Interaction between Angiotensin Type 1, Type 2, and Mas Receptors to Regulate Adult Neurogenesis in the Brain Ventricular–Subventricular Zone

The renin–angiotensin system (RAS), and particularly its angiotensin type-2 receptors (AT2), have been classically involved in processes of cell proliferation and maturation during development. However, the potential role of RAS in adult neurogenesis in the ventricular-subventricular zone (V-SVZ) and its aging-related alterations have not been investigated. In the present study, we analyzed the role of major RAS receptors on neurogenesis in the V-SVZ of adult mice and rats. In mice, we showed that the increase in proliferation of cells in this neurogenic niche was induced by activation of AT2 receptors but depended partially on the AT2-dependent antagonism of AT1 receptor expression, which restricted proliferation. Furthermore, we observed a functional dependence of AT2 receptor actions on Mas receptors. In rats, where the levels of the AT1 relative to those of AT2 receptor are much lower, pharmacological inhibition of the AT1 receptor alone was sufficient in increasing AT2 receptor levels and proliferation in the V-SVZ. Our data revealed that interactions between RAS receptors play a major role in the regulation of V-SVZ neurogenesis, particularly in proliferation, generation of neuroblasts, and migration to the olfactory bulb, both in young and aged brains, and suggest potential beneficial effects of RAS modulators on neurogenesis.


Introduction
New neurons destined to integrate into olfactory bulb (OB) circuits are continuously generated from neural stem cells (NSCs) in the ventricular-subventricular zone (V-SVZ) of the lateral ventricles in the adult rodent brain. These self-renewing NSCs exhibit features that are typical of mature astroglial

Laser Capture Microdissection (LCM)
Sections containing the mouse or rat V-SVZ were stained with a shortened protocol for neutral red (1% aqueous solution; 2 min at 4 • C). LCM was performed using a PALM MicroBeam system (Zeiss, Oberkochen, Germany) controlled by PALM Robo software [47,48]. V-SVZ cells (or striatal cells, used as a control) were visualized under bright-field microscopy at 40× magnification. Cell pools were selected and then cut and catapulted by laser pulses into an adhesive microtube cap (Zeiss). Cell pools were immediately lysed in RTL lysis buffer (Qiagen, Hilden, Germany) containing β-mercaptoethanol for 30 min at RT and stored at −80 • C.

RNA Extraction and RT-PCR
In microcaptured mouse and rat cells, total ribonucleic acid (RNA) extraction was performed using the RNeasy Micro kit (Qiagen), and in neurosphere cells was performed using TRIzol (Invitrogen, Thermo Fisher Scientific) according to the manufacturer's protocol. RNA was solubilized in 14 µl of RNAse-free water and transcribed into complementary DNA (cDNA) using the Moloney murine leukemia virus (M-MLV; 200U; Invitrogen, Thermo Fisher Scientific) reverse transcriptase (RT). For polymerase chain reaction (PCR), a reaction mix containing the BioTaq DNA polymerase (Bioline, London, United Kingdom), NH 4 reaction buffer, MgCl 2 , dNTPs (Invitrogen, Thermo Fisher Scientific), and the specific primer pairs (Sigma-Aldrich; see Table 1) were used. Primers were designed for each gene using NCBI Primer-BLAST. PCR was performed using a MultiGene Thermal Cycler (Labnet, Madrid, Spain). Finally, the PCR products were separated using agarose gel electrophoresis and were visualized with a UV detection system (Chemidoc XRS System, BioRad, Hercules, CA, USA). The mRNA expression of AT1 and AT2 receptors was determined by real-time quantitative RT-PCR as described in previous studies [56]. Briefly, PCR was performed using a real-time QuantStudio3 platform (Applied Biosystems, Foster City, CA, USA), the EvaGreen qPCR MasterMix (Applied Biological Materials Inc., Vancouver, Canada), and the primer sequences indicated in Table 1. The data were evaluated by the delta Ct method (2 -∆∆Ct ), where Ct is the cycle threshold. Expression of each gene was obtained as relative to the housekeeping (β-actin) transcripts, and the relative amount of mRNA was presented in the form of fold change over control. AGT was purified and detected in neurosphere growth culture media as previously described [57] with slight modifications. Culture media (1 mL) were collected, heated at 37 • C, and injected (20 µL/injection) into a Shimadzu HPLC system (Shimadzu, Kyoto, Japan). AGT was separated on a reverse phase column (Perkin Elmer Aquapore ODS 300, Perkin Elmer, Madrid, Spain) at 37 • C and a flow rate of 1.5 mL/min by using a two-step separation method consisting in an initial isocratic elution (5% of acetonitrile with 0.1% phosphoric acid for 5min) and then a linear gradient from 5% to 66.5% of acetonitrile with 0.1% phosphoric acid (30 min). The AGT fraction was visualized at 240 nm with a UV-VIS detector (Shimadzu SPD-20 AV, Shimadzu), collected in a fraction collector (Shimadzu FRC-1 A, Shimadzu), and dried in a vacuum concentrator (Savant ISS110, Thermo Fisher Scientific). The pellet was resuspended in RIPA buffer containing a protease inhibitor cocktail (Sigma-Aldrich) and the presence of AGT in the fraction was confirmed by WB analysis. Briefly, proteins were separated using 5%-10% Bis-Tris polyacrylamide gels and transferred to nitrocellulose membranes. The membranes were incubated overnight with a goat primary antibody against AGT (1:200; sc-7419; Santa Cruz Biotechnology) and then for 1 h with the HRP-conjugated secondary antibody donkey anti-goat IgG (1:2500; sc-2020; Santa Cruz Biotechnology). Immunoreactivity was detected with a Lumminata Crescendo HRP Chemiluminescent Kit (Millipore) and visualized with a chemiluminescence detection system (ChemiDoc XRS System, BioRad, Hercules, CA, USA). AT1 and AT2 protein levels were detected in mouse V-SVZ and striatum (i.e., positive control) by WB analysis (see above) using the corresponding primary antibodies from Santa Cruz Laboratories (1:200): goat polyclonal anti-AT1 (sc-31181) and rabbit polyclonal anti-AT2 (sc-9040), previously validated [41]. GAPDH (rabbit polyclonal antibody; 1:50000; G9545; Sigma-Aldrich) was used as a loading control.

