Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord neural stem/progenitors by ERK1/2, Akt, and PLCγ

Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord... ORIGINAL RESEARCH ARTICLE published: 27 August 2013 MOLECULAR NEUROSCIENCE doi: 10.3389/fnmol.2013.00023 Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord neural stem/progenitors by ERK1/2, Akt, and PLCγ 1 1 2 1 1 Wai Si Chan , Alexandra Sideris , Jhon J. Sutachan , Jose V. Montoya G , Thomas J. J. Blanck and Esperanza Recio-Pinto * Department of Anesthesiology, New York University Langone Medical Center, New York, NY, USA Departamento de Nutrición y Bioquímica, Pontificia Universidad Javeriana, Bogotá, Colombia Edited by: Proliferation of endogenous neural stem/progenitor cells (NSPCs) has been identified in Andreas Vlachos, Goethe University both normal and injured adult mammalian spinal cord. Yet the signaling mechanisms Frankfurt, Germany underlying the regulation of adult spinal cord NSPCs proliferation and commitment toward Reviewed by: a neuronal lineage remain undefined. In this study, the role of three growth factor-mediated Sebastian Jessberger, University of signaling pathways in proliferation and neuronal differentiation was examined. Adult spinal Zurich, Switzerland Li Zhang, National Institute on cord NSPCs were enriched in the presence of fibroblast growth factor 2 (FGF2). We Alcohol Abuse and Alcoholism-NIH, observed an increase in the number of cells expressing the microtubule-associated protein USA 2 (MAP2) over time, indicating neuronal differentiation in the culture. Inhibition of the *Correspondence: mitogen-activated protein kinase or extracellular signal-regulated kinase (ERK) kinase 1 Esperanza Recio-Pinto, Laboratory and 2/ERK 1 and 2 (MEK/ERK1/2) or the phosphoinositide 3-kinase (PI3K)/Akt pathways of Dr. Thomas J. J. Blanck, Department of Anesthesiology, New suppressed active proliferation in adult spinal cord NSPC cultures; whereas neuronal York University Langone Medical differentiation was negatively affected only when the ERK1/2 pathway was inhibited. Center, 180 Varick Street Room 673, Inhibition of the phospholipase Cγ (PLCγ) pathway did not affect proliferation or neuronal New York, NY 10014, USA differentiation. Finally, we demonstrated that the blockade of either the ERK1/2 or PLCγ e-mail: esperanza.recio-pinto@ nyumc.org signaling pathways reduced neurite branching of MAP2+ cells derived from the NSPC cultures. Many of the MAP2+ cells expressed synaptophysin and had a glutamatergic phenotype, indicating that over time adult spinal cord NSPCs had differentiated into mostly glutamatergic neurons. Our work provides new information regarding the contribution of these pathways to the proliferation and neuronal differentiation of NSPCs derived from adult spinal cord cultures, and emphasizes that the contribution of these pathways is dependent on the origin of the NSPCs. Keywords: Neuronal differentiation, ERK1/2, Akt, PLCγ, progenitors, spinal cord INTRODUCTION spinal cord, the question remains as to how these adult spinal cord Early studies of the adult mammalian central nervous system NSPCs are regulated. A better understanding of the basic biology identified the presence of neural stem/progenitor cells (NSPCs) in of these NSPCs will facilitate future attempts to manipulate these the adult spinal cord (Weiss et al., 1996; Shihabuddin et al., 1997). cells under pathological conditions. Subsequent research showed that proliferation of the adult NSPCs Unlike in the adult spinal cord, the occurrence of neurogenesis are stimulated in animal models of spinal cord injury (Johansson in the adult hippocampus has been firmly established (Alvarez- et al., 1999; Horner et al., 2000; Yamamoto et al., 2001; Danilov Buylla and Garcia-Verdugo, 2002; Ming and Song, 2005, 2011). et al., 2006). Some studies reported that the fate of the prolif- Astrocytes from this brain region have been shown to induce erating NSPCs is restricted to glial cell types in vivo due to the neurogenesis of adult hippocampal NSPCs via the Wnt signal- inhibitory microenvironment of the adult spinal cord (Johansson ing pathway (Song et al., 2002; Lie et al., 2005). Furthermore, et al., 1999; Shihabuddin et al., 2000; Horky et al., 2006; Yang diffusible factors from the neurovascular niche are reported to et al., 2006; Barnabe-Heider et al., 2010). Nonetheless, the con- stimulate neurogenesis in the adult subventricular zone (Palmer sensus is that adult spinal cord NSPCs are intrinsically multi- et al., 2000; Shen et al., 2004, 2008). On the other hand, astro- potent (i.e., they can also generate neurons), as demonstrated cytes from the adult spinal cord do not promote neurogenesis in by transplantation studies and neurosphere assays (Shihabuddin culture (Song et al., 2002). et al., 2000; Yamamoto et al., 2001). In fact, neurogenesis in the Growth factors, such as fibroblast growth factor 2 (FGF2), adult spinal cord has been detected in vivo in both pathological epidermal growth factor, nerve growth factor, and vascular (Danilov et al., 2006; Vessal et al., 2007) and normal conditions endothelial growth factor, can elicit a range of cellular responses (Shechter et al., 2007, 2010). However, as we gain more insight including cell proliferation, migration, differentiation, and cell into the existence and implications of neurogenesis in the adult death through various classes of receptor tyrosine kinases (RTKs) Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 1 Chan et al. Pathways regulating spinal cord NSPC (Hubbard and Till, 2000). The activation of these different RTKs nestin+ were mostly rounded/spindle-shaped. The morphology in turn induces the activation of several signal transduction of the cells co-labeled with nestin and NG2 was similar to that of pathways including the mitogen-activated protein kinase or extra- the nestin+ cells (Figure 1G,iii) . Based on the cell morphology, cellular signal-regulated kinase (ERK) kinase 1 and 2/ERK 1 and some of the EdU+ cells that were not immunopositive for nestin 2 (MEK/ERK1/2), phosphoinositide 3-kinase (PI3K)/Akt, and were likely NG2+ (not shown). Stage-specific embryonic anti- phospholipase Cγ (PLCγ)pathways (Huang et al., 2001; Mason, gen 1 (SSEA-1), a marker for undifferentiated NSPCs (Capela and 2007). Much effort has gone into elucidating the signaling cas- Temple, 2002; Sabourin et al., 2009), was observed in about one- cades involved in the neurogenesis of adult hippocampal NSPCs. tenth of the cells in culture at 3 DIV (Figure 1H). Taken together, For instance, both the MEK/ERK1/2 and PI3K/Akt pathways have the expression of these markers demonstrates the presence of been shown to be involved in regulating proliferation and self- uncommitted NSPCs at 1–4 DIV. renewal (Peltier et al., 2007; Ma et al., 2009), while the PLCγ pathway is reported to be essential for neuronal differentiation NEURONAL DIFFERENTIATION IN ADULT SPINAL CORD NSPC (Ma et al., 2009) of adult hippocampal NSPCs. Yet the role of CULTURES these signaling pathways in adult spinal cord NSPCs has not been Maintaining the expanded population of adult spinal cord NSPCs established. In this study, our aim was to assess the potential con- in medium containing FGF2 eventually resulted in neuronal tribution of the MEK/ERK1/2, PI3K/Akt, and PLCγ pathways in differentiation. At 0 DIV and 1 DIV, about 23 and 12% of proliferation and neuronal differentiation of NSPCs derived from the cells were doublecortin-positive (DCX+) immature neu- the adult rat spinal cord culture. Our findings reinforce the idea rons (Figure 2A) (the remainder of the cells are most likely that the contribution of signaling pathway activation in prolifera- nestin+ and/or NG2+ cells, see Figure 1G,ii). Over time, there tion and neuronal differentiation is unique in different cell types was an increase in the number of cells expressing the neuronal and is dependent on the source of NSPCs. marker, microtubule-associated protein 2 (MAP2) (Figure 2B i). The number of MAP2+ cells increased by nearly 4-fold by RESULTS 6DIV (Figure 2B,ii) and about 6-fold by 14 DIV (Figure 2C). ENRICHMENT OF ADULT SPINAL CORD-DERIVED NSPCs To demonstrate that some of these MAP2+ cells were newly generated, at 5 DIV proliferating cells were labeled with a 24 h The enriched adult spinal cord NSPCs were obtained by using a EdU pulse, and the culture was immunolabeled for MAP2 at cell isolation protocol that yielded about 1% of the total num- 6 DIV. After the pulse of EdU, about a quarter of all MAP2+ ber of cells in the spinal cord (see Materials and Methods) cells were found to be EdU+ (Figure 2D,i,ii). When FGF2 was (Figure 1A); these cells could proliferate as neurospheres (not not added to the culture medium there was no increase in the shown) or in adherent cultures in medium containing FGF2 number of MAP2+ cells nor an increase in total cell number (Figure 1B). Adherent cultures were used because we could assess (not shown). Antibodies against vesicular glutamate transporter proliferation in a shorter time frame thus avoiding possible phe- 1 (VGLUT1) and glutamic acid decarboxylase 67 (GAD-67) were notypic changes and/or selection of a subpopulation of NSPCs used to further characterize the neuronal phenotype of these as a result of long term cultivation and passaging. In addition, newly generated MAP2+ cells (i.e., whether they were excitatory adherent cultures in a single layer also facilitated identification or inhibitory neurons). We found that 42% of all the MAP2+ cells and quantification of immunolabeled cells with various markers. were VGLUT1+ (Figure 2E). Moreover, 96% of all the VGLUT1+ Under such condition, the total cell number increased by 1.7-fold cells were negative for GAD-67 (Figure 2F), suggesting that adult by 6 days in vitro (DIV) when compared to 0 DIV (Figure 1C). A spinal cord-derived NSPCs differentiate toward neurons with an thymidine analogue, EdU, was added to the culture to pulse-label excitatory phenotype. those cells undergoing cell division at 4 DIV, 6 h prior to fixa- tion (Figure 1D,i). During the 6-h pulse of EdU-labeling, 25% PROLIFERATION OF ADULT SPINAL CORD NSPC CULTURES REQUIRES of the cells were EdU+ (Figure 1D,ii). In addition, co-labeling of EdU and nestin, used here as an adult NSPC marker (Lendahl ACTIVATION OF ERK1/2 AND Akt et al., 1990; Johansson et al., 1999; Namiki and Tator, 1999; Fu To examinewhich of thesignaling pathways might beimportant et al., 2005), revealed that 32 ± 5% (n = 4) of nestin+ cells for proliferation, specific pharmacological inhibitors were each were EdU+ and that 46 ± 12% (n = 4) of EdU+ cells were applied to the culture for 4 days, to which EdU was also added nestin+ (Figure 1E). Immunostaining with nestin at 0, 1, and for 6 h on day 4. We found that treatment with 10 μM U0126 4 DIV showed an increase of nestin+ cells over time; with a (MEK1/2 inhibitor) or 10 μM LY294002 (PI3K inhibitor) sig- 12-fold increase in percentage by 4 DIV (Figure 1F). Cells were nificantly reduced the number of EdU+ cells compared to the also labeled with another progenitor marker, neural-glial antigen vehicle control (Figure 3A). This indicates that the activation of 2 (NG2), as studies have shown that NG2+ cells are prolifera- both MEK/ERK and PI3K/Akt signaling pathways contributed to tive and can give rise to neurons (Belachew et al., 