After birth, most of the mammalian brain presents a cellular environment that is refractory to the generation of new neurons under physiological conditions (Kempermann, 2015). However, adult neural stem cells (aNSCs), residing within spatially restricted microenvironments of the adult brain, referred to as “niches” (Schofield, 1978; Lim et al., 2007), maintain the capability to self-renew, proliferate and generate differentiated progeny, mainly neurons, and, infrequently, astrocytes (Bottes et al., 2021). The subventricular zone (SVZ) of the lateral ventricles and the subgranular zone (SGZ) of the dentate gyrus (DG) in the hippocampus (Gage, 2019) are two of the main neurogenic niches in the adult mammalian brain. Here, we will focus on the adult hippocampal niche, because of the functional relevance of the hippocampus to memory and cognition, its link to neurodevelopmental and mood disorders and neurodegenerative diseases, and its sensitivity to insults such as epileptic seizures, stroke, or traumatic brain injury (for systematic reviews on the functional significance of the adult hippocampal neurogenic niche, see Kempermann et al., 2018; Duque and Spector, 2019; Snyder, 2019; Lucassen et al., 2020; Babcock et al., 2021; Denoth-Lippuner and Jessberger, 2021).

Generally, stem cell niches are characterized by the ability to anatomically host stem cells and functionally control their development in vivo in order to adapt to organismal and environmental requirements (Alonso, 2009; Li and Guo, 2021). At the adult hippocampal neurogenic niche, this is achieved by a complex crosstalk between aNSCs and other cell types present (Figure 1; Vicidomini et al., 2020). More specifically, the milieu of the adult hippocampal neurogenic niche is tightly regulated by intracellular and extracellular signals (e.g. aging, stress, physical exercise, environmental enrichment, learning, pathological alterations, neurotrophins, growth factors, signaling pathways, extracellular matrix mediators and extracellular vesicles, vasculature-secreted factors and local circuit activitiy; for systematic reviews, see Balu and Lucki, 2009; Aimone et al., 2014; Baptista and Andrade, 2018; Bonafina et al., 2020; Losurdo and Grilli, 2020; Vicidomini et al., 2020). Thereby, a regulated balance between aNSC quiescence, proliferation, self-renewal, differentiation, and fate choice contributes to lifelong hippocampal plasticity and fitness, while preventing neurodegeneration and other pathologies (Faigle and Song, 2013; Fitzsimons et al., 2014; Yao et al., 2016; Lopatina et al., 2019; Toda et al., 2019; Vicidomini et al., 2020; Kim et al., 2021; Li and Guo, 2021). Hence, understanding the regulation of aNSCs and their crosstalk with other cell types within the hippocampal neurogenic niche is indispensable to fully map the mechanisms governing their fate determination under physiological conditions and in the diseased brain.

A considerable part of the regulatory complexity required for proper aNSC homeostasis in the adult hippocampal niche is executed by non-protein-coding RNAs (ncRNAs), which comprise a significant proportion of the transcriptional output of the eukaryotic genome, contributing to both housekeeping and key regulatory processes (Carninci et al., 2005; Dunham et al., 2012; Salta and De Strooper, 2012, 2017b; Salvatori et al., 2020). These functions are regarded as particularly relevant for cells such as aNSCs, which require fast adaptation of transcriptome and protein abundance, especially when transitioning through cellular states while undergoing fate decisions (Encinas and Fitzsimons, 2017). However, given that the majority of ncRNA species are still functionally uncharacterized, it remains unclear how the diverse ncRNA-mediated pathways regulate fate determination in aNSCs. A deeper understanding of how ncRNA functions impact aNSCs may provide a better mechanistic appreciation of physiological brain development, its plasticity, and the disorders arising from the deregulation of aNSCs. Such insights may also prove pivotal for the next generation of RNA-based therapeutics targeting abnormal gene expression in aNSCs, in order to prevent, counteract or correct disruption of their homeostasis, activation and differentiation.

