Hypoxia inducible factor-1α-driven signaling networks regulating adult neurogenesis via Notch, Wnt, and MAPK pathways

Abstract

The ability to detect and adapt to changes in oxygen availability is essential for survival, especially in the mammalian brain, where oxygen demand is high and tightly regulated. Hypoxia-inducible factor 1 (HIF-1) is a key transcriptional regulator of oxygen homeostasis that coordinates cellular responses to hypoxic stress. In the brain, HIF-1 is widely expressed and plays particularly important roles in neurogenic niches, where subtle variations in oxygen tension decisively influence cell fate. This narrative review systematically integrates and organizes current knowledge on the regulation of HIF-1 signaling in the brain under hypoxic conditions, with a focus on its role in mammalian neurogenesis. We highlight the temporal dynamics of HIF-1 activation and its biphasic actions following hypoxic or ischemic injury. Early activation of HIF-1 can exacerbate oxidative stress, inflammation, mitochondrial dysfunction, and neuronal apoptosis, while late activation promotes antioxidant defenses, angiogenesis, tissue repair, and neurogenesis. HIF-1α regulates the proliferation, maintenance, and differentiation of neural stem and progenitor cells through interconnected signaling pathways, such as the Wnt/β-catenin, Notch, and MAPK cascades, as well as by inducing erythropoietin and VEGF. This review provides a comprehensive framework for understanding the critical role of HIF-1 in mammalian neurogenesis and highlights future avenues for mechanistic and translational research.

Introduction

The ability to detect and respond to changes in oxygen availability is fundamental for survival, as oxygen is essential for energy production and cellular function. Consequently, organisms have evolved oxygen-sensing mechanisms that allow cells and tissues to adapt to hypoxia or hyperoxia, thereby preserving viability and maintaining physiological homeostasis [1]. Hypoxia-inducible factor 1 (HIF-1) is a heterodimeric transcription factor composed of the HIF-1α and HIF-1β subunits that regulates a wide range of embryonic and physiological processes, including glycolysis, angiogenesis, and erythropoiesis [2]. HIF-1β is constitutively expressed, while HIF-1α is regulated in an oxygen-dependent manner. Under normoxic conditions, HIF-1α is hydroxylated at two proline residues (Pro402 and Pro564 in the human protein) by prolyl hydroxylase domain protein 2 (PHD2), targeting it for polyubiquitination and subsequent degradation by the 26 S proteasome (Fig. 1). However, low oxygen levels inhibit the hydroxylation of HIF-1α, which is then translocated to the nucleus, where it dimerizes with HIF-1β to form the active HIF-1 complex [3]. Subsequently, the HIF-1 factor binds to cis-acting hypoxia response elements (HREs) in the promoters of target genes and induces the transcription of genes primarily involved in cellular adaptation to hypoxia [4]. In addition to low oxygen tension, HIF-1α expression and stabilization can also be induced by hormones, bacterial infections, and iron chelators, reflecting its interactions in broader stress response networks [5–7]. Also, it is important to note that HIF-1α signaling in neurons is regulated by the highly specialized neurogenic niche, where endothelial cells, astrocytes, microglia, and pericytes provide trophic factors and molecular signals that influence the activation and modulation of HIF-1α-dependent pathways and neuronal response [8]. Additional isoforms of the HIF-1α subunit, known as HIF-2α and HIF-3α, have been identified that share structural and biochemical similarities with HIF-1α and can form dimers with HIF-1β, although those are encoded by distinct genes [9]. It is well known that HIF-1α and HIF-1β mRNAs are ubiquitously expressed in mice and human tissues, including the adult brain, particularly within neurogenic niches where oxygen tension is tightly regulated and even small fluctuations can exert significant biological effects [10, 11]. Notably, HIF-2α mRNA is more highly expressed than HIF-1α in the adult mouse brain, including the neurogenic regions, with predominant expression in endothelial cells [12].

Fig. 1.

Oxygen-dependent regulation of HIF-1α. Under normoxic conditions, HIF-1α is hydroxylated at specific proline residues within its oxygen-dependent degradation domain by prolyl hydroxylase domain (PHD) proteins. This hydroxylation allows recognition by the von Hippel-Lindau (VHL) ubiquitin ligase E3 complex, leading to ubiquitination (Ub) and subsequent proteasomal degradation of HIF-1α. Conversely, under hypoxic conditions, prolyl hydroxylation is inhibited, allowing HIF-1α to stabilize and translocate to the nucleus, where it dimerizes with HIF-1β. The HIF-1 heterodimer then binds to hypoxia-sensitive elements (HREs) in the promoters of target genes and functions as a transcription factor to regulate hypoxia-induced gene expression

Numerous studies have documented the involvement of HIF-1 during neuroprotection, neurogenesis, and neurodegeneration in the mammalian brain [13]. HIF-1α-mediated neurogenesis involves multiple stages of central nervous system (CNS) biology, including proliferation, self-renewal, cell fate specification, maturation, and differentiation. In this sense, this narrative review offers a systematic integration and organization of current knowledge on the mechanisms that regulate HIF-1 in the brain under hypoxic conditions. Particular emphasis is placed on the signaling cascades activated by HIF-1α that contribute to the regulation of neurogenesis in adults. Furthermore, the temporal dynamics of HIF-1 activation are examined, highlighting how its effects can be detrimental or protective depending on the time elapsed since the hypoxic injury. Finally, potential therapeutic targets are discussed, and future lines of research are proposed.

Hypoxia induces increased expression of HIF-1α in the brain

HIF-1α expression is induced in all cell types under stress conditions such as hypoxia, although to varying degrees that probably mediate different molecular, biochemical, and cellular responses depending on the cell type involved [7, 14]. In response to hypoxia, several epigenetic changes occur, including histone acetylation, histone methylation remodeling, and chromatin reorganization, which together contribute to the transcriptional activity of HIF-1α [15]. In the normal mammalian brain, regional differences in oxygen availability correlate with baseline HIF-1α levels. Relatively low oxygen tensions have been reported in the cortex, striatum, thalamus, and CA1 and CA3 regions of the hippocampus, while comparatively higher levels are found in the dentate gyrus (DG), which contains the subgranular zone (SGZ), and in the subventricular zone (SVZ), the main neurogenic niches of the adult brain. However, oxygen tension in these regions remains lower than in most peripheral tissues under normoxic conditions [11, 16].

Following hypoxia, HIF-1α expression increases throughout the brain, with particularly pronounced elevations in the neocortex, DG, and SVZ, indicating region-specific regulation [11, 17, 18]. Notably, HIF-2α protein is ubiquitously expressed throughout the adult mouse brain under normoxic conditions and its protein levels remain largely unaltered following hypoxia insult in neuron-like cells, suggesting a differential regulation and functional role compared to HIF-1α [12, 19]. It is well known that adult neurogenesis capacity of both the SGZ and SVZ decreases with age, but the analysis of HIF-1α accumulation in the adult compared to young rat brain after 1 h ischemia exposure showed no significant differences. However, to address the direct effect of HIF-1α on age-dependent decreased neurogenesis more studies are required [20–22]. Immunohistochemical analyses show that hypoxia-induced upregulation of HIF-1α protein is prominent in neurons, as demonstrated by colocalization with neuronal nucleus antigen (NeuN, also known as RBFOX3), although expression is also detected in astrocytes, microglia, and pericytes [17, 23, 24]. NeuN is a widely used marker for identifying mature postmitotic neurons. Overall, these findings indicate that hypoxia induces space- and cell-type-dependent regulation of HIF-1α in the brain, suggesting that its functional outcomes are likely determined by both regional context and cell composition.

Dual and time-dependent actions of HIF-1α in the hypoxic adult brain

Hypoxia in the brain triggers oxidative stress, inflammation, mitochondrial dysfunction, and neuronal apoptosis, which ultimately contribute to stroke [13]. Paradoxically, hypoxia can also activate adaptive and protective responses against oxidative stress, excitotoxicity, and ischemic injury [25–27]. Thus, HIF-1α is a central regulator of both harmful and protective pathways, and its effects appear to be highly time-dependent.

