inhibitory molecules (Dkk-1 and Wif-1) and decrease Wnt-3 protein levels. Bisphenol-A treatment induces the expression of the regulator GSK-3β (glycogen synthase kinase 3 β) in the hippocampus and SVZ, consequently reducing the ratio of phospho-GSK-3β/GSK-3β and enhancing the phospho-β-catenin/β-catenin one. This process prevents β-catenin nuclear translocation and its interaction with nuclear (TCF/LEF) transcription factors ( Figure 3 ; Tiwari et al., 2015b ). It has been demonstrated that bisphenol-A not only affects the proliferation/differentiation balance of NSPCs but also the maturation of newborn neurons. Bisphenol-A treatment (200 μM for 5–7 days) of primary rat neuron cultures showed an increase of oxidative stress and impairment of neurite outgrowth ( Cho et al., 2018 ). This treatment also enhanced the intensity of DCX, a characteristic immature neuronal marker, indicating an inhibition or delay in the process of neuronal maturation ( Cho et al., 2018 ). Interestingly, thicker and longer neurites appear to be formed with bisphenol-A 100 μM treatment ( Kim et al., 2009 ). Other studies have shown that bisphenol-A blocks steroid induction of the dendritic spines in the hippocampus (CA1) and in the medial prefrontal cortex (mPFC) of rats ( MacLusky et al., 2005 ; Bowman et al., 2015 ) and primates ( Leranth et al., 2008 ). These effects consequently inhibit synaptogenesis ( MacLusky et al., 2005 ; Leranth et al., 2008 ; Bowman et al., 2015 ) and impair spatial memory in object placement tests ( Diaz Weinstein et al., 2013 ). Recent studies suggest that bisphenol-A can affect the migration and maturation of cortical inhibitory GABAergic interneurons. In this context, it has been shown that bisphenol-A can alter the perinatal chloride shift, a critical period during development where the cytosolic concentration of chloride downs from ∼100 to ∼20 mM ( Fiumelli and Woodin, 2007 ). A key element for the chloride shift is the upregulation of the chloride-extruding transporter molecule (KCC2) ( Rivera et al., 1999 ). High neuronal chloride is essential for interneuron precursors migration from ganglionic eminences to the dorsal region of the telencephalon (also known as pallium). Once in the cortex, upregulation of KCC2 is necessary for the reduction of interneuron motility through the drop of neuronal chloride, a process which is also a precondition for the inhibitory effects of the neurotransmitter GABA and glycine in the CNS ( Bortone and Polleux, 2009 ). A recent investigation has shown that treatments of organotypic cultures with bisphenol-A enhanced the repression of the KCC2 transcription, inducing an increment in the average speed and number of migrating neurons leaving the ganglionic eminence ( Djankpa et al., 2019 ). In that sense, bisphenol-A exposure during a sensitive time-point of neurodevelopment could disrupt the cytoarchitecture of the cortex by over-migration of interneurons ( Yeo et al., 2013 ). Additionally, neural repression of the KCC2 could affect synaptic maturation, compromising the function of inhibitory cortical neurons ( Yeo et al., 2013 ).

