Volume 247, Issue 1 p. 229-238
Patterns & Phenotypes
Free Access

HMGB2 expression is associated with transition from a quiescent to an activated state of adult neural stem cells

Ayaka Kimura

Ayaka Kimura

Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Search for more papers by this author
Taito Matsuda

Corresponding Author

Taito Matsuda

Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Correspondence to: Kinichi Nakashima and Taito Matsuda, Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, 812-8582, Japan. E-mail: [email protected]; [email protected]Search for more papers by this author
Atsuhiko Sakai

Atsuhiko Sakai

Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Search for more papers by this author
Naoya Murao

Naoya Murao

Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Search for more papers by this author
Kinichi Nakashima

Corresponding Author

Kinichi Nakashima

Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Correspondence to: Kinichi Nakashima and Taito Matsuda, Stem Cell Biology and Medicine, Department of Stem Cell Biology and Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka, 812-8582, Japan. E-mail: [email protected]; [email protected]Search for more papers by this author
First published: 02 August 2017
Citations: 22

Abstract

Background: Although quiescent neural stem cells (NSCs) in the adult hippocampus proliferate in response to neurogenic stimuli and subsequently give rise to new neurons continuously throughout life, misregulation of NSCs in pathological conditions, including aging, leads to the impairment of learning and memory. High mobility group B family 1 (HMGB1) and HMGB2, HMG family proteins that function as transcriptional activators through the modulation of chromatin structure, have been assumed to play some role in the regulation of adult NSCs; however, their precise functions and even expression patterns in the adult hippocampus remain elusive. Results: Here we show that expression of HMGB2 but not HMGB1 is restricted to the subset of NSCs and their progenitors. Furthermore, running, a well-known positive neurogenic stimulus, increased the proliferation of HMGB2-expressing cells, whereas aging was accompanied by a marked decrease in these cells. Intriguingly, HMGB2-expressing quiescent NSCs, which were shifted toward the proliferative state, were decreased as aging progressed. Conclusions: HMGB2 expression is strongly associated with transition from the quiescent to the proliferative state of NSCs, supporting the possibility that HMGB2 is involved in the regulation of adult neurogenesis and can be used as a novel marker to identify NSCs primed for activation in the adult hippocampus. Developmental Dynamics 247:229–238, 2018. © 2017 Wiley Periodicals, Inc.

Introduction

The adult mammalian brain retains slowly dividing quiescent neural stem cells (NSCs) that continuously generate new neurons in the subventricular zone (SVZ), and the subgranular zone (SGZ) of the hippocampal dentate gyrus (DG) (Mira et al., 2010; Ming and Song, 2011). As newly generated neurons in the DG integrate into the existing neuronal circuits (van Praag et al., 2002; Song et al., 2016) and influence learning and memory (Deng et al., 2009; Zhuo et al., 2016), adult neurogenesis is important for supporting proper brain functions. At the beginning of the process of adult neurogenesis in the DG, quiescent NSCs proliferate to maintain their pool and generate neural progenitor cells (NPCs) that have high proliferative capacity. These dividing NPCs then give rise to new neurons, which in turn become mature neurons in the DG (Urbán and Guillemot, 2014). Adult neurogenesis is dynamically regulated by many physiological stimuli, e.g., physical exercise is known to increase proliferation of NSCs and NPCs (NS/PCs) (van Praag et al., 1999). In contrast, various pathological insults, including aging, lead to significant reduction in the proliferation and numbers of NS/PCs (Kuhn et al., 1996; Encinas et al., 2011; Seib et al., 2013). Therefore, understanding the molecular mechanisms underlying adult NS/PCs regulation and neurogenesis should advance our knowledge of neural development and its plasticity, and eventually enable us to develop new strategies for regenerative medicine, including the treatment of brain disorders.

High mobility group (HMG) proteins are nonhistone chromatin proteins that contribute to the regulation of gene expression by increasing the accessibility of the chromatin for transcription factors (Thomas and Travers, 2001). They have been reported to play a key role in the differentiation of NSCs. For instance, HMG AT-hook 2 (Hmga2) is highly expressed in the ventricular zone of the embryonic brain, where NSCs reside (Sanosaka et al., 2008). Furthermore, HMGA1 and HMGA2 promote neuronal differentiation while they inhibit astrocyte differentiation of NSCs during embryonic stages (Kishi et al., 2012).

