Nuclear receptors in neural stem/progenitor cell homeostasis

Abstract

In the central nervous system, embryonic and adult neural stem/progenitor cells (NSCs) generate the enormous variety and huge numbers of neuronal and glial cells that provide structural and functional support in the brain and spinal cord. Over the last decades, nuclear receptors and their natural ligands have emerged as critical regulators of NSC homeostasis during embryonic development and adult life. Furthermore, substantial progress has been achieved towards elucidating the molecular mechanisms of nuclear receptors action in proliferative and differentiation capacities of NSCs. Aberrant expression or function of nuclear receptors in NSCs also contributes to the pathogenesis of various nervous system diseases. Here, we review recent advances in our understanding of the regulatory roles of steroid, non-steroid, and orphan nuclear receptors in NSC fate decisions. These studies establish nuclear receptors as key therapeutic targets in brain diseases.

Introduction

The mammalian central nervous system (CNS) is one of the most complex tissues. Generation of this complexity requires a well-synchronized action of extrinsic morphogenetic cues and intrinsic gene regulatory networks in neural stem/progenitor cell (NSC) fate decisions. NSCs are characterized by the ability to proliferate and generate neurons or macroglial cells, such as oligodendrocytes and astrocytes, during development. However, one of the breakthrough discoveries in neuroscience, a few decades ago, was that NSCs are not only restricted to embryonic development, but they are maintained throughout the whole adult life generating neural cells in many mammalian species including humans [1–4]. Abberant specification choices of NSCs during development or adulthood can result in severe developmental defects, abnormalities, neurological diseases, or cancers. Therefore, elucidation of regulatory factors that control NSC properties could offer significant insights into the treatment of CNS-related diseases and disorders. Towards this end, many lines of research strongly implicate nuclear receptors (NRs) in NSC homeostasis during embryonic and adult life. Most importantly, there is a wide spectrum of pharmacological compounds targeting NRs activity [5–11], raising the possibility of using these molecules in novel therapeutic approaches for neural diseases.

In general, NRs constitute a large superfamily of transcriptional regulators, including 49 members in mouse and 48 in human. Through their ability to regulate gene expression of many downstream effector genes, NRs are involved in various physiological functions, controlling development of several organs, as well as metabolism, reproduction, stem cell pluripotency, tissue, and cell homeostasis. The canonical structure of NRs is consisted of four distinct functional domains: (a) the ligand-independent AF-1 transactivation domain (also known as A/B domain); (b) the DNA-binding domain, a zinc-finger domain required for binding to response elements (DBD or C domain); (c) the hinge region (D domain) which contains the nuclear localization signal (NLS); and (d) the ligand-binding domain (LBD or E domain), a ligand-dependent AF-2 transactivation domain. Regarding their molecular mechanism of action, NRs bind to specific DNA response elements in the regulatory regions of target genes [7, 12]. In most cases, their ability to regulate transcription is initiated by the binding of specific ligands to LBD of the corresponding NRs.

Most NRs are expressed in the mammalian brain and have important roles in brain organogenesis and function. In addition, many of these NRs have been implicated in neurological, neurodegenerative, and psychiatric diseases, including Alzheimer’s disease (AD), Parkinson’s disease (PD), Huntington’s disease, amyotrophic lateral sclerosis (ALS), and schizophrenia [11, 13–16]. Despite their importance in nervous system and the fact that could be utilized as potential pharmacological targets for these diseases, there is only a limited number of articles that review the role of NRs in brain development and adult neurogenesis [17–21]. Here, we thoroughly discuss and present an updated view of the specific roles of steroid, non-steroid, and orphan NRs in NSC biology (Fig. 1; Table 1).

Fig. 1.

Schematic illustration of how specific NRs control NSC homeostasis. NRs affect the ability of NSCs to proliferate and self-renew (a), to differentiate into neurons (b) or glial cells, namely astrocytes (c) or oligodendrocytes (d). NRs that promote these fate decisions are indicated above the green arrows, while NRs that suppress them are below the red repression symbols. e embryonic NSCs, a adult NSCs

Table 1.

Overview of the roles of NRs in proliferation and differentiation properties in embryonic and adult NSCs

Members of nuclear receptors family	Nuclear receptor symbol	Functions		References
		Embryonic NSCs	Adult NSCs
Steroid hormone receptors
Thyroid hormone receptor α (TR-α)	NR1A1	Inhibition of proliferation and astrogliogenesis; induction of neuronal differentiation; regulation of oligodendrogenesis	Promotion of cell survival; induction of neuronal differentiation	[39–46, 49, 51–53]
Thyroid hormone receptor β (TR-β)	NR1A2	Regulation of neuronal differentiation; regulation of oligodendrogenesis	Inhibition of proliferation; regulation of gliogenesis	[38, 43–46, 54]
Vitamin D receptor (VDR)	NR1I1	Suppression of proliferation; induction of neuronal differentiation and maturation; specification of DA neurons	Regulation of proliferation; induction of differentiation into neurons and oligodendrocytes; enhancement of neuronal survival	[63–67, 70, 71]
Estrogen receptor α (ERα)	NR3A1	Regulation of proliferation; promotion of differentiation into DA neurons; induction of oligodendrogenesis	Enhancement of proliferation	[79–82, 85, 87]
Estrogen receptor β (ERβ)	NR3A2	Regulation of proliferation; promotion of neurogenesis and differentiation into DA neurons; induction of oligodendrogenesis; enhancement of neuronal survival	Enhancement of proliferation; stimulation of cerebellar Purkinje cells maturation	[79–83, 85, 87, 91]
Glucocorticoid receptor (GR)	NR3C1	Inhibition of proliferation; repression of neurogenesis; induction of apoptosis	Inhibition of proliferation; repression of neurogenesis; induction of apoptosis; induction of oligodendrogenesis	[103–112, 115–128]
Mineralocorticoid receptor (MR)	NR3C2	Promotion of cell survival; regulation of neurogenesis	Promotion of cell survival; regulation of neurogenesis; induction of astrogliogenesis	[112, 135–140]
Non-steroid hormone receptors
Retinoic acid receptor α (RAR-α)	NR1B1	Induction of gliogenesis; neuronal maturation	Neuronal maturation; regulation of neurogenesis; inhibition of proliferation; induction of gliogenesis	[155, 162, 164]
Retinoic acid receptor β (RAR-β)	NR1B2	Induction of proliferation and neuronal differentiation; neuronal maturation	Induction of proliferation and neuronal differentiation; neuronal maturation	[150–155, 162]
Retinoic acid receptor γ (RAR-γ)	NR1B3	Induction of gliogenesis		[155]
Retinoic X receptor α (RXR-α)	NR2B1	Regulation of neurogenesis	Regulation of neurogenesis	[173–176]
Retinoic X receptor β (RXR-β)	NR2B2	Regulation of neurogenesis	Regulation of neurogenesis	[173–176]
Retinoic X receptor γ (RXR-γ)	NR2B3	Regulation of neurogenesis	Maturation of oligodendrocytes; regulation of neurogenesis	[144, 171–176]
Peroxisome-proliferator-activated receptor α (PPAR-α)	NR1C1	Induction of astrogliogenesis; astrocyte maturation		[189]
Peroxisome-proliferator-activated receptor β/δ (PPAR-β/-δ)	NR1C2	Induction of gliogenesis; maturation of oligodendrocytes	Induction of proliferation; promotion of neuronal differentiation	[189, 190, 196]
Peroxisome-proliferator-activated receptor γ (PPAR-γ)	NR1C3	Induction of proliferation; promotion of cell survival; regulation of neurogenesis	Induction of proliferation; promotion of neuronal differentiation; neuronal maturation	[181–184, 186, 187, 195, 198]
Liver X receptor α (LXR-α)	NR1H3	Inhibition of proliferation; induction of neurogenesis	Promotion of cell survival; induction of neurogenesis	[9, 15, 218–221]
Liver X receptor β (LXR-β)	NR1H2	Inhibition of proliferation; induction of neurogenesis	Promotion of cell survival; induction of neurogenesis	[9, 15, 218–221]
Orphan receptors
TLX	NR2E1	Induction of proliferation; suppression of neuronal differentiation; generation and maturation of astrocytes	Induction of proliferation; promotion of neurogenesis; suppression of gliogenesis	[225–227, 231, 234–241, 243]
Chicken ovalbumin upstream promoter-transcription factor I (COUP-TFI)	NR2F1	Suppression of proliferation; induction of neurogenesis and neuronal maturation; promotion of oligodendrocyte differentiation	Suppression of proliferation; regulation of differentiation (retina); enhancement of neuronal survival	[275–282]
Chicken ovalbumin upstream promoter-transcription factor II (COUP-TFII)	NR2F2	Promotion of proliferation; induction of neuronal differentiation	Suppression of proliferation; regulation of differentiation (retina); enhancement of neuronal survival	[276, 277, 281, 282]
Nuclear receptor-related 1 protein (Nurr1)	NR4A2	Promotion of differentiation into DA neurons; survival of DA neurons; suppression of astrogliogenesis	Promotion of differentiation into DA neurons; survival of DA neurons	[292–298, 304]
Steroidogenic factor 1 (SF-1)	NR5A1	Promotion of cell-cycle exit; induction of neuronal differentiation	Induction of neuronal differentiation	[319, 320]
Liver receptor homolog 1 (LRH-1)	NR5A2	Suppression of proliferation; induction of neuronal differentiation; inhibition of astrogliogenesis		[329–331]
Germ cell nuclear factor (GCNF)	NR6A1	Suppression of proliferation; promotion of neuronal differentiation and maturation		[334, 341–343, 347]