Protein quantification showed an upregulation of HIF-1α at 4 and 8 h post-hypoxia, followed by a decrease at 16 and 24 h, and a subsequent increase at 48 and 72 h in the brain of newborn pigs [18, 28]. Similarly, oxygen and glucose deprivation (OGD) in cortical neuron cultures extracted from rat pups (0–1 days old) induced an upregulation of HIF-1α protein expression from 1 h to 12 h, a subsequent decrease between 16 h and 24 h, and a second increase in protein levels from 48 h to 96 h [17]. These data indicate that hypoxia in the mammalian brain induces an early increase in HIF-1α protein levels between the first hours, followed by a second rise 2 to 4 days later, corresponding to early and late phases, respectively; however, this temporal dynamic appeared to be context-dependent and requires further investigation. Interestingly, analysis of HIF-1α mRNA expression revealed upregulation only in the late phase, but not in the first few hours after middle cerebral artery occlusion (MCAO). These data suggest that the increase in HIF-1α is primarily due to greater protein stability, while the late phase reflects transcriptional induction of HIF-1α. Similarly, other studies have also shown that hypoxia-ischemia upregulates HIF-1α protein in the acute phase, while the late-phase response is characterized by increased HIF-1α mRNA expression [29, 30].

In adult rats, early administration of 2-methoxyestradiol (2ME2), a HIF-1α inhibitor, at 0.5 h after MCAO-induced hypoxia significantly reduced cell apoptosis and infarct volume, and increased the number of NeuN⁺ neurons in the brain, compared to animals subjected to MCAO alone. Conversely, later administration of 2ME2, 8 h after MCAO, induced greater cell apoptosis and reduced the number of NeuN⁺ neurons. Consistent with this finding, administration of 2ME2 at 0.5 h after MCAO improved motor function impairment and neurological damage, and decreased animal mortality, while administration at 8 h exacerbated neurological dysfunction and decreased animal survival [17]. Furthermore, in neonatal piglets, hypoxia induced marked neuronal loss during the early post-injury period, but caused significant recovery of neuronal populations within 48 h in the cortex and hippocampus [18]. In vitro induction of hypoxia in mesenchymal stem cells by cobalt (II) chloride (CoCl₂) decreased the number of BrdU⁺ (proliferating) neurons in the early phase, but induced an increase in the late phase [31]. Taken together, these observations illustrate that acute-phase upregulation of HIF-1α induces neuronal loss, behavioral deficits, and animal death, while late-phase upregulation of HIF-1α mediates neuronal protection, improves motor outcomes, and increases survival. MCAO-induced hypoxia also altered the integrity of the blood-brain barrier (BBB), as evidenced by increased brain water content and increased IgG permeability in the cortex of adult rats. Notably, early administration of 2ME2 attenuated BBB permeability, while late-phase treatment failed to counteract this protective effect [17]. In addition, pericyte-specific HIF-1α mutation-bearing mice subjected to MCAO showed a significant reduction in infarct volume, cerebral edema, and neuronal apoptosis, along with a decrease in IgG and Evans blue dye extravasation 3 days after MCAO (attenuation of BBB permeability), compared to wild-type MCAO animals [32]. Since vascular endothelial growth factor (VEGF) mediates angiogenesis, which increases vascular permeability, hypoxia-induced overexpression of HIF-1α in pericytes can boost VEGF expression, thereby contributing to disruption of the BBB and exacerbating ischemic brain injury [33, 34].

At the molecular level, hypoxia rapidly increased levels of NIP3-like protein X (NIX), which induces apoptosis in the brains of adult rats [35], returning to baseline levels 3 days later. In contrast, VEGF protein increased 24 h after hypoxic injury and remained elevated for at least 3 days. Early-phase administration of 2ME2 or HIF-1α-specific siRNA significantly reduced NIX expression and attenuated neuronal apoptosis; however, these effects were not observed when treatments were administered during the late phase. Meanwhile, both early- and late-phase administration of 2ME2 or siRNA significantly suppressed VEGF overexpression. Furthermore, the evaluation of the pro-death and pro-survival effects of HIF-1α revealed that NIX silencing by shRNA decreased neuronal apoptosis, while shRNA-mediated VEGF reduction increased apoptosis after OGD in in vitro cortical neuronal cultures [17]. Based on these observations, it is possible that HIF-1 contributes, at least in part, to early-phase neuronal apoptosis through NIX-dependent mechanisms, while late-phase HIF-1 signaling promotes neuroprotection through VEGF-mediated pathways. Overall, these findings indicate that HIF-1α could exert a dual, time-dependent role in the ischemic brain. This biphasic response underscores the importance of the temporal regulation of HIF-1α signaling in determining whether hypoxia triggers damage or facilitates adaptive repair mechanisms in the mammalian brain.

HIF-1α as a central regulator of adult neurogenesis: from stem cell maintenance to neuronal differentiation

Previous studies have shown that HIF-1α is continuously expressed and constitutively stabilized in mouse neural stem cell progenitor cells (NSPCs), including those located in the SGZ and SVZ [36, 37]. Consistently, exposure of isolated rat embryonic neural stem cells (NSCs) to hypoxic conditions increased HIF-1α protein levels from 3 h to 3 d and significantly improved their proliferative capacity compared to normoxic NSCs. Hypoxic NSC cultures also showed increased tyrosine hydroxylase (TH) expression and dopamine (DA) production, whereas NSCs infected with an adenoviral vector overexpressing HIF-1α showed even greater upregulation of TH expression and DA levels [38]. In addition, hypoxic rat embryonic NSC cultures showed increased BrdU incorporation and a higher proportion of Ki67⁺ cells (another marker of cell proliferation) and nestin⁺ cells (a marker of neural stem/progenitor cells), highlighting the role of HIF-1α in NSC proliferation and maintenance (Fig. 2A) [39]. Notably, although these findings were obtained from in vitro experimental systems, they collectively suggest a potential role for HIF-1α in the regulation of NSC maintenance, proliferation and differentiation under hypoxic conditions. To further support this function, bone marrow nestin⁺ mesenchymal stem cells (MSCs) isolated from adult male mice exhibited neuronal differentiation when exposed to chemically induced hypoxia with 100 µM CoCl₂, characterized by elongated, neurite-like bipolar processes. At the molecular level, hypoxia induction led to a progressive decrease in nestin mRNA and protein expression between 6 and 24 h, while Tuj1 (βIII-tubulin) mRNA and protein levels increased at 24 and 48 h, respectively. This finding indicates that the progenitor cells are undergoing differentiation toward a neuronal lineage. Hypoxia also increased the expression of the antiproliferative genes p21 and polycomb 3 homolog (PC3)/BTG2/TIS21 from 6 h to 24 h, while simultaneously decreasing cyclin D1 expression from 6 h to 3 d. Because cyclin D1 regulates cell cycle progression from G1 to S phase, these expression patterns suggest that hypoxia initially promotes proliferation, followed by a transition to cell cycle exit and neuronal differentiation. Treatment of adult mouse MSC cultures with CoCl₂ increased nuclear HIF-1α protein levels from 3 h to 3 d, which was accompanied by an increase in the number of MAP2⁺ neurons exhibiting functional activity, evidenced by calcium (Ca²⁺) influx in response to DA and glutamate [31]. Inhibition of HIF-1 binding to DNA by echinomycin prevented the morphological changes associated with neuronal differentiation, indicating that HIF-1 plays a key role in the differentiation of MSCs into neuron-like cells. Interestingly, neuronal differentiation was observed only when MSCs were cultured at a density of 2.5 to 3 × 10⁴ cells/cm², highlighting the dependence of cell density on HIF-1 mediation of neuronal differentiation (Fig. 2B) [31]. Moreover, in human striatal progenitor cell cultures, comprising a mixed population of neural stem cells, neuron/glia-restricted progenitors, and striatal neurons, treatment with CoCl₂ (400 µM) increased HIF-1α expression from 24 h to 72 h and raised the proportion of nestin⁺ cells at 48 h. However, CoCl₂ did not alter the expression of βIII-tubulin, NeuN, DARPP-32, or GFAP, suggesting that hypoxia improved the maintenance of precursors without promoting neuronal or glial lineage specification in this system [40]. Meanwhile, in human mesencephalic neuronal progenitor cells (NPCs), exposure to CoCl₂ increases the proportion of sub-G1 cells (from 5.1% to 9.6%), inhibits entry into S phase, and induces ROS-mediated damage in proliferating NPCs, while supporting dopaminergic differentiation of already compromised progenitors [41]. Taken together, these findings suggest that HIF-1α promotes the proliferation, maintenance, and differentiation of stem and progenitor cells. However, these effects are highly dependent on cell density, developmental stage, and the severity of hypoxia. In this context, the microenvironment plays a crucial role in determining whether HIF-1α activation favors precursor expansion, neuronal differentiation, or cell cycle arrest, thereby orchestrating lineage decision and neurogenic potential.