Other HMG proteins, HMG nucleosome-binding protein 1 (HMGN1), HMGN2, and HMGN3, are highly expressed in both NSCs and astrocytes, and have been reported to promote astrocyte differentiation of NSCs (Nagao et al., 2014). By using single-cell RNA-sequencing, a recent study showed that HMG mRNAs are also highly expressed in adult NSCs (Shin et al., 2015). In that study, among the HMG family members, Hmgb1 and Hmgb2 were listed as especially highly up-regulated genes during the activation of quiescent NSCs in the adult DG. However, their precise expression patterns and functions in the adult DG have yet to be elucidated, although HMGB2 has been reported to regulate NSCs proliferation in the SVZ of adult mice (Abraham et al., 2013a).

Considering the above findings, here we examined in detail the expression patterns of HMGB1 and HMGB2 during the process of adult hippocampal neurogenesis. In contrast to the homogeneous expression of HMGB1 in NS/PCs and neurons, we found that HMGB2 is restrictedly expressed in the subset of NS/PCs but not in neurons. We further show that running increased HMGB2-expressing cells whereas aging decreased them. Interestingly, HMGB2-expressing quiescent NSCs seemed to be primed toward an activated state, and these cells were reduced in aged mice. We also found that forced expression of HMGB2 in NS/PCs inhibited their cell cycle withdrawal caused by diazepam, which is known to induce quiescence in NS/PCs (Song et al., 2012). Taken together, these findings raise the possibility that HMGB2 participates in the activation of quiescent NSCs for the maintenance of proper neurogenesis in the adult hippocampus.

Results

HMGB1 and HMGB2 is Expressed in the Adult NS/PCs

We first confirmed the distribution of HMGB1- or HMGB2-expressing cells in the DG of 2-month-old mice by immunostaining. HMGB1 expression was homogenously observed in the entire granule cell layer (GCL) of the DG, while the expression of HMGB2 was enriched in the inner side of the GCL, where NS/PCs are known to reside (Fig. 1A–C). To clarify more precisely which types of cells express HMGB1 and HMGB2 in the DG, HMGB1 and HMGB2 were co-labeled with antibodies against glial fibrillary acidic protein (GFAP) and Nestin, markers for NSCs that show radial glia-like morphology (Seri et al., 2001). While HMGB1 was expressed in almost all Nestin- and GFAP-positive (Nestin+ GFAP+) NSCs (Fig. 2A,B), HMGB2 was expressed in 72% of Nestin+ GFAP+ NSCs (Fig. 2C,D). As Nestin+ GFAP+ NSCs are composed of both quiescent and activated NSCs with radial morphology, antibody against proliferating cell marker Ki67 was used to distinguish these two types of NSCs and to narrow down the HMGB2+ cell populations. Hereafter, we refer to Nestin+ cells with radial morphology as NSCs, unless otherwise mentioned.

Details are in the caption following the image

HMGB1 and HMGB2 expression in the adult hippocampal DG. A: Representative images of staining for HMGB1 (green) or HMGB2 (green) and Hoechst staining (gray, insets) in the DG. B: Relative intensity of fluorescence for HMGB1 or HMGB2-staining in (A). Color ranges from blue (low expression of HMGB1 or HMGB2) to green, yellow, and finally red (high expression of HMGB1 or HMGB2). C: The magnified images of the white dashed boxes in (B). White broken lines enclose granule cell layers. Scale bars = 100 μm.