2003; Tamura the proliferation of adult spinal cord-derived NSPCs. Consistent et al., 2007; Guo et al., 2010). At 1 DIV 28% of the cells were with this interpretation, we found that both treatments with immunopositive for NG2 (Figure 1G,i,ii) and 19% of the nestin+ U0126 and LY294002 for 4–6 days resulted in the suppression cells were also NG2+ at 1 DIV (Figure 1G,ii). NG2 and nestin of an increase in the total cell number (total Hoechst-stained coexpressing cells were also detected at 3 DIV (Figure 1G,iii). nuclei) (Figure 3B). In contrast, there was no indication that the The cells that were NG2+ were stellate-like, with short pro- PLCγ pathway was involved in NSPC proliferation, as 2.5 μM cesses extending from all directions; and the cells that were only U73122 (PLCγ inhibitor) did not significantly affect the number Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 2 Chan et al. Pathways regulating spinal cord NSPC FIGURE 1 | Enrichment of adult rat spinal cord neural stem/progenitor picture of the green and red channels showing EdU (green) and nestin (red) cells (NSPCs) in culture. (A) Dissociated adult spinal cord cells, 1 h after staining. Blue indicates Hoechst-stained nuclei. (F) Quantification of the isolation. (B) Spinal cord cells after 4 days in FGF2 (i) or in basal medium (ii). percentage of nestin+ cells showing that the population of NSPCs was (C) Adult spinal cord cells proliferated as the total cell number per coverslip enriched in culture medium containing FGF2 over time (20 random fields with ∗∗ increased over time in culture medium containing FGF2 [all the cells in each a40× objective per coverslip were analyzed, n = 4 coverslips; P < 0.01, † # coverslip were counted, n = 3 coverslips; P < 0.001 compared with 0 days P ≤ 0.0001, One-Way ANOVA Dunnett’s multiple comparison post-test). (G) in vitro (DIV), and between 4 DIV and 6 DIV ± FGF2, One-Way ANOVA Co-labeling of NG2 chondronitin sulfate proteoglycan (green) and nestin (red) Dunnett’s multiple comparison post-test]. (D) A thymidine analogue, EdU, at 1 DIV (i) and 3 DIV (iii). At 1 DIV, most of the NG2+ cells had distinct was added to culture medium at 4 DIV, 6 h before fixation. Proliferating morphology from nestin+ cells; while cells coexpressing both markers (actively dividing) cells were EdU-labeled (green) during the 6 h pulse (i). Blue were observed at 1 and 3 DIV. Blue indicates Hoechst-stained nuclei. (ii) indicates Hoechst-stained nuclei. (ii) Quantification of the percentage of Quantification of the percentage of nestin+, NG2+, and nestin+/NG2+ cells EdU+ cells (10 random fields with a 40× objective per coverslip were at 1 DIV (20 random fields with a 40× objective per coverslip were analyzed, analyzed, n = 3 coverslips; P < 0.005, unpaired t-test). (E) Co-labeling of n = 4 coverslips). (H) At 3 DIV, stage-specific embryonic antigen 1 (SSEA-1) EdU and nestin showing that some of the NSPCs (4 DIV) were actively (green), a cell surface carbohydrate epitope found on uncommitted NSPCs, dividing. EdU was added to culture medium for only 6 h before fixation at appeared to localize to the plasma membrane of cells with undifferentiated 4 DIV. (i) Proliferating (EdU+) cells (green) during the 6 h pulse. (ii) Merged morphology. All scale bars: 20 μm. of EdU+ cells or the total cell number when compared to the cells compared to the vehicle control. Z-VAD-FMK, a general cas- vehicle control (Figures 3A,B). TUNEL analysis was performed to pase inhibitor, did not decrease the number of TUNEL-positive address whether the inhibitors affected cell survival. None of the cells in the presence of any of the inhibitors (Figure 3C), indicat- inhibitors significantly increased the number of TUNEL-positive ing that the decrease in EdU+ cell and total cell number as a result Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 3 Chan et al. Pathways regulating spinal cord NSPC FIGURE 2 | Neuronal differentiation in adult spinal cord NSPC cultures. Examples of newly generated MAP2+ cells (green EdU+ nuclei and red (A) Immunostaining for doublecortin (DCX) at 0 DIV (top left panel) and 1 DIV MAP2+ somata) by 6 DIV. As EdU was added to the culture medium for a 24-h (bottom left panel). Blue indicates Hoechst-stained nuclei. (Right panel) period at 5 DIV, only those cells that were dividing then were labeled green. Quantification of the percentage of DCX+ cells at both time points (20 random Blue indicates Hoechst-stained nuclei. (ii) Quantification of the percentage of fields with a 40× objective per coverslip were analyzed, n = 4 coverslips). (B) EdU+/MAP2+ cells at 6 DIV (10 random fields per coverslips were analyzed, i, Microtubule-associated protein 2 (MAP2) staining (green) of adult rat spinal n = 8 coverslips from three experiments). (E) Immunostaining for vesicular cord cultures at 0 DIV, arrowheads denote MAP2+ cells (left), and at 6 DIV in glutamate transporter 1 (VGLUT1) and MAP2 revealed that 42% of the culture medium containing FGF2 (right). Blue indicates Hoechst-stained nuclei. MAP2+ neurons had a glutamatergic phenotype (n = 539 cells). Cultures (ii) Fold change of the number of MAP2+ cells at 6 DIV (n = 13 coverslips from were counterstained with Hoechst (blue). (F) The majority (96%) of VGLUT1+ five experiments; P < 0.0001, unpaired t-test). The data was normalized to cells were negative for glutamic acid decarboxylase 67 (GAD-67) at 6 DIV. (i) A the mean total number of cells at 0 DIV. (C) As the population of NSPCs was scatter plot of the fluorescent intensity of each individual cell (n = 554 cells). enriched, many of the cells differentiated toward a neuronal lineage as the Red and green lines denote “cut-offs” for VGLUT1 and GAD-67 number of MAP2+ cells increased over time (10 random fields with a 40× immunoreactivity, respectively. Cells with intensity values above each line ∗∗ objective per coverslip were analyzed, n = 3 coverslips; P < 0.01, were considered positively stained. (ii) Example of VGLUT1+/GAD-67- cells at P < 0.001, One-Way ANOVA Dunnett’s multiple comparison post-test). (D) i, 6 DIV. Cells were counterstained with Hoechst (blue). Scale bars: 20 μm. of the treatment with U0126 or LY294002 was not due to apopto- NEURONAL DIFFERENTIATION IN ADULT SPINAL CORD NSPC sis. The specificity of each of the pharmacological inhibitors used CULTURES IS ERK1/2-DEPENDENT was verified in the adult spinal cord culture enriched in NSPCs To evaluate whether each of the inhibitors might influence the by Western blot analysis. By using antibodies specific for ERK1/2, neuronal differentiation potential of NSPCs, we examined the Akt, and PLCγ1 and their phosphorylated isoforms, the ratio of number of MAP2+ neurons in each treatment. We found that phosphorylated protein to total protein (pERK1/2 to ERK1/2, neuronal differentiation of the adult spinal cord-derived NSPCs pAkt to Akt, and pPLCγ1to PLCγ1) in thepresenceofeachof was only suppressed by treatment with 10 μM U0126, inhibitor the inhibitor was measured. We confirmed that, at the concentra- of MEK and therefore ERK1/2 activation, such that there was tions used in the cultures, each of the three inhibitors specifically 40% less MAP2+ neuronsascompared to the DMSO control and significantly blocked the phosphorylation of ERK1/2, Akt, by 6 DIV (Figures 4A i,ii, B). Inhibition of the PI3K/Akt and and PLCγ1(Figure 3D). PLCγ pathways had no significant effect on the number of cells Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 4 Chan et al. Pathways regulating spinal cord NSPC FIGURE 3 | Cell proliferation in adult spinal cord NSPC cultures requires each of the inhibitors. None of the treatment significantly changed cell activation of ERK1/2 and Akt. (A) Quantification of the normalized number survival as compared to DMSO control, even with the addition of Z-VAD-FMK of EdU+ cells showed that inhibition of ERK1/2 (with MEK inhibitor, U0126, (20 μM), a general caspase inhibitor (20 random fields with a 20× objective 10 μM) and Akt (with PI3K inhibitor, LY294002, 10 μM) activation suppressed per coverslip were analyzed, n = 3 coverslips for all except for the first bar proliferation of adult spinal cord cells at 4 DIV; inhibition of PLCγ activation (DNase as a positive control for the TUNEL assay): n = 2 coverslips). (D) i–iii, (with U73122, 2.5 μM) had no effect (10 random fields with a 40× objective Western blot analysis showed that the MEK/ERK1/2, PI3K/Akt, and PLCγ ∗∗ † per coverslip were analyzed, n = 3 coverslips; P < 0.01, P < 0.001, pathways were activated in adult spinal cord cultures (6 DIV). The right panel One-Way ANOVA Bonferroni’s multiple comparison post-test). EdU was shows the ratios of phosphorylated protein to total protein (pERK1/2 to added to culture medium 6 h before fixation. (B) Normalized total number of ERK1/2, pAkt to Akt, and pPLCγ1toPLCγ1) in the presence of each of the Hoechst-stained nuclei to the DMSO control grouped showed that inhibition inhibitors. Each of the pharmacological inhibitors specifically blocked of both ERK1/2 and Akt activation decreased total number of cells at 4–6 DIV phosphorylation of its respective target protein when exposed to FGF2. (n = 9 coverslips, except LY294002: n = 8 coverslips, and U73122: n = 6 Quantified results of the blots generated using the ImageJ software # ∗∗ † coverslips, from three experiments; P ≤ 0.0001, One-Way ANOVA (n = 3 experiments; *P < 0.05, P < 0.01, P < 0.001, One-Way ANOVA Bonferroni’s multiple comparison post-test). All values were normalized to the Bonferroni’s multiple comparison post-test). (iv) Total PLCγ protein was pulled DMSO control group in each experiment. (C) Quantification of the number of down and then the membrane was blotted for tyrosine residues. Upper right TUNEL-positive cells at 4 DIV to determine cell survival after treatment with corner shows that PLCγ phosphorylation was blocked by 2.5 μM of U73122. expressing MAP2 (Figures 4A i,iii,iv, B). Taken together, this After 6 days, the culture medium containing FGF2 was supple- indicates that the MEK/ERK1/2 pathway had a crucial role in mented instead with neurite outgrowth-inducing factors (BDNF, neuronal differentiation. GDNF, and cAMP) for 7 more days (except for the NB-A/B27 control condition in which no exogenous growth factor was NEURITE BRANCHING OF MAP2+ NEURONS DERIVED FROM ADULT added) (Figure 5A). By 13 DIV, added growth factors doubled SPINAL CORD NSPC CULTURES REQUIRES ACTIVATION OF ERK1/2 AND the number of neurite branches in DMSO (vehicle) control from PLCγ 4.0 to 7.9 branches per cell; whereas the number of branches in Previous studies on PC12, adult rat dorsal root ganglion neu- NB-A/B27 control with no added growth factor remained sim- rons, and corticospinal motor neurons have reported that neurite ilar to that measured at 6 DIV (4.6 vs. 4.0 branches per cell) outgrowth involves growth factor-mediated ERK1/2 activation (Figure 5C; dotted line indicates level at 6 DIV). Hence, the addi- tion of exogenous factors (i.e., BDNF, GDNF, and cAMP) indeed (Hollis et al., 2009; Hashimoto and Ishima, 2011). To investi- gate whether the MEK/ERK1/2 pathway, as well as the PI3K/Akt promoted further neuronal differentiation of MAP2+ neurons, as indicated by the further increase in neurite outgrowth (com- and PLCγ signaling pathways played a role in neurite outgrowth, both neurite branching and the average length of the longest neu- pare