Steroid hormone nuclear receptors

Thyroid hormone receptors (TRs)

Thyroid hormone receptors are ligand-dependent nuclear receptors with two members, TR-α (NR1A1) and TR-β (NR1A2) [22]. The natural ligands of TRs are the thyroid hormones, 3,5,39-triiodo-l-thyronine (T3) and 3,5,39,59-tetraiodo-l-thyronine or thyroxine (T4), which are synthesized in thyroid gland [23, 24]. After activation, TRs bind to specific DNA regions that are called thyroid hormone response elements (TRE) as monomers, homodimers, heterodimers of TR-α with TR-β, or heterodimers with other factors such as Retinoic X Receptors (RXRs) [25–27]. TRs are expressed in almost every tissue [28, 29] and have great implication in metabolism, development, growth and function of many organs and tissues including liver, heart, bones, adipose tissue and CNS [30].

In CNS, TRs are expressed both in fetal and adult brain. In particular, TR-α is the major TR of the brain and is expressed in many areas of embryonic as well as adult CNS including hypothalamus, cerebellum, olfactory bulb, cerebral cortex, striatum and hippocampus. It is expressed in neurons and in some sub-populations of astrocytes and oligodendrocytes [31–33]. The expression of TR-β mainly begins in the post-natal period and is restricted in cerebral cortex, cerebellum, hypothalamus and hippocampus [32, 34, 35]. TRs have a significant role in fetal brain development. The thyroid hormone deficiency during development results in cognitive problems, hyperactivity, anxious behavior and locomotor disorders both in humans and rodents [36]. These defects are the result of an underdeveloped brain. In particular, the absence of thyroid hormones in the developing fetal brain of rats caused low proliferation, reduced cortical thickness and attenuated neurogenesis [37, 38]. Accordingly, the loss of TRs exhibits similar effects with the thyroid hormone deficiency. Loss of TR-α leads to dwarfism, metabolic disorders, underdeveloped brain and lethality in the neonates. In addition, TR-β deficiency affects learning, memory and development [39].

TRs are important regulators of NSC homeostasis (Fig. 1). Specifically, treatment of mouse embryonic telencephalic NSCs from E13.5 with T3, suppressed proliferation and induced neuronal differentiation via negative regulation of STAT3 signaling pathway. Knock-down of TR-α1 isoform counteracted the neurogenic effects of T3 on these NSCs [40]. In agreement, T3 is sufficient to induce dopaminergic (DA) neurogenesis in ex vivo cultured mouse embryonic NSCs (E13.5) from ventral midbrain. TR-α1 activation increases the expression levels of Otx-2, an important transcription factor in DA neurogenesis. Thyroid hormone deficiency leads to reduced DA neurons in ventral midbrain area, causing motor disorders [41]. Moreover, TR-α1 activation promotes retinoic acid (RA)-inducible neuronal differentiation of mouse embryonic stem cells (ESCs), a phenotype which is abolished in ESCs deficient for TR-α1 [42]. Furthermore, TR-α1 is necessary for proper post-natal cerebellum development in rodents. Dominant negative mutation of this receptor reduced proliferation and migration of granule cells and negatively affected the development of Purkinje cells as well as maturation of Bergman glia [43]. Interestingly, TRs have also been implicated in the regulation of gliogenesis. Treatment with T3 of rat embryonic NSCs from E16 induced a gliogenic phenotype [44]. In addition, both TR-α and TR-β are implicated in differentiation of neonatal rat NSCs into oligodendrocyte progenitors and their subsequent maturation into oligodendrocytes through the regulation of aTf (apotransferrin), a key regulator of oligodendrogenesis. On the other hand, TR-α1 antagonist decreased the number of mature oligodendrocytes [45, 46]. Furthermore, a TR-β agonist, GC-1, promoted the maturation of oligodendrocyte precursor cells into mature and functional oligodendrocytes both in mice and human [47].

TRs are also expressed in adult hippocampal NSCs and are correlated with the regulation of adult neurogenesis. In particular, chemically-induced thyroid hormone deficiency in rats results in decreased survival in newborn cells, attenuated neuronal differentiation and delayed neuronal maturation in the hippocampal area [35, 48]. Similarly, expression of mutant TR-α1 with low ligand affinity decreased adult hippocampal neurogenesis and cell survival in mice [49]. In contrast, knock-out of the other member of TRs, TR-β, induced hippocampal NSC proliferation without affecting differentiation [50]. TR-α1 is also expressed in the NSCs of subventricular zone (SVZ) in mice and is implicated in neurogenesis. The lack of TR-α1 mimics the antineurogenic effects of hypothyroidism on NSCs of SVZ [51]. Mechanistically, TR-α1 directly represses the expression of Sox2, a stemness marker of NSCs and induces neuronal differentiation. Conversely, there is an increase in Sox2 expression in the NSCs lacking TR-α1 [52]. Furthermore, TRs exert an important role in adult gliogenesis, both astrogliogenesis and oligendrodrogenesis. Ablation of TR-α1 in mice causes deficient astrocyte maturation in various regions of cerebellum, suggesting that TR-α1 is required to mediate the actions of thyroid hormones in cerebellar astrocyte maturation [53]. In addition, hyperthyroidism in rats leads NSCs to oligodendrogenic fate and proper oligodendrocyte precursor cells maturation compared to normal or hypothyroidic rats [54]. In conclusion, TRs exert their fundamental role in CNS in various developmental stages and cell populations. Based on their implication in proliferation, survival and differentiation of embryonic and adult NSCs, TRs are promising targets for treatment of neuropathological diseases.

Vitamin D receptor (VDR)

The last few years increasing evidence implicate Vitamin D receptor (VDR; NR1I1), a steroid hormone receptor, in NSC homeostasis and brain development. VDR is activated by 1,25-dihydroxyvitamin D₃ (1,25(OH)₂D₃), the biologically most active metabolite of vitamin D₃, leading to its heterodimerization with RXR. Subsequently, the heterodimer binds to the VDR response element (VDRE) on the promoters of target genes and regulates their transcription. This action is facilitated by the recruitment of co-activators and co-repressors in the targeted promoters [55, 56]. Interestingly, VDR is widely distributed in CNS. It is detected in developing brain from E11.5 in mouse and E12 in rats. In adult brain, VDR is localized in several areas including substantia nigra, cortex, hippocampus, hypothalamus, cerebellum, olfactory system and basal ganglia, while it is also expressed in peripheral nervous system. In particular, VDR is found in both neurons and glia in adult rat and human brains [57–60]. Remarkably, vitamin D acting through VDR is important in several processes, such as calcium homeostasis, bone integrity and brain development. Its deficiency has been linked with schizophrenia, Parkinson’s disease, Alzheimer’s disease, autism and depression [60, 61].

During rat brain development, high levels of VDR are detected from E15 to E22, a period that cell proliferation is reduced and apoptosis is increased, suggesting the involvement of VDR in these processes [62]. In agreement, an animal model for developmental vitamin D (DVD) deficiency in rodents displayed increased mitosis and reduced apoptosis both in embryonic and neonatal brain [63, 64]. Consequently, brain morphology of these neonatal rats was also altered, exhibiting larger and longer hemispheres, thinner cortex and increased volume of lateral ventricles [63]. Furthermore, NSCs isolated from SVZ of DVD-deficient neonatal brain exhibited increased proliferating properties as compared to controls [65]. In addition, these DVD-deficient neonates demonstrated decreased levels of the neurotrophic factors, Glial-derived neurotrophic factor (GDNF) and Nerve growth factor (NGF), as well as neurotrophin receptor p75^NTR, all of which are important in growth and survival of developing neurons in the brain, further indicating the neurotrophic and neuroprotective role of vitamin D [63]. Vitamin D₃ via VDR is also involved in differentiation and maturation of neuronal cells during brain development. In particular, in an embryonic hippocampal cell line, vitamin D₃ promotes differentiation through down-regulation of proliferation-associated genes (cyclin D1 and PCNA) and induction of NGF and STAT-3 that stimulate axonal and neurite outgrowth [66]. Moreover, in DVD-deficient embryos Nurr1 and p57Kip2 which are markers for DA phenotype, were significantly reduced, suggesting a crucial role of vitamin D₃ in specification of DA neurons [67].