We observed that 74% of Nestin+ Ki67- quiescent NSCs expressed HMGB2, whereas almost all Nestin+ Ki67+ activated NSCs expressed HMGB2 (Fig. 2E,F), suggesting that quiescent NSCs may start to express HMGB2 before they shift toward an actual activated state. To confirm this possibility, we also monitored proliferating cells by injection of bromodeoxyuridine (BrdU) into 2-month-old mice once a day for 7 days. We found that among Nestin+ NSCs, all of the BrdU+ cells were HMGB2+ cells, although only 30% of all HMGB2+ Nestin+ NSCs were BrdU+ (Fig. 2G–I). In other words, HMGB2- Nestin+ NSCs were not labeled with BrdU at all (Fig. 2G–I), and thus it seems most likely that HMGB2+ cells are NSCs which are primed toward their activated state.

Details are in the caption following the image

HMGB1 and HMGB2 expression in NSCs in the adult DG. A: Representative image of staining for HMGB1 (green), Nestin (red), and GFAP (blue). Arrowheads indicate HMGB1+ Nestin+ GFAP+ cells. B: Percent of HMGB1+ cells among Nestin+ GFAP+ NSCs. C: Representative image of staining for HMGB2 (green), Nestin (red), and GFAP (blue). Yellow arrowhead and white arrowhead indicate HMGB2+ Nestin+ GFAP+ cell and HMGB2- Nestin+ GFAP+ cell, respectively. D: Percent of HMGB2+ cells among Nestin+ GFAP+ NSCs. E: Representative merged image of staining for HMGB2 (red), Nestin (cyan), and Ki67 (green). Yellow arrowhead and white arrowhead indicate HMGB2+ Nestin+ Ki67- cell and HMGB2+ Nestin+ Ki67+ cell, respectively. F: Percent of HMGB2+ cells among Nestin+ Ki67- or Ki67+ cells. G: Representative image of staining for BrdU (red), Nestin (cyan), and HMGB2 (green) in NSCs. Arrowheads indicate BrdU+ Nestin+ HMGB2+ cells. H: Representative image of staining for BrdU (red), Nestin (cyan), and HMGB2 (green) staining in NSCs. Arrowhead indicates BrdU- Nestin+ HMGB2+ cell. I: Percent of BrdU+ cells among Nestin+ HMGB2- or Nestin+ HMGB2+ cells. Scale bars = 20 μm. n = 3, error bars represent SEM.

Next, to investigate which stages of NS/PCs during neuronal differentiation express HMGB1 and HMGB2 in the neurogenic area, we stained the brain sections with anti-HMGB1 or -HMGB2 antibody together with antibodies for Ki67 and an immature neuron marker, doublecortin (DCX). Most of the Ki67+ DCX- cells, Ki67+ DCX+ cells, and Ki67- DCX+ cells expressed both HMGB1 (Fig. 3A,B) and HMGB2 (Fig. 3C,D). However, we found that NeuN+ mature neurons did not express HMGB2 (Fig. 3G,H) although they expressed HMGB1 (Fig. 3E,F). Taken together, our findings indicate that HMGB2 expression is increased in NSCs during their activation and persists until the early neuronal maturation phase (Fig. 3I). In contrast, HMGB1 expression is continuous during the whole process of neuronal differentiation (Fig. 3I).

Details are in the caption following the image

HMGB1 and HMGB2 expression in NPCs, immature and mature neurons in the adult DG. A: Representative image of staining for HMGB1 (green), DCX (red), and Ki67 (cyan) in the adult DG. B: Percent of HMGB1+ cells among Ki67+ DCX- cells, Ki67+ DCX+ cells, and Ki67- DCX+ cells. C: Representative image of staining for HMGB2 (green), DCX (red), and Ki67 (cyan) in the DG. D: Percent of HMGB2+ cells among Ki67+ DCX- cells, Ki67+ DCX+ cells, and Ki67- DCX+ cells. E: Representative image of staining for HMGB1 (green) and NeuN (red) in the DG. F: Percent of HMGB1+ cells among NeuN+ cells. G: Representative merged image of staining for HMGB2 (green) and NeuN (red) in the DG. H: Percent of HMGB2+ cells among NeuN+ cells. Scale bar = 20 μm. n = 3, error bars represent SEM. I: Schematic illustration of expression of cell type specific markers, HMGB1, and HMGB2 during the process of adult neurogenesis in the DG.