The beneficial or detrimental effects of estrogens on NSPCs biology and neurogenesis, and their effects on neuronal function and survival represent a challenging research field. The data discussed here summarize some of the roles of endogenous estrogens, estrogen-based therapies, and estrogen-like molecules on brain homeostasis, particularly in neurogenesis, gliogenesis, and regulation of the neuroprotection/neurodegeneration balance. The neuroprotective role of estrogens implies a close regulation of processes such as (i) anti-oxidant activity ( Kumar et al., 2010 ), (ii) DNA repair ( Bethea et al., 2016 ), (iii) synthesis of growth factors ( Scharfman and MacLusky, 2006 ), (iv) synaptic plasticity ( McEwen and Milner, 2007 ), and (v) the blood flow to the brain ( Chambliss et al., 2000 ). These mechanisms are vital for maintaining neuronal survival and counteract the increased risk of cognitive decline and neurodegenerative disorders associated with aging ( Zhao and Brinton, 2006 ; Engler-Chiurazzi et al., 2017 ). In fact, women with premature menopause who do not receive estrogen treatment show an increased risk of developing Alzheimer’s disease ( Shuster et al., 2010 ); while epidemiological reports support that HRT acts as a neuroprotective factor in post-menopausal women ( Song et al., 2020 ). The study of molecular and cellular mechanisms underlying estrogen-mediated neuroprotection represents a field of research with exciting projections since its impact transcends to a series of processes that regulate nervous tissue homeostasis. Therefore, phytoestrogens and their metabolites, which have been shown to replicate the benefits of endogenous estrogens, and estrogen-based therapies, are considered promising approaches for preventing the onset and progression of neurodegenerative disorders. By contrast, xenoestrogens commonly found in polymers of plastic polycarbonates, such as mineral water bottles and lacquer coating in food cans, can act as endocrine disruptors that significantly alter estrogen-mediated signaling pathways ( Brotons et al., 1995 ; Wagner and Oehlmann, 2009 ). This is key since bisphenol-A molecules are joined by ester bonds, which are disrupted by high temperatures or pH chances, and subsequently released on water and food ( Brotons et al., 1995 ; Wagner and Oehlmann, 2009 ), thus tempting to speculate on the consequences of chronic exposure to these molecules, especially in post-menopausal women. The published data show that estrogen-like molecules can mimic or block endogenous estrogen’s biological effects through a series of non-genomic responses ( Park et al., 2012 ; Kumar et al., 2016 ; Jahan et al., 2018 ). Phytoestrogens are mostly beneficial to the processes of proliferation and differentiation of NSPCs; however, some detrimental effects have also been reported. A possible explanation for this dichotomy could be given by diverse factors, such as the volume of consumption, time of exposure, genetic background, and degree of development. It is essential to consider that most in vitro studies have evaluated the effects of phytoestrogens in an isolated manner, leaving aside the fact that a set of molecules induces the effects, and not only a single one. Also, an important part of their activity can be attributed to the generation of diverse microbiome-derived metabolites ( Setchell et al., 2002 ; Branca and Lorenzetti, 2005 ), some of which are not generated by the entire population ( Setchell et al., 2002 ; Setchell and Lydeking-Olsen, 2003 ; Legette et al., 2014 ). So, it is expected that a similar set and quantity of phytoestrogens can induce dichotomic effects depending on the individual’s idiosyncrasy. Besides, the short half-life prevents bioaccumulation and may restrict their activity in a given period ( Yang et al., 2012 ; Legette et al., 2014 ), mimicking endogenous estrogens. Endocrine-disrupting chemicals such as bisphenol-A mostly induce detrimental effects in NSPCs, reducing their self-renewal ability and cell potency. Although the potential human health risks associated with these chemicals represent a significant public health concern, some controversy has been generated by their relatively low biological potency and affinity for classical and non-classical ERs ( Rider et al., 2010 ). In contrast to phytoestrogens, these chemicals tend to bioaccumulate and induce persistent and refractory effects, even though their environmental concentrations oscillate in a relatively low range ( Liu et al., 2012 ). Clinical and preclinical evidence supports the fact that the maternal transfer of bisphenol-A can induce long-term effects in diverse systems, including the CNS. Actually, exposure to bisphenol-A during gestation has been shown to produce glucose intolerance, insulin resistance, and a reduction in the number of pancreatic β-cells ( Alonso-Magdalena et al., 2010 ). In the same direction, it has been reported that bisphenol-A generates apoptosis in some cerebral nuclei and variations in the expression of ERs in others ( Kubo et al., 2001 ; Rebuli et al., 2014 ), leading to consider its potential for inducing changes in the cellularity and/or functionality of diverse areas of the brain or even reduce the sensitivity of others to endogenous estrogen. Both in vitro and in vivo studies reveal that bisphenol-A exposure during postnatal development alters ER phosphorylation and its translocation to the nucleus in the hippocampus ( Xu et al., 2014 ). Additional studies indicate that bisphenol-A affects estrogen signaling and induces severe alterations in the proliferation and differentiation processes of NSPCs ( Tiwari et al., 2016 ). These chemicals can induce modifications in the brain’s function by disrupting molecular and cellular mechanisms associated with neurogenesis, gliogenesis, neuronal survival, and brain repair. The data mentioned above consider several findings in both embryonic and adult NSPCs, as well as OPCs. It is noteworthy that the CNS-residing NSPCs show many biochemical and morphological differences throughout life, but the origin of adult NSPCs -type B cells- has been traced from early neurodevelopmental stages ( Merkle et al., 2004 ). Elegant imaging studies have proven that radial glial cells can divide asymmetrically to generate type B cells that remain quiescent until they are reactivated in adulthood, without affecting the prenatal neurogenesis; a process that highlights the efficiency of the mechanisms of cellular specification ( Merkle et al., 2004 ; Kriegstein and Alvarez-Buylla, 2009 ; Kheirbek, 2015 ). Also, the origin of OPCs is traced early in the formation of the mammalian CNS and comprises multiple and sequential waves of production of OPCs that end in colonization and positioning in the cortex and spinal cord ( Kessaris et al., 2006 ; Kriegstein and Alvarez-Buylla, 2009 ). Exposure to these types of endocrine chemical disruptors during pregnancy could alter the proliferative activity and cell potency in both populations of NSPCs, but their effects could be reflected only later in life. Future research focusing on the importance of these chemicals as factors that predispose to developing diverse neurological conditions or even their impact on ongoing illness is needed.