VDR is also involved in adult neurogenesis (Fig. 1). Adult brain of DVD-deficient rats displays larger lateral ventricles, while reintroduction of vitamin D in their diet at birth is able to ameliorate this effect [68]. These adult rats are characterized by altered behavior, including spontaneous hyperlocomotion, a symptom analogous to schizophrenia in humans, as well as deficits in learning and memory [60]. Significantly, adult hippocampal neurogenesis is impaired in DVD-deficient adult rats and this defect can be reversed by haloperidol, an inverse agonist of dopamine [69]. In addition, 1α-hydroxylase knock-out mice that develop deficiency in 1,25(OH)₂D₃ due to inability to synthesize this active metabolite of vitamin D₃, exhibit increased proliferation of progenitors cells in adult dentate gyrus (DG), increased apoptosis and decreased survival of newly born neurons. In this study, it was also found that the enhanced proliferation in the absence of 1,25(OH)₂D₃ was mediated by increased Ca²⁺ influx via L-VGCC, whereas reduced survival of neurons was due to down-regulation of NGF [70]. Recently, a study in adult murine NSCs showed that these cells express VDR and its levels were increased after 1,25(OH)₂D₃ treatment. Through this action, 1,25(OH)₂D₃ was able to promote proliferation of NSCs and enhance their differentiation into neurons and oligodendrocytes but not astrocytes. In these cells the expression of neurotrophic factors, Neurotrophin-3 (NT-3), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF) and GDNF, was elevated, which may explain why vitamin D₃ is sufficient to regulate NSC homeostasis [71].

Taken together, these data highlight the key role of VDR and vitamin D₃ in developing and adult brain via regulation of proliferation and differentiation of NSCs as well as neurotrophic and neuroprotective actions that sustain the survival of neuronal cells. Furthermore, deficiency in vitamin D experiments have demonstrated its association with neuropsychiatric disorders and neurodegenerative diseases [60, 61], suggesting the necessity for further investigation to fully understand its functions.

Estrogen receptors (ERs)

Estrogen receptors (ERs) are members of NR3A family of nuclear receptors and also of steroid hormone receptors. The nuclear ERs subtypes are ERα (NR3A1) and ERβ (NR3A2), which are also detected in membrane and cytoplasm. In particular, estrogen hormones activate ERs to regulate the transcription of downstream target genes, either directly via their dimerization and binding to estrogen response element (ERE) of DNA or indirectly through interaction with co-factors such as AP-1 and Sp-1. Some of the effects of estrogen via ERs are so rapid that cannot be explained by activation of transcription and protein synthesis. These effects are referred as non genomic and are mediated through membrane associated ERs [72, 73]. ERs are expressed in many tissues including uterus, breast, prostate, cardiovascular system and brain. These NRs are found in the developing as well as adult brain, specifically in pituitary, hypothalamus, hippocampus, cerebral cortex, cerebellum, amygdala and spinal cord. ERα and ERβ exhibit distinct differences in their expression patterns in CNS [74, 75].

ERs have an important role in reproduction, breast cancer and cognition [76–78]. They have also been implicated in the development of CNS. Significantly, Brannvall and co-workers showed that activated ERs by 17β-estradiol (E2) regulate proliferation of embryonic and adult NSCs. Their effect on proliferation of embryonic NSCs depends on the presence of EGF and is mediated via upregulation of cyclin-dependent kinase inhibitor, p21^Cip1. In addition, E2 increases the differentiation of embryonic NSCs to neurons at the expense of glial cell differentiation [79]. Further studies have shown that E2 or bisphenol A (BPA), an estrogenic-like endocrine-disrupting chemical, promote the proliferation and oligodendroglia differentiation of rat embryonic NSCs in the absence of mitogens (EGF, FGF-2) or differentiation factors [80, 81]. Moreover, it was found that E2 acts through ERs in human embryonic NSCs to induce their differentiation into DA neurons [82]. Regarding cortical morphogenesis during brain development, ERβ knock-out mice display thinner cortex at E18.5, due to abnormal migration of neurons and increased apoptosis, suggesting a significant role of ERβ in neuronal differentiation and survival in late corticogenesis [83]. On the contrary, ERα was detected in early stages of corticogenesis, indicating a potential role in brain development [84].

ERs also influence adult hippocampal neurogenesis through their activation by E2 [75]. It has been demonstrated that ERα and ERβ enhance proliferation and possibly affect differentiation of adult NSCs in the DG [85–87]. In addition, ERβ has been found to regulate hippocampal-dependent spatial learning, while ERβ knock-out mice failed to perform hippocampus-dependent spatial tasks. ERβ regulates this process by modulating synaptic plasticity and adult neurogenesis of hippocampus [75, 86–88]. Both ERs, upon activation, exert neuroprotective functions in CA1 neurons against global ischemia in rodents, via their ability to inhibit apoptosis [75, 89]. Furthermore, it is reported that ERs may have a role in cerebellum development. Kim and co-workers have shown that C17.2 cells (immortalized neural progenitor cells from neonatal mouse cerebellum) treated with BPA, displayed a significant decrease in their proliferation, indicating a role of ERs in early neonatal cerebellum formation [90]. In vivo and in vitro studies suggest that ERβ affects differentiation and maturation of cerebellar Purkinje cells through stimulation of dendritic growth and spine [91]. To sum up, nuclear ERs are involved in CNS development and function but further studies are essential for clarifying their specific mechanisms of action.

Glucocorticoid receptor (GR)

Glucocorticoid receptor (NR3C1) is a member of the steroid hormone nuclear receptors which includes the isoforms GRα, GRβ, GRγ, GRA, and GRP which are the result of alternative splicing of the same gene [92, 93]. Its physical ligands are cortisol in humans and corticosterone in rodents [94]. The release of glucocorticoids in blood circulation after their production in adrenal glands is under the control of inflammatory signals and stress-induced hypothalamus–pituitary–adrenals (HPA) axis activation [95]. GR activation involves glucocorticoid (GC) binding, dimerization of the complex and binding to a glucocorticoid response element (GRE) [96]. GR can also bind DNA elements as monomer after protein–protein interactions with other transcription factors. GR also exerts its actions via non genomic mechanisms [97]. GR is globally expressed in all organs, tissues and cell types [98, 99], and has major role in many biological functions such as development, reproduction, metabolism, immune response and cardiovascular functions [100–102].

GR has a significant role in CNS development. High dose of a synthetic GR agonist in pregnancy leads to nervous system-related pathology such as cognitive and motor disorders in humans [103, 104]. GR is expressed from E11.5 in NSCs of dorsal and ventral telencephalon in rodents. In particular, GR is detected in radial glial cells and intermediate progenitor cells [105]. GR activation negatively regulates fetal brain development. The synthetic GR agonist dexamethasone (DEX), represses the migration of cortical post-mitotic neurons in rodents [106]. In embryonic NSCs, GR exerts an antiproliferative role. In particular, DEX or corticosterone was able to suppress proliferation via reduction of cyclin D1 levels in mouse NSCs. This phenotype was abolished by the GR antagonist mifepristone [107]. Furthermore, DEX is able to induce the expression of senescence markers and negative regulators of cell cycle in mouse embryonic NSCs by GR activation [108]. Beyond the regulation of cell cycle, GR activation induces apoptosis in NSCs. Specifically, the use of DEX in rat neonatal hippocampal NSCs shows an apoptotic phenotype which is repressed by pretreatment of the cells with mifepristone [109]. It has also been implicated in the suppression of Gap Junction Intracellular Communication (GJIC), which is pivotal for NSCs’ survival [110]. In mouse NSCs, the GR-dependent apoptotic effect is mediated through the regulation of the Puma pro-apoptotic gene of Bcl-2 protein family [111]. Along these lines, in human NSCs, activation of GR leads to low proliferative activity and suppression of neurogenesis. Particularly, DEX treatment in human embryonic NSCs reduces proliferation and neuronal differentiation by inducing the expression of Dkk1, a direct suppressor of Wnt signaling. Furthermore, in human fetus derived hippocampal NSC cell line, HPC03A/07, GR negatively affects Shh and TGFb-SMAD2/3 signaling pathways, through which it exerts its antiproliferative and antineurogenic actions [112].