Voluntary Running Increases HMGB2+ Proliferating NPCs

Voluntary running is known to increase proliferation of NPCs (van Praag et al., 1999; Naylor et al., 2008). To examine whether HMGB2+ cells respond to this positive neurogenic stimulus, we provided mice with a running wheel for 2 weeks when they reached 8 weeks old and observed a statistically significant higher number of Ki67+ cells in the runner compared with control mice (Fig. 4A,B), as shown in a previous report (Garrett et al., 2012). Furthermore, the number of double-positive Ki67+ HMGB2+ cells in the DG was significantly increased in the runner compared with control mice (Fig. 4A,B). To assess whether proliferation of NSCs was also elevated in the runners, we counted the number of Ki67+ proliferating HMGB2+ NSCs and found no statistically significant difference between control and runner mice, although there was a trend toward a higher number in the latter (Fig. 4C,D). In addition, there was no significant difference between the percent HMGB2+ quiescent NSCs or the percent HMGB2+ proliferating NSCs in control compared with runner mice (Fig. 4C,E). These data suggest that voluntary running expands the population of HMGB2+ proliferating NPCs without affecting the transition of quiescent NSCs toward their activated state.

Details are in the caption following the image

Increase of HMGB2+ cells in response to running stimulus in the adult DG. A: Representative images of staining for Hoechst (gray, inset), Ki67 (red, left), HMGB2 (green, middle), and merged images (right) in the DG of control (top) and runner mice (bottom). B: The number of Ki67+ cells (left), HMGB2+ cells (middle), and Ki67+ HMGB2+ cells (right) in the DG of control and runner mice. C: Representative images of staining for HMGB2 (red), Nestin (blue), and Ki67 (green) in the DG of control (top) and runner mice (bottom). D: The number of Nestin+ Ki67+ HMGB2+ cells in the DG of control and runner mice. E: The percent of HMGB2- or HMGB2+ cells among Nestin+ Ki67- cells (left) and the percent of HMGB2- or HMGB2+ cells among Nestin+ Ki67+ cells (right) in the DG of control and runner mice. Scale bars = 20 μm. n = 5 in each group. Error bars represent SEM. *P < 0.05 (Student's t-test).

HMGB2 Expression is Suppressed in Quiescent NSCs During Aging

We next assessed the behavior of HMGB2+ NS/PCs in response to a negative regulator of neurogenesis. Neurogenesis is greatly reduced by aging (Seki, 2002; Encinas et al., 2011), and this reduction is paralleled by decreased proliferation and disappearance of putative NSCs (Artegiani and Calegari, 2012). We analyzed the expression of HMGB2 in the DG at four different time points (postnatal day [P] 5, P21, 2 months, and 6 months). We observed continuous expression of HMGB2 in the DG; however, the number of HMGB2+ cells as well as Ki67+ cells was gradually decreased over time (Fig. 5A,C).

Details are in the caption following the image

Reduction of HMGB2+ cells during aging. A: Representative images of staining for Ki67 (red, left), HMGB2 (green, middle), and merged images (right) in the DG of P5 (top), P21 (second from top), 2-month-old (second from bottom), and 6-month-old (bottom) mice. Nuclei were also stained (gray, insets). Scale bars = 100 μm. B: Representative images of staining for HMGB2 (red), Nestin (blue), and Ki67 (green) in the DG of 2-month-old (top) and 6-month-old (bottom) mice. Scale bars = 20 μm. C: The number of Ki67+ cells, HMGB2+ cells, and Ki67+ HMGB2+ cells in the DG of 2-month-old and 6-month-old mice. D: The number of Nestin+ Ki67- HMGB2- cells (left) and the number of Nestin+ Ki67- HMGB2+ cells (right) in the DG of 2-month-old and 6-month-old mice. E: The percent of HMGB2- or HMGB2+ cells among Nestin+ Ki67- cells (left) and the percent of HMGB2- or HMGB2+ cells among Nestin+ Ki67+ cells (right) in the DG of 2-month-old and 6-month-old mice. n = 4 in each group. Error bars represent SEM. *P < 0.05 (Student's t-test).