Statistical Analysis
Two-group comparisons were carried out by using Student's t-test. Multiple comparisons were analyzed by one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test. The normality of populations and homogeneity of variances were tested before each analysis of variance. All culture data were obtained from at least three separated experiments and normalized to the values of the control group of the same batch (100%). The results are presented as means ± standard error of the means (SEM). Differences were considered statistically significant at p < 0.05. All statistical analyses were performed with SigmaPlot 11.0 (Systat Software Inc., San Jose, CA, USA).

AT2 Receptors Mediated Promoting Effects of Ang II on Neurosphere Formation
As a first approach to the analysis of local RAS in the V-SVZ murine niche, we isolated subventricular cells from young adult mice and plated them as single cells in serum-free growth medium containing mitogens EGF and FGF-2 to produce neurospheres, floating clonal aggregates that can be expanded through subculture ( Figure 1A). Neurospheres can be formed by NSCs and NPCs and constitute an ideal system to evaluate the action of signaling pathways in proliferation and self-renewal [49]. The endogenous expression of AGT, AT1, and AT2 receptors by neurosphere cells was confirmed in cell lysates by PCR ( Figure 1B). Moreover, the presence of the angiotensinogen (ANG) protein in the medium conditioned by the cells during 5 days was demonstrated by combining HPLC-based extraction and WB [57] ( Figure 1C). Because the results indicated the possibility of RAS actions in neurospheres, we next seeded single cells dissociated from neurospheres in medium with and without 100 nM Ang II combined with concurrent administration of AT1 and AT2 receptor antagonists, 1 µM. In particular, we used peptide antagonist ZD7155 (ZD from now on) and non-peptide antagonist candesartan to block AT1 receptors and peptide PD123319 (PD from now on) to block AT2 receptors. AT1 receptor inhibition did not produce any change in the numbers of neurospheres, but we found reduced numbers of neurospheres in the cultures treated with the AT2 receptor inhibitor, independently of the exogenous addition of Ang II, indicating that the AT2 receptor mediated promoting effects of endogenous and exogenously added Ang II in neurosphere formation ( Figure 1D). We did not find, however, any significant differences in sphere diameters among treatments, suggesting that AT2 plays a role in neurosphere initiation but not growth ( Figure 1E).
To directly test that AT2 receptors indeed promote neurosphere formation, we plated individual neurosphere cells in the presence of two different specific agonists. Treatment with agonist CGP42112A (CGP from now on) induced an increase in the number of spheres generated, which was blocked by the AT2 receptor antagonist PD, and similar results were obtained with the non-peptide AT2 receptor agonist C21 ( Figure 1F). To analyze whether the AT2 receptor-mediated positive effects on neurosphere numbers included effects on self-renewal, we re-plated cells obtained from neurospheres that had been grown in the presence of the AT2 receptor agonists and antagonists in growth medium without any specific treatment and evaluated neurosphere formation. In the cultures in which we had withdrawn the AT2 receptor modulators (i.e., derived from neurospheres pre-treated with AT2 agonist; 'pre-treated' condition), we could obtain information about the type of cell division that was promoted by the drugs when they were present ('treated' condition; Figure 1F) [49]. We observed reduced proportions of new neurospheres formed by cells that had been expanded in their previous passage in the presence of AT2 receptor agonists ( Figure 1G). These data suggested that AT2 receptor activation induces asymmetrical division of neurosphere-forming cells, an effect that results in increased numbers of neurospheres that are, however, less clonogenic (i.e., asymmetrical divisions lead to a loss of stemness capacity and reduced ability to generate new neurospheres, i.e., tertiary neurospheres).  Representative bands for angiotensinogen (ANG), Ang II type-1 (AT1) and type-2 (AT2) receptors, and β-actin obtained by RT-PCR in neurospheres (NF). Homogenates of striatum (ST) were used as a positive control. (C) ANG was detected in neurosphere culture medium by HPLC and visualized by western blot. ANG (250 µg/mL) was used as a positive control (C+). Bar graphs showing the number (D) and diameter (E) of neurospheres after treatment with angiotensin II (Ang II), AT1 receptor antagonist (ZD7155 or candesartan), and AT2 receptor antagonist (PD123319). Histograms showing the number of neurospheres after treatment with AT2 receptor agonists (CGP42112A or C21) and the AT2 antagonist PD123319 ((F); treatment) or in cultures derived from neurospheres pre-treated with AT2 agonists and reseeded in the absence of any treatment ((G); neurospheres derived from cultures pre-treated with AT2 agonists; pre-= previously treated with). All culture data were obtained from at least three separated experiments. Data are means ± standard error of the mean (SEM). * p < 0.05 relative to control (untreated) group (one-way ANOVA and Bonferroni post hoc test.). SEM = standard error of the mean. ANOVA = analysis of variance. bp = base pairs. Scale bar: 150 µm.