Figure 2B with Figure 5B). We found that neurite branching of MAP2+ neurons was significantly suppressed when activa- rite of MAP2+ neurons were quantified in cultures treated with the corresponding inhibitors and vehicle control (Figures 5A,B). tion of ERK1/2 and PLCγ was blocked (5.2 and 6.1 branches per Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 5 Chan et al. Pathways regulating spinal cord NSPC and were therefore more differentiated (Figure 5D). By adding EdU to the culture medium at 5 DIV for 24 h, followed by a switch to the medium with neurite-inducing factors for 7 days (as in Figure 5A), we found that some of the MAP2+ cells with pro- cesses that came in contact with neighboring MAP2+ cells were also EdU+, once again confirming that they were newly gener- ated neurons (Figure 5E). Co-labeling of GAD-67 and VGLUT1 revealed that 63.3% of the cells were only VGLUT1+ at 14 DIV, while 97.4% of the VGLUT1+ cells were negative for GAD-67 (n = 428 cells from two coverslips) (Figures 5F,G). This suggests that many of the cells in culture remained glutamatergic since 6DIV (Figure 2E). DISCUSSION Since the role of the MEK/ERK1/2, PI3K/Akt, and PLCγ signaling pathways in adult spinal cord NSPCs has not been established, we assessed the contribution of these pathways to the prolifera- tion and neuronal differentiation of NSPCs derived from adult rat spinal cord cultures. Proliferation was evaluated using the thymidine analogue, EdU; while neuronal differentiation was demonstrated through a time course of MAP2 staining and neurite outgrowth, as well as by the appearance of synapto- physin in the neurites. First, our data demonstrate that both the MEK/ERK1/2 and PI3K/Akt pathways are essential for pro- liferation of adult spinal cord NSPCs. Second, we show that the activation of ERK1/2 is necessary for directing adult spinal cord NSPCs toward a neuronal fate. Finally, our data indicate that neu- rite branching is dependent on the activation of the MEK/ERK1/2 and PLCγ pathways. Currently there are only few publications that describe the FIGURE 4 | Neuronal differentiation in adult spinal cord NSPC cultures stem cell niche in the adult spinal cord (Sabourin et al., 2009; is ERK1/2-dependent. (A) Immunostaining for MAP2 in adult spinal cord Hugnot and Franzen, 2011). Among those studies, cells that NSPC cultures after 6 DIV in vehicle control (DMSO) (i), U0126 (10 μM) (ii), express the immature neuronal marker, DCX were detected LY294002 (10 μM) (iii), and U73122 (2.5 μM) (iv). Compared to the vehicle control (i), the number of MAP2+ cells was noticeably less when the throughout the ependymal region in the adult spinal cord, where cells were treated with the MEK inhibitor, U0126 (ii). Blue represents occasional proliferating nestin+ cells were also found (Hamilton Hoechst-stained nuclei. Scale bar: 20 μm. (B) Quantification of the et al., 2009; Sabourin et al., 2009; Hugnot and Franzen, 2011). normalized number of MAP2+ cells in adult spinal cord NSPC cultures In addition, proliferating NG2+ cells found throughout the (6 DIV). By 6 days, there was a similar increase in MAP2+ cell number as parenchyma in the adult spinal cord have been reported to coex- shown in Figure 2. Only U0126 (which inhibited ERK1/2 activation) significantly inhibited the increase of the number of MAP2+ cells after 6 press DCX (Shechter et al., 2010), even though conventionally days. LY294002 (10 μM) and U73122 (2.5 μM) had no significant effect on NG2 is a marker for oligodendrocyte progenitor cells. Together neuronal differentiation (10 random fields with a 40× objective per coverslip with in vivo studies which describes coexpression of NG2 and were analyzed, n = 6 coverslips, except LY294002: n = 5 coverslips and † # DCX in the adult neocortex (Tamura et al., 2007)and in the U73122: n = 3 coverslips; P < 0.001, P ≤ 0.0001, One-Way ANOVA adult piriform cortex (Guo et al., 2010), these findings suggest Bonferroni’s multiple comparison post-test). All values are normalized to the DMSO control in each experiment. that at least a subpopulation of NG2 cells can be less committed to the oligodendrocyte lineage than previously thought and has the potential to give rise to neurons. cell, respectively) but not when activation of Akt was blocked Nestin, a NSPC marker (Lendahl et al., 1990; Johansson et al., (7.7 branches per cell). It is interesting to note that while neu- 1999; Namiki and Tator, 1999; Fu et al., 2005), was used in this rite branching was ERK1/2- and PLCγ-dependent, the inhibition study to characterize the expansion of adult spinal cord NSPCs in of these two signaling pathways in turn resulted in an increase culture medium containing FGF2. We showed that by 4 DIV 50% in neurite length of the MAP2+ neurons (from 91 μmat 6DIV of the cells were nestin+ and 30% were MAP2+.Based on the to 130.4 μm and 125.4, respectively) (Figure 5C). The ability of experiments done at early time points (at 0–3 DIV), the remain- these MAP2+ cells to extend neurites also indicates that neurite ing population of the cells are most likely NG2, DCX, and/or elongation of these cells is not ERK1/2- and PLCγ-dependent. SSEA-1 cells. The presence of cells expressing various progeni- The expression of synaptophysin in MAP2+ neurons indicates tor markers (nestin, NG2, and SSEA-1) indicate the presence of that those MAP2+ neurons in the control condition began to different uncommitted progenitor cells with a range of poten- establish synapses with neighboring cells after 2 weeks of culture tials, and hence, the term “neural stem/progenitor cells (NSPCs).” Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 6 Chan et al. Pathways regulating spinal cord NSPC FIGURE 5 | Neurites branching requires ERK1/2 and PLCγ activation; the neurite length of the MAP2+ neurons (U0126: 130.4 ± 5.6 μm, n = 83 some neurons form synapses and are glutamatergic. (A) Schematic of cells; U73122: 125.4 ± 4.4 μm, n = 96 cells). PI3K inhibitor, LY294002 experimental protocol used in the neurite studies. FGF2 was present in (10 μM), had no effect on neurite branching or elongation (n = 98 cells). culture to expand nestin+ NSPCs (0–6 DIV). Culture medium containing Inhibitor(s) or vehicle control (DMSO) was added to culture from 6 DIV, FGF2 was then replaced with BDNF, GDNF (GFs), and cAMP to promote except in the last condition (NB-A/B27) in which the culture was further differentiation of MAP2+ neurons (6–13 DIV). (B) A typical MAP2+ maintained in basal medium only with no growth factor treatment (For ∗∗ † neuron derived from an adult spinal cord NSPC culture (13 DIV). The each condition, three coverslips were analyzed; P < 0.01, P < 0.001, number of neurite branches of each MAP2+ neuron in each treatment was One-Way ANOVA Bonferroni’s multiple comparison post-test). (D) MAP2+ analyzed by scoring those branches that intersected the perimeter of a (red) neurons displayed synapse formation with one another, as evidenced 30 μm-circle. The mean neurite length was quantified by measuring the by co-labeling with anti-synaptophysin (syn, green) (13 DIV). Scale bar: length of the longest neurite of each MAP2+ neuron in each treatment. 5 μm. (E) Evidence of newly generated neurons (cells with green EdU+ Scale bar: 20 μm. (C) Quantification of neurite branching (solid bars) and nuclei and white MAP2+ labeling) at 13 DIV with processes that came in neurite length (patterned bars) of MAP2+ cells (13 DIV). Dotted gray line contact with neighboring cells. EdU was added to culture medium at 5 DIV represents average measurements (left Y-axis: number of neurite branches; for 24hr, followed by a switch of culture medium as described in (A). Scale right Y-axis: mean neurite length) at 6 DIV before growth factor treatment. bar: 20 μm. (F) i, Co-labeling of GAD-67 (green) and VGLUT1 (red) of adult By 13 DIV, added growth factors increased neurite branches in DMSO spinal cord NSPC cultures at 14 DIV showed that the majority of the cells control from 4.0 ± 0.4(at 6DIV) to7.9 ± 0.3 (at 13 DIV) (n = 25 and 87 were only VGLUT1+. (ii) Immunostaining negative control in the same cells, respectively). The number of branches in NB-A/B27 control with no culture condition as (i) at 14 DIV. Blue represents Hoechst-stained nuclei. added growth factor (4.6 ± 0.3 at 13 DIV; n = 94 cells) remained similar to Insets show phase contrast pictures of the same field. Scale bar: 20 μm. that at 6 DIV (4.0 ± 0.4; n = 25 cells). Neurite branching of MAP2+ (G) Scatter plot of the fluorescent intensity of each individual cell analyzed, neurons was significantly suppressed when the activation of ERK1/2 and which showed that 63.3% of the cells were VGLUT1+ at 14 DIV; while PLCγ was respectively blocked by 10 μM U0126 (5.2 ± 0.3at13DIV; 97.4% of the VGLUT1+ cells were negative for GAD-67 (n = 428 cells from n = 83 cells) and 2.5 μM U73122 (6.1 ± 0.2 at 13 DIV; n = 96 cells). The two coverslips were analyzed). Red and green lines denote the “cut-off” blockage of the activation of ERK1/2 and PLCγ resulted in an increase in for VGLUT1 and GAD-67 immunoreactivity, respectively. Moreover, we can infer from our data that by 14 DIV essentially With the use of proliferative markers, studies in blastocyst- all of the NSPCs in our study had undergone neuronal differ- derived embryonic stem cells (ESCs) (Li et al., 2007)and in entiation as suggested by the expression of the mature neuronal primary cultures of embyronic (E12.5-E13.5) cortical progen- marker, MAP2, as well as synaptophysin and VGLUT1. itor cells (Barnabe-Heider and Miller, 2003) indicate that the Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 7 Chan et al. Pathways regulating spinal cord NSPC activation of Akt, but not ERK1/2, is required for proliferation. In interact. Our cultures were enriched for NSPCs, but we cannot contrast, the activation of either Akt or ERK1/2 has been shown to rule out the presence of other adult spinal cord cells particularly promote proliferation in adult hippocampal (Peltier et al., 2007; at early DIV. However, we found that by 14 DIV all the cells were Ma et al., 2009) and in adult subventricular zone (Torroglosa MAP2+, indicating that all of the adult spinal cord NSPCs present et al., 2007; Lao et al., 2013) neural stem cell cultures. Similarly, in in the culture underwent neuronal differentiation. adult spinal cord NSPC cultures, we have shown that the inhibi- Collectively, these findings suggest that while the PI3K/Akt tion of either the PI3K/Akt pathway or the MEK/ERK1/2 pathway, pathway may be distinctly pivotal for proliferation of ESCs and but not the PLCγ pathway, markedly reduced cell division (pro- embryonic cortical progenitor cells, proliferation of adult neu- liferation). Past studies using different cell lines and primary ral stem and progenitor cells involves both the PI3K/Akt and neurons have suggested that cell survival involves the activation MEK/ERK1/2 pathways in a non-overlapping manner (Table 1). of these two pathways (Xia et al., 1995; Kennedy et al., 1997; Xue The downstream targets of these two pathways regulating pro- et al., 2000; Yamaguchi and Wang, 2001). Nonetheless, we have liferation in adult spinal cord NSPCs remain to be determined. shown that the inhibition of either the PI3K/Akt pathway or the However, recent studies of proliferation of adult neural stem cells MEK/ERK1/2 pathway did not induce additional cell