Expression and function of Ndel1 during the differentiation of neural stem cells induced by hippocampal exosomes

Background In the brain of adult mammals, neural stem cells persist in the subventricular zone of the lateral ventricle and the subgranular zone of the dentate gyrus, which are specialized niches with proliferative capacity. Most neural stem cells are in a quiescent state, but in response to extrinsic stimuli, they can exit from quiescence and become reactivated to produce new neurons, so neural stem cells are considered to be a potential source for cell replacement therapy of many nervous system diseases. We characterized the expression of Ndel1 during the differentiation of neural stem cells induced by hippocampus exosomes, and assessed the effect of Ndel1 on neural stem cells differentiation Methods Hippocampal exosomes were isolated extracted, and co-cultured with neural cells. Western blot, ow cytometry, and immunouorescence analyses were used to analyze expression of neuronal markers. Further, utilizing high-throughput RNA sequencing technology, we found that nudE neurodevelopment protein 1-like 1 was signicantly up-regulated in exosomes derived from denervated hippocampus, and then characterized its mechanism and function during neural stem cells differentiation by qRT-PCR, western blot, ow cytometry, and immunouorescence analyses. system diseases. These results indicated that deafferentation led to changes in the hippocampal expression of molecules that regulated NSCs differentiation. However, it remains unknown whether deafferentation of the hippocampal exosomes could promote the differentiation of NSCs. Our results revealed that deafferentation of hippocampal exosomes co-cultured with NSCs could promote neuronal differentiation. Subsequently, we found that nuclear distribution protein like 1 (Ndel1) was signicantly upregulated and highly expressed in the nervous system. Additionally, we found that Ndel1 enhanced spatial learning and hippocampal neurogenesis in rats after mbria fornix (FF) transection in vivo. These ndings revealed a novel mechanism and identied specic targets for treating CNS diseases. Modied Eagle Medium (DMEM) and F-12 mixture (both, Gibco, Grand Island, USA) containing 2% B27 (Gibco), 20 ng/mL epidermal growth factor (EGF; Sigma-Aldrich, St. Louis, MO, and 20 ng/mL broblast growth factor 2 (bFGF; Sigma-Aldrich). Cells were passaged every 6 d to obtain neurospheres that originated from a single primary cell. For in vitro differentiation, cell suspensions were plated with DMEM/F-12 medium supplemented with 2% B27 and 2% fetal bovine serum (FBS, Gibco). For the mixed co-culture experiments, isolated exosomes were mixed with NSCs and processed in different ways after cocultivation.


Introduction
The hippocampus originates from the medial pallium of the dorsal telencephalon and plays important roles in learning, memory, and affective behaviors (Yang et al., 2017). The subgranular zone of the hippocampal dentate gyrus (DG) is one of the stem-cell-containing niches in the adult mammalian brain (Akers et al., 2018). This thin band between the granule cell layer and the hilus provides a unique microenvironment for the adult neural stem cell (NSC) population (Goncalves et al., 2016).
Heterogeneous pools of NSCs in the adult mammalian brain are the source of new neurons that contribute to brain maintenance and regeneration (Llorens-Bobadilla et al., 2015). Most adult NSCs are quiescent and show a low metabolic rate and a high sensitivity to their microenvironment (Urban et al., 2019). The balance of NSC activation and quiescence, as well as the induction of lineage-speci c transcription factors, may contribute to the generation of neuronal or glial progeny cells (Harris and Guillemot, 2019).
Exosomes are nano-sized extracellular vesicles secreted by a variety of cell types that have been proven to be important intercellular messengers and exhibit molecular pro les that re ect normal and disease states (Malm et al., 2016). A recent study revealed that exosomes in the brain can play critical roles in Our previous research showed that the deafferent hippocampus provided a supportive microenvironment for the survival, migration, and neuronal differentiation of endogenous hippocampal and implanted NSCs. Importantly, extracts from the denervated hippocampus promoted more NSCs to differentiate into neurons and their subsequent in vitro maturation (Zhang et al., 2007;Zhang et al., 2009). These results indicated that deafferentation led to changes in the hippocampal expression of molecules that regulated NSCs differentiation. However, it remains unknown whether deafferentation of the hippocampal exosomes could promote the differentiation of NSCs. Our results revealed that deafferentation of hippocampal exosomes co-cultured with NSCs could promote neuronal differentiation. Subsequently, we found that nuclear distribution protein like 1 (Ndel1) was signi cantly upregulated and highly expressed in the nervous system. Additionally, we found that Ndel1 enhanced spatial learning and hippocampal neurogenesis in rats after mbria fornix (FF) transection in vivo. These ndings revealed a novel mechanism and identi ed speci c targets for treating CNS diseases.