AT2 Receptor Agonist C21 Stimulated Activity in the V-SVZ Niche
Our data indicated that AT2 receptor activation induces proliferation of neurosphere-forming cells in the V-SVZ. We, therefore, decided to analyze the in vivo effects of the pharmacological activation of the AT2 receptor using its non-peptide agonist C21. We analyzed the cell composition of V-SVZs dissected from WT mice treated with C21 using combinations of multiple markers and FACS technology [46,[58][59][60]. After discarding non-neurogenic cells, the remaining cell pool was sub-divided using a panel of markers that included antibodies to GLAST (astroglial cells), CD24 (neuroblasts), and CD9 (higher in NSCs than in astrocytes), as well as fluorescently-labeled EGF, which labels activated cells within the lineage, as previously described [45]. We classified the cells into EGFR + proliferating and EGFR-postmitotic GLAST-CD24 high neuroblasts, EGFR + proliferating NPCs (GLAST-CD24 low and GLAST + CD24 high cells), and CD9 high GLAST + CD24 low/neg NSCs [45] (Figure 2A,B). After a 15 day treatment with C21, we found a significant increase in the total population of NSCs and of NPCs ( Figure 2C). In line with these data, we observed higher proportions of neuroblasts arriving at the OB ( Figure 2C). The C21 effect was transient, as we could not observe any signs of increased activation one month after the treatment ( Figure 2D). The data together further indicated that pharmacological activation of the AT2 receptor increased V-SVZ activity, and that the effect was present during a chronic treatment (i.e., 15 days) but did not continue for a long period if the treatment was stopped. pre-treated with AT2 agonists; pre-= previously treated with). All culture data were obtained from at least three separated experiments. Data are means ± standard error of the mean (SEM). * p < 0.05 relative to control (untreated) group (one-way ANOVA and Bonferroni post hoc test.). SEM = standard error of the mean. ANOVA = analysis of variance. bp = base pairs. Scale bar: 150 μm.

AT2 Receptor Agonist C21 Stimulated Activity in the V-SVZ Niche
Our data indicated that AT2 receptor activation induces proliferation of neurosphere-forming cells in the V-SVZ. We, therefore, decided to analyze the in vivo effects of the pharmacological activation of the AT2 receptor using its non-peptide agonist C21. We analyzed the cell composition of V-SVZs dissected from WT mice treated with C21 using combinations of multiple markers and FACS technology [46,[58][59][60]. After discarding non-neurogenic cells, the remaining cell pool was subdivided using a panel of markers that included antibodies to GLAST (astroglial cells), CD24 (neuroblasts), and CD9 (higher in NSCs than in astrocytes), as well as fluorescently-labeled EGF, which labels activated cells within the lineage, as previously described [45]. We classified the cells into EGFR + proliferating and EGFRpostmitotic GLAST -CD24 high neuroblasts, EGFR + proliferating NPCs (GLAST -CD24 low and GLAST + CD24 high cells), and CD9 high GLAST + CD24 low/neg NSCs [45] ( Figure  2A,B). After a 15 day treatment with C21, we found a significant increase in the total population of NSCs and of NPCs ( Figure 2C). In line with these data, we observed higher proportions of neuroblasts arriving at the OB ( Figure 2C). The C21 effect was transient, as we could not observe any signs of increased activation one month after the treatment ( Figure 2D). The data together further indicated that pharmacological activation of the AT2 receptor increased V-SVZ activity, and that the effect was present during a chronic treatment (i.e., 15 days) but did not continue for a long period if the treatment was stopped.

AT1 Receptor-Dependent Restraining Action on Adult Neurogenesis Was Counteracted by Stimulatory AT2 Receptor Activity
Although AT2 receptors appeared to play a significant role in the activation of neurosphere-initiating cells by endogenous Ang II, complex interactions between AT1 and AT2 receptors were observed in different tissues and, therefore, we turned to mice that carry specific deletions in the genes coding for AT1 and AT2 receptors to dissect individual and combined functions in vivo. AT1-KO and AT2-KO homozygous mutant mice and WT littermates were injected i.p. with BrdU (four pulses at 50 mg/kg, every 2 h) and sacrificed 24 h after the last injection to detect proliferating cells. Consistent with our in vitro data, we did not observe any significant change in the number of BrdU-ir and DCX-ir cells in AT1-KO mice, whereas both parameters were significantly reduced in AT2-KO mice relative to control WT mice ( Figure 3A-D). These data further indicated that the AT2 receptor is a NSC positive regulator and demonstrated its essential role in vivo.