We also observed that both the number of HMGB2+ and that of HMGB2- quiescent NSCs were dramatically decreased in 6-month-old mice compared with 2-month-old mice, coinciding with the reduction of the proportion of HMGB2+ cells among quiescent NSCs (Fig. 5B,D,E). Interestingly, the proportion of HMGB2+ cells among active NSCs was not changed between 2-month-old and 6-month-old mice (Fig. 5E). These findings suggest that HMGB2 is suppressed in quiescent NSCs during aging, which may impair the transition from a quiescent to an activated state of NSCs, resulting in failure of maintaining a quiescent NSCs pool even though HMGB2+ NSCs proliferate normally regardless of age.

HMGB2 Participates in the Activation of aNS/PCs

The neurotransmitter γ-aminobutyric acid (GABA) has been shown to control the activation of quiescent NSCs (Song et al., 2012). In that study, the authors found that quiescence of NSCs in the adult SGZ is promoted by the administration of diazepam, which specifically enhances the response of γ2-subunit-containing GABAA receptors to GABA. In accordance with this notion, diazepam treatment decreased the proliferation of adult hippocampus-derived NS/PCs (aNS/PCs) in vitro (Fig. 6A,B).

Details are in the caption following the image

Inhibition of diazepam-induced quiescence of aNS/PCs by HMGB2 expression. A: Representative images of staining for Hoechst (gray, inset), FLAG (cyan, upper left), GFP (green, upper right), EdU (red, bottom left), and merged images (bottom right) of control (left side) and FLAG-HMGB2-expressing aNS/PCs (right side). Arrowheads indicate GFP+ EdU+ cells. Scale bars = 50 μm. B: The percent of EdU+ GFP+ cells among GFP+ cells. Gray bars and black bar indicates control retrovirus and FLAG-HMGB2-expressing retrovirus infected cells, respectively. Middle and right bars represent diazepam-treated group. n = 4 experiments. Error bars represent ± SEM. *P < 0.05 (Tukey's test).

Next, to examine whether HMGB2 expression influences the activation of NS/PCs, we expressed HMGB2 together with GFP in aNS/PCs in vitro and subsequently stimulated the cells with diazepam. We found that, in the presence of diazepam, the proliferation rate of aNS/PCs overexpressing HMGB2 was higher than that of control diazepam-treated aNS/PCs (Fig. 6A,B), indicating that HMGB2 inhibits the diazepam-induced quiescence of aNS/PCs. These results suggest that HMGB2 participates in the activation of aNS/PCs, possibly for the maintenance of homeostatic neurogenesis in the adult hippocampus.

Discussion

In this study, we have shown the ubiquitous expression of HMGB1 and the restricted expression of HMGB2 in NS/PCs and their progeny in the adult DG. These results may highlight the importance of HMGB2 for the regulation of neurogenesis. However, HMGB1 and HMGB2 have a highly conserved amino acid sequence (>80%), and are known to share similar functions (Thomas and Travers, 2001). For instance, HMGB1 and HMGB2 have been demonstrated to play an important role in inflammation following tissue injury as damage-associated molecular pattern molecules (Lee et al., 2014), although HMGB2 also has various other functions in differentiation programs, including spermatogenesis, erythropoiesis and chondrogenesis (Ronfani et al., 2001; Laurent et al., 2010; Taniguchi et al., 2011). Because HMGB1 is also expressed in NSCs, further investigation using mice deficient for HMGB2 and/or deficient for HMGB1 will be worthwhile to gain a better understanding of exactly how these two molecules participate in adult neurogenesis.

Quiescent NSCs in the adult hippocampus display a radial glia-like morphology with a long process that extends through the GCL (Seri et al., 2001). They are activated or maintain quiescence in response to extrinsic stimuli (Song et al., 2012) or intrinsic signals (Mira et al., 2010). Dysregulation of these NSC behaviors may contribute to pathogenesis of various neuropsychiatric disorders, such as depression (Schoenfeld and Cameron, 2015; Noguchi et al., 2016; Siopi et al., 2016) and schizophrenia (Kim et al., 2012; Yu et al., 2014). Previous work revealed two morphotypes of quiescent NSCs, namely, Type α cells with a relatively long radial process and Type β cells with a shorter one (Gebera et al., 2016), suggesting the existence of heterogeneous cell populations even among quiescent NSCs.