Estrogens are cholesterol-derived sex hormones that play a fundamental role in many physiological processes ( Kaludjerovic and Ward, 2012 ). The three main forms of estrogens: estrone, estradiol, and estriol, are produced endogenously by different organs and tissues, such as ovaries, testis, adipose tissue, and adrenal cortex. Estrogens are also used as part of pharmacological compounds in contraceptive therapies for pre-menopausal women and hormonal replacement therapies for post-menopausal women. These molecules have a strong affinity for the estrogen receptors (ERs) ERα, ERβ, and G-protein coupled ER1 (GPER1), through which, after binding, induce a direct or indirect genomic signaling to change the expression pattern of several genes ( Fuentes and Silveyra, 2019 ). Additionally, some exogenous estrogen-like molecules, such as phytoestrogens and xenoestrogens, also have ER-binding capabilities ( Nadal et al., 2018 ). By binding to these receptors, estrogens and estrogen-like molecules modulate cell proliferation ( Gompel et al., 2004 ; Pisano et al., 2017 ; Molina et al., 2020 ), differentiation ( Bassler, 1970 ; Grubbs et al., 1985 ; Imanishi et al., 2005 ), and survival ( Weldon et al., 2004 ; Kishi et al., 2005 ; Yu et al., 2012 ) processes, by different downstream pathways. Furthermore, since estrogen receptors are ubiquitously located, estrogen-mediated signals modulate the biology of several tissues and organs, including the central nervous system. Estrogens play a pivotal role in the adult and developing brain ( Brannvall et al., 2002 ; Morissette et al., 2008 ), exerting neuroprotective effects under homeostatic and pathologic conditions. In that sense, estrogens can enhance anti-oxidant activity ( Bellanti et al., 2013 ; Rekkas et al., 2014 ) and induce DNA repair ( Bethea et al., 2016 ), two processes required to modulate the adverse effects of oxidative stress in the cell. Moreover, estrogens increase the expression of different growth factors and anti-apoptotic molecules, promoting cell survival and reducing neurodegenerative processes ( Singh et al., 1995 ; Solum and Handa, 2002 ). They also positively modulate dendritic spines and axonal growth, synaptic transmission, and plasticity, processes required to maintain and enhance cognitive and memory performance ( Wallace et al., 2006 ; Luine and Frankfurt, 2013 ). Neurodegenerative diseases are characterized by increased oxidative stress ( Starkov, 2008 ), cell death, and concomitant loss of synapses; in this context, estrogen supplementation can prevent or reduce these negative features. Several studies have shown the beneficial effects of estrogens in reducing the progression and ameliorating the symptoms of Alzheimer’s and Parkinson’s disease ( Kumar et al., 2010 ; Vail and Roepke, 2019 ). Besides, estrogens are essential in maintaining the balance between proliferation and differentiation of embryonic and adult neural stem/progenitor cells (NSPCs), thus modulating neurogenic processes. Interestingly, classical (ERα and ERβ) and non-classical (GPER-1) estrogen receptors are expressed by cortical and hippocampal NSPCs ( Brannvall et al., 2002 ; Waldron et al., 2010 ; Zhang et al., 2019 ; Zhong et al., 2019 ), and also by oligodendrocyte progenitor cells (OPCs) and post-mitotic oligodendrocytes, playing a central role in oligodendrogenesis and myelination ( Kumar et al., 2010 ; Okada et al., 2010 ). Here, we review the current knowledge on the role of estrogens and estrogen-like molecules in the neuro/gliogenic processes and the regulation of neuronal survival and neurodegeneration.

FB-B was employed by company Cells for Cells (Chile). The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.