Although GR is a negative regulator of NSCs’ survival, there is evidence that some substances promote NSCs’ survival and neurogenesis via GR activation mechanisms. In particular, co-treatment with the antidepressant sertraline and DEX enhanced the proliferative and neurogenic capacities in human-derived multipotent NSC cell line. On the other hand, sertraline had no effect when NSCs were treated with GR antagonist [8]. Furthermore, ginsenoside Rg1, a steroidal saponin, induces neuronal differentiation via GR signal transduction in mouse ESCs, while the blockade of GR with an antagonist did not promote the initial effects of Rg1 [113]. These data indicate that GR function in embryonic NSCs is not totally understood and more mechanistic insights are required towards this direction.

In addition, GR is expressed in adult brain and has a significant role in its function [114, 115]. As a major effector in stress response, GR has a pivotal role in mood, cognition, and behavior [96, 116]. It is significant that chronic administration of corticosterone is characterized by a depression-like phenotype in rodents [117]. GR is also expressed in hippocampal NSCs and its activation suppresses proliferation and neurogenesis [118]. More specifically, adult rodents treated with corticosterone and DEX showed reduced proliferative activity in the DG of hippocampus. A very important observation was that after adrenalectomy or chemical removal of glucocorticoids, there was a significant increase in the number of newborn neurons in the hippocampal area [119–122]. Moreover, there was increased neurogenesis in the hippocampus after the knock-down of GR in mice, indicating that antineurogenic effects of glucocorticoids are mediated by GR activation [123]. Similar to embryonic NSCs, GR promotes cell death in adult NSCs via the induction of another member of pro-apoptotic Bcl-2 protein family, namely, Bax [124]. In addition, DEX induces the expression of p21^Cip1, a critical cell-cycle regulator, in mouse hippocampal NSCs [125]. Another important observation is that GR activation has neurodegenerative effects on spinal cord. Treatment of adult mice with methylprednisolone, a GR agonist, reduced the number of NSCs in the spinal cord, both in vitro and in vivo [126].

Despite the antiproliferative role of GR in adult CNS and NSCs, a correlation between GR activation and oligodendrogenesis has also been reported. GR is expressed in oligodendrocyte precursor cells of many brain regions including hippocampus, somatosensory cortex, basolateral amygdala, corpus callosum and external capsule. The same expression levels were observed in mature oligodendrocytes in these regions [127]. In agreement with these observations, another study revealed a role of GR in oligodendrogenesis. It was observed that acute stress in adult rats increased the number of oligodendrocytes and decreased the number of neurons in DG of hippocampus. The same effect was observed with the administration of corticosterone. Consistently, when these NSCs where cultured and treated with corticosterone, there was an increase in the Olig2 positive cells [128]. Collectively, these experiments render GR one of the most critical regulators of embryonic and adult neurogenesis as well as CNS function.

Mineralocorticoid receptor (MR)

Mineralocorticoid receptor (NR3C2) is a steroid hormone nuclear receptor with Mineralocorticoid aldosterone as its specific natural ligand. The mechanism of MR actions shares similarities with GR actions. MR binds to the same DNA response elements with GR and is also activated by glucocorticoids. In fact, MR has higher affinity for glucocorticoids than GR [129]. Therefore, the conversion of cortisol or corticosterone into inactive steroid cortisone is essential for MR to exert its mineralocorticoid-dependent actions [130]. MR controls many physiological functions [131] and is expressed in many tissues and cell types including kidney, adipose tissue, endothelium, macrophages, and CNS [132].

In CNS, MR shows a restricted expression in the area of hippocampus and amygdala [133]. The concentration of mineralocorticoids (aldosterone) is very low in these regions and in brain generally, and thus, MR activation is instead mediated by glucocorticoids [134]. Activation of MR in these brain regions has a neuroprotective role, in contrast to the neurodegenerative effects of GR activation [135]. This neuroprotective activity directly affects mood, behavior, and cognitive functions [136]. In particular, inactivation of MR in the limbic regions of the mouse brain caused learning and behavioral disorders, probably due to hippocampal dysfunction [137]. In addition, treatment of ex vivo cultured hippocampal neurons with aldosterone was able to counteract the apoptotic effect of DEX-induced GR activation. On the other hand, the use of MR antagonist was able to mimic the apoptotic action of GR activation [138].

The expression of MR in hippocampus raises the possibility of a potential role for this NR in adult NSC homeostasis. Indeed, although MR is poorly expressed in NSCs at proliferative state, it is significantly increased in differentiation state [18]. Furthermore, administration of aldosterone in adrenalectomized mice increased the proliferation of adult NSCs in the dentate gyrus of hippocampus [139]. In addition, genetic deletion of MR, caused a significant impairment of adult neurogenesis and decreased size of granule layer in dentate gyrus of hippocampus, followed by neuropathological phenotypes [140]. Besides the involvement of MR in regulating neurogenesis, recent data indicate an additional correlation with astrogliogenesis. Activation of MR was able to induce an astrogliogenic phenotype in human embryonic NSCs. This is the first indication that associates MR activation with astrogliogenesis [112].

Non-steroid hormone nuclear receptors

Retinoic acid receptors (RARs)

Retinoic acid receptors are ligand-dependent nuclear receptors and their natural ligand is retinoic acid (RA), a metabolite of vitamin A. This NR family is comprised of three members, RAR-α (NR1B1), RAR-β (NR1B2), and RAR-γ (NR1B3). After activation, RARs form heterodimers with RXRs and bind to specific DNA sequences that are called Retinoic Acid Response Elements (RAREs). RA signaling exerts a fundamental role in development, cell growth, and differentiation in many organs and tissues, including CNS. Knock-out mice models for RARs exhibit growth retardation, malformations in many organs, sterility, and early post-natal lethality [141]. RA signaling is correlated with CNS development and function. RARs exert a critical role in anteroposterior and dorsoventral patterning of neural plate and neural tube. RA signaling is involved in the early development of hindbrain and spinal cord. Moreover, RARs have great implication in differentiation of many neural progenitor cell populations [142]. Recent studies also associate RA signaling with blood brain barrier (BBB) formation. In particular, activation of RAR-β induces the expression of specific markers for BBB in endothelial cells that are responsible for its formation [143]. Beyond CNS development, RARs have a significant role in CNS functions. Specifically, RA signaling regulates synaptic plasticity and consequently behavioral and learning abilities [144]. Furthermore, RAR impairments are correlated with anxiety, schizophrenia, locomotor disorders, and neurodegenerative disorders such as Parkinson's and Alzheimer's disease [13, 145, 146]. Along these lines, amyloid beta (Αβ) reduces RA synthesis and specifically represses RAR-α signaling. On the other hand, specific RAR-α agonist ameliorates cognitive deficits in Alzheimer's disease mouse models via the induction of anti-inflammatory signals, amyloid beta (Aβ) clearance, and tau protein dephosphorylation [147].

All members of RARs are expressed in the embryonic CNS and their activation significantly controls proliferation and differentiation of NSCs. RA signaling is necessary for the proper development of embryonic forebrain. In particular, RA deficiency in mice causes reduction in the expression of genes involved in ventral forebrain development, impaired response to Sonic hedgehog (Shh) signal, and attenuated FGF signaling. Consistently, it is observed defective telencephalic vesicle outgrowth and overall thinning of the forebrain neuroepithelium [148]. From mechanistic point of view, RAR activation induces the expression of JMJD3, a member of the jumonji C family of putative histone demethylases, and thus promotes the neuronal fate of forebrain NSCs [149]. In striatal area, ablation of RAR-β reduces proliferation and differentiation of embryonic progenitors via negative regulation of cyclin E2 and proneural marker Ascl1 (Mash1), resulting in abnormal development and psychomotor disorders [150]. In agreement, Rarb ^−/− mice exhibit low proliferative activity and premature neuronal differentiation in striatal NSCs due to repression of fgf3 and induction of Meis1 expression, an important regulator of neurogenesis [151]. Moreover, RA is essential for the maintenance of NSC population in the olfactory bulb of chick and mouse embryo. RA deficiency reduced the expression of neuronal precursor markers such as Ascl1 and Ngn1 [152]. Embryonic spinal cord is another CNS region that RARs exert a pivotal role in differentiation and especially in motor neuronal fate of NSCs [153, 154]. RAR-β activation by specific agonists promotes differentiation of NSCs into motor neuron progenitors via induction of islet-1. Although, RAR-α induces gliogenesis of NSCs, it is also essential for the maintenance of motor neuron progenitors and their maturation into functional motor neurons. Finally, RAR-γ activation triggers NSCs differentiation in astrocytes and oligodendrocytes. Interestingly, treatment with RA mimics the effects of RAR-β activation [155]. The inextricable correlation between RA signaling and neural development is revealed by its action on ESC neuronal differentiation. In particular, treatment of mouse ESCs with RA induces the proneural marker Pax6 and thus leads these cells to a neuronal fate [156]. Recent studies define Protein Arginine Methyl transferase 1 (PRMT1) and 8 (PRMT8) as critical rheostats of RA response. Both PRMT1 and PRMT8 are key regulators of ESC differentiation into neurons [157]. Moreover, RA and Shh agonist promote ESC differentiation into spinal cord motor neurons. RARs negatively regulate histone 3 methylation status and lead to further activation of Hox1–Hox5 genes domain in these cells [158]. Human ESCs are also affected by RA-induced neuronal differentiation. Treatment of human ESCs and iPSCs with RA induces a motor neuron phenotype via repression of GLI3, a negative regulator of Shh signaling pathway [159]. RAR activation is also implicated in the neuronal fate of many pluripotent embryonic carcinoma cell lines [160].