death in the from other regions have identified additional pathways as well as adult spinal cord cultures when compared to the vehicle control. crucial downstream targets of the PI3K/Akt and MEK/ERK1/2 Hence, the reduction in cell proliferation when each of these two pathways, including the Wnt signaling pathway (Gage, 2010), pathways was inhibited was not due to apoptosis. In other words, the interleukin/Janus kinase/signal transducers and activators of both the PI3K/Akt and MEK/ERK1/2 pathways are crucial for transcription pathway (Gomez-Nicola et al., 2011), mammalian inducing cell proliferation of adult spinal cord NSPCs. target of rapamycin (Paliouras et al., 2012), and cyclin-dependent In the adult brain, the hippocampal subgranular zone (SGZ) kinase inhibitor (Marques-Torrejon et al., 2013). and the subventricular zone (SVZ) contain at least two popula- With respect to neural lineage differentiation, we have shown tions of stem/progenitor cells that are nestin+ (Doetsch et al., in cultures of adult spinal cord NSPCs activation of the 1997; Fukuda et al., 2003). Type 1 and B cells (in the SGZ and MEK/ERK1/2 (but not PI3K/Akt or PLCγ) pathway is required SVZ, respectively), which are also GFAP+ (glial fibrillary acidic for neuronal differentiation. On the other hand, it has been protein); and type 2 and C cells (in the SGZ and SVZ, respec- reported that, in adult hippocampal neural stem cell cultures, tively) which are GFAP- and occasionally DCX+.Type 1 and neuronal differentiation is blocked by ERK1/2 activation and B cells, respectively, give rise to Type 2 and C cells (Suh et al., is induced through activation of PLCγ (Ma et al., 2009). Our 2009). The proliferation of both of these cell types involves Akt findings provide evidence that, unlike in the adult hippocam- and ERK1/2 pathways (Lao et al., 2013). However, in the adult pal neural stem cells, but similar to mouse embryonic stem cells spinal cord, the populations of stem/progenitor cells are much (Li et al., 2006), ERK1/2 signaling promotes neuronal differentia- less well-characterized; they appear to be located at the dorsal area tion in adult spinal cord NSPC cultures (Table 1). The inhibition of the ependymal zone in which proliferating cells, nestin+ cells, of ERK1/2 activation likely prevents the existing/remaining adult GFAP+ cells, and BLBP+ (brain-lipid-binding protein) cells have spinal cord NSPCs from exiting cell cycle and from proceeding been detected (Shechter et al., 2007; Meletis et al., 2008; Hamilton to differentiate into MAP2+ neurons. Although the activation et al., 2009; Sabourin et al., 2009). In this area there are also of the MEK/ERK1/2 pathway has been implicated in promoting DCX+ cells (Shechter et al., 2007). Additional characterization neuronal survival (Xue et al., 2000), as mentioned earlier the inhi- is required to determine the number and type of progenitor pop- bition of the MEK/ERK1/2 pathway did not result in an increase ulations in the adult spinal cord, as well as how these populations of cell death. This implies that the decrease in the number of Table 1 | Contribution of ERK1/2, Akt, PLCγ to proliferation and neuronal differentiation of stem/progenitor cells. Stem or progenitor cell origin Proliferation Neuronal differentiation Publications MEK/ERK1/2 PI3K/Akt PLCγ MEK/ERK1/2 PI3K/Akt PLCγ Embryonic stem cells (human) No Yes n/d n/d n/d n/d Li et al., 2007 Embryonic stem cells (mouse) n/d n/d n/d Yes n/d n/d Li et al., 2006 E12.5 – E13.5 cortical progenitors (mouse) No Yes n/d Yes n/d n/d Barnabe-Heider and Miller, 2003 Adult hippocampal neural progenitor cells (rat) No Yes n/d n/d Inhibitory n/d Peltier et al., 2007 Adult hippocampal neural progenitor cells (rat) Yes n/d n/d Inhibitory n/d Yes Ma et al., 2009 Adult SVZ neurospheres (mouse) n/d Yes n/d n/d n/d n/d Torroglosa et al., 2007 Adult SVZ neurospheres (mouse) Yes Yes n/d n/d n/d n/d Lao et al., 2013 Adult spinal cord progenitor cells (rat) Yes Yes No Yes No No Present study No: activation of the pathway is not involved. Yes: activation of the pathway is involved. Inhibitory: activation of the pathway is inhibitory. n/d: not determined in the study. Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 8 Chan et al. Pathways regulating spinal cord NSPC MAP2+ neurons we observed after ERK1/2 inhibition was not as to understand the signaling mechanism that is unique to the due to an increase in neuronal death but rather a suppression cell type of interest. of neuronal differentiation. We have also shown in our NSPC As mentioned in the introduction, adult spinal cord NSPCs are cultures that the inhibition of ERK1/2 and Akt activation signif- stimulated to proliferate in response to injury (Johansson et al., icantly, and to the same extent, suppressed active proliferation, 1999; Horner et al., 2000; Yamamoto et al., 2001; Danilov et al., yet only the inhibition of ERK1/2 activation suppressed neuronal 2006). This surge of cell proliferation, however, produces mostly differentiation (Table 1; Figures 3A,B). Therefore, it is unlikely reactive astrocytes (Johansson et al., 1999; Shihabuddin et al., that the suppression in neuronal differentiation as a result of 2000; Horky et al., 2006; Yang et al., 2006). There is also support- the ERK1/2 pathway inhibition was simply due to a decrease in ing evidence that there is an upregulation of FGF2 expression by proliferation. the reactive astrocytes after spinal cord or spinal nerve injuries In addition to studying the role of the signaling pathways in (Madiai et al., 2003; Qi et al., 2003), despite the lack of neuronal neuronal differentiation, we have also investigated their role in regeneration. Intriguingly, Miura et al. have reported partial func- neurite outgrowth in the newly generated MAP2+ cells. Our tional recovery in adult rats infected with a constitutively active findings indicate that both the MEK/ERK1/2 and PLCγ path- form of MEK1 (activator of ERK) following spinal cord injury ways act to modulate cytoskeletal dynamics of the MAP2+ cells, (Miura et al., 2000). While they attributed the functional recov- as both MEK and PLCγ inhibitors, respectively, reduced neurite ery to axonal regeneration, neurogenesis of the endogenous spinal branching of these cells. Reminiscent of previous neurite studies cord NSPCs was not examined. Based on our study, it is likely (Mattson et al., 1988; Costantini and Isacson, 2000), the length of that the constitutively active MEK/ERK signaling pathway may the longest neurite increases in conditions where neurite branch- have also promoted neuronal differentiation of those endogenous ing is compromised. These findings suggest that the MEK/ERK1/2 NSPCs, resulting in the replenishment of lost neurons following and PLCγ pathways may have a specific effect on neurite complex- injury to the spinal cord. As we gain more knowledge regarding ity rather than neurite elongation. During neuronal development, the regulation of NSPCs in the adult spinal cord, more research to part of neurite outgrowth includes neurite elongation. Elongation better understand such a complex system is necessary to facilitate is dependent on the dynamics of cytoskeleton, which consists of future attempts to manipulate these NSPCs in pathological con- microfilaments, intermediate filaments, and microtubules (MTs). ditions. To conclude, we found in our study that the activation of As it turns out, several microtubule-binding proteins, includ- the MEK/ERK1/2 pathway and the PI3K/Akt pathway (but not the ing MAP2, can affect the polymerization and stability of MTs PLCγ pathway) are essential for proliferation of adult rat spinal (Sanchez et al., 2000). Based on previous studies a model for neu- cord NSPCs; and that the activation of the MEK/ERK1/2 path- rite elongation and branching has been already proposed (Hely way(butnot the PI3K/Aktor the PLCγ pathway) is required for et al., 2001; Kiddie et al., 2005). When it is phosphorylated, MAP2 neuronal differentiation of the adult spinal cord NSPCs. Together has low affinity for MTs. This decrease in MAP2 association with with other studies, our findings reinforce the idea that the con- MTs in turn decreases MT stability which results in a decrease tribution of signaling pathway activation in proliferation and in neurite elongation while facilitates neurite branching (Hely neuronal differentiation is unique in different cell types and is et al., 2001; Kiddie et al., 2005). On the other hand, when MAP2 dependent on the source of NSPCs. is de-phosphorylated, this favors the binding of MAP2 to MTs, MATERIALS AND METHODS which leads to MTs bundling and polymerization and therefore promotes neurite elongation while decreases neurite branching PRIMARY CULTURES (Hely et al., 2001; Kiddie et al., 2005). Our results, therefore, can All animals were used in accordance with guidelines approved by be explainedasfollows. AsMAP2is a known substrate formany the New York University Langone Medical Center Institutional protein kinases, including ERKs and protein kinase C (which is Animal Care and Use Committee. Adult male sprague-Dawley activated by PLCγ), it is likely that the inhibition of ERK1/2 and rats weighing 260–390 g were used. Animals were anesthetized PLCγ activation in fact promoted MAP2 dephosphorylation and and perfused with artificial cerebrospinal fluid (CSF) as described hence its binding to MTs, resulting in an increase in neurite length previously (Montoya et al., 2009). After perfusion and decapi- (elongation) and a decrease in neurite branching. tation of the animal, a 20 gauge needle and syringe filled with By comparing the findings in our study with those in oth- 10 mL of cold CSF was inserted into the sacral canal to extrude ers, there is a recurring theme that the effect of signaling the spinal cord. Microdissection of the spinal cord into approx- pathway activation is cellular context-dependent (Schlessinger, imately 1 mm pieces was carried out in cold hibernate A (HA) 2000). Along these lines, it is worth mentioning that while it (Brain Bits, Springfield, IL) supplemented with 2% B27 [self- has been suggested that the PI3K/Akt pathway can antagonize prepared; (Brewer et al., 1993)], GlutaMAX (0.5 mM), penicillin the MEK/ERK1/2 pathway (Mason, 2007), and vice versa (in (100 U/mL) and streptomycin (100 μg/mL) (all purchased from NIH3T3 cells) (Hayashi et al., 2008), we do not have any evi- GIBCO, Carlsbad, CA). The dissected spinal cord tissue was dence that supports such crosstalk between this two pathways in digested in papain (36 U/mL, Worthington, Lakewood, NJ) and our enriched adult spinal cord NSPC culture. For instance, from DNase I (0.02% w/v, Worthington) in a 15-mL conical tube for our Western blot results, we did not detect the potentiation of 5 min at room temperature (RT), then at 30 Cfor 25min. The Akt phosphorylation (vs. DMSO control) in response to ERK1/2 digested tissue was kept on ice and allowed to settle by gravity inactivation, or vice versa. Collectively, these data indicate that (∼1 min). Supernatant was discarded and the pellet was resus- even well-established pathways warrant further investigation so pended with 2 mL of HA containing 0.02% DNase I. Using three Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 9 Chan et al. Pathways regulating spinal cord NSPC fire-polished glass pipettes of decreasing diameter, the cell sus- IMMUNOCYTOCHEMISTRY pension was triturated with each no more than ten times. Before Adult rat spinal cord cells were fixed in 4% paraformalde- switching to another pipette, the suspension was allowed to settle hyde (15710, Electron Microscopy Sciences, Hatfield, PA) for (as above), and the supernatant was kept in another conical tube, 15 min and permeabilized in 0.3% Triton X-100 (X100RS, Sigma- while 2 mL of fresh HA (containing 0.02% DNase I) was added to Aldrich) [or 0.1% Tween 20 (P5927, Sigma-Aldrich) for mouse the suspension. At the end of the trituration procedure when all anti-stage-specific embryonic antigen 1 (SSEA-1)] for 10 min three glass pipettes were used, a cell suspension with a total vol- at RT. To label proliferating cells, 5-ethynyl-2 -deoxyuridine ume of 6 mL was layered on top of 5 mL of 6% OptiPrep (Accurate (EdU) (1.5–3 μM, from the Click-iT EdU Imaging Kit, C10337, Chemical and Scientific Corporation, Westbury, NY) gradient in Invitrogen), a modified thymidine analogue, was applied to a 50-mL conical tube. The gradient was centrifuged at 822 × gfor the culture (for 24 h, unless otherwise stated) before fixation. 