However, distinguishing among heterogeneous quiescent NSCs has been difficult so far using specific marker proteins, and functional differences among them are not understood. In the present study, we demonstrated that quiescent NSCs can be classified into at least two groups: HMGB2-expressing and HMGB2-nonexpressing cells, which show proliferative and nonproliferative properties, respectively. Although we have not examined whether these populations correspond to Type α or Type β cells, this question warrants further investigation. However, our finding that HMGB2-expressing aNS/PCs remain relatively active even in the presence of a quiescence-inducing factor, diazepam (Fig. 6A,B), raises the possibility that HMGB2 expression elicits the activation of quiescent NSCs.

In support of this, HMGB2 was reported to promote the proliferation of satellite cells, which are quiescent stem cells in the muscle, orchestrating muscle regeneration (Zhou et al., 2016). In that report, Insulin-like growth factor-2 mRNA-binding protein (IGF2BP2), a member of the family of IGF2 mRNA-binding proteins known to bind to various RNAs such as IGF2 for enhancement of its translation, was shown to be a downstream target of HMGB2 and to increase the production of proteins encoded by cell cycle-related genes to evoke the activation of quiescent satellite cells. In addition, a recent study has shown that the behavior of NSCs is affected by IGF2BP2 (Jiang et al., 2017).

Considering these observations, we suggest that HMGB2 expression can distinguish between proliferative and nonproliferative quiescent NSC populations as a relevant marker protein and may mediate the activation of quiescent NSCs in the adult hippocampus, possibly through downstream targets such as IGF2BP2. Of note, IGF2BP2 has also been identified as a direct target of HMGA2 in proliferating myoblasts, fibroblasts and mouse embryos (Brants et al., 2004; Cleynen et al., 2007; Li et al., 2012). Because HMGA2 is also known to regulate the adult NSCs in the SVZ (Nishino et al., 2008), we must await future investigations to determine whether HMGB2 and other HMG-family proteins, including HMGA2, cooperatively modulate the behavior of quiescent NSCs.

In contrast with our findings, it has been reported that HMGB2 deficiency leads to an increase in the proliferation of embryonic NSCs and of adult NSCs in the SVZ (Abraham et al., 2013a, 2013b). NSCs located in distinct areas, such as the adult SVZ and SGZ, have been suggested to exhibit different properties. For instance, endothelial cell-derived IGF2 contributes to NSC maintenance in the SVZ but not in the SGZ, although IGF2 is expressed in both regions (Ferrón et al., 2015). Moreover, RNA-binding protein Fragile X Relative Protein 2 specifically controls the NSCs in the SGZ but not in the SVZ (Guo et al., ). Therefore, it is conceivable that HMGB2 has context-dependent functions in distinct neurogenic areas.

In the present study, we also studied the behavior of HMGB2+ cells in the DG in response to both positive and negative neurogenic regulations. Increased proliferative activity of HMGB2+ cells upon running was observed, in line with data from immunohistochemistry of other NS/PC markers such as Ki67 (Fig. 4B) (Kee et al., 2002), Sox2 (Steiner et al., 2006). However, we did not find a selective activation of HMGB2+ radial NSCs such as that shown in Hes5-GFP mice (Lugert et al., 2010), although the number of HMGB2+ proliferating radial NSCs tended to increase (Fig. 4D). Because running preferentially promotes the proliferation of nonradial NPCs (Suh et al., 2007; Klempin et al., 2013; Farioli-Vecchioli et al., 2014), these facts suggest that voluntary running expands the population of HMGB2+ nonradial NPCs.