Here, we review recent discoveries reporting the regulation of cell intrinsic programs by small ncRNAs (sncRNAs) in several immature cell types of the adult hippocampal neurogenic niche, we discuss the mechanisms by which they impact fate determination of these cell populations and how, sometimes, sncRNAs may interact with each other to act as intercellular signals and regulate neighboring cells within the hippocampal neurogenic niche.

In the adult hippocampus, aNSCs occupy a narrow area between the hilus and the granular cell layer of the DG, known as SGZ (Kempermann et al., 2004), which forms part of the neurogenic niche (Figure 1). ANSCs and their immature progeny in the adult hippocampal neurogenic niche have often been classified using divergent nomenclature, which we summarize in Table 1. ANSCs are also known as “radial glia-like cells” (RGLs), due to their similarity to embryonic radial glial NSCs, and “quiescent neural stem cells” or “type 1 cells” (Table 1; Kempermann et al., 2004; Encinas et al., 2011). ANSCs are maintained in a reversible cell cycle arrest state, termed quiescence. Daughter cells generated by initial mitotic divisions of activated aNSCs exhibit reduced self-renewal and increased lineage restriction and proliferative potential (Berg et al., 2018). These highly proliferative progeny are functionally classified as “(transiently) amplifying (neural) progenitors” (herein collectively termed neural progenitor cells, “NPCs”), “intermediate progenitors”, or based on marker expression, as “type 2a cells” with a glial phenotype, and “type 2b cells” with a neuronal phenotype (Table 1). NPCs generate immature neuroblasts, also termed “type 3 cells,” which exit the cell cycle to become immature neurons first and, over a period of several weeks, newborn mature dentate granule neurons (“granule cells”, DGCs; Table 1; Encinas et al., 2011; Kempermann, 2015; Berg et al., 2018). Oligodendrocyte precursor cells (OPCs) are also present in the adult DG, yet to a significantly lesser extent, and not found to be restricted to the subgranular zone (Encinas et al., 2011). In addition, neuron-glial antigen 2 (NG2, a chondroitin sulfate proteoglycan)-expressing progenitor cells are present and continue to proliferate in the adult CNS (Ffrench-Constant and Raff, 1986; Wolswijk and Noble, 1989), and may represent a distinct class of proliferating glial cells in the adult hippocampus (Klempin et al., 2011; Schouten et al., 2020). Yet, their intrinsic stem cell potential and functional contribution are not well-characterized, with some studies reporting that NG2-expressing cells are multipotent progenitors that generate astrocytes and mature neurons in the postnatal hippocampus (Belachew et al., 2003), while others suggest that they may be OPCs (Dawson et al., 2000). Of note, single-cell transcriptomic profiling studies in the mouse DG have demonstrated that cells of the neurogenic lineage may co-exist in a heterogeneous continuum of cell states (Shin et al., 2015; Artegiani et al., 2017). A similar heterogeneity has been reported for oligodendrocytic and astrocytic precursors in several brain regions, including, among others, cortical, subcortical, and hippocampal subfields (Steindler and Laywell, 2003; Marques et al., 2016).