Fig. 2.

Hypoxia-induced HIF-1α signaling drives neurogenesis in neural stem cells. (A) Hypoxia increased HIF-1α protein levels, accompanied with enhanced BrdU incorporation and a higher proportion of Ki67⁺ and nestin⁺ cells, indicating elevated proliferative activity in neural stem cells (NSCs). Hypoxia also upregulated tyrosine hydroxylase (TH) expression and increased dopamine (DA) levels in NSC, consistent with the induction of neuronal differentiation. (B) In nestin⁺ mesenchymal stem cells (MSCs), hypoxia induced the formation of elongated bipolar neurite-like processes and increased both mRNA and protein levels of βIII-tubulin (Tuj1), along with the emergence of MAP2⁺ neuronal cells. Functionally, these cells exhibited calcium (Ca²⁺) influx in response to dopamine stimulation. Additionally, hypoxia modulated cell cycle-related gene expression, increasing p21 levels while decreasing Cyclin D1, a profile consistent with cell-cycle exit and progression toward neuronal differentiation. Upward arrows indicate an increase, whereas downward arrows indicate a decrease. Created in BioRender. Cornelio, S. (2026) https://BioRender.com/41ds7gu

Another aspect to consider is that HIF-1α promotes a metabolic reprogramming characterized by enhanced glycolysis and reduced mitochondrial oxidative phosphorylation, resulting in altered intracellular levels of metabolites that function as essential cofactors for chromatin remodeling and transcriptional regulation [4]. In neural stem and progenitor cells, these hypoxia-induced metabolic changes may directly influence epigenetic processes, including histone acetylation, DNA methylation, and chromatin accessibility, thereby modulating stem cell maintenance, proliferation, and lineage commitment [16].

Regarding HIF-2α, a study using NSCs isolated from wild type and HIF-2α knockout mice has shown that HIF-2α silencing impairs cell migration. Also, under OGD, neuronal death was increased in wild type but not significantly affected in HIF-2α knockout cells, suggesting HIF-2α mediates different cellular processes than HIF-1α in NSCs. Notably, HIF-1α protein upregulation was higher in HIF-2α knockout cells compared with wild type, indicating that HIF isoforms compensate each other [42]. In parallel, comparative analysis of neuronal-like PC12 and NT2 cells and the epithelial cancer cell line MCF7 demonstrated basal levels of HIF-2α protein across all cell types. Hypoxia significantly upregulated the mRNA expression of HIF-2α in PC12 and NT2 cells, whereas it was reduced in MCF7 cells; however, protein levels increased after 4 h of hypoxia exposure in all cell lines. In contrast, HIF-3α mRNA levels remained relatively stable in PC12 and NT2 cells, showing minimal response to hypoxia. At the protein level, HIF-3α increased after 4 h of hypoxia in PC12 cells, decreased in NT2 cells, and was not detected in MCF7 cells [19]. These findings indicate isoform-specific and cell type dependent regulation of HIF-α subunits, indicating that their roles in adult neurogenesis are likely distinct but remain incompletely understood.

HIF-1α plays a fundamental role during neurogenesis in the hippocampus

In adult rats, hypoxia significantly increased HIF-1α expression along with increased BrdU incorporation in the DG of the hippocampus, indicating increased proliferative activity [11]. Similarly, traumatic brain injury (TBI) induced marked upregulation of HIF-1α expression 8 h post-injury, which correlated with a higher number of newly generated cells labeled with BrdU and DCX in the hippocampus [43]. Consistent with these findings, genetic inactivation of HIF-1α in the SGZ of adult mice resulted in a significant reduction in newly generated NeuN⁺ cells in the DG 45 days post-inactivation, and impaired performance in spatial and contextual pattern discrimination task [44]. Furthermore, conditional HIF-1α deletion in NSCs, and the application of tamoxifen, which inactivates HIF-1α produced by other cell types, resulted in a 50% reduction in the number of NeuN⁺ cells in the DG and negatively affected contextual fear discrimination learning, an impairment strongly associated with reduced neurogenesis in the DG of adult rats. Taken together, these findings indicate that HIF-1α plays a critical role in regulating adult DG neurogenesis and associated cognitive functions [44]. Furthermore, administration of FG4592, a PHD inhibitor, to adult rats subjected to stress injury significantly increased hippocampal expression of HIF-1α and its downstream target genes EPO and VEGF at both the mRNA and protein levels. The FG4592-induced upregulation of HIF-1α was associated with increased dendritic spine density and enhanced cell proliferation. Additionally, stress injury induced a reduction in key synaptic structural proteins, including PSD95 and Homer1, an effect reversed by FG4592 treatment, which was accompanied by increased expression of BDNF and phosphorylated CREB (p-CREB). Consistent with these in vivo observations, cultured hippocampal neurons isolated from neonatal rats treated with FG4592 exhibited a significant increase in dendritic spine density, along with elevated expression of HIF-1α, EPO, VEGF, BDNF, and PSD95 [45]. These findings further support the role of HIF-1α activity in promoting neuritogenesis through the activation of downstream target genes associated with neuronal plasticity and survival.

Mechanistically, both EPO and VEGF, transcriptional target genes of HIF-1, contribute to neurite extension. EPO has previously been reported to mediate neuritogenesis and neurite growth in vitro, an effect that appears to involve activation of the JAK2/STAT3 signaling pathway [46]. Similarly, in vitro studies have reported that VEGF mediates neurite growth through activation of Rho/ROK signaling pathways [47, 48]. Although BDNF is not a direct transcriptional target of HIF-1, evidence indicates that in neuronal cells, VEGF and EPO can activate CREB, which in turn induces BDNF [49–51]. Notably, BDNF is a key regulator of neurogenesis and neuritogenesis in the DG of the hippocampus [52]. Taken together, in vivo evidence suggests that HIF-1α plays a critical role in regulating adult hippocampal neurogenesis and neuroplasticity, whereas complementary in vitro studies suggest that these effects may involve EPO, VEGF, and the CREB–BDNF signaling axis.

HIF-1α plays a critical role during neurogenesis in the SVZ

Consistent with observations made in the SGZ, hypoxia increased HIF-1α expression and enhanced BrdU uptake by NSCs in the SVZ of adult rats [11]. In contrast, adult rats subjected to MCAO and treated with edaravone, a HIF-1α inhibitor, showed a decrease in HIF-1α and VEGF expression, along with fewer DCX⁺ and BrdU⁺ cells, compared to hypoxic animals given saline solution [53]. Additional in vivo evidence further implicates peroxynitrite (ONOO⁻) signaling in this process, as rats subjected to 1 h of MCAO followed by 7 days of reperfusion showed upregulation of 3-NT, a marker of ONOO⁻ formation, along with elevated nestin expression and increased BrdU incorporation in the SVZ. These findings strongly support a central role for HIF-1 in regulating adult neurogenesis in the SVZ and suggest that ONOO⁻ production contributes to this process under hypoxic conditions. Consistently, analysis of cultured rat embryo-derived neural stem cells (NSCs) revealed that hypoxia induced increased ONOO⁻ production, which in turn enhanced BrdU incorporation and a higher proportion of Ki67⁺ and nestin⁺ cells. These effects were abolished by treatment with the ONOO⁻ decomposition catalysts FeTMPyP and FeTPPS. Hypoxic exposure of the NSCs also increased both the number and diameter of the neurospheres, as well as the number of BrdU⁺/nestin⁺ and Ki67⁺ cells; however, these effects were also counteracted by the application of the ONOO⁻ decomposition catalyst [39]. Furthermore, exogenous application of ONOO⁻ to cultured rat embryonic NSCs increased BrdU incorporation and Ki67 immunoreactivity, as well as the number and diameter of neurospheres. Remarkably low concentrations of ONOO⁻ (20 nM) were sufficient to induce these effects, with no significant differences between 20 nM and 200 nM. Treatment with the ONOO⁻ donor SIN1 (10 µM) similarly increased the number of Tuj-1⁺ and MAP2⁺ cells, as well as Tuj-1⁺/DCX⁺ cells, indicating enhanced proliferation and differentiation. Importantly, exogenous ONOO⁻ also increased HIF-1α expression, while pharmacological inhibition of HIF-1α attenuated ONOO⁻-induced neural stem cell proliferation, suggesting that ONOO⁻ acts at least partially through HIF-1α; however, these effects are highly dependent on cellular context and redox balance [39]. Collectively, these findings indicate that HIF-1α is a central regulator of neurogenesis in the SVZ under hypoxic conditions and suggest that ONOO⁻ signaling contributes to this process. Complementary in vitro studies further indicate that ONOO−modulates proliferation and neuronal differentiation, at least partially through HIF-1α-dependent mechanisms, although these effects appear to be strongly influenced by cellular context and redox balance. However, further studies are needed to determine whether ONOO⁻ similarly influences neurogenesis in the hippocampal SGZ.