Animals and surgery
Pregnant Sprague-Dawley rats, 1-day-old neonatal Sprague-Dawley rats, and adult Sprague-Dawley rats (weighing 220-250 g) were obtained from the Experimental Animal Center of Nantong University FF transections were performed as described by Hefti (Hefti, 1986). Brie y, after chlorpent anesthesia (2 mL/kg body weight, intraperitoneal), adult SD rats were transferred to the stereotaxic apparatus, and then FF transection was performed with a wire knife at the CA1 layer of the dorsal hippocampus, at coordinates of bregma: AP = 1.4, ML = 1.0 and AP = 1.4, ML = 4.0, depth 5.6 mm. There were no restrictions on the sex of the experimental animals.

Exosome isolation
Seven days following FF transection, deafferented and normal hippocampi were quickly dissected, trypsinized, and homogenized into ice-cold phosphate-buffered saline (PBS). Exosomes were precipitated using Total Exosome Isolation reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Homogenates were centrifuged at 2,000 × g at 4 °C for 30 min to remove cells and debris, and then supernatants were passed through a 0.22-µm lter to remove extracellular vesicles larger than exosomes. The supernatants were transferred to a new tube without disturbing the pellet and mixed with 0.5 volumes of Total Exosome Isolation reagent and incubated overnight at 4 °C. The mixture was then centrifuged at 10,000 × g for 30 min, and the supernatant was decanted, while the exosome pellet was resuspended into 100 µL PBS.

Cell culture
The isolation, culture, and differentiation of NSCs were performed as previously described with some modi cations (He et al., 2018). Brie y, pregnant SD rats were anesthetized, and the embryos were removed by cesarean section. Hippocampi were dissected from embryonic day 14.5 (E14.5) embryos, and were then mechanically dissociated into a single-cell suspension. After centrifugation and resuspension, the cell suspensions were plated into asks with a 1:1 Dulbecco's Modi ed Eagle Medium (DMEM) and Ham F-12 mixture (both, Gibco, Grand Island, NY, USA) containing 2% B27 (Gibco), 20 ng/mL epidermal growth factor (EGF; Sigma-Aldrich, St. Louis, MO, USA), and 20 ng/mL broblast growth factor 2 (bFGF; Sigma-Aldrich). Cells were passaged every 6 d to obtain neurospheres that originated from a single primary cell. For in vitro differentiation, cell suspensions were plated with DMEM/F-12 medium supplemented with 2% B27 and 2% fetal bovine serum (FBS, Gibco). For the mixed co-culture experiments, isolated exosomes were mixed with NSCs and processed in different ways after cocultivation.
Primary neurons were isolated using standard methods, as previously described (Xing et al., 2014).
Brie y, hippocampi were dissected from E14.5 embryos, and the resultant single cell suspensions were diluted in serum-free neurobasal medium (Gibco) containing 2% B27 and 0.5 mM L-glutamine (Gibco). The cells were then seeded onto plates precoated with poly-D-lysine. Half of the medium was replaced every 3 d.
Primary astrocytes were derived from cerebral cortices of 1-day-old neonatal rats as previously described (Xing et al., 2014). Brie y, dissociated cortical cells were suspended in DMEM/F-12 containing 10% FBS and plated in asks. After 3-4 d, the heterogenous primary cells were orbitally shaken to remove microglia and oligodendrocytes. Astrocytes were dissociated by trypsinization and then replated into asks.