In contrast to running, aging decreased HMGB2+ cells (Fig. 5C). We also observed the reduction of both HMGB2+ quiescent and HMGB2+ proliferative radial NSCs during aging (Fig. 5D). It has been reported that impaired proliferation of radial NSCs during aging leads to a decrease in the number of nonradial NPCs and newly generated neurons (Encinas et al., 2011). These results reinforce the possibility that HMGB2 primes the change of a quiescent state toward an activated state in radial NSCs, and reduction of these primed cells leads to attenuation of neurogenesis during aging. However, we cannot completely exclude the possibility that HMGB2 expression impairs proliferation of NSCs during aging through the induction of senescence associated secretory phenotype gene expression, as shown in senescent lung cells in vitro (Aird et al., 2016).

We have shown here that anti-HMGB2 immunostaining labels NS/PCs and their progenies that react dynamically to positive and negative neurogenic stimuli in the adult hippocampus. Our findings also suggested that HMGB2-expressing quiescent NSCs may be primed toward the proliferative state, and these cells are reduced during aging. Last but not least, these findings indicate that HMGB2 should be a novel marker for analyzing the behavior of quiescent NSCs and subsequent neurogenic changes for further dissection and better understanding of the complex regulation of NSCs in the adult hippocampus.

Experimental Procedures

Mice

All efforts were made to minimize animal suffering and to reduce the number of animals used. P5, P21, 2 months, and 6 months old C57BL/6N mice were used for this study. Mice were housed under 12-hr light/dark cycle conditions and could access food and water ad libitum. To label the dividing cells, BrdU (50 mg/kg; Sigma-Aldrich, B5002) dissolved in saline was injected intraperitoneally into 2-month-old mice daily for 1 week. Mice were killed 24 hr after the last BrdU injection. For the voluntary running experiments, a Fast Trac amber with Mouse Igloo (Animec, K-3250) was placed in the home cage. These mice were killed 2 weeks later. All experiments were performed in accordance with Guidelines for Proper Conduct of Animal Experiments (Science Council of Japan) and Related Activities in Academic Research Institutions under the jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology of Japan.

Immunohistochemistry

Mice were deeply anesthetized and perfused with phosphate buffered saline (PBS) followed by 4% paraformaldehyde (PFA). Brains were removed and postfixed overnight in 4% PFA at 4 °C. For cryoprotection, the brains were transferred to 15% sucrose in PBS overnight at 4 °C, and then transferred to 30% sucrose in PBS overnight at 4 °C. These brains were embedded in optimal cutting temperature compound (Tissue Tek, Sakura Finetek, 25608-930) and stored at -80 °C. Frozen brains were serially sectioned at 40 μm thickness. For immunohistochemistry, sections were washed with PBS and incubated with blocking buffer (3% fetal bovine serum [FBS] and 0.1% Triton X-100 in PBS) for 1 hr at room temperature. Sections were then incubated with blocking buffer containing primary antibodies overnight at 4 °C. These sections were washed with PBS 3 times and incubated with blocking buffer containing secondary antibodies for 2 hr at room temperature.

Primary antibodies used in this study were as follows: rabbit anti-HMGB1 (1:500; abcam, ab18256); rabbit anti-HMGB2 (1:500; abcam, ab67282); chicken anti-Nestin (1:500; Aves Labs, NES); mouse anti-Nestin (1:500; Millipore, MAB353); chicken anti-GFAP (1:500; Millipore, AB5541); mouse anti-Ki67 (1:500; BD Biosciences, 550609); goat anti-DCX (1:200; Santa Cruz Biotechnology, sc-8066); mouse anti-NeuN (1:500; Millipore, MAB377); rat anti-BrdU (1:500; Ab D Serotec, OBT0030). Secondary antibodies used in this study were as follows: CF488 donkey anti-mouse IgG (H+L), highly cross-adsorbed (1:500; Biotium, 20014); CF543 donkey anti-rabbit IgG (H+L), highly cross-adsorbed (1:500; Biotium, 20038); CF647 goat anti-chicken IgY (H+L), highly cross-adsorbed (1:500; Biotium, 20044); CF647 donkey anti-goat IgG (H+L), highly cross-adsorbed (1:500; Biotium, 20048); CF647 donkey anti-rabbit IgG (H+L), highly cross-adsorbed (1:500; Biotium, 20047); and CF568 donkey anti-rat IgG (H+L), highly cross-adsorbed (1:500; Biotium, 20092). Hoechst 33258 (1:500; Nacalai Tesque) was used for nuclear staining.