Signaling pathways activated by HIF-1 during adult neurogenesis

Wnt/β-catenin pathway regulates HIF-1α-mediated neurogenesis

The Wnt/β-catenin signaling pathway is an evolutionarily conserved pathway in the animal kingdom, from cnidarians to mammals, that regulates numerous biological processes during development, such as cell fate determination, establishment of the body axis, and organ formation. Furthermore, in adulthood, this pathway continues to play important roles such as tissue homeostasis, cell renewal, and regeneration [54]. The Wnt/β-catenin signaling pathway consists of Wnt ligands released by a secretory cell, their affinity receptors on the cytoplasmic membrane of the receiving cell, and signal transducers within the receiving cell [55]. Wnt ligand binds to the coreceptors known as Frizzled (FZD) family receptor and a member of the low-density lipoprotein receptor–related protein family, LRP5 or LRP6. The central effector of this pathway, β-catenin is phosphorylated by GSK3β which forms a destruction complex with casein kinase 1 alpha (CK1α), axis inhibition protein (AXIN), and adenomatous polyposis coli (APC). Thereafter, β-catenin phosphorylated at the N-terminal serine and threonine residues triggers its ubiquitination and degradation by the proteasome [55]. The binding of Wnt to LRP5/6 and FZD mediates the formation of a heteroreceptor and induces phosphorylation of LRP5/6 monomers, which then recruit the Disheveled (DVL) signal transduction complex to the membrane and subsequently oligomerize. The destruction complex then assembles a signalosome by binding to the heteroreceptor and DVL, mediating the inhibition of β-catenin phosphorylation, which accumulates in the cytoplasm and translocates to the nucleus [56]. In the nucleus, β-catenin binds to the T cell factor (TCF)/ lymphoid enhancer-binding factor (LEF) family of transcription factors and induces the expression of its target genes. The Wnt/β-catenin pathway is well established as a key regulator of neurogenesis in adult mammals; however, a comprehensive integration of current evidence linking HIF-1α with the subsequent activation of the Wnt/β-catenin pathway during adult neurogenesis is still needed (Fig. 3). In embryonic murine stem cell (ESC) cultures transfected with a luciferase reporter construct containing TCF-binding sites (TOP-Flash gene), hypoxia significantly increased luciferase activity, indicating activation of the Wnt/β-catenin signaling pathway. Silencing β-catenin using siRNA under hypoxic conditions abolished TOP-Flash activity, confirming the requirement of β-catenin for this effect. Consistent with this finding, hypoxia increased β-catenin protein levels, as well as nuclear levels of TCF-1 and LEF-1 proteins, accompanied by increased transcription of the Wnt target genes Axin2 and Dickkopf-4 (Dkk-4) [57]. Hypoxia also increases endothelial cell (EC) proliferation, while treatment with Dickkopf-1, an inhibitor of the Wnt/β-catenin pathway, markedly reduced both EC proliferation and TOP-Flash reporter activity. Endothelial cells derived from HIF-1α knockout mice showed no significant changes in TOP-Flash activity, reduced β-catenin protein levels, and reduced expression of the Wnt target genes Dkk-4, Lef-1, and Tcf-1 mRNA compared to wild-type ECs. Interestingly, wild-type ECs differentiated into neuronal precursors using N2B27 medium, characterized by β-tubIII and DCX expression, showed no TOP-Flash activity or induction of Lef-1 and Tcf-1 expression under hypoxic conditions [57]. These findings indicate that HIF-1α is a critical activator of the Wnt/β-catenin signaling pathway in undifferentiated stem cells under in vitro hypoxic conditions, thereby promoting proliferative responses. This pathway ceases to respond to hypoxia once the cells differentiate into the neuronal lineage.

Fig. 3.

Wnt/β-catenin signaling and its role in adult neurogenesis under hypoxic conditions. In the canonical pathway, secreted Wnt ligands bind to Frizzled (FZD) receptors and the co-receptors LRP5/6 on the target cell membrane. In the absence of Wnt, β-catenin is continuously phosphorylated by the destruction complex composed of GSK3β, CK1α, AXIN, and APC, leading to its ubiquitination and proteasomal degradation. Upon Wnt binding, LRP5/6 phosphorylation and Dishevelled (DVL) recruitment inhibit the destruction complex, allowing β-catenin to accumulate in the cytoplasm and translocate into the nucleus, where it associates with TCF/LEF transcription factors to activate Wnt target genes. In adult neurogenic niches, such as the subgranular zone (SGZ) of the dentate gyrus and the subventricular zone (SVZ), hypoxia activates HIF-1α, which in turn stimulates Wnt/β-catenin signaling. This activation enhances the expression of β-catenin, LEF-1, TCF-1, and canonical Wnt target genes (e.g., Axin2, Dkk4), promoting neural stem and progenitor cell proliferation

Evaluation of the Wnt/β-catenin signaling pathway in adult mice using the brain β-catenin-activated reporter transgene BAT-GAL revealed that low oxygen levels significantly increased BAT-GAL expression in the DG of the SGZ. Conversely, genetic deletion of HIF-1α resulted in decreased expression of BAT-GAL, Dkk-4, Lef-1, and Axin2, as well as a reduction in the number of DCX⁺ and BrdU⁺ neurons in the SGZ. Pharmacological inhibition of GSK-3β by SB 216,763, or lentiviral expression of a constitutively active form of β-catenin in HIF-1α-deficient mice, rescues the neurogenesis defects caused by HIF-1α deletion, suggesting that β-catenin acts after HIF-1α. Furthermore, adult mice exposed to hypoxia for 6 h, 24 h, and 72 h showed increased expression of HIF-1α, β-catenin, and phosphorylated Dvl3 proteins, along with an increased number of Ki67⁺ cells in the hippocampal SGZ, with the most pronounced effects observed at 24 h. Analysis of BrdU incorporation and DCX immunostaining in the 72 h hypoxia group revealed an increase in BrdU⁺ cells but not DCX⁺/BrdU⁺, suggesting that hypoxia promotes neuronal proliferation but not early neuronal differentiation in this time interval [57]. Similar patterns were observed in a Parkinson’s disease model, where hypoxia also enhanced the proliferative response through the Wnt/β-catenin signaling pathway in the hippocampus of adult mice [58]. Furthermore, studies in adult rats demonstrated that hypoxia significantly increased the mRNA expression of HIF-1α, Lef-1, and Tcf-1, along with increased BrdU incorporation in the SVZ. Taken together, these in vivo findings indicate that HIF-1α activates neurogenic processes via the Wnt/β-catenin pathway in both neurogenic niches, the SGZ and the SVZ [11].

In contrast to these proliferative effects, treatment of glioblastoma (GBM) stem cell cultures with recombinant Wnt3 or transfection with the constitutively active form of β-catenin promoted neuronal differentiation rather than proliferation. This was evidenced by increased βIII-tubulin expression and a consequent reduction in Nestin and Ki67 expression. Both treatments also increased the expression of the neuronal differentiation markers NeuroD1 and Neurog1, further confirming the induction of a neuronal lineage program. Concurrently, a significant increase in the expression of p21, a cell cycle arrest marker, and a marked reduction in the formation of neurospheres and proliferating CD133⁺ cells were observed, suggesting suppression of stem cell self-renewal capacity [59]. Consistent with this pattern, additional studies have reported a role for Wnt3a in inducing differentiation in other cell lineages, including hematopoietic and mesodermal cells [60–62]. Collectively, these findings indicate that the effects of Wnt/β-catenin signaling are highly context-dependent. While adult in vivo studies support a role for HIF-1α-associated Wnt/β-catenin activation in precursor proliferation within neurogenic niches, complementary in vitro evidence suggests that specific Wnt ligands, such as Wnt3, promote neuronal differentiation and cell cycle arrest, highlighting the need for further studies to define how distinct Wnt isoforms differentially regulate proliferation and differentiation.