Besides aNSCs and their progeny, the adult hippocampal niche is a dynamic structure containing multiple other cell types (Figure 1). The functional intercellular crosstalk among the niche-resident cells allows for proper maintenance of the aNSC pool, progenitor proliferation, differentiation, survival, and maturation (Vicidomini et al., 2020). Of note, aNSCs and NPCs can regulate their own homeostasis, as well as the differentiation and maturation of their progeny, through autocrine and paracrine signals (Schultheiß et al., 2013; Vicidomini et al., 2020; Shariq et al., 2021). Similarly, hippocampal niche-resident astrocytes have been shown to provide both structural and functional support to aNSCs and their progeny, and to control cell proliferation, differentiation, migration, and synaptic integration of adult-born granule neurons (Bonafina et al., 2020). Moreover, microglia, the brain-resident immune cells, actively regulate multiple phases of neurogenesis (Butovsky et al., 2006; Ekdahl et al., 2009; Woodbury et al., 2015; Pérez-Rodríguez et al., 2021). In addition, blood vessels localize within the neurogenic niche and their capillary loops are found in close proximity to aNSCs and dividing NPCs (Licht and Keshet, 2015). As such, the vascular compartment can impact neurogenesis through direct contact with, and paracrine signaling from, endothelial and mural cells that make up blood vessels, or indirectly, via transporting systemic signals (in the form of soluble factors) from blood circulation into the neurogenic niche (Fitzsimons et al., 2013; Lee et al., 2013; Villeda et al., 2014; Licht and Keshet, 2015; Smith et al., 2015; Bátiz et al., 2016; Schouten et al., 2020). Last, mature hippocampal DGCs can modulate aNSC homeostasis via cell autonomously secreted factors and by recruiting circuit mechanisms involving, among others, local inhibitory interneurons and mossy cells (Gobeske et al., 2009; Ma et al., 2009; Jang et al., 2013; Brooker et al., 2017).

How this melting pot of cells and signals is orchestrated to eventually direct aNSCs and other niche-resident progenitors towards quiescence, neurogenesis, astrogliogenesis, and oligodendrogliogenesis is of particular importance for understanding the functional relevance of these processes in both the healthy and the diseased adult brain. In the next sections, we will focus on the pivotal regulatory role of sncRNAs as cell-intrinsic or -extrinsic signals and will critically discuss the emerging evidence supporting distinct key functions for these small but mighty regulators in the adult hippocampal niche.

NcRNAs are RNA molecules not translated into proteins. They are traditionally classified based on their average size into small (<200 nucleotides) and long (>200 nucleotides) species (Hombach and Kretz, 2016; Salvatori et al., 2020). These molecules are highly versatile and exhibit a very diverse mechanistic repertoire, ranging from regulation of chromatin state, transcription, epigenetic events, alternative splicing, and RNA interference (RNAi)-mediated gene silencing, to effects on mRNA stability, translational regulation, and sequestration of RNA and proteins from their natural targets in the intracellular space (Schouten et al., 2012; Salta and De Strooper, 2017b; Figure 2). In this review, we focus on regulatory sncRNAs (Figure 2), and in particular on microRNAs (miRNAs), which are the most studied sncRNA subclass with well-documented and emerging roles in the regulation of aNSCs and their progeny. We also discuss the functions of PIWI-interacting RNAs (piRNAs), which are rapidly emerging as important regulators of somatic stem cells. For further information on long ncRNAs in the central nervous system (CNS), readers are invited to consider several previous reviews that cover the topic in more detail (Salta and De Strooper, 2017b; Zhang et al., 2019b).

MiRNAs are a class of small (19–22 nucleotides in length) single-stranded ncRNAs. AGO proteins associate with miRNAs and other sncRNAs acting in RNA-based silencing mechanisms by inhibiting protein synthesis and altering RNA stability (Hutvagner and Simard, 2008; Figure 2). Their functions are particularly important for the fine-tuning of gene expression in brain development, physiology, and homeostasis (Saba and Schratt, 2010; Barca-Mayo and De Pietri Tonelli, 2014; Zhang et al., 2019b). Of note, miRNAs may repress individual targets only to a limited extent (Baek et al., 2008; Selbach et al., 2008). Thereby, their regulatory potential may often rely on the synergistic action of several miRNAs over distinct sets of transcripts, which in aNSCs often encode proteins that share similar or converging biological functions (Barca-Mayo and De Pietri Tonelli, 2014; Schouten et al., 2015; Pons-Espinal et al., 2017).