RARs are key players in adult neurogenesis. Specifically, both RAR-α and RAR-β are expressed in the adult forebrain [161]. Activation of RAR-β induces Shh signaling and synergistically promote proliferation and neuronal precursor identity of adult SVZ NSCs. Furthermore, RAR-α, via FGF signaling, promotes these neural precursor into migrating neuroblasts. Interestingly, RAR-α and Shh signaling decrease NSC proliferation and promote a gliogenic phenotype in vitro [162]. RA signaling also controls adult hippocampal neurogenesis. NSCs treated with RA exhibited a neurogenic phenotype, via induction of the negative cell-cycle regulator p21^Cip1 and the neuronal precursor marker NeuroD [163]. Recent studies reveal a fundamental role of RAR-α in neurogenesis. More specifically, RAR-α agonist promoted the neuronal fate of adult hippocampal NSCs, whereas antagonist attenuated RA-induced neuronal differentiation [164]. RA signaling has also neuroprotective action during aging. In particular, RAR agonist Am80 induces proliferation and cell survival in the hippocampus of aged mice. Am80 promotes the expression of ADAM10, which facilitates Notch1 cleavage, resulting in increased levels of Hes5, which is important for the maintenance of adult NSC identity. ADAM10 also inhibits the formation of Αβ via generation of amyloid precursor protein, suggesting a neuroprotective function. Consequently, Am80 administration in these mice ameliorated the age-related cognitive deficits [165]. Furthermore, RAR activation exerts a determinant role in neuronal fate of adult pluripotent cells. RAR-β induces neuronal markers such as Nestin, beta tubulin III, and tau in mesenchymal stem cells (MSCs), which is a bone-marrow-derived cell population [166]. In conclusion, RARs regulate many developmental and functional processes in CNS and are promising tools in regenerative medicine.

Retinoic X receptors (RXRs)

Retinoic X receptors share a significant homology with RARs; however, they have critical differences in structure and ligand binding. Similar to RARs, RXR family includes three members, RXR-α (NR2B1), RXR-β (NR2B2), and RXR-γ (NR2B3). The natural ligands of RXRs are the isoform of all-trans RA, 9-cis RA, docosahexaenoic acid (DHA), oleic acid, and phytanic acid. RXRs exert their actions as heterodimer partners of many NR families. They mainly form heterodimers with RARs and bind to RAREs. These NRs are also able to form heterodimers with TRs, PPARs, LXRs, Nurr1, and Nurr77. They have pronounced implications in development, growth, function, and homeostasis of many organs [167–169]. RXRs are expressed in almost every tissue of developing and adult mammals. In particular, RXR-α and RXR-β exhibit ubiquitous expression, whereas RXR-γ expression is restricted in muscle tissues and brain [170].

One of the major target tissues of RXR action is CNS. RXRs are implicated in the proper development and function of specific neural populations. Mice lacking RXR-γ1 displayed reduced choline acetyltransferase activity in the striatum and altered response to dopamine antagonists [171]. Moreover, genetic ablation of RXR-γ in mice depleted hippocampal long-term depression, which affects synaptic plasticity and exerts an essential role in cognitive functions [144]. RXR-γ is also implicated in maturation of oligodendrocytes. Specifically, knock-down or pharmacological inhibition of this member of RXRs repressed differentiation of precursor cells into adult oligodendrocytes, whereas 9-cis RA increased remyelination in cerebral cells [172]. RXRs exert a neuroprotective role in neurodegenerative diseases. Particularly, bexarotene (Bxt), a specific RXR agonist, attenuates neurodegenerative phenotype in animal models of Amyotrophic Lateral Sclerosis, Parkinson’s and Alzheimer’s diseases [173–175]. Besides the neuroprotective role of RXRs, recent data correlate RXR activation with neurogenesis. Bxt increases proliferation of DG NSCs in a mouse model for Alzheimer’s disease. RXR activation also promotes the expression of neuronal markers in embryonic and adult NSCs, suggesting a positive effect on neurogenesis [176].

Peroxisome-proliferator-activated receptors (PPARs)

Peroxisome-proliferator-activated receptors belong to nuclear receptors superfamily and include three members, PPAR-α (NR1C1), PPAR-β/-δ (NR1C2), and PPAR-γ (NR1C3). PPARs are expressed in many tissues and organs including heart, kidney, pancreas, spleen and CNS. These NRs have been implicated in metabolic, developmental and other functional processes [177]. The mechanism that PPARs exert their actions is initiated via ligand binding. These NRs have various ligands such as fatty acids, arachidonic acid metabolites (eicosanoids) and hypolipidemic drugs [178]. The next step in the activation cascade is the formation of heterodimers with RXRs and binding to peroxisome-proliferator DNA response elements (PPREs) [179].

All three members of PPARs are expressed in the developing CNS. In rodents, PPAR-β and PPAR-γ expression in the CNS starts from E11.5 and E13.5, respectively. PPAR-α is detected at E13.5, but its expression is decreased until E15.5 [180]. The best studied member of PPARs is PPAR-γ, and its activation exerts a protective role in NSCs. PPAR-γ induces growth of mouse embryonic NSCs via EGF-ERK pathway activation. Conversely, genetic ablation of this NR inhibited cell growth in NSCs, a phenotype that was mimicked by PPAR-γ antagonist [181]. In addition, PPAR-γ counteracts the inhibitory effect of advanced glycated ends (AGEs) on NSC growth. It has been shown that AGEs are associated with the pathogenesis of diabetic cognitive deficits and neurodegeneration. Interestingly, under conditions of PPAR-γ knock-down, AGEs were not able to inhibit growth of NSCs, suggesting that PPAR-γ could be a potential pharmacological target for AGE-related neurodegenerative conditions [182, 183]. PPAR-γ is also correlated with the neuronal differentiation of embryonic NSCs. PPAR-γ activation, mediated by 15-deoxy-PGJ2, leads to an excessive neuronal phenotype in mouse embryonic NSCs through the regulation of MAP kinase, JNK [184]. In agreement, the PPAR-γ activation by pioglitazone rescued the antineurogenic effect of Pex11β (peroxisomal membrane elongation factor) knock-down on mouse ESCs [185]. The protective actions of PPAR-γ activation were also observed in human NSCs. In particular, rosiglitazone was able to inhibit the apoptotic phenotype in human NSCs treated with TNFα. This effect was abolished by PPAR-γ antagonist, indicating that this NR is required for NSC survival. More specifically, PPAR-γ activation induces the expression of important regulators of mitochondrial biogenesis and function, such as AMPK, SIRT1, and PGC-1α. Rosiglitazone is also able to abolish the apoptotic phenotype of Αβ in human NSCs by normalization of oxidative stress and mitochondrial function [186, 187]. PPAR-γ is further implicated in the neuronal fate of human ESCs. PPAR-γ inactivation in RA-treated human ESCs resulted in reduced expression levels for neural precursor markers, Sox1 and Pax6 [188]. Although PPAR-γ is the best characterized member of PPARs in NSC homeostasis, the other two members have also been implicated in CNS development. Particularly, PPAR-β and PPAR-α are expressed during early astroglial differentiation, while the expression of PPAR-α remains in the mature astrocytes and might be involved in the regulation of metabolism [189]. In addition, PPAR-β promotes maturation of oligodendrocytes. Agonists for this NR induce the differentiation of oligodendrocytes both in mixed glial cultures and in oligodendrocytes-enriched cultures, through the induction of myelin basic protein and proteolipid protein [190]. Interestingly, activation of PPAR-β also promotes the neuronal differentiation of mouse ESCs, indicating the major role of PPARs in CNS development [191]. Beyond NSCs, PPARs exert a protective role in mature neurons. Specifically, PPAR-γ activation in embryonic hippocampal neurons inhibits the oxidative stress-mediated apoptosis via the induction of anti-apoptotic protein Bcl-2 [192].