Adult ischemic rats treated with delayed hyperbaric oxygen therapy (HBOT) showed elevated levels of reactive oxygen species (ROS) and HIF-1α, along with increased expression of β-catenin and its coregulators TCF-1 and LEF-1, 14 days post-treatment. Conversely, administration of the HIF-1α inhibitor 2ME2 or the ROS scavenger N-acetylcysteine ​​(NAC) prevented the overexpression of ROS, HIF-1α, and β-catenin. Furthermore, delayed HBOT promoted neuronal proliferation and differentiation, as evidenced by increased immunoreactivity for neurogenin-1 and synapsin-1, as well as the incorporation of BDU 42 days post-treatment. These effects were abolished by administration of NAC, 2ME2, or the β-catenin inhibitor PKF, indicating that HBO-induced neurogenesis is mediated through activation of the ROS/HIF-1α/β-catenin signaling axis [63]. Consistent with these in vivo observations, complementary in vitro studies using rat embryo-derived neural stem cells demonstrated that ONOO⁻ treatment increased HIF-1α expression, activated Lef-1, and promoted nuclear translocation of β-catenin. These molecular changes were accompanied by increased BrdU incorporation and enhanced immunoreactivity for Ki67, nestin, Tuj-1, MAP2, and DCX, indicating increased self-renewal, proliferation, and differentiation. Similarly, C17.2 murine neuronal progenitor cells transfected with the luciferase-based TCF-sensitive reporter TOP-Flash showed a significant increase in reporter activity following ONOO⁻ exposure, which correlated with a greater neurogenic response. Importantly, treatment with FeTMPyP and FeTPPS PDCs abolished TOP-Flash activity. Consistently, hypoxic exposure of C17.2 cells increases the expression of β-catenin and its downstream target Lef-1, while treatment with PDCs prevented these effects [39]. Notably, cytotoxicity assessment using lactate dehydrogenase (LDH) release revealed that low concentrations of ONOO⁻ (1 µM) did not induce LDH release but did promote neurogenesis. Conversely, higher concentrations of ONOO⁻ (5 µM) significantly increased LDH release and reduced cell viability [39]. Previously, it has been well established that ONOO⁻ induced the production of ROS [64]. Taken together, the available in vivo evidence suggests that reactive oxygen and nitrogen species contribute to hypoxia-induced neurogenic responses through activation of the HIF-1α/Wnt/β-catenin pathway. Complementary in vitro studies further indicate that these effects are highly dependent on redox balance and concentration, promoting proliferation and differentiation at low oxidative levels while inducing cytotoxicity under excessive oxidative stress.

Within hypoxic zones, HIF-1α shifts NSC metabolism toward anaerobic glycolysis by upregulating glycolytic enzymes and inhibiting mitochondrial activity, an essential state for preserving the stem cell pool and preventing oxidative stress [65]. This metabolic control is physically linked to epigenetic regulation through the stabilization of β-catenin. HIF-1α interacts with the β-catenin/TCF complex to recruit histone acetyltransferases such as p300/CBP, which remodels chromatin to activate pro-neurogenic genes such as Sox2 and NeuroD1 [57, 66].

Notch signaling pathway regulated neurogenesis mediated by HIF-1α

The Notch signaling pathway is essential throughout development for regulating the precise determination of cell fate in vertebrates and invertebrates. In the mammalian CNS, Notch signaling is crucial for the proliferation and differentiation of neural stem cells during development and for neurogenesis in adulthood [67]. This signaling pathway involves interactions between receptors, ligands, transcriptional regulators, and transcriptional modifications. Four Notch receptors have been identified: NOTCH1, NOTCH2, NOTCH3, and NOTCH4. These receptors are transmembrane glycoproteins composed of an extracellular domain (NECD), a transmembrane domain (TM), and an intracellular domain (NICD). The NECD domain contains epidermal growth factor (EGF)-like repeats, which facilitate ligand binding, and a negative regulatory region (NRR) that prevents proteolytic cleavage of the receptor. For example, mammalian Notch ligands are divided into two subtypes: delta-like receptors (DLL1, DLL3, and DLL4) and jagged-like receptors (JAG1 and JAG2), which are also transmembrane proteins. Canonical Notch signaling is initiated by the S1 proteolytic cleavage of unprocessed, single-pass transmembrane Notch receptors by furin-like proteases in the trans-Golgi network, forming the NECD and TM+NICD domains. Activated receptors are then transported to the cytoplasmic membrane, where they interact with Notch ligands expressed on neighboring cells. Ligand binding induces a conformational change that destabilizes the receptor’s NRR region, exposing the S2 cleavage site. Subsequent cleavage by an ADAM family metalloprotease leads to the release of the NECD domain. Finally, intramembrane cleavage at the S3 site by γ-secretase releases the NICD domain, which is subsequently translocated to the nucleus and binds to the transcription factor RBPJ, displacing corepressors and recruiting coactivators to the promoter region of Notch target genes [68].

Evidence from in vivo and in vitro studies suggests that HIF-1 regulates the fate of neural stem and progenitor cells through modulation of the Notch signaling pathway (Fig. 4). In the DG of adult rats, induction of status epilepticus (SE) increased the expression of HIF-1α along with Notch and its downstream effector Hairy and enhancer of split 1 (Hes1). However, administration of 2ME2 inhibited the upregulation of Notch and Hes1, suggesting that HIF-1 induced components of the Notch signaling pathway in the DG. Furthermore, SE increased the number of BrdU-labeled neurons in the DG, while treatment with either 2ME2 or DAPT, a γ-secretase inhibitor, suppressed neurogenesis, supporting a role for HIF-1α-mediated activation of Notch signaling in adult DG neurogenesis [60]. As further evidence of this association, NSCs from the DG of adult rats subjected to SE differentiated into both mature neurons and astrocytes; however, inhibition of HIF-1α or Notch selectively reduced astrocytic differentiation, while increasing neuronal differentiation [69]. These findings are consistent with previous reports demonstrating that inhibition of the Notch pathway promoted neuronal differentiation [61]. Mechanistically, the immunoprecipitation assay demonstrates a physical interaction between HIF-1α and the NICD domain in the DG, providing direct evidence of crosstalk between hypoxia and Notch signaling [64]. These findings are consistent with previous reports which demonstrate that Notch pathway inhibition promoted neuronal differentiation [70]. Complementary in vitro studies using 3-day-old HIF-1α knockout murine embryonic stem cells (ESCs) further support a critical role of HIF-1α in the prevention of premature neuronal differentiation. HIF-1α deficient ESCs exhibited increased expression of NSC markers LewisX and nestin, as well as the NPC marker Pax6, accompanied by reduced levels of the pluripotency marker Nanog. In addition, knockout ESCs expressed the Notch ligand Dll1, a marker of neuronal commitment, which was absent in wild-type ESCs. Consistent with enhanced lineage commitment, colony-forming assays revealed a 36% reduction in the neurosphere formation in HIF-1α knockout cultures, indicating impaired self-renewal capacity. Prolonged culture of knockout neurospheres resulted in accelerated differentiation, characterized by increased expression of βIII-tubulin and GLAST, along with extensive neurite outgrowth. In contrast, wild-type neurospheres showed increased HIF-1α protein levels between 2d and 3d of culture, whereas knockout neurospheres displayed reduced expression of Hes1 while Notch remained unchanged. However, DAPT treatment did not further reduce Hes1 levels, suggesting that in NSCs HIF-1α is required for effective activation of Notch downstream transcriptional program rather than receptor expression itself. Consistent with enhanced neuronal differentiation, HIF-1α deficient cells exhibited increase expression of NeuroD1, a transcriptional factor essential for neuronal differentiation initiation. Similar results were observed in neural epithelial cells derived from mouse embryonic tissue, suggesting that this mechanism is conserved in NSCs populations [71, 72].

Fig. 4.