13 min at 4 C. After centrifugation, the white myelin layer and the To detect EdU-labeled cells, a click-iT reaction between the top half of the supernatant were discarded. After gentle mixing, Alexa Fluor 488 azide and the modified EdU alkyne was car- the suspension was brought up to 20 mL with HA and was allowed ried out according to the instructions of the company. This to slowly pass through a 70 μm nylon mesh. The cell suspension reaction was carried out prior to immunostaining procedures was spun one last time at 480 × g for 5 min. The isolation protocol with various primary antibodies. Mouse anti-nestin (∼1 μg/mL, yielded ∼1% of the total cell number (∼110,000 cells isolated) in Rat-401-concentrate, Developmental Studies Hybridoma Bank, the spinal cord, which was calculated based on the DNA content Iowa City, IA; 2.5 μg/mL, MAB353, Millipore) and mouse anti- (PureLink Genomic DNA kits, K1820–01, Invitrogen, Carlsbad, SSEA-1 (∼1 μg/mL, MC-480, Developmental Studies Hybridoma CA) of the final pellet and previous studies of amount of DNA Bank) were used as markers for undifferentiated adult neu- per cell (Santen and Agranoff, 1963) and totalcellnumberin ral stem/progenitor cells. Rabbit anti-NG2 chondroitin sulfate the adult rat spinal cord (Bjugn, 1993). The final pellet was proteoglycan (0.67 μg/mL, AB5320, Millipore) to label neu- resuspended with 400 μL of NB-A/B27 [Neurobasal A (GIBCO) ral progenitor cells. Primary antibodies used to label neu- supplemented with 2% B27, GlutaMAX (0.5 mM), penicillin rons were mouse anti-microtubule associated protein-2a and (100 U/mL) and streptomycin (100 μg/mL)]. Ten microliter of -2b (0.4–0.67 μg/mL, MAP2, clone AP20, MAB3418, Millipore, cell suspension (∼3000 cells) was plated on poly-D-lysine- Temecular, CA), goat anti-doublecortin (0.8 μg/mL, DCX, sc- coated glass coverslips (50 μg/mL, P6407, Sigma-Aldrich, St. 8066, Santa Cruz Biotechnology, Santa Cruz, CA), mouse anti- Louis, MO) in 4-well plates. After 1 h in 5% CO ,37 C, glutamic acid decarboxylase (0.67 μg/mL, GAD-67, MAB5406, 500 μL of pre-warmed NB-A/B27 containing 20 ng/mL of FGF2 Millipore), and rabbit anti-vesicular glutamate transporter (13256-029, Invitrogen) was added to each well. Adult spinal 1 (1:2000 polyclonal antiserum, VGLUT1, 135302, Synaptic cord cultures were treated with 10 μM U0126 (Cat#662005), Systems, Germany). Rabbit anti-synaptophysin 1 (1:150 poly- 10 μM LY294002 (Cat#440202), 2.5 μM U73122 (Cat#662035; clonal antiserum, 101002, Synaptic Sytems) was used to iden- all from Calbiochem, San Diego, CA), or vehicle control DMSO tify synapse formation. Ten percent normal goat serum (G9023, (dimethyl sulfoxide, D2650, Sigma-Aldrich). For longer culture Sigma-Aldrich) was used to block non-specific binding. Cultures period, NB-A/B27 from 6 to 13 days in vitro were supple- were incubated overnight at 4 C with primary antibodies diluted mented with recombinant human brain-derived neurotrophic in 2% goat (or, forDCX,donkey) seruminPBS. Species- factor (BNDF, 1 ng/mL, 10909010, Invitrogen), recombinant appropriate IgG, IgG1, and IgG2a conjugated with Alexa Fluor human glial-derived neurotrophic factor (GDNF, 0.1 ng/mL, 488, 546, or 555 (1:800–1:2000, Molecular Probes, Eugene, OR) 10907012, Invitrogen), and membrane permeable cAMP ana- were used as secondary antibodies. After discarding the buffer log [8-(4-chlorophylthio)adenosine 3 ,5 -cyclic monophosphate containing the primary antibodies and extensive rinsing (at least sodium salt, 125 μM, C3912, Sigma-Aldrich]. Culture medium 5 rinses, 5 min each), the secondary antibodies were added and was changed every other day in all experiments. incubated for 1 h at RT. Cells were counterstained with Hoechst In Western blots, the PLC inhibitor U73122 inhibited PLCγ1 33342 for 15 min (6 μg/mL). Coverslips were mounted with activation at 10 μM in our cultures. Due to its cytotoxic effect Aqua-Poly/Mount (Polysciences, Warrington, PA). Negative con- in longer exposure (6 DIV), the concentration of U73122 was trols were prepared by omitting the primary antibodies. Rabbit eventually lowered to 2.5 μM. At 2.5 μM, U73122 still signifi- and mouse whole molecule IgG (Jackson Immuno Research, West cantly and specifically inhibited PLCγ1 activation while it did Grove, PA) at comparable concentrations to primary antibodies not induce a significant increase in cell death as compared to the used for immunostaining were also used as alternative negative vehicle control (see results). This U73122 concentration is similar controls. Appropriate cell types or tissues were used as immunos- to its reported IC value of 1–2.1 μM (product information, taining controls to test the specificity of the primary antibodies Cat#662035 Calbiochem). The MEK inhibitor U0126 was tested used for immunocytochemistry in this study (Figure 6). at various concentrations in our cultures, resulting in a decrease For TUNEL labeling, adult spinal cord cells were cultured for in the number EdU+,MAP2+,and EdU+/MAP2+ cells as 4 days in FGF2-containing NB-A/B27, treated with each of the the concentrations of U0126 was increased (not shown). At inhibitors (or DMSO), and fixed as described above. To prevent the highest dosage we tested, U0126 (10 μM) did not cause an apoptosis, 20 μM Z-VAD-FMK (Cat#219007, EMD Biosciences, increase in cell death when compared to vehicle control and San Diego, CA), a general caspase inhibitor, was used. Click-iT its action was specific (see Western blot results), hence this TUNEL Alexa Fluor 488 Imaging Assay (C10245, Invitrogen) was concentration was used for subsequent experiments. used to detect apoptotic cells. The TUNEL reaction, based on a Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 10 Chan et al. Pathways regulating spinal cord NSPC FIGURE 6 | Immunostaining controls of primary antibodies used in the the primary antibody omitted. (iii) Composite of an adult spinal cord section study. As positive controls, primary antibodies used in the study were tested taken at a lower magnification. Prominent MAP2 staining (white) was in different cell types or tissues where they have been detected in previous observed only in the gray matter [similar to a previous study studies. (A), Immunostaining of mouse monoclonal nestin (green) in (Suzuki-Yamamoto et al., 2000)]. Scale bar: 50 μm. (D), Co-labeling of rabbit hippocampal progenitor cell culture from embryonic day (E) 18 rat embryos polyclonal synaptophysin 1 (green) and mouse monoclonal MAP2 (red) in rat (1 DIV), where the majority of the cells were nestin+ (i) [previous reported in postnatal day 7 hippocampal culture (37 DIV). (i) MAP2+ neurons formed (Mistry et al., 2002)]. (ii) Negative control where the primary antibody was synapses (expressed synaptophysin) with neighboring neurons. (ii) Negative omitted. Scale bar: 20 μm. (B) i,ii, Immunostaining of mouse monoclonal control where the primary antibody was omitted. Whole molecule rabbit and GAD-67 (green) in an adult rat spinal cord section. (i) Fluorescent image mouse IgG at concentrations identical to those of the two primary antibodies showing that GAD-67 immunoreactivity was localized in lamina II of the were used to show specificity of the primary antibodies. Blue represents dorsal horn, which matches the typical staining pattern of GAD-67 in the Hoechst-stained nuclei. Scale bar: 20 μm. (E), Immunostaining of rabbit spinal cord (Edgerton et al., 2001). (ii) Phase contrast image of the same polyclonal VGLUT1 (red) in rat E18 hippocampal culture (22 DIV). (i) Negative dorsal horn field. Scale bar: 20 μm. (B) iii,iv, Immunostaining of rat postnatal control where the primary antibody was omitted. (ii) Only those cells with day 7 hippocampal culture (16 DIV) using mouse monoclonal GAD-67 neuronal morphology were immunoreactive for VGLUT1. (F) i, antibody. (iii) GAD-67 staining (green). (iv) Negative control in which the Immunostaining of goat polyclonal DCX (magenta) in an adult mouse brain primary antibody was omitted. Scale bar: 20 μm. (C), Immunostaining of section. DCX+ immature neurons were detected in the dentate gyrus of the mouse monoclonal MAP2 in adult rat spinal cord sections. (i) MAP2 staining hippocampus. (ii) Negative control with the primary antibody omitted. Blue (red) in ventral horn neurons. Scale bar: 20 μm. (ii) Negative control where represents Hoechst-stained nuclei. Scale bar: 20 μm. copper catalyzed reaction between the Alexa Fluor azide and a then exposed toFGF2 (20ng/mL) for30min afterwhich the modified dUTP alkyne, was carried out according to the company medium was discarded and the cells were rinsed once with ice- manual. DNase I was provided in the company kit as a positive cold PBS. The following steps were carried out on ice or at control of the TUNEL assay. 