Cross-Regulation of AT1 and AT2 Receptors Underlies Ang II Effects on Neurosphere-Forming Cells
Because our data suggested complex interactions between AT1 and AT2 receptors in the V-SVZ, and because a number of studies in different cells and tissues have shown reciprocal regulation between AT1 and AT2 receptors [61][62][63], we next set up to evaluate expression levels of these two receptors. We visually selected cells of the V-SVZ in lightly stained brain sections of WT, as well in We next employed the brain-permeable AT1 receptor antagonist candesartan and the AT2 receptor agonist C21. First, we treated young WT mice with the AT2 agonist C21 or vehicle for 15 days, and on the last day of treatment the animals received BrdU as described above. Specific activation of the AT2 receptor with the C21 agonist increased the number of BrdU-ir and DCX-ir cells in the V-SVZ relative to vehicle-treated WT mice, as expected ( Figure 3E,F). In line with our previous results, no significant changes were observed when candesartan was administered to WT mice, however, interestingly, the reductions observed in the AT2-KO mice were partially restored by the AT1 receptor inhibitor ( Figure 3E,F). These data indicated that AT1 receptors inhibited proliferation in the V-SVZ, and that this inhibitory effect was normally blocked by the AT2 receptor activity (i.e., the AT1 inhibitory effect is mostly blocked by AT2 in basal conditions).

Cross-Regulation of AT1 and AT2 Receptors Underlies Ang II Effects on Neurosphere-Forming Cells
Because our data suggested complex interactions between AT1 and AT2 receptors in the V-SVZ, and because a number of studies in different cells and tissues have shown reciprocal regulation between AT1 and AT2 receptors [61][62][63], we next set up to evaluate expression levels of these two receptors. We visually selected cells of the V-SVZ in lightly stained brain sections of WT, as well in AT1-KO and AT2-KO mice, which were subsequently isolated by laser-mediated capture. RT-PCR detection of AT1 and AT2 receptors with specific primers (see Table 1) in the cell pools of AT1-KO mice or of WT mice that had been treated with candesartan revealed a significant decrease in Agtr2 mRNA expression, indicating that AT1 receptor activity upregulates the expression of AT2 receptors in the V-SVZ ( Figure 4A). In contrast, cells from AT2-KO mice showed a significant increase in Agtr1 mRNA expression, indicating that AT2 receptors downregulate expression of AT1 receptors in this niche ( Figure 4A). In line with these results, treatment of WT neurosphere cultures with the AT2 antagonist PD increased the levels of Agtr1 mRNA, whereas the AT1 antagonist candesartan reduced Agtr2 mRNA levels ( Figure 4B).
We then decided to test in vitro the pharmacological conditions used in vivo by using our angiotensin receptor modulators in neurosphere cultures obtained from AT1 and AT2 receptor-deficient mice. We dissected the V-SVZ of WT, AT2-KO, and AT1-KO young mice; obtained homogenates; and seeded equal numbers of isolated cells in growth medium to determine primary neurosphere yield. We found that AT1-KO tissue yielded similar numbers of neurospheres as WT tissue, whereas the samples from AT2-KO produced significantly fewer clones, a further indication of the role of AT2 receptors in mediating the stimulatory effects of Ang II on the V-SVZ ( Figure 4C). Despite these changes, no significant differences were detected size of the clones in any of the genotypes (diameter, in µm ± SEM: 106.8 ± 2.8 in WT, 106.0 ± 3.3 in AT1-KO, and 108.8 ± 2.9 in AT2-KO primary spheres). To analyze the effects of Ang II mediated by each receptor, we next plated AT1-KO and AT2-KO cells in regular neurosphere medium with and without 100 nM Ang II and measured the number of clones formed after 5 days. In line with the pharmacological profile shown before, AT1-KO cultures did not respond to exogenously added Ang II; however, inhibition of the AT2 receptor with PD did not result in reduced neurosphere formation as in WT cultures ( Figure 4D). Significantly fewer neurospheres were formed in the AT2-KO cultures treated with Ang II, however, interestingly, AT1 receptor antagonist ZD restored neurosphere generation to untreated levels ( Figure 4D). These data indicated that AT1 receptor set a basal inhibition of proliferation that was antagonized by the AT2 receptor, likely through its repressive effects on AT1 receptor levels.
The data together suggest that endogenous and exogenously administered Ang II stimulated neurosphere formation through a complex balance of AT1 and AT2 receptor activities. AT2 receptor activation stimulated proliferation, whereas AT1 receptor activation mediated a basal inhibition of Ang II on proliferation that was counteracted by a repressive action of AT2 receptor-dependent signaling on Agtr1 expression. In turn, AT1 activity stimulated expression of Agtr2, generating a negative feedback which set a basal level of Ang II-dependent regulation of neurogenesis.
The data together suggest that endogenous and exogenously administered Ang II stimulated neurosphere formation through a complex balance of AT1 and AT2 receptor activities. AT2 receptor activation stimulated proliferation, whereas AT1 receptor activation mediated a basal inhibition of Ang II on proliferation that was counteracted by a repressive action of AT2 receptor-dependent signaling on Agtr1 expression. In turn, AT1 activity stimulated expression of Agtr2, generating a negative feedback which set a basal level of Ang II-dependent regulation of neurogenesis.  (B) AT1 expression was higher in neurospheres treated with the AT2 antagonist PD123319 than in controls, and AT2 expression was lower in neurospheres treated with the AT1 antagonist candesartan than in controls. The number of primary neurospheres obtained from WT, AT1-KO, and AT2-KO mice is shown in (C). (D) Bar graphs showing the number of neurospheres derived from AT1-KO or AT2-KO mice after the corresponding treatments. All culture data were obtained from at least three separate experiments. Data are means ± SEM. * p < 0.05 relative to controls (Student's t-test or one-way ANOVA and Bonferroni post hoc test). Ang II = angiotensin II; CGP42112A = AT2 receptor agonist; ZD7155 = AT1 receptor antagonist.