PiRNAs are a class of animal-specific single-stranded sncRNAs defined by their ability to associate with PIWI proteins, a subclass of Dicer-binding AGO proteins (Doyle et al., 2012). PiRNAs are longer (21–35 nucleotides in length) than miRNAs and endogenous siRNAs, and unlike the former classes, they are processed in a Dicer-independent manner, bear 2′-O-methyl-modified 3′ termini and only function through binding with PIWI-clade proteins, rather than AGO-clade proteins (Figure 2). Although much less studied than miRNAs, piRNAs are present in the mammalian CNS, while nearly 30,000 neuronal piRNAs have been identified in the mouse brain throughout development, suggesting that they are expressed and dynamically regulated during brain development (Ghoshes et al., 2016). In addition, several studies have shown that piRNAs might regulate neuronal function and synaptic plasticity in the mammalian brain (Wakisaka and Imai, 2019).

The disruption of miRNA and piRNA expression has been linked to a multitude of brain pathologies, including neurodegenerative diseases (Salta and De Strooper, 2012, 2017b; Wakisaka and Imai, 2019; Wu and Kuo, 2020; Huang and Wong, 2021), psychiatric (Yoshino and Dwivedi, 2020) and neurodevelopmental disorders (Barca-Mayo and De Pietri Tonelli, 2014; Zhang et al., 2019d). In agreement with their general involvement in health and disease, accumulating evidence links sncRNAs to aNSC regulation under both physiological (Figure 3) and pathological conditions (Table 2). In the following sections, we discuss the existing evidence demonstrating how sncRNA species, particularly miRNAs and piRNAs, can impact aNSC maintenance, fate determination, differentiation, and maturation of their progeny.

Mounting evidence has shown that some of the PIWI proteins and Piwi-interacting RNAs (piRNAs; Figure 2) are expressed in the CNS of different animal models, and participate in the processes of neurodevelopment, neurotransmission, transient focal ischemia and neurodegeneration (Kaur et al., 2018; Kim, 2019; Huang and Wong, 2021). In particular, PIWIL1 (MIWI in the mouse) and piRNAs have been linked to neuronal injury and axonal regeneration in rodents (Phay et al., 2018; Sohn et al., 2019) and C. elegans (Kim et al., 2018), but the underlying mechanisms are still obscure. In the mouse developing cerebral cortex, the knockdown of MIWI resulted in a multipolar morphology of neurons and a larger number of primary neurites, suggesting that the piRNA pathway can regulate neuronal migration and maturation during development (Zhao et al., 2016). The expression of PIWIL2 (MILI in mouse) protein has also been reported in the adult mouse brain, with hypomethylation of transposons and behavioral defects observed in MILI knockout mice (Nandi et al., 2016). A more recent study has described behavioral impairment in mice upon hippocampal MILI and/or MIWI depletion. Knockdown of MILI results in hyperactivity, whereas knockdown of both MILI and MIWI leads to enhanced contextual fear memory without affecting anxiety (Leighton et al., 2019).

The aforementioned studies and the majority of those which investigated the possible roles of the piRNA pathway in mammalian CNS, have focused on neurons (Dharap et al., 2011; Lee et al., 2011; Ghoshes et al., 2016; Nandi et al., 2016; Zhao et al., 2016; Leighton et al., 2019). However, as piRNA levels in these cells are low compared to germline cells, the proposed function of the piRNA pathway in synaptic plasticity and memory control (if any) is debated.

Interestingly, the hippocampus shows the highest number of unique piRNA sequences and piRNA transcripts (5494 piRNAs) in the adult brain (Perera et al., 2019). In agreement with this observation, it was recently demonstrated that the piRNA pathway has a crucial role in maintaining the homeostasis and the fate of hippocampal NSCs (Gasperini et al., 2021). In particular, the expression of MILI and piRNAs was found to be enriched in NPCs of the postnatal mouse hippocampus, compared to their differentiated progeny, and piRNA functions were shown to be essential for neurogenesis. Moreover, inhibition of the piRNA pathway in adult NPCs leads to senescence-associated inflammation and an increase in reactive astrogliogenesis (Gasperini et al., 2021), a condition which has been linked to age-related impairment of neurogenesis and neurodegeneration.