In adult CNS, all members of PPARs are expressed in many brain areas, both in neuronal and glial cell populations [193]. PPARs are also expressed in adult NSCs exerting an important role in the regulation of adult neurogenesis [194]. In particular, PPAR-γ activation by pioglitazone increased the size of SVZ in adult rats by inducing proliferation and neuronal differentiation. Similar effects were also reproduced in ex vivo cultured adult NSCs [195]. In addition, activation of PPAR-β/-δ in the DG of adult mice was sufficient to increase the proliferative activity and neuronal differentiation of NSCs. Consequently, it was also able to ameliorate the depressive behavior induced by chronic mild stress [196]. Furthermore, activation of PPAR-β/-δ counteracted the defects of MPTP-induced Parkinsonic phenotype and brain ischemic injury [197]. It was also observed that activated PPAR-γ promoted the survival of neuronal precursors after spinal cord injury [198]. To sum up, these data and the fact that functional impairment of PPARs is correlated with defects in CNS development and function [199–201], render PPARs as important pharmacological targets for the treatment of neurodegenerative diseases.

Liver X receptors (LXRs)

The liver X receptors (LXRs) are ligand-dependent transcription factors and include two members, LXR-α (NR1H3) and LXR-β (NR1H2) which have 78% similarity within their LBDs [202–204]. Their natural ligands are derivatives of cholesterol and are called oxysterols [205, 206]. Their mechanism of action involves the binding of the ligand and the formation of heterodimers with the RXR. The heterodimer binds to specific response elements within promoters of target genes and regulates their expression [207]. The best characterized roles of LXRs are the cholesterol homeostasis [208–211] and the immune response regulation [212–214], despite the fact that LXRs and their natural ligands are detected in many other tissues, including CNS [215–217]. The deletion of both LXRs in mice led to an increase of self-renewal properties of NSCs and consequently a reduction in neurogenesis, resulting in decreased DA neurons at birth. Oxysterol ligands, through the activation of LXRs, increased the number of DA neurons in mouse ESCs, whereas the opposite effect took place in the absence of LXRs. Similarly, oxysterol treatment of human ESCs during DA differentiation increased neurogenesis as well as the number of mature DA neurons and reduced proliferating progenitor cells [218]. In addition, two endogenous midbrain ligands of LXRs, 3α,7α,12α-trihydroxy-5β-cholan-24-oic acid (CA), a cholic acid that activates LXRs, and 24,25-EC, seem to selectively regulate the development of distinct midbrain neuronal populations. CA increased survival and neurogenesis of Brn3a-positive red nucleus neurons, whereas 24,25-EC promoted DA neurogenesis [219]. Another interesting observation is that the administration of GW3965 (LXR agonist) in an adult mouse model for Alzheimer’s disease promoted the survival of NSCs in the subgranular zone (SGZ) of the DG [15]. Furthermore, deletion of LXRs in adult mice showed accumulation of lipids in the brain and loss of adult spinal cord motor neurons and ventral midbrain DA neurons [220]. Lack of Lxrα or Lxrβ genes increased amyloid deposition in the amyloid precursor protein (APP) transgenic mice model of Alzheimer’s disease [221]. These neurodegenerative effects may be the result of destabilization of cholesterol homeostasis in CNS due to the lack of LXRs. Moreover, use of synthetic agonists of LXRs in Alzheimer’s disease mouse model improved the pathophysiology of the disease [9]. These results make LXRs highly important regulators of neurogenesis and neuronal survival.

Orphan nuclear receptors

TLX

TLX is an orphan nuclear receptor, also known as NR2E1 with no identified ligand. This NR is found in both vertebrates and invertebrates and its homologue in Drosophila is tailless (tll) gene [222, 223]. Mouse TLX is detected in the developing brain from E8.5 until E13.5 and also in the developing retina from E11.5 until E17.5 [224–226]. TLX has high expression levels in adult brain and specifically in cortex, SGZ of the DG, and SVZ of lateral ventricle [227, 228]. Significantly, TLX has critical functions in regulating the proliferation and differentiation of NSCs and hence brain and retina development as well as behavioral phenotypes (Fig. 1; Table 1) [19].

In particular, TLX mutant mice have no obvious defects during early development, though adult mice display reduced thickness of their cortical layers, reduced cerebral hemispheres, defects in the eye, olfactory bulb and hippocampal DG [229, 230]. Roy and co-workers have shown that these defects are connected with altered proliferation versus differentiation decisions of neural progenitors during embryonic development, mainly via its key role in controlling timing of these processes in the embryonic cortex [231]. Accordingly, in TLX-null mice, the cyclin-dependent kinase inhibitor p21^Cip1 was upregulated, whereas cyclin D1 was downregulated [232]. These regulatory actions may explain the TLX role in controlling the timing of differentiation. In addition, mouse TLX has a crucial role in retinogenesis. TLX-deficient mice exhibited retinal and optic nerve degeneration resulting in visual impairment [226, 233]. Specifically, TLX regulates the proliferation of retina progenitor cells via PTEN-cyclin D1 pathway and is significant for astrocyte development through its interaction with Shh and BMP signaling pathway which activate Pax2 expression [225, 226, 234].

Apart from TLX functions in embryonic development, TLX is also a pivotal regulator of NSCs in adult neurogenesis. This orphan receptor is essential for the proliferation and self-renewal of adult NSCs, for the repression of glial differentiation, and also for activation of neuronal differentiation [227, 235–238]. In particular, TLX silencing in adult mouse brain displayed reduction of NSC proliferation and defects in spatial learning [239]. In molecular terms, TLX controls NSC proliferation through p53 and via interaction with histone deacetylases (HDACs) which repress p21^Cip1 and Pten expression [237, 240]. In addition, TLX activates Wnt/b-catenin signaling and binds to Oct3/4 upon hypoxia to stimulate the proliferation of adult NSCs [241, 242]. Regarding differentiation of NSCs in adult brain, TLX contributes to the generation of new neurons. Ιt promotes the expression of Ascl1 and DCX proneural genes, resulting in neuronal differentiation of hippocampal NSCs and suppresses BMP-SMAD pathway to regulate the post-natal astrogliogenesis [236, 243]. Interestingly, TLX through promotion of hippocampal neurogenesis enhances learning and memory [244].

Although TLX function in brain development is well characterized, until recently there was little evidence about its regulation. Recent studies have shown that Sox2 binds to TLX and enhances its expression and activity [245, 246]. In addition, IL-1β negatively regulates TLX, causing the decrease of proliferation of hippocampal neural precursor cells and the reduction of neurogenesis [247, 248]. It was also found that miRNAs regulate post-transcriptionally various neurogenic genes, including TLX. MiR-9, miR-137, miR-219, miR-378, let-7b and let-7d were shown to modulate TLX expression in embryonic and adult brain, thus regulating proliferation and differentiation of NSCs [249–253].

Significantly, mutations in mouse TLX caused aggressive behavior, lack of normal maternal instincts, fearfulness, and learning disabilities, suggesting TLX connection with human neurological disorders [229, 239, 254]. In fact, mutations in human TLX are associated with bipolar disorder, microcephaly, and schizophrenia [16, 255]. Furthermore, TLX plays an important role in CNS tumorigenesis. Increased TLX expression was observed in several types of human brain tumors, such as glioblastoma, neuroblastoma, astrocytoma, and ependymomas [256–261]. In particular, TLX was found to induce the genesis and expansion of glioblastoma cells and is correlated with poor prognosis of glioblastoma patients [256, 257, 262, 263]. Furthermore, TLX contributes to the progression of neuroblastoma by increasing the self-renewal and migration of neuroblastoma cells and it is also linked with poor survival [258, 264]. Collectively, these observations render TLX as an ideal candidate target gene for therapeutic purposes.

Chicken ovalbumin upstream promoter-transcription factors (COUP-TFs)

Chicken ovalbumin upstream promoter-transcription factors (COUP-TFs) are orphan nuclear receptors and members of the steroid/thyroid hormone receptor superfamily. The most studied members of this family are COUP-TFI (or NR2F1) and COUP-TFII (or NR2F2) [265]. COUP-TFs were first found to bind to the promoter of ovalbumin and activate its transcription. In general, they activate or repress the transcription of several genes either directly through binding to specific response elements or indirectly via competition or interaction with other receptors or co-repressors [266]. Notably, it was reported to repress the transcription of TRs, RARs, RXRs, and VDRs which are important for brain development, implying a regulatory role of COUP-TFs in this developmental process [267, 268]. In fact, COUP-TFs have been established to be crucial for differentiation of NSCs, neuronal migration, and eye morphogenesis. Moreover, COUP-TF homologues exhibit distinct and overlapping expression patterns in developing nervous system [269]. In adult, COUP-TFI is detected in cerebrum, striatum, olfactory bulb, hypothalamus, and spinal cord and its functions are well characterized. On the other hand, COUP-TFII is expressed in midbrain, hypothalamus, eye, cerebellum, and spinal cord, and although COUP-TFII knock-out mice die in E9.5, its role in development is poorly understood [270–273].