4 C: cells were lysed directly in Petri dishes by adding 50μlof lysis buffer: RIPA (20188, Millipore) supplemented with 0.1% IMMUNOBLOTTING AND IMMUNOPRECIPITATION (v/v) SDS and 1X phosphatase inhibitor cocktail (78420, Thermo Adult spinal cord cells were cultured with FGF2 for 6 days in Scientific, Rockford, IL). Solubilized cell lysate was collected to one side of the dish by scraping. Lysate was then transferred 35 mm Petri dishes. Cells were rinsed with NB-A/B27 four times, once every hour for 4 h to remove traces of FGF2, followed by to Eppendorf tubes and rocked for 5 min. Supernatant from each sample was collected after a 13-min spin at 13k rpm. Total a 1 h-pre-treatment with one of the following inhibitors, 10 μM protein concentrations were determined using the Lowry assay. U0126, 10 μM LY294002, 2.5 μM U73122, or DMSO. Cells were Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 11 Chan et al. Pathways regulating spinal cord NSPC Various volumes of samples containing 50 μg of total protein were IMAGE ACQUISITION AND ANALYSIS lyophilized and resuspended with 20 μL of sample buffer. After Fluorescent images were captured using Carl Zeiss Axiovert 200 boiling the samples for 5 min, proteins were separated by SDS- M. For each experiment, ten (using a 40× objective) to twenty PAGE and transferred to PVDF membranes (Amersham, UK) (using a 20× or 40× objective) random fields of each of the cov- for 3 h (100 V). Membranes were blocked in 5% BSA in Tris- erslips in each treatment were acquired. For total cell numbers, buffered saline, 0.1% Tween 20 (TBST, all from Sigma-Aldrich) Hoechst-stained nuclei were counted using a 10× objective over for 1 h at RT, and then they were incubated overnight at 4 C the entire coverslip. The exposure times of the fluorescent chan- with primary antibodies diluted in 4% BSA in TBST. Membranes nels (DAPI, GFP, and Rhodamine) was determined using negative were first probed with antibodies against phospho-proteins. Then controls where either the primary antibody was omitted or the the membranes were stripped (Restore Plus, 46430, Thermo whole molecule IgG of matching species was added in place of the Scientific) and reprobed with antibodies against total pro- primary antibody. The fluorescent intensities of individual cells teins. Primary antibodies used were mouse anti-phospho-p44/42 were analyzed using the AxioVision software. Cells were scored MAPK (ERK1/2) (1:2000, Thr202/Tyr204, 9106, Cell Signaling as immune-positive when their intensities were at least three Technology, Danvers, MA), rabbit anti-phospho-Akt (1:1000, standard deviations above the average values of their correspond- Ser-473, 9271, Cell Signaling), rabbit anti-phospho-PLCγ1 ing negative controls. Alternatively, the background fluorescence (1:700, Thy-783, 2821, Cell Signaling Technology), rabbit anti- from each negative control was subtracted from the cells and flu- p44/42 MAPK (1:1000, 9102, Cell Signaling Technology), mouse orescence above the background was considered as positive based anti-Akt (1:1000, 610860, BD Biosciences), and mouse anti- on the RGB histogram generated by AxioVision. All results are PLCγ1 (1:300, sc-7290, Santa Cruz Biotechnology). Membranes represented as mean + SEM. Statistical significance was deter- were incubated with HRP-conjugated goat anti-rabbit (1:5000, mined by unpaired t-test when comparing two groups. When sc-2004, Santa Cruz Biotechnology) or anti-mouse (1:5000, sc- comparing three or more groups One-Way ANOVA was used. 2005, Santa Cruz Biotechnology) secondary antibodies diluted in Bonferroni’s multiple comparison post-test was used when com- 2.5% BSA in TBST. Membranes were developed with SuperSignal paring three to five groups and Dunnett’s for five or more groups. ∗ ∗∗ § † # chemiluminescent substrates (West Pico, 34077, West Dura, P < 0.05, P < 0.01, P < 0.005, P < 0.001, P ≤ 0.0001. 37071, Thermo Scientific). Blots were quantified using the gel analyzer function in the ImageJ software. ACKNOWLEDGMENTS For immunoprecipitation, anti-PLCγ-coated Dynabeads The authors wish to thank Dr. Jin Zhang and Lourdes A. Martin (Dynabeads Protein G, 100.03D, Invitrogen) was incubated with Hernandez for their technical assistance and Joey Ting for his crit- solubilized cell lysate containing 40 μg of total protein overnight ical comments on the manuscript. The Rat-401 antibody devel- at 4 C. The lysate-antibody-Dynabeads mixture was washed three oped by Susan Hockfield and the MC-480 (SSEA-1) antibody times with coldPBS andthen resuspended with 20 μLof sample developed by David Solter were obtained from the Developmental buffer for subsequent Western blot analysis as described previ- Studies Hybridoma Bank developed under the auspices of the ously. Monoclonal phosphotyrosine antibody (1:1500 pTyr, PY99, NICHD and maintained by The University of Iowa, Department sc-7020) was purchased from Santa Cruz Biotechnology. of Biology, Iowa City, IA 52242. Bjugn, R. (1993). The use of the opti- Exp. Neurol. 164, 60–70. doi: Region-specific growth proper- REFERENCES Alvarez-Buylla, A., and Garcia- cal disector to estimate the num- 10.1006/exnr.2000.7417 ties and trophic requirements of Verdugo, J. M. (2002). Neurogenesis ber of neurons, glial and endothe- Danilov, A. I.,Covacu, R.,Moe, M. brain- and spinal cord-derived rat in adult subventricular zone. lial cells in the spinal cord of C., Langmoen, I. A., Johansson, embryonic neural precursor cells. J. Neurosci. 22, 629–634. the mouse–with a comparative note C. B., Olsson, T., et al. (2006). Neuroscience 135, 851–862. doi: Barnabe-Heider, F., Goritz, C., on the rat spinal cord. Brain Res. Neurogenesis in the adult spinal 10.1016/j.neuroscience.2005.05.049 Sabelstrom, H., Takebayashi, 627, 25–33. doi: 10.1016/0006-8993 cord in an experimental model of Fukuda, S., Kato, F., Tozuka, Y., H., Pfrieger, F. W., Meletis, K., et al. (93)90744-8 multiple sclerosis. Eur. J. Neurosci. Yamaguchi, M., Miyamoto, Y., and (2010). Origin of new glial cells Brewer,G.J., Torricelli, J. R., Evege,E. 23, 394–400. doi: 10.1111/j.1460- Hisatsune, T. (2003). Two distinct in intact and injured adult spinal K., and Price, P. J. (1993). Optimized 9568.2005.04563.x subpopulations of nestin-positive cord. Cell Stem Cell 7, 470–482. doi: survival of hippocampal neurons Doetsch, F., Garcia-Verdugo, J. M., and cells in adult mouse dentate gyrus. 10.1016/j.stem.2010.07.014 in B27-supplemented Neurobasal, a Alvarez-Buylla, A. (1997). Cellular J. Neurosci. 23, 9357–9366. Barnabe-Heider, F., and Miller, F. D. new serum-free medium combina- composition and three-dimensional Gage, F. H. (2010). Molecular and cellu- (2003). Endogenously produced tion. J. Neurosci. Res. 35, 567–576. organization of the subventricu- lar mechanisms contributing to the neurotrophins regulate survival and doi: 10.1002/jnr.490350513 lar germinal zone in the adult regulation, proliferation and differ- differentiation of cortical progeni- Capela, A., and Temple, S. (2002). mammalian brain. J. Neurosci. 17, entiation of neural stem cells in the tors via distinct signaling pathways. LeX/ssea-1 is expressed by adult 5046–5061. adult dentate gyrus. Keio J. Med. 59, J. Neurosci. 23, 5149–5160. mouse CNS stem cells, identify- Edgerton, V. R., de Leon, R. D., 79–83. doi: 10.2302/kjm.59.79 Belachew, S., Chittajallu, R., Aguirre, A. ing them as nonependymal. Neuron Harkema, S. J., Hodgson, J. A., Gomez-Nicola, D., Valle-Argos, A.,Yuan, X.,Kirby,M., Anderson, 35, 865–875. doi: 10.1016/S0896- London, N., Reinkensmeyer, B., Pallas-Bazarra, N., and S., et al. (2003). Postnatal NG2 6273(02)00835-8 D. J., et al. (2001). Retraining Nieto-Sampedro, M. (2011). proteoglycan-expressing progenitor Costantini, L. C., and Isacson, O. the injured spinal cord. J. Interleukin-15 regulates prolif- cells are intrinsically multipotent (2000). Immunophilin ligands and Physiol. 533, 15–22. doi: 10.1111/ eration and self-renewal of adult and generate functional neurons. GDNF enhance neurite branching j.1469-7793.2001.0015b.x neural stem cells. Mol. Biol. Cell 22, J. Cell Biol. 161, 169–186. doi: or elongation from developing Fu,S. L., Ma, Z.W., Yin, L.,Iannotti, 1960–1970. doi: 10.1091/mbc.E11- 10.1083/jcb.200210110 dopamine neurons in culture. C., Lu, P. H., and Xu, X. M. (2005). 01-0053 Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 12 Chan et al. Pathways regulating spinal cord NSPC Guo, F., Maeda, Y., Ma, J., Xu, Rev. Biochem. 69, 373–398. doi: Madiai,F., Hussain,S.R., Goettl, neurite outgrowth of adult J., Horiuchi, M., Miers, L., 10.1146/annurev.biochem.69.1.373 V. M.,Burry, R.W., Stephens, spinal cord motor neurons. et al. (2010). Pyramidal neu- Hugnot, J. P., and Franzen, R. (2011). R. L. Jr., and Hackshaw, K. V. Exp. Neurol. 220, 303–315. doi: rons are generated from The spinal cord ependymal region: (2003). Upregulation of FGF-2 10.1016/j.expneurol.2009.09.003 oligodendroglial progenitor a stem cell niche in the caudal cen- in reactive spinal cord astrocytes Namiki, J., and Tator, C. H. (1999). cells in adult piriform cortex. tral nervous system. Front. Biosci. following unilateral lumbar spinal Cell proliferation and nestin J. Neurosci. 30, 12036–12049. doi: 16, 1044–1059. doi: 10.2741/3734 nerve ligation. Exp. Brain Res. 148, expression in the ependyma of the 10.1523/JNEUROSCI.1360-10.2010 Johansson, C. B., Momma, S., Clarke, 366–376. adult rat spinal cord after injury. Hamilton, L.K., Truong,M. K., D. L., Risling, M., Lendahl, U., and Marques-Torrejon,M.A., Porlan, E., J. Neuropathol. Exp. Neurol. 58, Bednarczyk, M. R., Aumont, Frisen, J. (1999). Identification of Banito, A., Gomez-Ibarlucea, 489–498. doi: 10.1097/00005072- A., and Fernandes, K. J. (2009). a neural stem cell in the adult E., Lopez-Contreras, A. J., 199905000-00008 Cellular organization of the central mammalian central nervous system. Fernandez-Capetillo, O., et al. Paliouras, G. N., Hamilton, L. K., canal ependymal zone, a niche Cell 96, 25–34. doi: 10.1016/S0092- (2013). Cyclin-dependent kinase Aumont, A., Joppe, S. E., Barnabe- of latent neural stem cells in the 8674(00)80956-3 inhibitor p21 controls adult Heider, F., and Fernandes, K. adult mammalian spinal cord. Kennedy, S. G., Wagner, A. J., Conzen, neural stem cell expansion by J. (2012). Mammalian target Neuroscience 164, 1044–1056. doi: S. D.,Jordan, J.,Bellacosa, A., regulating Sox2 gene expression. of rapamycin signaling is a 10.1016/j.neuroscience.2009.09.006 Tsichlis, P. N., et al. (1997). The Cell Stem Cell 12, 88–100. doi: key regulator of the transit- Hashimoto, K., and Ishima, T. PI 3-kinase/Akt signaling pathway 10.1016/j.stem.2012.12.001 amplifying progenitor pool in (2011). Neurite outgrowth delivers an anti-apoptotic sig- Mason, I. (2007). Initiation to end the adult and aging forebrain. mediated by translation elon- nal. Genes Dev. 11, 701–713. doi: point: the multiple roles of fibrob- J. Neurosci. 32, 15012–15026. doi: gation factor eEF1A1: a target 10.1101/gad.11.6.701 last growth factors in neural 10.1523/JNEUROSCI.2248-12.2012 for antiplatelet agent cilosta- Kiddie, G., Mclean, D., Van Ooyen, A., development. Nat. Rev. Neurosci. 8, Palmer,T.D., Willhoite,A. R., and zol. PLoS ONE 6:e17431. doi: and Graham, B. (2005). Biologically 583–596. doi: 10.1038/nrn2189 Gage, F. H. (2000). Vascular niche 10.1371/journal.pone.0017431 plausible models of neurite out- Mattson, M. P., Guthrie, P. B., and for adult hippocampal neurogene- Hayashi, H., Tsuchiya, Y., Nakayama, growth. Prog. Brain Res. 147, Kater, S. B. (1988). Components of sis. J. Comp. Neurol. 425, 479–494. K.,Satoh, T.,and Nishida, E. 67–80. doi: 10.1016/S0079-6123 neurite outgrowth that determine Peltier, J., O’neill, A., and Schaffer, (2008). Down-regulation of the PI3- (04)47006-X neuronal cytoarchitecture: influence D. V. (2007). PI3K/Akt and kinase/Akt pathway by ERK MAP Lao, C. L., Lu, C. S., and Chen, J. C. of calcium and the growth substrate. CREB regulate adult neural kinase in growth factor signal- (2013). Dopamine D(3) recep- J. Neurosci. Res. 20, 331–345. doi: hippocampal progenitor prolif- ing. Genes Cells 13, 941–947. doi: tor activation promotes neural 10.1002/jnr.490200307 eration and differentiation. Dev. 10.1111/j.1365-2443.2008.01218.x stem/progenitor cell proliferation Meletis, K., Barnabe-Heider, F., Carlen, Neurobiol. 