Notch signaling and its role in regulating adult neurogenesis under hypoxic conditions. Canonical Notch signaling is activated by direct cell-to-cell contact, whereby transmembrane Notch receptors (NOTCH1-4) on one cell bind to Delta (DLL1/3/4) or Jagged (JAG1/2) ligands on an adjacent cell. Ligand binding induces a series of proteolytic cleavages that culminate in the release of the γ-secretase-dependent intracellular Notch domain (NICD). The NICD subsequently translocates to the nucleus, where it interacts with the CSL/RBPJ transcriptional complex to induce the expression of Notch target genes, such as HES, HEY, and FABP7. This transcriptional program promotes the maintenance of neural stem and progenitor cells (NSCs) in a proliferative and undifferentiated state. Under hypoxic conditions, stabilization of HIF-1α leads to the activation of hypoxia-responsive genes such as VEGF, PGK1, and ET1 via binding to hypoxia-responsive elements (HREs). HIF-1α also potentiates Notch signaling by interacting with and stabilizing NICD, thereby enhancing its transcriptional activity. This convergence between HIF-1α and Notch signaling promotes sustained activation of Notch-dependent gene expression, which supports the maintenance of neural stem cells and regulates the balance between proliferation and differentiation during adult neurogenesis

Additional in vitro studies reinforce the role of HIF-1α-Notch signaling in the regulation of proliferation and differentiation balance in NSCs. Pharmacological blockade of Notch by DAPT in human GBM stem cell cultures reduced Ki67⁺ proliferating cells while increased βIII-tubulin neurons, suggesting a change from proliferation toward differentiation [59]. Similarly, NSCs isolated from the embryonic rat cortex maintain an undifferentiated and proliferative state in the presence of fibroblast growth factor 2 (FGF-2), a known activator of Notch signaling, whereas FGF-2 withdrawal promotes differentiation. However, under hypoxia without FGF-2, neuronal differentiation was suppressed and proliferation was enhanced. Importantly, inhibition of Notch signaling with the γ-secretase inhibitor L-685,458 fully rescued neuronal differentiation under hypoxia, as evidenced by increased numbers of Tuj-1⁺ cells. At the molecular level, hypoxic exposure of NSC cultures upregulated Hes1 expression and enhanced Notch receptor cleavage within 4 h, leading to increased release and stabilization of NICD. Subsequently, NICD was shown to interact with HIF-1α, thereby promoting the transcription of the Notch target genes [73]. Consistently, forced expression of mutated NICD in the developing chick CNS induced a significant increase in differentiated neurons expressing insulin gene enhancer proteins Isl1/2 [24]. This model is consistent with broader evidence indicating that Notch signaling maintains progenitor identity, suppresses neuronal differentiation, and promotes glial decision during development [74]. Taken together, these findings indicate that HIF-1α-mediated Notch activation regulates the balance between proliferation and differentiation in neural progenitor populations. Adult in vivo studies support a role for this pathway in maintaining neurogenesis within the DG, whereas complementary embryonic and in vitro models suggest that HIF-1α-dependent Notch activation suppresses premature neuronal differentiation and preserves progenitor identity under hypoxic conditions.

Crosstalk between Wnt and Notch signaling during-HIF-1α mediated adult neurogenesis

It has been reported that Wnt/β-catenin and Notch signaling pathways exhibit extensive crosstalk in the regulation of neurogenesis. Accordingly, the following section integrated current evidence demonstrating how HIF-1α factor modulates this interplay, thereby coordinating Wnt and Notch dependent mechanisms that regulate neurogenesis. In vitro studies in human GBM stem-like cells demonstrated that Wnt3a treatment increased Numb and NumbL, endogenous Notch inhibitors, through direct β-catenin binding to Numb promoter, thereby reducing NICD levels and driving neuronal differentiation, especially under hypoxia [59]. Other studies have demonstrated that β-catenin recruitment to Numb promoter leads to increase Numb expression, which in turn suppresses Notch pathway activity [75, 76]. In addition, activation of both Wnt and Notch pathways mediated by HIF-1α has also been reported in mouse embryonic stem cells exposed to hypoxic conditions [77]. Wnt3 or constitutive activation of β-catenin inhibited Notch transcription specifically in CD133⁺ human GBM stem-like cells, leading to reduced proliferation and increased neuronal differentiation evidenced by upregulated βIII-tubulin expression. Consistently, in vivo, GBM cells transplanted into the Wnt rich Zebrafish brain displayed neuronal reprogramming, showing neurite outgrowth, reduced nestin, and increased MAP2, βIII-tubulin, NeuroD1 and Neurog1 neuronal marker levels. On the contrary, GBM transplantation into the Zebrafish brain overexpressing Dickkopf-1, a Wnt inhibitor, maintained high proliferative rates evidenced by high nestin and Ki67 levels and prevented neuronal differentiation [59]. Another study conducted in human NPC cultures demonstrate a functional crosstalk between the Wnt and Notch signaling pathways, which regulate cell proliferation and neuronal differentiation. Notably, activation of Wnt exerts a dominant effect over Notch inhibition, indicating that Wnt/β-catenin signaling regulate early NPCs differentiation and maintained cell renewal. However, modulation of these pathways also alters brain regional identity of the NPCs, highlighting the need for further investigation to clarify how Wnt and Notch interactions influence fate specification and regional differentiation during neurogenesis [78]. Furthermore, HIF-1α, β-catenin, and NICD co-recruit the histone acetyltransferase p300/CBP to remodel chromatin, facilitating the transition from quiescent progenitors to active neuroblasts [79]. Collectively, these findings suggest that Wnt/β-catenin signaling promotes neuronal differentiation and cell-cycle exit in neural stem/progenitor cells, at least partially through suppression of Notch activity, although these effects appear to be highly dependent on cellular context and developmental state.

MAPK pathway: downstream regulation mediated by HIF-1α via EPO/VEGF

Mitogen activated protein kinase (MAPK) is evolutionary conserved signaling involved in important cellular functions such as cellular proliferation, migration, differentiation, and senescence; however, various cancers and degenerative diseases are also modulated by this pathway. MAPK signaling comprises of 7 different cascades, including 4 conventional cascades known as extracellular signal-regulated kinase (ERK) 1/2, c-Jun N-terminal kinase (JNK) 1/2/3, p38, and ERK5, as well as 3 non-conventional cascades ERK3/4, ERK7/8, and Nemo-like kinase (NLK) [72]. Different receptor tyrosine kinase (RTK) activates MAPK, including vascular endothelial growth factor receptor (VEGFR), fibroblast growth factor receptor (FGFR) and epidermal growth factor receptor (EGFR); however, Erythropoietin receptor (EPOR) which is not RTK can also activate the MAPK pathway [80]. The binding of extracellular activators such as cytokines, growth factors and hormones to membrane receptors induces the transmission of the signaling cascade through the three major kinases: MAPK, mitogen-activated protein kinase kinase (MAPKK), and mitogen-activated protein kinase kinase kinase (MAPKKK) [81]. It is well known that MAPK signaling regulates neurogenesis in adult mammals; however, the crosstalk between HIF-1α and MAPK pathway during adult neurogenesis has not been integrated [82].

It is well established that HIF-1 factor regulates the expression of EPO and VEGF in an oxygen dependent manner, a process that has also been well characterized in the CNS, including the SGZ and SVZ. In this section, we integrated the available evidence to date indicating that hypoxia induced-neurogenesis may be mediated by HIF-1 through the transcriptional induction of EPO and VEGF, which subsequently activate MAPK signaling pathways and promote neurogenic processes. Consistent with this hypothesis, several in vitro studies have demonstrated that hypoxia activates HIF-1α together with EPO/VEGF-associated neurogenic mechanisms. In human striatal precursor cells, CoCl₂-induced hypoxia caused a significant accumulation of HIF-1α protein and increased VEGF mRNA and protein levels at 24, 48 and 72 h, accompanied by enhanced human striatal precursor cell proliferation [40]. Similarly, CoCl₂ exposure of MSC cultures increased HIF-1α nuclear levels from 3 h up to 3d and upregulated EPO, VEGF and p21 mRNA expression between 6 and 24 h, with corresponding increased in protein levels detected at 3d. Notably, these molecular changes were associated with differentiation of MSCs into neuron-like cells [31]. Also, in a study using an immortalized human NSC line, hypoxic exposure increased the proportion of βIII-tubulin⁺ and MAP2⁺ neurons, as well as GalC⁺ oligodendrocytes, indicating that hypoxia promotes NSC differentiation [47]. Moreover, reoxygenation following hypoxia exposure further promotes proliferation and differentiation of NSCs isolated from rat pups. At 48 h of reoxygenation, PCNA immunoreactive cells were detected in the CA1 region, while nestin⁺ cells increased by 24 h. Western blot analysis confirmed an elevated PCNA and nestin protein expression at 24 and 48 h after hypoxia-reoxygenation. Importantly, administration of ERK inhibitor U0126 1 h prior to hypoxia-reoxygenation markedly reduced PCNA expression and abolished BrdU incorporation, which was otherwise increased under hypoxia reoxygenation condition [83]. Collectively, these in vitro findings suggest that hypoxia-driven stabilization of HIF-1α promotes neurogenic responses through induction of EPO and VEGF, which subsequently activate MAPK/ERK signaling pathways. Furthermore, the enhanced proliferation and differentiation observed during reoxygenation, together with their sensitivity to ERK inhibition, support a central role for the HIF-1α–EPO/VEGF–MAPK/ERK axis in mediating hypoxia-associated neurogenic processes in cultured neural cells.