COUP-TFs are significantly involved in neurogenesis. Naka and co-workers showed that COUP-TFs induce differentiation of embryonic NSCs and promote gliogenesis via epigenetic modifications [274]. With gain- and loss-of-function experiments in mouse embryos, it was shown that COUP-TFI represses proliferation and induces neural differentiation of cortical ventricular zone’s progenitors and SVZ’s basal progenitors through repression of MEK/ERK, PI3K/Akt, and Wnt/beta-catenin signaling [275]. Recently, Zhou and co-workers found that both COUP-TFI and COUP-TFII decrease the proliferation of mouse SVZ progenitors through the regulation of Asc11 proneural gene and are responsible for migration of neuroblasts and survival of the SVZ cells [276]. In addition, COUP-TFII was found to regulate cerebellum development. In mouse post-natal cerebellum, it promotes proliferation of granule cell progenitors and protects them from apoptosis through binding and activation of the IGF-1 promoter [277]. Furthermore, COUP-TFI mutant mice display defects in oligodendrocyte differentiation as well as axon myelination. Additional observations indicate that the involvement of COUP-TFI in these processes could be explained by the positive regulation of SCIP/Oct-6/Tst-1 transcription factor [278]. COUP-TFI also contributes significantly to maturation of differentiated neural cells. COUP-TFI-deficient mice die 2 days after birth and display atypical neuronal axonal projection, guidance, and arborization [279]. It is also involved in the development of forebrain via its role in neuronal migration and neurite outgrowth [280]. Importantly, COUP-TFs are critical for mouse eye development. COUP-TF knock-out mice exhibit congenital ocular coloboma and microphthalmia. Apparently, these NRs are required for the proper differentiation of mouse neural progenitor cells in the developing eye through direct regulation of Pax6 and Otx2 genes [281, 282]. Concluding, COUP-TFs expression in specific areas of developing or adult brain contributes to the formation of these areas through regulation of proliferation and cell fate determination of NSCs.

Nuclear receptor-related 1 protein (Nurr1)

The orphan nuclear receptor Nurr1, also known as NR4A2, is a member of nuclear receptor superfamily. The other two members of NR4A family are Nur77 (NR4A1) and Nor1 (NR4A3). They are all encoded by immediate-early genes and their transcription can be induced by physiological cell-specific stimuli such as cAMP, inflammatory signals, hormones, calcium, growth factors, and neurotransmitters [283]. In particular, Nurr1 activates the transcription of specific genes through binding to their regulatory DNA elements. Nurr1 binds to NGFI-B response element (NBRE) as monomer or homodimer or to Nur response element as homodimer and promotes the TH (tyrosine hydroxylase) and DAT (dopamine transporter) expression [284–286]. It has also been found to bind as monomer, homodimer, or heterodimer to RXR [287]. In mice, Nurr1 expression appears at E10.5 in the ventral midbrain. In adult brain, Nurr1 expression is reduced, but is detected in high levels in substantia nigra, cerebrum, cerebellum, olfactory bulb, and hypothalamus [288–290]. Remarkably, Nurr1 has been reported to have a significant role in CNS development and precisely in the differentiation of neural progenitor cells, especially in DA system.

Nurr1 mutant mice die soon after birth and are not able to develop DA neurons in midbrain, while heterozygous mice display decreased levels of dopamine [291]. Significantly, Nurr1 overexpression in rat neural progenitor cells promotes their differentiation and generates DA neurons, whereas mouse and human neural progenitor cells require the co-overexpression of Nurr1 and Fox2 to differentiate into functional DA neurons [292, 293]. Further studies have shown that Nurr1 determines the DA phenotype of neurons by transcriptionally regulating genes responsible for synthesis, packaging and transport of dopamine such as TH, DAT, AADC (aromatic l-amino acid decarboxylase), and VMAT2 (vesicular monoamine transporter-2) [294, 295]. In particular, Nurr1 recruits Pitx3 (pituitary homeobox 3) to reduce its association with co-repressor SMRT, leading to activation of Nurr1 target genes that are pivotal for complete determination of DA neuronal fate [296]. Another study in mouse NSC line has showed that Nurr1 combined with FGF-8 and Shh are able to specify the DA neuronal fate of these cells [297]. Moreover, Nurr1 is essential for axonal growth, mitochondrial function, maintenance, and survival of DA neurons in adult midbrain [298–300]. In mice, Nurr1 deletion in late maturing or adult striatal DA neurons has led to loss of these neurons and neuronal degeneration which associate Nurr1 with neurodegenerative diseases [298]. Specifically, Nurr1 regulates mitochondrial survival through repression of p53 tumor suppressor and reduction of pro-apoptotic Bax gene [301]. Furthermore, this NR regulates the expression of Ret, a receptor of GDNF which protects midbrain DA neurons from apoptosis, cell death and neurotoxicity [302, 303]. Finally, Nurr1 might have a role in the suppression of astrocytic differentiation. Overexpression of Nurr1 in rat embryonic cortical NSCs induced neuronal differentiation and repression of astrogliogenesis. This effect of Nurr1 was accompanied with the upregulation of neurogenic neurotrophins BDNF, GDNF and NT-3 as well as down-regulation of astrogenic factors LIF (leukemia inhibitory factor) and CNTF [304].

In conclusion, Nurr1 nuclear receptor is associated with determination of cell fate and survival of DA neurons and its deletion or mutations lead to loss of DA neurons. For that reason, it has been linked to Parkinson’s disease (PD). A number of human genetic studies have identified mutations of Nurr1 gene that reduce its expression in PD patients [305, 306]. In addition, postmortem expression of Nurr1 in PD patient midbrains was found to be reduced, especially in Lewy body-containing neurons, indicating a protective role of Nurr1 [307, 308]. In addition, Nurr1 when overexpressed in mouse NSCs has a neuroprotective role against neurotoxic stimuli and can generate functional dopamine neurons [292, 309]. Taken together, these data and considering the recently discovered Nurr1-activating chemical compounds, suggest Nurr1 as a very promising therapeutic target for PD [11].

Steroidogenic factor 1 (SF-1)

Steroidogenic factor 1 (SF-1), also known as NR5A1 or adrenal 4-binding protein (Ad4BP), is an orphan NR that is homologous to Drosophila Fushi tarazu factor 1 (Ftz-F1). SF-1 binds to DNA of target genes as a monomer and regulates their transcription. In addition, the transcriptional activity of this NR is modulated by protein–protein interactions or post-translational modifications [310]. SF-1 is considered as an orphan NR, due to the fact that there is no ligand for SF-1 identified in vivo. However, it is found that hydroxysterols and phospholipids can function as ligands of SF-1 [311, 312]. It is expressed in adrenal cortex, hypothalamus and pituitary in developing brain [313, 314]. Interestingly, SF-1 is significant for development and function of endocrine system, as well as of hypothalamus–pituitary–gonad/adrenal axis [315]. Its functions are inferred from experiments with SF-1 knock-out mice which bear defects in adrenal gland, gonads, and pituitary gonadotropes [313, 316, 317]. They also exhibit disorganized ventromedial hypothalamic (VMH) neurons and increased anxiety-like behavior [313, 318]. Specifically, SF-1 regulates terminal differentiation of ventromedial hypothalamic nucleus (VMN) precursors, while SF-1 deletion leads to inhibition of differentiation, altered expression of early and late neuronal markers, and loss of neuronal projections [319]. Moreover, a recent study was focused on the role of SF-1 in the developing neocortex. SF-1 knock-out mouse embryos display significant impairments in neurogenesis. In particular, the deletion of SF-1 causes inhibition of cell-cycle exit of NSCs and increased numbers of apical progenitor cells. SF-1 modulates these processes via direct or indirect regulation of estrogen synthetase Cyp19a1 and ER [320]. To sum up, until now, SF-1 is found to be involved in the development of VMH and neocortex through regulation of proliferation and differentiation of NSCs and further investigation may unravel new functions of this orphan receptor.

NR5A2/LRH-1

NR5A2 is the other member of the NR5A subfamily, also known as LRH-1 (liver receptor homolog 1) with homology to Drosophila Ftz-F1. Similarly to SF-1, NR5A2 binds as monomer to DNA and activates or represses transcription of its target genes [321]. Although it is a constitutively active orphan NR, it was discovered that specific phospholipids can act as ligands of NR5A2 [10, 312]. In addition, NR5A2 has a predominant role in development of several tissues, steroidogenesis, embryogenesis, and neurogenesis. It is expressed during mouse embryogenesis and detected in several areas of adult mouse brain, such as cortex, hippocampus, and hypothalamus [322–324].