Stimulatory Effects of AT2 Receptor Activation on Young and Aged NSC Involved MasR
Because activation of the AT2 receptor can stimulate the proliferation of NSCs in young mice, we decided to analyze whether AT2 receptor agonists could also improve the capacity of aged NSCs in generating neurospheres in vitro as a way to evaluate the potential of AT2 activation in elderly mice. In line with previous reports [64], V-SVZs isolated from aged mice yielded reduced numbers of primary neurospheres relative to the young control group ( Figure 5A). However, there was no significant difference in neurosphere diameter between groups (in µm ± SEM: 103.4 ± 3.2 in cultures derived from young mice and 100.4 ± 4.3 in cultures derived from aged mice). When primary cultures from elderly mice were subcultured, treatment with the AT2 receptor agonist CGP induced a significant increase in the number of generated spheres, which was similar to that observed in young untreated cultures. This increase was blocked by treatment with the AT2 receptor antagonist PD ( Figure 5B). When CGP-treated secondary neurospheres generated from aged mice were reseeded in the absence of any treatment (i.e., neurospheres derived from cultures pre-treated with AT2 agonists; 'pre-treated') the number of spheres formed showed a reduction in comparison with control aged group ( Figure 5C), indicating that, as in young cultures, the increase in proliferation was accompanied by reduced self-renewal.

AT1 Receptor Involvement in the RAS Actions on Adult Neurogenesis in Rat
We next analyzed expression of Ang II receptors in young adult rats. Detection with specific antibodies revealed immunoreactivity for AT1 and AT2 receptors in cells of the adult rat V-SVZ ( Figure 6A,B). Double-immunofluorescence detections indicated that most DCX-ir neuroblasts had AT1 and AT2 receptors and that some scattered GFAP-ir cells, likely representing NSCs and/or nonneurogenic astrocytes, were also positive for both receptors ( Figure 6A,B). The presence of AT1 and AT2 receptors in the rat V-SVZ was confirmed by RT-PCR ( Figure 6C). RT-PCR detection with specific primers in cell pools isolated by LCM from the V-SVZ of young adult rats revealed expression of both receptors at higher levels than in mice ( Figure 7A,B). However, the AT1/AT2 ratio was around one in young rats, whereas mouse tissue had three times more AT1 than AT2 receptor mRNA levels ( Figure 7C). WB analysis showed the presence of AT1 and AT2 receptor proteins in the mouse V-SVZ ( Figure 7D). Furthermore, we observed that rats treated with the AT1 receptor inhibitor candesartan showed a marked increase in Atgr2 expression in the V-SVZ ( Figure 7E), which is consistent with our previous observations in the striatum of candesartan-treated rats [63] but opposite to our Several new components of RAS, such as elements of the Ang 1-7/MasR axis, have recently been discovered [11]. Different data have suggested the cooperative role of MasR with the AT2 receptor in RAS neuroprotective actions [55]. However, the role of Ang 1-7/MasR components in V-SVZ neurogenesis has not been investigated. To address this issue, we treated neurospheres from young and aged mice with Ang 1-7 and with the antagonist of the anti-inflammatory MasR A779. Treatment with Ang 1-7 or with the MasR inhibitor by itself did not significantly affect the number of neurospheres generated in any of the two age conditions. Interestingly, the A779 antagonist prevented the CGP-induced increase in the number of spheres both in young and aged cultures ( Figure 5D), suggesting potential interactions between AT2 and MasR receptors.