Despite the growing interest in brain-related piRNAs, the unavailability of proper methods and tools to investigate the piRNA pathway outside gonads, and therefore the lack of a fully characterized pathway in neuronal tissues, complicates research in this field. Indeed, most of the existing studies did not provide conclusive evidence that piRNA-dependent, rather than piRNA-independent, functions of these proteins drove the various phenotypes. Functions of PIWI proteins were shown to regulate chromatin structure/stability, histone/DNA methylation, DNA repair transcription, and RNA turnover and translation in gonads (Czech et al., 2018; Ozata et al., 2019), but these mechanisms are poorly investigated in CNS. Because PIWI protein functions have been attributed roles in several CNS diseases, including neurodevelopmental and neurodegenerative diseases and various cancers, a better understanding of these processes in aNSCs entails important translational implications.

With respect to additional classes of sncRNAs, at present very little is known about their involvement in the regulation of aNSC homeostasis, activation, and fate determination. Interestingly, a double-stranded RNA (dsRNA) molecule (Figure 2) expressed at the SGZ of DG, was shown to regulate fate determination of hippocampal aNSCs in vitro (Kuwabara et al., 2004). This dsRNA bares a recognition motif for NRSF/REST, a major negative transcriptional regulator of neuronal differentiation. Interaction between the dsRNA and the NRSF/REST machinery results in the transition from aNSCs to a neuronal (but not glial) fate, while the effects seem to be mediated by a dsRNA/protein interplay (Figure 3).

The complexity of the biological processes involved in aNSC homeostasis, including activation, self-renewal, and fate determination, requires tight control of numerous multilayered regulatory programs. SncRNAs, as other ncRNA species, regulate gene expression at the transcriptional and post-transcriptional level and are additionally considered to be the main players in epigenetic regulation.

In this review, we have summarized evidence indicating that sncRNAs, and primarily microRNAs, play major regulatory roles at virtually all stages of aNSC homeostasis in both health and disease. We have discussed in detail the role of individual miRNAs, in the regulation of key targets crucial for aNSC quiescence, and for their activation, which can eventually result in neurogenesis, astrogliogenesis or oligodendrogliogenesis. Moreover, we surveyed miRNA roles in the functional crosstalk between aNSCs and other relevant cell types in the adult hippocampal neurogenic niche, in particular microglia. However, with the advent of novel technologies, we have only recently begun to unveil previously hidden layers of the biology of sncRNAs. For instance, tRNA-derived fragments are a new class of sncRNAs with noncanonical biological functions unrelated to protein synthesis (Torres et al., 2019); they are produced from tRNAs by specific cleavage events catalyzed by ribonucleases, such as Dicer; and have been shown to increase during disease, where they seem to regulate stress responses, proteostasis, and neuronal survival (Fagan et al., 2021). Interestingly, tRNA-derived piRNAs, putatively requiring PIWI protein function in somatic cells, have also been identified (Keam et al., 2014). How these and other newly identified sncRNAs may independently or in synergy (with miRNAs, piRNAs, or other ncRNA species) impact phenomena related to aNSCs remains elusive.

Further contributing to our understanding of the complexity of sncRNA-mediated regulation, the functional crosstalk across and within different classes of sncRNAs has been shown to enable precise regulation of several brain functions (Mehta et al., 2021). Functional pleiotropy (each miRNA may simultaneously regulate multiple targets) and convergence onto shared targets within distinct pathways (multiple miRNAs may simultaneously regulate the same target) are two key elements of miRNA biology (Barca-Mayo and De Pietri Tonelli, 2014; Walgrave et al., 2021b) with functional relevance in AHN (for a review, see Stappert et al., 2018). Systematic interrogation of the intricate targetomes of single miRNAs or of synergistic miRNA networks in the context of aNSC regulation will enable a more accurate profiling of their regulatory repertoire and, hence, of their potential therapeutic use as mediators of regeneration of distinct cell types in the adult brain. Even though these interactions may also be critical for maintaining the heterogeneity of the adult hippocampal niche by determining the balance between neurogenesis and astrogliogenesis (Pons-Espinal et al., 2017), the majority of such mechanistic aspects have not been experimentally addressed in the context of AHN, apart from a few exceptions (Schouten et al., 2015; Jin et al., 2016; Pons-Espinal et al., 2017; Wang et al., 2017b; Bielefeld et al., 2019; Pan et al., 2019).