After sections were washed with PBS 3 times and mounted, image photos were taken using a Zeiss LSM 700 confocal microscope and processed using Photoshop.

For BrdU staining, sections were incubated with 2N HCl for 15 min at 37 °C before blocking. Ki67 staining was performed after antigen retrieval using target retrieval solution (DAKO) for 15 min at 105 °C.

Cell Counts

Cell counts were determined by using every 6th hemisphere section from rostral to caudal which contained the DG. The number of marker-positive cells of every 6th section was counted, and the counted number was multiplied by 6 to calculate the total number of cells in the brain.

Cell Culture

aNS/PCs derived from rat hippocampus were plated on poly-L-ornithine/laminin-coated dishes in Dulbecco's modified eagle's medium (DMEM)/F-12 containing N2 supplement (Gibco, 7502048) and 20 ng/mL of basic fibroblast growth factor (bFGF; PeproTech, 100-18B), under 5% CO2 at 37 °C. Plat-E cells were maintained in DMEM with 10% FBS (Biowest, S1820) under 5% CO2 at 37 °C.

Retroviral Constructs, Preparation of Retrovirus, and Virus Infection

HMGB2 sequence was cloned into pMY vector (pMY-IRES-EGFP), which was a gift from Dr. T. Kitamura (Kitamura et al., 2003). To identify HMGB2 expression, we inserted FLAG sequence (DYKDDDDK) at the N-terminus of the HMGB2 sequence (pMY-FLAG-HMGB2-IRES-EGFP). To prepare retrovirus, The Plat-E packaging cells were transfected with retrovirus constructs, and then culture supernatants were collected 48 hr after transfection.

Proliferation Assay

aNS/PCs derived from rat hippocampus were passaged at 2 days after viral infection. To induce quiescence of these cells, we added 60 μM diazepam (Wako, 045-18901) and cultured the cells for 2 days in the presence of bFGF. To monitor the cell proliferation, 10 μM EdU (5-ethynyl-2′-deoxyuridine; Click-iT EdU Imaging Kit, Invitrogen, C10338) was added to the medium 30 min before fixation.

Immunocytochemistry

The cells were fixed with 4% PFA for 20 min, and then they were washed with PBS 3 times and incubated with blocking buffer for 30 min at room temperature. EdU staining was done before incubation with primary antibody according to the manufacturer's instructions. Cells were then incubated with blocking buffer containing primary antibody for 1.5 hr at room temperature. The cells were then washed with PBS 3 times and incubated with blocking buffer containing secondary antibody for 1.5 hr at room temperature. Primary antibodies used in this study were as follows: mouse anti-FLAG (1:500; Sigma, F1804); chicken anti-green fluorescent protein (GFP; 1:500; Aves Laboratories, GFP-1020). Secondary antibodies used in this study were as follows: CF488A donkey anti-chicken IgY (H+L), highly cross-adsorbed (1:500; Biotium, 20166); CF647 donkey anti-mouse IgG (H+L), highly cross-adsorbed (1:500; Biotium, 20046). Hoechst 33258 (1:500; Nacalai Tesque) was used for nuclear staining. Image photos were taken using a Leica AF600 fluorescence microscope and processed using Photoshop.

Statistical Analysis

The unpaired, two-tailed t-test was used. In Figure 6B, Tukey's test was used following one-way analysis of variance. All data are presented as mean + SEM.

Acknowledgments

T.M. was supported by a Grant-in-Aid for JSPS Fellows and K.N. was funded by Grants-in-Aid for Scientific Research on Innovative Areas. We thank Dr. F.H. Gage for aNS/PCs derived from rat hippocampus; Dr. T. Kitamura for pMY vector and Plat-E cells; Drs. T. Imamura, and S. Katada for valuable discussions; Y. Nakagawa for excellent secretarial assistance; and Dr. E. Nakajima for proofreading the manuscript. We appreciate the technical assistance from The Research Support Center, Research Center for Human Disease Modeling, Kyushu University Graduate School of Medical Sciences.