AT1 Receptor Involvement in the RAS Actions on Adult Neurogenesis in Rat
We next analyzed expression of Ang II receptors in young adult rats. Detection with specific antibodies revealed immunoreactivity for AT1 and AT2 receptors in cells of the adult rat V-SVZ ( Figure 6A,B). Double-immunofluorescence detections indicated that most DCX-ir neuroblasts had AT1 and AT2 receptors and that some scattered GFAP-ir cells, likely representing NSCs and/or non-neurogenic astrocytes, were also positive for both receptors (Figure 6A,B). The presence of AT1 and AT2 receptors in the rat V-SVZ was confirmed by RT-PCR ( Figure 6C). RT-PCR detection with specific primers in cell pools isolated by LCM from the V-SVZ of young adult rats revealed expression of both receptors at higher levels than in mice ( Figure 7A,B). However, the AT1/AT2 ratio was around one in young rats, whereas mouse tissue had three times more AT1 than AT2 receptor mRNA levels ( Figure 7C). WB analysis showed the presence of AT1 and AT2 receptor proteins in the mouse V-SVZ ( Figure 7D). Furthermore, we observed that rats treated with the AT1 receptor inhibitor candesartan showed a marked increase in Atgr2 expression in the V-SVZ ( Figure 7E), which is consistent with our previous observations in the striatum of candesartan-treated rats [63] but opposite to our observations in the murine V-SVZ ( Figure 4A,B). Interestingly, we observed that the V-SVZ of aged animals showed an increase in AT1/AT2 ratio relative to young controls ( Figure 7F), which is consistent with our previous observations in the striatum and substantia nigra of aged rats and mice [62,65,66]. The present data indicate the possibility that RAS manipulations (e.g., candesartan) could exert different effects on adult neurogenesis in different species (e.g., rats and mice).
To study the potential role of endogenous Ang II on adult neurogenesis in the rat V-SVZ, we first orally-treated young and aged rats with the AT1 antagonist candesartan or vehicle for 15 days [32]. Due to its non-peptide nature, candesartan efficiently permeates the brain where it can block effects of Ang II at doses that have little effect on blood pressure [67]. At the last day of treatment, rats were injected i.p. with the traceable nucleoside BrdU (four pulses at 50 mg/kg, every 2 h) and sacrificed 24 h after the last injection to detect proliferating cells. We found a significant increase in the number of BrdU-ir cells in the V-SVZ of young animals treated with candesartan relative to vehicle-treated control animals ( Figure 7G,H). As previously observed [9], the overall number of BrdU-ir cells was lower in untreated aged animals than in young animals, but they were increased by the treatment with candesartan compared to untreated animals of equivalent age ( Figure 7G,H). Similar results were obtained both in young and aged rats when the number of DCX-ir neuroblasts was analyzed ( Figure 7I,J). These results indicated that the AT1 receptor restrains proliferation and AT2 receptor expression in the rat V-SVZ. Together, our work demonstrates effects of Ang II on the V-SVZ niche under physiological conditions in both rats and mice, but highlights that the actions of the RAS in adult neurogenesis may be mechanistically different depending on the species under study. observations in the murine V-SVZ ( Figure 4A,B). Interestingly, we observed that the V-SVZ of aged animals showed an increase in AT1/AT2 ratio relative to young controls ( Figure 7F), which is consistent with our previous observations in the striatum and substantia nigra of aged rats and mice [62,65,66]. The present data indicate the possibility that RAS manipulations (e.g., candesartan) could exert different effects on adult neurogenesis in different species (e.g., rats and mice).
To study the potential role of endogenous Ang II on adult neurogenesis in the rat V-SVZ, we first orally-treated young and aged rats with the AT1 antagonist candesartan or vehicle for 15 days [32]. Due to its non-peptide nature, candesartan efficiently permeates the brain where it can block effects of Ang II at doses that have little effect on blood pressure [67]. At the last day of treatment, rats were injected i.p. with the traceable nucleoside BrdU (four pulses at 50 mg/kg, every 2 h) and sacrificed 24 h after the last injection to detect proliferating cells. We found a significant increase in the number of BrdU-ir cells in the V-SVZ of young animals treated with candesartan relative to vehicle-treated control animals ( Figure 7G,H). As previously observed [9], the overall number of BrdU-ir cells was lower in untreated aged animals than in young animals, but they were increased by the treatment with candesartan compared to untreated animals of equivalent age ( Figure 7G,H). Similar results were obtained both in young and aged rats when the number of DCX-ir neuroblasts was analyzed ( Figure 7I,J). These results indicated that the AT1 receptor restrains proliferation and AT2 receptor expression in the rat V-SVZ. Together, our work demonstrates effects of Ang II on the V-SVZ niche under physiological conditions in both rats and mice, but highlights that the actions of the RAS in adult neurogenesis may be mechanistically different depending on the species under study.