As an additional layer of intercellular communication, aNSCs and other cell types in the niche are capable of releasing and taking up extracellular vesicles like many other cell types in the mammalian brain (Lopez-Verrilli and Court, 2013; Luarte et al., 2016; Vogel et al., 2018). Extracellular vesicles can be divided into several categories depending on size, among which exosomes have been mostly implicated in the regulation of the aNSC niche. Exosomal content varies greatly based on the cell type from which they are released and the (patho)physiological state of the cell. In general, exosomes contain a plethora of non-coding RNAs, including miRNAs, piRNAs, tRNAs, long non-coding RNAs (lncRNAs), as well as mRNAs, lipids, and transcription factors (Valadi et al., 2007; O’Brien et al., 2020). Exosomes in the neurogenic niche have been implicated in all stages of neurogenesis, ranging from the regulation of aNSC activation (Cheng et al., 2021) to synaptic pruning of mature neuronal spines through microglia (Bahrini et al., 2015). An example of exosome-mediated intercellular signaling and synergy between multiple miRNAs is the regulation of neuroplasticity and functional recovery after stroke by the miR-17-92 cluster. Exosomes extracted from multipotent mesenchymal stromal cells and artificially enriched in miRNAs from the miR-17-92 cluster, increased oligodendrogliogenesis, neurogenesis, neurite remodeling, and neuronal dendrite plasticity in the ischemic boundary zone after stroke, possibly by activating the PI3K/protein kinase B/mechanistic target of rapamycin/glycogen synthase kinase 3β signaling pathway (Xin et al., 2017). In addition, intercellular communication via exosomal miRNAs has been shown to impact brain physiology and a diverse range of pathologies (Xia et al., 2019). More recently, endogenous sncRNAs, termed “glycoRNAs”, were found to bear glycan conjugations and to primarily localize on the cell membrane, putatively acting as signals in the intercellular communication and further contributing to intricate cellular interaction pathways (Flynn et al., 2021).

An issue confounding the current understanding of hippocampal aNSC biology is that a big part of our knowledge on the sncRNA-mediated regulation of aNSCs is derived from in vitro systems and from studies in embryogenesis or in the adult SVZ niche. Yet, evidence from heterotopical transplantation experiments demonstrated the significance of the niche microenvironment in fate choice heterogeneity of aNSCs (Beyer et al., 2020), suggesting that findings not derived from the adult SGZ should not be extrapolated to the adult DG without further experimental validation. In addition, given the context-specificity of several miRNA regulatory networks, extrapolating such observations to putative effects in AHN should be done with caution.

Another pivotal limitation is the vast knowledge gap with respect to whether the coding and non-coding mechanisms regulating aNSC quiescence, activation, and fate choice in the adult rodent brain are also conserved in humans. Studies in human systems in vitro and high-resolution cellular and molecular profiling of the adult hippocampal neurogenic niche in the human brain will shed more light on the existence of aNSCs, their putative heterogeneity, and their regulation in health and disease.

In conclusion, the intricate sncRNA landscape of mammalian cells has only recently started to be appreciated. Understanding the complex sncRNA-mediated regulatory mechanisms involved in aNSC homeostasis, including self-renewal, activation, and fate commitment will be key to decoding the regulatory determinants that can be used to leverage aNSC biology in brain diseases. Further studies focusing on the adult hippocampal neurogenic niche and considering the elaborate sncRNA functional communication both among them and with target pathways will be critical to profile the full regulatory potential of sncRNAs in this regard, as well as their future applications in the next generation of (RNA) therapies, particularly targeting age-related pathologies.

CF and ES conceived the article. AP, GT, OA, PB, CG, DD, CF, and ES discussed, wrote and corrected the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.