67, 1348–1361. doi: Hely, T. A., Graham, B., and Ooyen, through AKT and ERK1/2 path- M.,Evergren, E.,Tomilin,N., 10.1002/dneu.20506 A. V. (2001). A computational ways and expands type-B and -C Shupliakov, O., et al. (2008). Spinal Qi,M.L., Wakabayashi, Y., Haro, H., model of dendrite elongation cells in adult subventricular zone. cord injury reveals multilineage and Shinomiya, K. (2003). Changes and branching based on MAP2 Glia 61, 475–489. doi: 10.1002/ differentiation of ependymal in FGF-2 expression in the distal phosphorylation. J. Theor. Biol. glia.22449 cells. PLoS Biol. 6:e182. doi: spinal cord stump after complete 210, 375–384. doi: 10.1006/jtbi. Lendahl, U., Zimmerman, L. B., and 10.1371/journal.pbio.0060182 cord transection: a comparison 2001.2314 Mckay, R. D. (1990). CNS stem Ming, G. L., and Song, H. (2005). Adult between infant and adult rats. Spine Hollis, E. R. 2nd., Jamshidi, P., Low, cells express a new class of inter- neurogenesis in the mammalian (Phila Pa 1976) 28, 1934–1940. K., Blesch, A., and Tuszynski, M. H. mediate filament protein. Cell 60, central nervous system. Annu. Rev. doi: 10.1097/01.BRS.0000083323. (2009). Induction of corticospinal 585–595. doi: 10.1016/0092-8674 Neurosci. 28, 223–250. doi: 10.1146/ 38962.2A regeneration by lentiviral trkB- (90)90662-X annurev.neuro.28.051804.101459 Sabourin,J. C., Ackema,K.B., induced Erk activation. Proc. Natl. Li, J., Wang, G., Wang, C., Zhao, Y., Ming, G. L., and Song, H. (2011). Ohayon, D., Guichet, P. O., Perrin, Acad.Sci.U.S.A. 106, 7215–7220. Zhang, H., Tan, Z., et al. (2007). Adult neurogenesis in the F. E., Garces, A., et al. (2009). doi: 10.1073/pnas.0810624106 MEK/ERK signaling contributes mammalian brain: significant A mesenchymal-like ZEB1(+) Horky, L. L., Galimi, F., Gage, F. H., to the maintenance of human answers and significant ques- niche harbors dorsal radial glial and Horner, P. J. (2006). Fate embryonic stem cell self-renewal. tions. Neuron 70, 687–702. doi: fibrillary acidic protein-positive of endogenous stem/progenitor Differentiation 75, 299–307. doi: 10.1016/j.neuron.2011.05.001 stem cells in the spinal cord. Stem cells following spinal cord injury. 10.1111/j.1432-0436.2006.00143.x Mistry, S. K., Keefer, E. W., Cells 27, 2722–2733. doi: 10.1002/ J. Comp. Neurol. 498, 525–538. doi: Li,Z., Theus, M. H.,and Wei, L. Cunningham, B. A., Edelman, stem.226 10.1002/cne.21065 (2006). Role of ERK 1/2 signaling G. M., and Crossin, K. L. (2002). Sanchez, C., Diaz-Nido, J., and Avila, Horner, P. J., Power, A. E., in neuronal differentiation of cul- Cultured rat hippocampal neural J. (2000). Phosphorylation of Kempermann, G., Kuhn, H. G., tured embryonic stem cells. Dev. progenitors generate spontaneously microtubule-associated protein Palmer,T.D., Winkler, J., etal. Growth Differ. 48, 513–523. doi: active neural networks. Proc. Natl. 2 (MAP2) and its relevance for (2000). Proliferation and dif- 10.1111/j.1440-169X.2006.00889.x Acad.Sci.U.S.A. 99, 1621–1626. the regulation of the neuronal ferentiation of progenitor cells Lie, D. C., Colamarino, S. A., Song, H. doi: 10.1073/pnas.022646599 cytoskeleton function. Prog. throughout the intact adult J., Desire, L., Mira, H., Consiglio, Miura, T., Tanaka, S., Seichi, A., Neurobiol. 61, 133–168. doi: rat spinal cord. J. Neurosci. 20, A., et al. (2005). Wnt signalling reg- Arai, M., Goto, T., Katagiri, H., 10.1016/S0301-0082(99)00046-5 2218–2228. ulates adult hippocampal neuroge- et al. (2000). Partial functional Santen, R. J., and Agranoff, B. W. Huang, C. C., Lo, S. W., and Hsu, K. nesis. Nature 437, 1370–1375. doi: recovery of paraplegic rat by (1963). Studies on the estimation of S. (2001). Presynaptic mechanisms 10.1038/nature04108 adenovirus-mediated gene delivery deoxyribonucleic acid in rat brain. underlying cannabinoid inhibition Ma,D.K., Ponnusamy,K., Song, M. of constitutively active MEK1. Biochim. Biophys. Acta 72, 251–262. of excitatory synaptic transmission R., Ming, G. L., and Song, H. Exp. Neurol. 166, 115–126. doi: doi: 10.1016/0926-6550(63)90339-6 in rat striatal neurons. J. Physiol. (2009). Molecular genetic analysis 10.1006/exnr.2000.7493 Schlessinger, J. (2000). Cell signaling 532, 731–748. doi: 10.1111/j.1469- of FGFR1 signalling reveals distinct Montoya, G. J., Sutachan, J. J., Chan, by receptor tyrosine kinases. Cell 7793.2001.0731e.x roles of MAPK and PLCgamma1 W. S., Sideris, A., Blanck, T. 103, 211–225. doi: 10.1016/S0092- Hubbard,S.R., andTill, J. H. activation for self-renewal of adult J., and Recio-Pinto, E. (2009). 8674(00)00114-8 (2000). Protein tyrosine kinase neural stem cells. Mol. Brain 2, 16. Muscle-conditioned media and Shechter, R., Baruch, K., Schwartz, M., structure and function. Annu. doi: 10.1186/1756-6606-2-16 cAMP promote survival and and Rolls, A. (2010). Touch gives Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 13 Chan et al. Pathways regulating spinal cord NSPC new life: mechanosensation modu- stem cells. Nature 417, 39–44. doi: Eur. J. Neurosci. 26, 2777–2794. doi: (2006). Endogenous neurogenesis lates spinal cord adult neurogene- 10.1038/417039a 10.1111/j.1460-9568.2007.05871.x replaces oligodendrocytes and sis. Mol. Psychiatry 16, 342–352. doi: Suh, H.,Deng, W.,and Gage, F.H. Weiss, D. J.,Fried, G.W., Chancellor, astrocytes after primate spinal cord 10.1038/mp.2010.116 (2009). Signaling in adult neuro- M. B., Herbison, G. J., Ditunno, J. injury. J. Neurosci. 26, 2157–2166. Shechter, R., Ziv, Y., and Schwartz, genesis. Annu. Rev. Cell Dev. Biol. F. Jr., and Staas, W. E. Jr. (1996). doi: 10.1523/JNEUROSCI.4070- M. (2007). New GABAergic 25, 253–275. doi: 10.1146/annurev. Spinal cord injury and bladder 05.2005 interneurons supported by myelin- cellbio.042308.113256 recovery. Arch. Phys. Med. Rehabil. specific T cells are formed Suzuki-Yamamoto, T., Toida, K., 77, 1133–1135. doi: 10.1016/S0003- Conflict of Interest Statement: The in intact adult spinal cord. Tsuruo, Y., Watanabe, K., 9993(96)90135-5 authors declare that the research Stem Cells 25, 2277–2282. doi: and Ishimura, K. (2000). Xia, Z., Dickens, M., Raingeaud, J., was conducted in the absence of any 10.1634/stemcells.2006-0705 Immunocytochemical localiza- Davis, R. J., and Greenberg, M. E. commercial or financial relationships Shen, Q., Goderie, S. K., Jin, L., tion of lung-type prostaglandin (1995). Opposing effects of ERK that could be construed as a potential Karanth, N., Sun, Y., Abramova, F synthase in the rat spinal cord. and JNK-p38 MAP kinases on conflict of interest. N., et al. (2004). Endothelial Brain Res. 877, 391–395. doi: apoptosis. Science 270, 1326–1331. cells stimulate self-renewal and 10.1016/S0006-8993(00)02709-8 doi: 10.1126/science.270.5240.1326 Received: 17 April 2013; accepted: 01 expand neurogenesis of neural stem Tamura,Y., Kataoka, Y., Cui,Y., Xue, L., Murray, J. H., and August 2013; published online: 27 August cells. Science 304, 1338–1340. doi: Takamori, Y., Watanabe, Y., Tolkovsky, A. M. (2000). The 2013. 10.1126/science.1095505 and Yamada, H. (2007). Multi- Ras/phosphatidylinositol 3-kinase Citation: Chan WS, Sideris A, Sutachan Shen, Q., Wang, Y., Kokovay, E., Lin, directional differentiation and Ras/ERK pathways function JJ, Montoya G JV, Blanck TJJ and Recio- G.,Chuang, S. M.,Goderie,S. of doublecortin- and NG2- as independent survival modules Pinto E (2013) Differential regulation K., et al. (2008). Adult SVZ stem immunopositive progenitor cells each of which inhibits a distinct of proliferation and neuronal differen- cells lie in a vascular niche: a in the adult rat neocortex in vivo. apoptotic signaling pathway in tiation in adult rat spinal cord neural quantitative analysis of niche cell- Eur. J. Neurosci. 25, 3489–3498. doi: sympathetic neurons. J. Biol. stem/progenitors by ERK1/2, Akt, and cell interactions. Cell Stem Cell 10.1111/j.1460-9568.2007.05617.x Chem. 275, 8817–8824. doi: PLCγ. Front. Mol. Neurosci. 6:23. doi: 3, 289–300. doi: 10.1016/j.stem. Torroglosa, A., Murillo-Carretero, 10.1074/jbc.275.12.8817 10.3389/fnmol.2013.00023 2008.07.026 M., Romero-Grimaldi, C., Yamaguchi, H., and Wang, H. G. This article was submitted to the journal Shihabuddin, L. S., Horner, P. J., Ray, Matarredona, E. R., Campos-Caro, (2001). The protein kinase PKB/Akt Frontiers in Molecular Neuroscience. J., and Gage, F. H. (2000). Adult A., and Estrada, C. (2007). Nitric regulates cell survival and apopto- Copyright © 2013 Chan, Sideris, spinal cord stem cells generate neu- oxide decreases subventricular zone sis by inhibiting Bax conformational Sutachan, Montoya G, Blanck and rons after transplantation in the stem cell proliferation by inhibi- change. Oncogene 20, 7779–7786. Recio-Pinto. This is an open-access adult dentate gyrus. J. Neurosci. 20, tion of epidermal growth factor doi: 10.1038/sj.onc.1204984 article distributed under the terms of 8727–8735. receptor and phosphoinositide- Yamamoto, S., Yamamoto, N., the Creative Commons Attribution Shihabuddin, L. S., Ray, J., and Gage, 3-kinase/Akt pathway. Stem Cells Kitamura, T., Nakamura, K., and License (CC BY). The use, distribution F. H. (1997). FGF-2 is sufficient 25, 88–97. doi: 10.1634/stemcells. Nakafuku, M. (2001). Proliferation or reproduction in other forums is per- to isolate progenitors found in 2006-0131 of parenchymal neural pro- mitted, provided the original author(s) the adult mammalian spinal cord. Vessal, M.,Aycock, A.,Garton, M. T., genitors in response to injury or licensor are credited and that the Exp. Neurol. 148, 577–586. doi: Ciferri, M., and Darian-Smith, C. in the adult rat spinal cord. original publication in this journal 10.1006/exnr.1997.6697 (2007). Adult neurogenesis in pri- Exp. Neurol. 172, 115–127. doi: is cited, in accordance with accepted Song, H., Stevens, C. F., and Gage, mate and rodent spinal cord: com- 10.1006/exnr.2001.7798 academic practice. No use, distribution F. H. (2002). Astroglia induce paring a cervical dorsal rhizotomy Yang, H., Lu, P., Mckay, H. M., Bernot, or reproduction is permitted which does neurogenesis from adult neural with a dorsal column transection. T., Keirstead, H., Steward, O., et al. not comply with these terms. Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 14 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Frontiers in Molecular Neuroscience Unpaywall

Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord neural stem/progenitors by ERK1/2, Akt, and PLCγ

Download PDF

Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord neural stem/progenitors by ERK1/2, Akt, and PLCγ

Abstract

ORIGINAL RESEARCH ARTICLE published: 27 August 2013 MOLECULAR NEUROSCIENCE doi: 10.3389/fnmol.2013.00023 Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord neural stem/progenitors by ERK1/2, Akt, and PLCγ 1 1 2 1 1 Wai Si Chan , Alexandra Sideris , Jhon J. Sutachan , Jose V. Montoya G , Thomas J. J. Blanck and Esperanza Recio-Pinto * Department of Anesthesiology, New York University Langone Medical Center, New York, NY, USA Departamento de Nutrición y Bioquímica, Pontificia Universidad Javeriana, Bogotá, Colombia Edited by: Proliferation of endogenous neural stem/progenitor cells (NSPCs) has been identified in Andreas Vlachos, Goethe University both normal and injured adult mammalian spinal cord. Yet the signaling mechanisms Frankfurt, Germany underlying the regulation of adult spinal cord NSPCs proliferation and commitment toward Reviewed by: a neuronal lineage remain undefined. In this study, the role of three growth factor-mediated Sebastian Jessberger, University of signaling pathways in proliferation and neuronal differentiation was examined. Adult spinal Zurich, Switzerland Li Zhang, National Institute on cord NSPCs were enriched in the presence of fibroblast growth factor 2 (FGF2). We Alcohol Abuse and Alcoholism-NIH, observed an increase in the number of cells expressing the microtubule-associated protein USA 2 (MAP2) over time, indicating

Loading next page...