Discussion
In the present study, we have shown the presence of AT1 and AT2 receptors in the adult rat and mouse V-SVZ, and in mouse V-SVZ-derived stem/progenitor cells grown as neurospheres. We observed that both AT1 and AT2 receptors were involved in regulation of proliferation of V-SVZ cells ( Figure 8A), and that the expression of AT1 and AT2 receptors (i.e., AT1/AT2 ratio) may be modified by aging, drugs such as AT1 receptor antagonists (candesartan), or species differences ( Figure 8B). The results are consistent with previous studies showing immunoreactivity for AT1 and AT2 receptors in aggregates of precursor cells obtained from developing brain tissue [19], and with a time-dependent increase in AT1 and AT2 receptor expression in cultured rat NSCs derived from embryonic and adult hippocampus [68,69]. Using confocal microscopy, we observed that AT1 and AT2 receptors were mostly expressed in V-SVZ neuroblasts. Moreover, some GFAP-positive cells, mainly located in the zone lining the striatum, showed co-expression of both receptors. On the basis of this location, these cells may probably correspond to non-neurogenic astrocytes, also called niche astrocytes or type-B2 cells [70]. In addition, the presence of AT1 and AT2 receptors in our cell culture system suggests that both receptors were located on NSCs or type-B1 cells, which were also positive for GFAP, as well as their transit-amplifying progenitors that can form neurospheres [71]. The present results also show that AGT, which is the precursor molecule for Ang II, was released from NSC/precursor cells, as the effects were also observed in the absence of Ang II administration. In the brain, astrocytes are the major source of AGT, and NSC/precursor cells have characteristics of astrocytes [72,73].

Discussion
In the present study, we have shown the presence of AT1 and AT2 receptors in the adult rat and mouse V-SVZ, and in mouse V-SVZ-derived stem/progenitor cells grown as neurospheres. We observed that both AT1 and AT2 receptors were involved in regulation of proliferation of V-SVZ cells ( Figure 8A), and that the expression of AT1 and AT2 receptors (i.e., AT1/AT2 ratio) may be modified by aging, drugs such as AT1 receptor antagonists (candesartan), or species differences ( Figure 8B). The results are consistent with previous studies showing immunoreactivity for AT1 and AT2 receptors in aggregates of precursor cells obtained from developing brain tissue [19], and with a timedependent increase in AT1 and AT2 receptor expression in cultured rat NSCs derived from embryonic and adult hippocampus [68,69]. Using confocal microscopy, we observed that AT1 and AT2 receptors were mostly expressed in V-SVZ neuroblasts. Moreover, some GFAP-positive cells, mainly located in the zone lining the striatum, showed co-expression of both receptors. On the basis of this location, these cells may probably correspond to non-neurogenic astrocytes, also called niche astrocytes or type-B2 cells [70]. In addition, the presence of AT1 and AT2 receptors in our cell culture system suggests that both receptors were located on NSCs or type-B1 cells, which were also positive for GFAP, as well as their transit-amplifying progenitors that can form neurospheres [71]. The present results also show that AGT, which is the precursor molecule for Ang II, was released from NSC/precursor cells, as the effects were also observed in the absence of Ang II administration. In the brain, astrocytes are the major source of AGT, and NSC/precursor cells have characteristics of astrocytes [72,73].  Ang II, via AT1 receptors, set a basal inhibition of proliferation that was antagonized by the AT2 receptors, which set a basal level of Ang II-dependent regulation of neurogenesis. In addition, AT2 receptor activity may exert an additional direct stimulatory effect on proliferation, which was modulated by Mas receptors (MasR). (B) Aging, candesartan (or AT1 deletion), and PD (or AT2 deletion) regulate AT1/AT2 expression ratio. Aging and PD (or AT2 deletion) increased ratio and candesartan (or AT1 deletion) increased ratio in mice (i.e., decreased AT2 expression) and decreased ratio in rats (i.e., increased AT2 expression). Candesartan = AT1 receptor antagonist; PD123319 = AT2 receptor antagonist.
The results also show that AT1 and AT2 receptors played a major role in NSC proliferation and/or generation of neuroblasts in the V-SVZ. Most of known actions of Ang II in the adult brain were mediated by AT1 receptors [74]. Local Ang II, via AT1 receptors, has been shown to enhance oxidative stress damage and neuroinflammation contributing to the progression of neurodegeneration [75,76]. However, effects of AT1 on cell proliferation and neurogenesis are controversial. Ang II, through AT1 receptors, is a major activator of the NADPH-oxidase complex, which is a primary source of ROS [77], and enhanced ROS generation plays a recognized role in cell death and pathogenesis. However, previous studies have also shown that ROS derived from NADPH-oxidase activation contribute to the self-renewal [78] and lineage determination in NSCs [79], and that AT1 receptors have an important role in progenitor cell proliferation [80]. Other studies have observed that blockage of AT1 receptors with different antagonists such as candesartan, valsartan, or losartan induced proliferation and neurogenesis in the dentate gyrus [25,26,81], but also that treatment with AT1 antagonists or angiotensin converting enzyme inhibitors diminishes neurogenesis [27,28,[82][83][84], or that AT1 blockade does not affect hippocampal neurogenesis [85,86]. There are also controversial results on the effects of AT2 receptors. Activation of AT2 receptors with the selective agonist CGP reportedly induced an increase in BrdU-positive cells in models of traumatic brain [87,88] and proliferation of hippocampal NSCs in vitro [68]. However, antiproliferative effects of AT2 receptors have also been suggested [89][90][91].