In primary hippocampal tissue cultures derived from rat pups subjected to 6 h of hypoxia followed by reoxygenation, ERK1 activation exhibited a dynamic and region-specific pattern. In the stratum moleculare, phosphorylated ERK increased by 24 h of reoxygenation and declined by 96 h. This temporal activation correlated with enhanced proliferative activity, as evidenced by increased numbers of PCNA⁺ and BrdU⁺ cells in the stratum moleculare at 24 h, followed by a progressive increase in the stratum radiatum and CA1 region at 96 h of reoxygenation. In contrast, normoxic control cultures showed PCNA⁺ and BrdU⁺ cells only sporadically, primarily within the CA1 region. Consistent with a role for ERK signaling in post-hypoxic proliferation, hypoxia-reoxygenated cultures exhibited abundant BrdU⁺ cells after 6d, whereas pharmacological inhibition of ERK1 with U0126 markedly reduced BrdU incorporation. These findings indicate that ERK signaling is required for the induction of cell proliferation following reoxygenation [84]. In parallel, hypoxia induced early changes in neural progenitor cell distribution in primary tissue cultures. Clusters of nestin⁺ cells were detected as early as 4 h of hypoxia and extended into the stratum moleculare adjacent to the rat dentate granule cell layer by 24 h of reoxygenation. By 96 h, these cells became further dispersed into the stratum radiatum and oriented toward the CA1 region, suggesting active progenitor migration. In contrast to normoxic controls, reoxygenated cultures displayed numerous cells co-labeled with BrdU in the nucleus and doublecortin (DCX) in the cytoplasm, indicating ongoing neurogenesis. Notably, DCX⁺ cells were observed both subjacent to and within the CA1 region following reoxygenation, whereas they were sporadic in normoxic cultures. Double immunolabeling with BrdU and the astrocytic marker S100β further demonstrated that a subset of proliferating cells acquired an astroglia phenotype in both the stratum moleculare and CA1 region of rat primary tissue cultures [84]. Together, these observations indicate that hypoxia followed by reoxygenation induces a robust regenerative response characterized by ERK-dependent progenitor proliferation, migration, and differentiation. ERK activation occurred rapidly during hypoxia and was reactivated at 24 h of reoxygenation, coinciding with the onset of detectable neurogenesis. The subsequent decline in ERK phosphorylation after prolonged reoxygenation may reflect reduced afferent signaling, diminished availability of trophic factors, or a programmed downregulation of neurogenic signaling.

Intermittent hypoxia-induced neurogenesis mediated by HIF-1α-dependent activation of the MAPK pathway

To date, several reports have demonstrated that intermittent hypoxia (IH) exerts beneficial effects as a promising strategy for the prevention and mitigation of neurological disorders such as neurovascular diseases. In this context, IH enhances resistance to acute ischemic and hypoxia stress by the induction of neuroprotection, repair and neuroregeneration effects [85]. In this section we summarize the current evidence supporting IH-induced neurogenesis mediated by HIF-1α-dependent activation of the MAPK signaling pathway. Neurospheres and MASC cultures derived from the SVZ of neonatal mice exhibited higher HIF-1α levels and enhanced proliferative activity under IH compared with normoxic conditions, as evidenced by an increase of Ki67⁺ cells. Contrary, neutralizing antibodies against EPO and VEGF partially reduced proliferation in hypoxic cultures without affecting normoxic cells, indicating that both factors contribute to IH-induced proliferation. Additionally, neurospheres from both normoxic and hypoxic animals developed processes-bearing morphologies, whereas EPO or VEGF neutralizing antibodies truncate these processes, although IH exposure neurospheres retain a higher proportion of processes bearing cells [86]. Complementing these in vitro studies, IH increased phosphorylated MAPK (pMAPK) levels in the hippocampus of adult rats subjected to MCAO, as well as in normoxic control animals, 14d after injury. Notably, MCAO rats exposed to 7 cycles of IH exhibited significantly higher levels of HIF-1α, pMAPK and c-Fos compared to animals exposed to a single IH episode, indicating a positive correlation between the extent of IH exposure activation of MAPK signaling pathway [87]. Other works have documented that c-Fos expression is regulated by pMAPK which is also implicated in learning and memory [88, 89]. Consistent with these findings, IH intervention increased the number of BrdU⁺/c-Fos⁺ cells in both sham-operated and MCAO animals, whereas ischemic animals without IH depicted a significant reduction in double positive cells. Moreover, a strong correlation was observed between the number of BrdU⁺/c-Fos⁺ cells and improved spatial memory performance. By contrast, AZT application, an inhibitor of neurogenesis, significantly decreased the number of c-Fos/BrdU cells in hypoxic and sham groups [87]. The parallel increase of pMAPK and c-Fos expression following IH suggests that hypoxia activates MAPK pathway, which then induced c-Fos expression and neurogenesis, which improve learning and memory. Supporting this interpretation, increased number of IH exposures resulted in progressive elevations of HIF-1α, pMAPK, and c-Fos in ischemic animals, indicating a dose dependent activation of HIF-1α in response to IH. Additionally, administration of AZT to rats subjected to MCAO combined with IH impaired both neurogenesis and spatial learning and memory, compared to normal and sham operated rats. Furthermore, IH following MCAO increased c-Fos in the hippocampus relative to ischemic controls; however, no significant differences were observed between MCAO plus HI and normoxic plus IH groups. In contrast, MCAO plus HI group treated with AZT showed decreased levels of c-Fos protein and BrdU incorporation compared to hypoxic animals [87]. Collectively, these findings suggest that HIF-1α mediates c-Fos expression through activation of MAPK signaling, thereby promoting neurogenesis and improving spatial learning and memory following ischemic injury.

Therapeutic targeting of HIF-1α

Several pharmacological agents targeting HIF-1α have progressed into clinical trials, primarily in the field of oncology, although none have yet achieved broad clinical approval. These compounds interfere with the HIF-1 pathway at distinct regulatory levels and have been assessed mainly for safety and preliminary efficacy. For instance, EZN-2208, a PEGylated derivative of SN-38, suppresses HIF-1α mRNA expression and was evaluated in phase I/II trials in patients with refractory solid tumors. EZN-2968, a locked nucleic acid (LNA) antisense oligonucleotide designed to inhibit HIF-1α protein synthesis, demonstrated target engagement and disease stabilization in a subset of patients in phase I studies. NLG207, a nanoparticle formulation of camptothecin with HIF-1 inhibitory activity, has been investigated in phase II trials for advanced prostate cancer. In addition, several approved anticancer agents, including bortezomib and topoisomerase inhibitors (irinotecan and topotecan), indirectly suppress HIF-1 activity in vivo [90].

Despite this progress, no HIF-1 inhibitor has been evaluated in interventional clinical trials for stroke according to records from ClinicalTrials.gov. Current stroke-related studies involving HIF pathways are limited to observational or biomarker-based approaches. For example, the registered study NCT02297984 examines circulating neuroglobin and HIF-1α levels in patients with acute ischemic stroke but does not involve therapeutic modulation of HIF-1.