Transfection, lentiviral transduction, and injection
Prior to transfection or transduction, cells were cultured in plates overnight. Cells were transfected with the miR-107-3p/NC mimic or miR-107-3p/NC inhibitor (Ribobio, Guangzhou, China) using Lipofectamine 3000 (Invitrogen) according to the manufacturer's instructions. Cells were transduced with lentivirus that was constructed by Genechem Company (Shanghai, China), including lentiviral vectors Nedl1 (abbreviated as LV-Nedl1 and LV-Ndel1i), corresponding to the negative control lentiviruses (LV-NC and LV-NCi) following the manufacturer's instructions. Green uorescence expression was then observed under a uorescence microscope (Axio Scope A1, Zeiss, Oberkochen, Germany). The cells were cultured with lentivirus for 12 h to obtain the best infection complex value, after which the lentivirus was removed and replaced with fresh medium.
In total, 60 SD rats were used for lentivirus injections into the hippocampus. Brie y, after chlorpent anesthesia, adult SD rats were transferred to the stereotaxic apparatus. On day 7 after FF transection, injections of virus into the left and right hippocampal DG at two points were performed at the following coordinates: 3.6 mm to bregma, 1.39 mm to the right or left of the midline, and 3.9 mm in depth. Five µL of virus were loaded into an internal cannula needle with cannula tubing connected to a Hamilton syringe mounted onto a microinjection pump (Harvard Apparatus, Dover, MA, USA). The speed of the injection was 0.5 µL/min. The needle was kept in position for an additional 10 min after completing the injection, and then was slowly retrieved from the brain.

Statistical analysis
Statistical analyses were mainly conducted using GraphPad Prism 6.0 (GraphPad Software Inc., San Diego, CA, USA). Differences between two groups were compared using an unpaired Student two-tailed ttest, and differences among multiple groups were analyzed by one-way ANOVA. The results were considered statistically signi cant when * P < 0.05; ** P < 0.01; *** P < 0.001.

Effects of hippocampal exosomes on NSC differentiation
To identify the isolated exosomes, we applied transmission electron microscopy. As shown in Figure S1A, hippocampal-derived exosomes were lightly stained and had diameters within 30-200 nm. To con rm that these exosomes could be transferred to cells, we co-cultured CM-Dil-labeled exosomes with NSCs. After incubation with exosomes, the CM-Dil uorescence signal was observed in most NSCs ( Figure S1B). As shown in Fig. 1A, western blotting showed that Tuj1 was signi cantly upregulated in the transected group. Similarly, ow cytometric analysis and immuno uorescence staining showed that there were more Tuj1-positive cells in the transected group than in controls (Fig. 1B-1E). Our results also revealed that exosomes derived from deafferented hippocampi facilitated neuronal differentiation of NSCs.
High-throughput functional screening for differentially expressed mRNAs To identify and characterize the differentially expressed exosomal mRNAs, RNA-seq was implemented in three pairs of hippocampal exosomes. When we set the lter criteria to be fold-change ≥ 2 and a P-value < 0.05, we found 770 differentially expressed mRNAs, among which 764 were upregulated and six were downregulated in hippocampal exosomes (Table S2). The heat map of differentially expressed genes is shown in Fig. 2A. Next, a bioinformatics analysis was performed to characterize the mRNA pro le of hippocampal exosomes. Gene ontology (GO) analyses suggested the differentially expressed genes were associated with protein transport, gene expression, cellular metabolic processes, and other important functions (Fig. 2B). Pathway analyses suggested that oxidative phosphorylation, spliceosome, and Ubiquitin-mediated proteolysis were most enriched among the differentially expressed genes (Fig. 2C). Figure 2D presents the relationships between enriched pathways.

Identi cation and characteristics of Ndel1
Among the upregulated mRNAs, we focused on Nedl1, which was enriched in neuron projection development, microtubule cytoskeleton organization, nervous system development, and central nervous system neuron axonogenesis according to GO analysis. As shown in Figs. 3A and 3B, differential expression of Nedl1 was consistent with the trends observed using RNA sequencing. Furthermore, after being co-cultured with exosomes, we found that Nedl1 expression was increased in the transected group (Figs. 3C and 3D). To explore the Ndel1 expression pattern, we extracted RNA from tissues derived from ectoderm (cerebrum, cerebellum, brain stem, and hippocampus), mesoderm (heart and muscle), and endoderm (liver), and then performed a RT-qPCR analysis. As shown in Fig. 3E, Ndel1 was signi cantly overexpressed in nervous tissues compared with other tissues. Additionally, Ndel1 showed its highest expression in NSCs, followed by neurons, and minimally in astrocytes (Fig. 3F). We then examined the expression pattern of Ndel1 in the hippocampus by immunohistochemistry. The results showed that Ndel1 was more highly localized to the somata of some polymorph layer cells, but was also expressed in the granular layer of the DG (Fig. 3G). Seven days after FF injury, we found that the number of Nedl1positive cells had increased in the denervated hippocampus (Figs. 3H). These data suggested that Ndel1 played an important role in neurogenesis.