Our results suggest that most of the aforementioned controversial results may be explained by the complex interactions between AT1 and AT2 receptors and possible differences in the AT1/AT2 ratio in different regions and species. In AT2-KO mice, which show increased expression of AT1 in the V-SVZ, we observed a marked decrease in proliferation and generation of neuroblasts, which was inhibited by treatment with the AT1 antagonist candesartan, suggesting that a decrease in AT2 activity and an increase in AT1 activity inhibit proliferation and generation of neuroblasts. This was confirmed by treating WT mice with the AT2 agonist C21, which increased proliferation and generation of neuroblasts. Interestingly, AT1-KO mice or WT mice treated with the AT1 blocker candesartan, which also showed downregulation of AT2 receptors, did not show significant changes in proliferation and generation of neuroblasts. A basal inhibition of proliferation by AT1 receptors without any direct effect of AT2 receptors (i.e., beyond inhibition of AT1 activity) should lead to higher levels of proliferation in AT1-KO mice relative to WT mice. These similar levels may be explained by an additional direct positive effect of AT2 receptors on proliferation, which is inhibited by the downregulation of AT2 receptors observed in AT1-KO and candesartan-treated mice or neurosphere cultures. In summary, both downregulation of AT1 and upregulation of AT2 contribute to the increase in V-SVZ proliferation and generation of neuroblasts.
The results obtained in cultures of neurospheres were consistent with that observed in mice. Activation of AT2 receptors led to an increase in the generation of neurospheres. On the contrary, inhibition of AT2 led to a decrease in generation of neurospheres, and no significant changes were observed after blocking AT1 (which also leads to AT2 downregulation in neurospheres) as observed in neurospheres from AT1-KO mice or candesartan or ZD-treated WT mouse neurospheres. We also observed that the generation of neurospheres was reduced after previous stimulation of AT2 receptors, suggesting a possible loss of stemness.
The rat V-SVZ and mouse neurospheres from aged animals showed a significant decrease in proliferation and neurogenesis, which was inhibited by AT2 agonists or AT1 antagonists and exacerbated by AT2 antagonists. These results are consistent with the aforementioned mechanisms, as we observed an increase in AT1/AT2 expression ratio in the V-SVZ of aged animals relative to young controls, as previously observed in different types of cells [92], and particularly in the striatum and substantia nigra [62,65,66]. The present results are consistent with previous studies showing that neurogenesis continues throughout life, although its rate declines with increasing age in rodents and non-human primates [64], and with previous studies showing a progressive reduction in the production of progenitors throughout the mouse adult life and a maintenance of NSC pools [6,7,93]. Interestingly, our results reveal that the reduction in proliferation and generation of neuroblasts and neuropheres observed in the V-SVZ of aged animals may be counteracted by acting on the local RAS, suggesting that old NSCs show proliferation and differentiation capacity after adequate stimulation, which is consistent with other recent observations [94].
In cultures of neurospheres, we also investigated a possible role of Ang1-7/MasR. We administered Ang 1-7 to cultures derived from young and aged mice, and no significant change was detected. However, treatment of cultures with the MasR antagonist A779 blocked the AT2 agonist-induced increase in the number of spheres. This suggests a functional dependence between AT2 and MasR for the AT2-induced generation of neurospheres. Accumulating evidence shows that both receptors are co-localized, form heterodimers, and have different functional interactions in different cells and experimental conditions [27,[95][96][97], and recent data from our laboratory (unpublished) showed that young AT2-KO mice show a reduced expression of MasR in the substantia nigra and striatum, which may also contribute to the decrease in proliferation and neurogenesis observed in the V-SVZ of AT2-KO mice. Therefore, although RAS receptors have been associated with different cellular and physiological responses, their role in neurogenesis is complex, as shown by the present results that reveal the presence of AT1, AT2, and MasR interactions possibly both in physiological or pathological situations.

Conclusions
The present data revealed that the RAS played a major role in the regulation of adult neurogenesis both in young and aged brains. Ang II, via AT1 receptors, set a basal inhibition of V-SVZ proliferation that was antagonized by AT2 receptors. In addition, AT2 receptor activity may exert an additional direct stimulatory effect on proliferation. AT1/AT2 expression and ratio were affected by aging, which resulted in a decrease of neurogenesis. However, AT1 and AT2 receptor expression and neurogenesis may be modulated by therapies with agonists and antagonist of these receptors. Furthermore, RAS may interact with other modulators of neurogenesis. It is well-known that dopamine modulates neurogenesis in the V-SVZ [98], and counterregulatory interactions between Ang II and dopamine receptors have been demonstrated in the striatum [57,66] and other tissues [99,100]. Therefore, it seems plausible that dopamine regulates proliferation and neurogenesis by modulating RAS activity. Our results also suggest the potential beneficial effects of RAS modulators on neurogenesis. However, AT1/AT2 expression, ratio, and responses to drugs may vary depending on the species. Future studies are required to further clarify the effects of manipulation of the brain RAS in humans and pathological conditions associated with neurogenesis impairments, and to provide a solid basis for the use of RAS manipulation in such diseases.