In mice, deletion of NR5A2 results in embryonic lethality by E6.5–E7.5, suggesting its significant role in early embryogenesis [325]. Furthermore, NR5A2 is required for the maintenance of pluripotency of ESCs in mouse epiblast through binding and activation of Oct4 [326]. Subsequently, Wagner and co-workers found that Wnt/beta-catenin signaling through the direct regulation of NR5A2 receptor is able to control the expression of pluripotency genes in ESCs [327]. Moreover, it was shown that it can replace Oct4 in the reprogramming of murine somatic cells into iPSCs (induced pluripotent stem cells) [328]. On the other hand, little was known about the role of NR5A2 in NSC homeostasis and CNS development. We have recently unraveled its fundamental role in embryonic neurogenesis. Accordingly, we demonstrated that NR5A2 expression is correlated with the neuronal lineage. It is detected in high levels in differentiated neuronal cells (bIII-Tubulin+ and NeuN+) and is poorly expressed in NSCs (Nestin+, Pax6+) at several stages of developing mouse spinal cord and telencephalon. Gain- and loss-of-function experiments in vitro and in vivo reveal that NR5A2 suppresses proliferation of embryonic NSCs, induces their neuronal differentiation, and inhibits astrogliogenesis. In particular, NR5A2 induces cell-cycle exit via binding and activation of cyclin-dependent kinase inhibitors p16^Ink4a and p15^Ink4b, promotes neuronal differentiation through direct regulation of Prox1, and blocks astrogliogenesis by negatively regulating Notch1 and JAK/STAT signaling pathways [329, 330]. In addition, upstream regulation of NR5A2 expression is mediated by proneural genes Neurog2 and Ascl1 (Mash1) that promote its expression, as well as by Notch1 and JAK/STAT signaling that repress NR5A2 to permit initiation of astrogliogenesis. In agreement with our observations, it was recently shown that NR5A2 together with RAR-γ or their agonists facilitate the capacity of neurogenic factors, Ascl1, Brn2, and Neurog2, to reprogram mouse fibroblasts into functional and mature neurons [331]. Collectively, these data manifest the critical function of NR5A2 in embryonic neural differentiation and raise the question whether it also plays any role in adult brain neurogenesis. The discovery of small molecules as potent ligands and pharmacological agonists of NR5A2 together with its action in NSC homeostasis suggest that NR5A2 could be a promising target gene for NSC-based treatments of CNS-related diseases.

Germ cell nuclear factor (GCNF)

Germ cell nuclear factor (GCNF), also called RTR (retinoid receptor-related testis-associated receptor), or NR6A1 is a member of the nuclear receptor family. GCNF binds to DNA as a homodimer or oligomer and represses transcription of target genes through recruitment of co-regulators, such as N-CoR (nuclear receptor co-repressor) [332–335]. GCNF is expressed in mouse embryos and in the developing nervous system, although in adults is mostly expressed during gametogenesis [336–340]. GCNF has a critical role in embryonic survival and development, since GCNF knock-out mice embryos die around E10.5 [337]. This orphan NR is also important for neuronal differentiation. It represses the expression of the pluripotent genes Oct4 and Nanog in embryonic and adult NSCs and in RA-induced murine ESCs, leading to the induction of neuronal fate [334, 341–344]. Mechanistically, GCNF recruits DNA methylation complexes to the promoter of Oct4 to silence its expression [345, 346]. Moreover, global analysis of gene expression in undifferentiated and differentiated human ESCs with RA revealed that this orphan receptor regulates a significant number of extra pluripotency genes [344]. These experiments may indicate a general role of GCNF in suppressing stemness and induction of differentiation in NSCs via the regulation of pluripotency genes. In addition, another study showed that the expression levels of GCNF are significant for the expression of MAP2 (microtubule-associated protein 2) and Syp (synaptic vesicle protein synaptophysin) proteins and, consequently, have a critical role in the differentiation and maturation of NSCs [347]. Furthermore, ablation of GCNF in mouse embryos causes abnormal forebrain and midbrain development [348]. Finally, a recent study highlighted the negative role of this NR in proliferation of ESCs by showing that GCNF regulates indirect the activation of cyclin D1 expression in ESCs through the inhibition of its suppressor, Mir302a [349]. It would be interesting to investigate whether GCNF regulates also cyclin D1 in NSCs, considering the key function of this cyclin in promoting NSC proliferation and inhibiting neuronal differentiation.

Concluding remarks and future prospects

NRs have fundamental roles in NSC homeostasis. They critically regulate NSCs proliferation and self-renewal properties as well as neural cell fate acquisition. In particular, many NRs control the maintenance of NSCs identity and their differentiation via induction or repression of neurogenesis or gliogenesis. Apart from NRs actions in differentiation of NSCs, they are also involved in cell maturation, leading to functional neural cell subtypes. Interestingly, some members of NR family display distinct and sometimes opposing roles in embryonic and adult NSCs, indicating their differential association with diverse signaling pathways in a context-dependent manner (Fig. 1; Table 1). Considering that NRs activity can be modulated by specific agonists/antagonists and that a large subset of FDA-approved drugs target NRs, elucidation of their tissue- and time-specific molecular mechanisms in NSCs will significantly advance novel therapeutic strategies for the treatment of neuro-developmental and adult CNS-related disorders. Therefore, it is of paramount importance to further investigate the utilization of these compounds in enhancing brain’s regenerative potential in many pathological conditions, including dementia, depression, CNS lesions, and other neurological diseases.

Abbreviations

AADC

Aromatic l-amino acid decarboxylase

Aβ

Amyloid beta

AD

Alzheimer’s disease

AGE

Advanced glycated end

ALS

Amyotrophic lateral sclerosis

APP

Amyloid precursor protein

BBB

Blood-brain barrier

BDNF

Brain-derived neurotrophic factor

BPA

Bisphenol A

Bxt

Bexarotene

CA

3α,7α,12α-Trihydroxy-5β-cholan-24-oic acid

CNS

Central nervous system

CNTF

Ciliary neurotrophic factor

COUP-TF

Chicken ovalbumin upstream promoter-transcription factor

DA

Dopaminergic

DAT

Dopamine transporter

DBD

DNA binding domain

DEX

Dexamethasone

DG

Dentate gyrus

DHA

Docosahexaenoic acid

DVD

Developmental vitamin D

E2

17β-Estradiol

ER

Estrogen receptor

ERE

Estrogen response element

ESCs

Embryonic stem cells

Ftz-F1

Fushi tarazu factor 1

GC

Glucocorticoid

GCNF

Germ cell nuclear factor

GDNF

Glial-derived neurotrophic factor

GJIC

Gap junction intracellular communication

GR

Glucocorticoid receptor

GRE

Glucocorticoid response element

HDAC

Histone deacetylase

HPA

Hypothalamus–pituitary–adrenals

iPSCs

Induced pluripotent stem cells

LBD

Ligand-binding domain

LRH-1

Liver receptor homolog 1

LXR

Liver X receptor

MAP2

Microtubule-associated protein 2

MR

Minerolocorticoid receptor

MSCs

Mesenchymal stem cells

NBRE

NGFI-B response element

N-CoR

Nuclear receptor co-repressor

NGF

Nerve growth factor

NLS

Nuclear localization signal

NR

Nuclear receptor

NSCs

Neural stem/progenitor cells

NT-3

Neurotrophin-3

Nurr1

Nuclear receptor-related 1 protein

PD

Parkinson’s disease

Pex11β

Peroxisomal membrane elongation factor

Pitx3

Pituitary homeobox 3

PPAR

Peroxisome-proliferator-activated receptor

PPRE

Peroxisome-proliferator DNA response element

PRMT1

Protein arginine methyl transferase 1

PRMT8

Protein arginine methyl transferase 8

RA

Retinoic acid

RAR

Retinoic acid receptor

RARE

Retinoic acid response element

RXR

Retinoic X receptor

SF-1

Steroidogenic factor 1

SGZ

Subgranular zone

Shh

Sonic hedgehog

SVZ

Subventricular zone

Syp

Synaptic vesicle protein synaptophysin

T3

3,5,39-Triiodo-l-thyronine

T4

3,5,39,59-Tetraiodo-l-thyronine

TH

Tyrosine hydroxylase

TR

Thyroid hormone receptor

TRE

Thyroid hormone response element

VDR

Vitamin D receptor

VDRE

VDR response element

VMAT2

Vesicular monoamine transporter-2

VMH

Ventromedial hypothalamic

VMN

Ventromedial hypothalamic nucleus