Effects of Nedl1 on NSC differentiation
To examine the precise functions of Ndel1 in NSCs, we transfected NSCs with lentiviral vectors encoding Ndel1 ( Figure S1C-S1F). To explore whether Ndel1 regulated NSC differentiation, we measured the expression levels of two commonly used nerve-speci c molecules, Map2 and Neurod1. The results showed that Ndel1 upregulation promoted Map2 and Neurod1 expression. Knocking down Ndel1 had the opposite effect (Fig. 4A). Western blotting showed that overexpressing Ndel1 notably increased Tuj1 expression. Conversely, knocking down Ndel1 induced decreased Tuj1 expression (Fig. 4B). Consistent with these results, ow cytometry and immuno uorescence revealed that overexpressing Ndel1 notably increased the number of Tuj1-positive cells. Conversely, knocking down Ndel1 induced a decrease in Tuj1 expression (Figs. 4C-4F). Together, these results implied that Ndel1 promoted the neuronal differentiation of NSCs.
To probe the underlying molecular mechanisms of Ndel1, we rst used three algorithms (miRWalk, TargetScan, and miRDB) to predict potential upstream miRNA of Ndel1. For all three algorithms, miR-107-3p was the commonly predicted target. We also found that miR-107-3p exhibited high expression in nervous tissues (Fig. 5A). To further investigate the potential biological function of miR-107-3p, we constructed a miR-107-3p mimic and inhibitor. qRT-PCR results showed that miR-107-3p expression was signi cantly upregulated and downregulated in NSCs transfected with the miR-107-3p mimic and inhibitor, respectively ( Figure S1G). Next, we investigated the impact of miR-107-3p on Ndel1 expression by qRT-PCR and western blot. The results showed that overexpression and knockdown of miR-107-3p resulted in the downregulation and upregulation of Ndel1 in NSCs, respectively (Figs. 5B and 5C).
Nedl1 enhanced hippocampal neurogenesis in vivo after FF transection.
The Morris water maze test was performed in the last 5 days before sacri ce (35 d post injury) to evaluate spatial learning. Compared with rats in the LV-Ndel1 group, the escape latency of rats in the PBS and LV-NC rats to reach the platform was signi cantly longer ( Fig. 6A and 6B). Furthermore, LV-Ndel1 group rats crossed the platform more frequently (Figs. 6C and 6D). GFP detection in the hippocampus proved that the lentivirus successfully infected target tissues ( Figure S1G). As shown in Fig. 6E and 6F, Tuj1 was signi cantly upregulated. Thus, Ndel1 expression was associated with signi cantly improved learning and memory ability and enhanced neurogenesis in the hippocampus of adult rats following FF transection.

Discussion
The discovery of NSCs in the adult brain provides evidence that the CNS may have the potential to repair insults by generating new neurons (Zhu et al., 2018). NSCs are self-renewing and multipotent cells with the potential to differentiate into neurons, astrocytes, and oligodendrocytes (Huang and Zhang, 2019 MiRNAs are a class of small noncoding RNAs that either prevent translation or promote the degradation of speci c targets by binding to target sequences usually located in the 3 -UTR (Correia de Sousa et al., 2019). To explore the potential molecular mechanism of Ndel1, we used three algorithms to predict miRNAs that could bind Ndel1, which identi ed miR-107-3p. There are almost no reports on the relationship between miR-107 and NSCs differentiation, and to date, most studies on miR-107 have been related to cancer. A growing body of evidence indicates that aberrant miR-107 expression plays a key role in cancers, including breast cancer (Luo et  Our study found that miR-107-3p was highly expressed in nervous tissues; moreover, we found that Ndel1 was directly regulated by miR-107-3p. Subsequently, overexpression of miR-107-3p suppressed Ndel1 expression and inhibited the differentiation of NSCs into neurons.

Conclusions
Our results revealed that deafferentation of the hippocampal exosomes co-cultured with NSCs could promote them to differentiate into neurons. Hence, we identi ed that Ndel1 was signi cantly up-regulated and highly expressed in the nervous system. In addition, these results suggested that miR-107-3p may regulate NSCs differentiation by targeting Ndel1. With a better understanding of endogenous NSCs under normal and pathological conditions, we may be able to employ endogenous NSCs for neuroregeneration in the future.

Declarations
Ethics approval and consent to participate This study was approved by the Institutional Review Board of the Medical School of of Nantong University.

Consent for publication
All authors agree to publish this manuscript.

Availability of data and materials
Not applicable.