On the other hand, the most advanced pharmacological activators of hypoxia-inducible factor 1 (HIF-1) are HIF–prolyl hydroxylase inhibitors (HIF-PHIs). This drug class includes roxadustat, daprodustat, vadadustat, molidustat, and enarodustat. These agents have advanced through phase II and phase III clinical trials, and several are approved in multiple countries for the treatment of anemia associated with chronic kidney disease (CKD). Their therapeutic efficacy is based on HIF-mediated transcriptional activation of EPO, genes involved in iron metabolism, and angiogenic factors such as VEGF [91]. Beyond HIF-PHIs, indirect activation of HIF-1 has been reported with other pharmacological approaches, including iron chelators (e.g., deferoxamine), which inhibit PHD enzyme activity, nitric oxide donors, and metabolic modulators that alter mitochondrial oxygen consumption. However, these strategies remain largely experimental and have not progressed to advanced clinical evaluation as HIF-1–targeted therapies [92]. Beneficial effects have been reported across multiple disease models, including stroke and TBI, as well as neurodegenerative disorders such as Alzheimer’s and Parkinson’s diseases. Table 1 summarizes the key findings from these experimental models.

Table 1.

HIF-1α activators studied in CNS disorders

HIF-1α Activator	Targets (Disease)	Stage	Key Effects	References
Dimethyloxalylglycine (DMOG)	TBI (rodents) Stroke (rodents)	Preclinical	↑HIF-1α, ↑VEGF/EPO/PDK1, ↑Akt signaling; ↓neuronal death, ↓lesion volume, ↑angiogenesis; improved motor & memory post-TBI. Stroke studies show ↓infarct size and ↑recovery.	[94, 95]
Roxadustat (FG-4592)	PD (cell models)	Preclinical	↑HIF-1α, ↑HO-1 and PGC-1α; ↓α-synuclein toxicity, ↓ROS production, improved mitochondrial function.	[96]
Deferoxamine (DFOA)	AD models (mice)	Preclinical	Iron chelation, ↑HIF-1α, ↓synaptic loss and improves memory.	[97]
M30	AD models (mice)	Preclinical	↑HIF-1α, ↑neuronal survival pathways, reported to ameliorate Aβ pathology and cognition in AD transgenic mice.	[98]
S-nitrosoglutathione (GSNO)	TBI (rodents)	Preclinical	Stabilizes HIF-1α via S-nitrosylation. Markedly stimulated neuroregeneration and functional recovery.	[99]

Importantly, to date, no HIF-1 activator has been tested in interventional clinical trials for stroke, hypoxic–ischemic brain injury, or neuroregeneration, although HIF-1 activation is clinically validated in hematological disorders via HIF-PHIs [93].

Conclusions and future perspectives

This manuscript describes the biphasic role of HIF-1 signaling in the brain in response to hypoxia arising from different pathological conditions, including stroke. In the early phase after ischemia (approximately 4–8 h), HIF-1 activity is largely detrimental, promoting oxidative stress, inflammation, mitochondrial dysfunction, and neuronal apoptosis in response to hypoxia. In contrast, during the late phase (approximately 48–72 h), HIF-1 signaling becomes protective, exerting antioxidant effects and activating VEGF–mediated pathways that support tissue repair. Beyond acute injury responses, HIF-1α also regulates neurogenesis by promoting the proliferation, maintenance, and differentiation of neural stem and progenitor cells. These effects are mediated through the activation of interconnected signaling pathways, including Wnt/β-catenin and Notch, as well as through the induction of EPO and VEGF, which subsequently engage MAPK signaling to support neuronal development and regeneration. However, it is important to note that, in many cases, these pathways appear to be co-regulated under hypoxic conditions rather than being strictly organized in a linear hierarchy downstream of HIF-1α, and further studies are required to define their precise mechanistic relationships.

Future research on HIF-1 in mammalian neurogenesis is expected to focus on defining how oxygen sensing integrates with intrinsic developmental programs that govern the fate of neural stem and progenitor cells. In this context, greater attention should be paid to the contribution of other hypoxia-inducible factor isoforms, particularly HIF-2α and HIF-3α, which may play distinct or compensatory roles in neural stem cell biology. Clarifying the specific functions of each isoform, as well as their cell type-dependent expression patterns, will be crucial for a more complete understanding of hypoxia signaling in the brain and for the development of more targeted therapeutic strategies.

Another important area for future research is the interaction between HIF-1α-mediated metabolic reprogramming and epigenetic regulation during neurogenesis. HIF-1α is a key regulator of cellular metabolism under hypoxic conditions, as it promotes glycolytic pathways and alters mitochondrial function. These metabolic changes are increasingly recognized as modulators of epigenetic patterns, influencing chromatin accessibility and gene expression programs that determine stem cell fate. A deeper integration of metabolic and epigenetic mechanisms will be crucial for understanding how hypoxia shapes long-term cellular identity and neurogenic potential.

HIF-1α signaling extends beyond neurons to multiple components of the neurovascular unit, including endothelial cells and pericytes, where its activation directly contributes to disruption of the blood-brain barrier and the propagation of inflammatory responses. Importantly, HIF-1α exerts distinct—and often opposing—effects on neuronal cell populations, including neurons, astrocytes, and microglia, underscoring its highly context-dependent and cell-type-specific regulatory functions. This marked cellular heterogeneity is a fundamental determinant of HIF-1α’s function in the brain and must be explicitly considered when interpreting experimental data. Failure to account for these differences risks oversimplifying its biological effects and may hinder the development of effective therapeutic strategies.

With this in mind, future research could make use of new experimental models and high-resolution analytical approaches. Human brain organoids and three-dimensional culture systems that allow precise control of oxygen gradients will enable more accurate investigation of HIF-1-dependent mechanisms in a developmental context that closely mimics in vivo conditions. In parallel, single-cell and spatial multi-omic technologies will facilitate the identification of cell-type-specific HIF-1 transcriptional programs and their dynamic regulation within hypoxic neurogenic niches. There is growing evidence pointing to the fundamental role of HIF-1-mediated metabolic reprogramming in shaping the fate of neural stem and progenitor cells through interactions with epigenetic regulatory mechanisms.

Finally, despite robust preclinical evidence, the translation of HIF-1-targeted therapies into clinical applications for neurological disorders remains limited. To address this gap, a deeper understanding of the systemic effects of HIF modulation, potential off-target consequences, and the interaction of HIF-1 signaling with other key regulatory pathways, such as mTOR, AMPK, and inflammatory networks, is required. Furthermore, factors such as aging, which profoundly influence neurogenesis and the response to hypoxia, should be incorporated into future studies. Advances in experimental models, including human brain organoids and spatial multi-omics technologies, will facilitate the identification of cell-type-specific HIF-1 transcriptional programs and their dynamic regulation within hypoxic niches. Collectively, these efforts can help bridge the gap between experimental findings and therapeutic development, ultimately driving the use of hypoxia-based strategies for brain repair and regeneration.

This review has several limitations. First, most of the evidence analyzed comes from adult animal models, particularly rodent models with hypoxia, ischemia, or traumatic injuries, while information on HIF-1α during embryonic and developmental neurogenesis is under-reported. Second, there is considerable heterogeneity among the experimental paradigms used to study hypoxia-induced neurogenesis. Differences in oxygen concentration, duration of hypoxic exposure, reoxygenation intervals, injury models, animal age, and brain regions analyzed make direct comparisons between studies difficult and may contribute to seemingly divergent findings regarding proliferation, differentiation, neuronal survival, or neuroprotection. Third, most of the findings come from preclinical studies, while clinical evidence evaluating HIF-targeted therapies in neuroregeneration or ischemic brain injury remains scarce. Consequently, the translational relevance of several proposed mechanisms still requires validation in human systems. Despite these limitations, this review provides a comprehensive framework that organizes the current evidence on HIF-1α-mediated neurogenesis under hypoxic conditions. By systematically linking hypoxic signaling with the Wnt/β-catenin, Notch, MAPK, EPO, and VEGF pathways, the manuscript helps to elucidate the dual and context-dependent roles of HIF-1α in the regulation of NSCs, neuronal survival, and regenerative responses. Furthermore, the review identifies important conceptual and translational gaps that can guide future mechanistic and therapeutic studies in this field.

Author contributions

SC-M: Conceptualization, Formal analysis, Investigation, Data Curation, Writing - Original Draft, Writing - Review & Editing. FAL-R: Investigation, Visualization, Writing - Review & Editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.