 
/lp/unpaywall/differential-regulation-of-proliferation-and-neuronal-differentiation-1I1Dy4B4Sb

References (153)

Publisher
Unpaywall
ISSN
1662-5099
DOI
10.3389/fnmol.2013.00023
Publisher site
See Article on Publisher Site

Abstract

ORIGINAL RESEARCH ARTICLE published: 27 August 2013 MOLECULAR NEUROSCIENCE doi: 10.3389/fnmol.2013.00023 Differential regulation of proliferation and neuronal differentiation in adult rat spinal cord neural stem/progenitors by ERK1/2, Akt, and PLCγ 1 1 2 1 1 Wai Si Chan , Alexandra Sideris , Jhon J. Sutachan , Jose V. Montoya G , Thomas J. J. Blanck and Esperanza Recio-Pinto * Department of Anesthesiology, New York University Langone Medical Center, New York, NY, USA Departamento de Nutrición y Bioquímica, Pontificia Universidad Javeriana, Bogotá, Colombia Edited by: Proliferation of endogenous neural stem/progenitor cells (NSPCs) has been identified in Andreas Vlachos, Goethe University both normal and injured adult mammalian spinal cord. Yet the signaling mechanisms Frankfurt, Germany underlying the regulation of adult spinal cord NSPCs proliferation and commitment toward Reviewed by: a neuronal lineage remain undefined. In this study, the role of three growth factor-mediated Sebastian Jessberger, University of signaling pathways in proliferation and neuronal differentiation was examined. Adult spinal Zurich, Switzerland Li Zhang, National Institute on cord NSPCs were enriched in the presence of fibroblast growth factor 2 (FGF2). We Alcohol Abuse and Alcoholism-NIH, observed an increase in the number of cells expressing the microtubule-associated protein USA 2 (MAP2) over time, indicating neuronal differentiation in the culture. Inhibition of the *Correspondence: mitogen-activated protein kinase or extracellular signal-regulated kinase (ERK) kinase 1 Esperanza Recio-Pinto, Laboratory and 2/ERK 1 and 2 (MEK/ERK1/2) or the phosphoinositide 3-kinase (PI3K)/Akt pathways of Dr. Thomas J. J. Blanck, Department of Anesthesiology, New suppressed active proliferation in adult spinal cord NSPC cultures; whereas neuronal York University Langone Medical differentiation was negatively affected only when the ERK1/2 pathway was inhibited. Center, 180 Varick Street Room 673, Inhibition of the phospholipase Cγ (PLCγ) pathway did not affect proliferation or neuronal New York, NY 10014, USA differentiation. Finally, we demonstrated that the blockade of either the ERK1/2 or PLCγ e-mail: esperanza.recio-pinto@ nyumc.org signaling pathways reduced neurite branching of MAP2+ cells derived from the NSPC cultures. Many of the MAP2+ cells expressed synaptophysin and had a glutamatergic phenotype, indicating that over time adult spinal cord NSPCs had differentiated into mostly glutamatergic neurons. Our work provides new information regarding the contribution of these pathways to the proliferation and neuronal differentiation of NSPCs derived from adult spinal cord cultures, and emphasizes that the contribution of these pathways is dependent on the origin of the NSPCs. Keywords: Neuronal differentiation, ERK1/2, Akt, PLCγ, progenitors, spinal cord INTRODUCTION spinal cord, the question remains as to how these adult spinal cord Early studies of the adult mammalian central nervous system NSPCs are regulated. A better understanding of the basic biology identified the presence of neural stem/progenitor cells (NSPCs) in of these NSPCs will facilitate future attempts to manipulate these the adult spinal cord (Weiss et al., 1996; Shihabuddin et al., 1997). cells under pathological conditions. Subsequent research showed that proliferation of the adult NSPCs Unlike in the adult spinal cord, the occurrence of neurogenesis are stimulated in animal models of spinal cord injury (Johansson in the adult hippocampus has been firmly established (Alvarez- et al., 1999; Horner et al., 2000; Yamamoto et al., 2001; Danilov Buylla and Garcia-Verdugo, 2002; Ming and Song, 2005, 2011). et al., 2006). Some studies reported that the fate of the prolif- Astrocytes from this brain region have been shown to induce erating NSPCs is restricted to glial cell types in vivo due to the neurogenesis of adult hippocampal NSPCs via the Wnt signal- inhibitory microenvironment of the adult spinal cord (Johansson ing pathway (Song et al., 2002; Lie et al., 2005). Furthermore, et al., 1999; Shihabuddin et al., 2000; Horky et al., 2006; Yang diffusible factors from the neurovascular niche are reported to et al., 2006; Barnabe-Heider et al., 2010). Nonetheless, the con- stimulate neurogenesis in the adult subventricular zone (Palmer sensus is that adult spinal cord NSPCs are intrinsically multi- et al., 2000; Shen et al., 2004, 2008). On the other hand, astro- potent (i.e., they can also generate neurons), as demonstrated cytes from the adult spinal cord do not promote neurogenesis in by transplantation studies and neurosphere assays (Shihabuddin culture (Song et al., 2002). et al., 2000; Yamamoto et al., 2001). In fact, neurogenesis in the Growth factors, such as fibroblast growth factor 2 (FGF2), adult spinal cord has been detected in vivo in both pathological epidermal growth factor, nerve growth factor, and vascular (Danilov et al., 2006; Vessal et al., 2007) and normal conditions endothelial growth factor, can elicit a range of cellular responses (Shechter et al., 2007, 2010). However, as we gain more insight including cell proliferation, migration, differentiation, and cell into the existence and implications of neurogenesis in the adult death through various classes of receptor tyrosine kinases (RTKs) Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 1 Chan et al. Pathways regulating spinal cord NSPC (Hubbard and Till, 2000). The activation of these different RTKs nestin+ were mostly rounded/spindle-shaped. The morphology in turn induces the activation of several signal transduction of the cells co-labeled with nestin and NG2 was similar to that of pathways including the mitogen-activated protein kinase or extra- the nestin+ cells (Figure 1G,iii) . Based on the cell morphology, cellular signal-regulated kinase (ERK) kinase 1 and 2/ERK 1 and some of the EdU+ cells that were not immunopositive for nestin 2 (MEK/ERK1/2), phosphoinositide 3-kinase (PI3K)/Akt, and were likely NG2+ (not shown). Stage-specific embryonic anti- phospholipase Cγ (PLCγ)pathways (Huang et al., 2001; Mason, gen 1 (SSEA-1), a marker for undifferentiated NSPCs (Capela and 2007). Much effort has gone into elucidating the signaling cas- Temple, 2002; Sabourin et al., 2009), was observed in about one- cades involved in the neurogenesis of adult hippocampal NSPCs. tenth of the cells in culture at 3 DIV (Figure 1H). Taken together, For instance, both the MEK/ERK1/2 and PI3K/Akt pathways have the expression of these markers demonstrates the presence of been shown to be involved in regulating proliferation and self- uncommitted NSPCs at 1–4 DIV. renewal (Peltier et al., 2007; Ma et al., 2009), while the PLCγ pathway is reported to be essential for neuronal differentiation NEURONAL DIFFERENTIATION IN ADULT SPINAL CORD NSPC (Ma et al., 2009) of adult hippocampal NSPCs. Yet the role of CULTURES these signaling pathways in adult spinal cord NSPCs has not been Maintaining the expanded population of adult spinal cord NSPCs established. In this study, our aim was to assess the potential con- in medium containing FGF2 eventually resulted in neuronal tribution of the MEK/ERK1/2, PI3K/Akt, and PLCγ pathways in differentiation. At 0 DIV and 1 DIV, about 23 and 12% of proliferation and neuronal differentiation of NSPCs derived from the cells were doublecortin-positive (DCX+) immature neu- the adult rat spinal cord culture. Our findings reinforce the idea rons (Figure 2A) (the remainder of the cells are most likely that the contribution of signaling pathway activation in prolifera- nestin+ and/or NG2+ cells, see Figure 1G,ii). Over time, there tion and neuronal differentiation is unique in different cell types was an increase in the number of cells expressing the neuronal and is dependent on the source of NSPCs. marker, microtubule-associated protein 2 (MAP2) (Figure 2B i). The number of MAP2+ cells increased by nearly 4-fold by RESULTS 6DIV (Figure 2B,ii) and about 6-fold by 14 DIV (Figure 2C). ENRICHMENT OF ADULT SPINAL CORD-DERIVED NSPCs To demonstrate that some of these MAP2+ cells were newly generated, at 5 DIV proliferating cells were labeled with a 24 h The enriched adult spinal cord NSPCs were obtained by using a EdU pulse, and the culture was immunolabeled for MAP2 at cell isolation protocol that yielded about 1% of the total num- 6 DIV. After the pulse of EdU, about a quarter of all MAP2+ ber of cells in the spinal cord (see Materials and Methods) cells were found to be EdU+ (Figure 2D,i,ii). When FGF2 was (Figure 1A); these cells could proliferate as neurospheres (not not added to the culture medium there was no increase in the shown) or in adherent cultures in medium containing FGF2 number of MAP2+ cells nor an increase in total cell number (Figure 1B). Adherent cultures were used because we could assess (not shown). Antibodies against vesicular glutamate transporter proliferation in a shorter time frame thus avoiding possible phe- 1 (VGLUT1) and glutamic acid decarboxylase 67 (GAD-67) were notypic changes and/or selection of a subpopulation of NSPCs used to further characterize the neuronal phenotype of these as a result of long term cultivation and passaging. In addition, newly generated MAP2+ cells (i.e., whether they were excitatory adherent cultures in a single layer also facilitated identification or inhibitory neurons). We found that 42% of all the MAP2+ cells and quantification of immunolabeled cells with various markers. were VGLUT1+ (Figure 2E). Moreover, 96% of all the VGLUT1+ Under such condition, the total cell number increased by 1.7-fold cells were negative for GAD-67 (Figure 2F), suggesting that adult by 6 days in vitro (DIV) when compared to 0 DIV (Figure 1C). A spinal cord-derived NSPCs differentiate toward neurons with an thymidine analogue, EdU, was added to the culture to pulse-label excitatory phenotype. those cells undergoing cell division at 4 DIV, 6 h prior to fixa- tion (Figure 1D,i). During the 6-h pulse of EdU-labeling, 25% PROLIFERATION OF ADULT SPINAL CORD NSPC CULTURES REQUIRES of the cells were EdU+ (Figure 1D,ii). In addition, co-labeling of EdU and nestin, used here as an adult NSPC marker (Lendahl ACTIVATION OF ERK1/2 AND Akt et al., 1990; Johansson et al., 1999; Namiki and Tator, 1999; Fu To examinewhich of thesignaling pathways might beimportant et al., 2005), revealed that 32 ± 5% (n = 4) of nestin+ cells for proliferation, specific pharmacological inhibitors were each were EdU+ and that 46 ± 12% (n = 4) of EdU+ cells were applied to the culture for 4 days, to which EdU was also added nestin+ (Figure 1E). Immunostaining with nestin at 0, 1, and for 6 h on day 4. We found that treatment with 10 μM U0126 4 DIV showed an increase of nestin+ cells over time; with a (MEK1/2 inhibitor) or 10 μM LY294002 (PI3K inhibitor) sig- 12-fold increase in percentage by 4 DIV (Figure 1F). Cells were nificantly reduced the number of EdU+ cells compared to the also labeled with another progenitor marker, neural-glial antigen vehicle control (Figure 3A). This indicates that the activation of 2 (NG2), as studies have shown that NG2+ cells are prolifera- both MEK/ERK and PI3K/Akt signaling pathways contributed to tive and can give rise to neurons (Belachew et al., 2003; Tamura the proliferation of adult spinal cord-derived NSPCs. Consistent et al., 2007; Guo et al., 2010). At 1 DIV 28% of the cells were with this interpretation, we found that both treatments with immunopositive for NG2 (Figure 1G,i,ii) and 19% of the nestin+ U0126 and LY294002 for 4–6 days resulted in the suppression cells were also NG2+ at 1 DIV (Figure 1G,ii). NG2 and nestin of an increase in the total cell number (total Hoechst-stained coexpressing cells were also detected at 3 DIV (Figure 1G,iii). nuclei) (Figure 3B). In contrast, there was no indication that the The cells that were NG2+ were stellate-like, with short pro- PLCγ pathway was involved in NSPC proliferation, as 2.5 μM cesses extending from all directions; and the cells that were only U73122 (PLCγ inhibitor) did not significantly affect the number Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 2 Chan et al. Pathways regulating spinal cord NSPC FIGURE 1 | Enrichment of adult rat spinal cord neural stem/progenitor picture of the green and red channels showing EdU (green) and nestin (red) cells (NSPCs) in culture. (A) Dissociated adult spinal cord cells, 1 h after staining. Blue indicates Hoechst-stained nuclei. (F) Quantification of the isolation. (B) Spinal cord cells after 4 days in FGF2 (i) or in basal medium (ii). percentage of nestin+ cells showing that the population of NSPCs was (C) Adult spinal cord cells proliferated as the total cell number per coverslip enriched in culture medium containing FGF2 over time (20 random fields with ∗∗ increased over time in culture medium containing FGF2 [all the cells in each a40× objective per coverslip were analyzed, n = 4 coverslips; P < 0.01, † # coverslip were counted, n = 3 coverslips; P < 0.001 compared with 0 days P ≤ 0.0001, One-Way ANOVA Dunnett’s multiple comparison post-test). (G) in vitro (DIV), and between 4 DIV and 6 DIV ± FGF2, One-Way ANOVA Co-labeling of NG2 chondronitin sulfate proteoglycan (green) and nestin (red) Dunnett’s multiple comparison post-test]. (D) A thymidine analogue, EdU, at 1 DIV (i) and 3 DIV (iii). At 1 DIV, most of the NG2+ cells had distinct was added to culture medium at 4 DIV, 6 h before fixation. Proliferating morphology from nestin+ cells; while cells coexpressing both markers (actively dividing) cells were EdU-labeled (green) during the 6 h pulse (i). Blue were observed at 1 and 3 DIV. Blue indicates Hoechst-stained nuclei. (ii) indicates Hoechst-stained nuclei. (ii) Quantification of the percentage of Quantification of the percentage of nestin+, NG2+, and nestin+/NG2+ cells EdU+ cells (10 random fields with a 40× objective per coverslip were at 1 DIV (20 random fields with a 40× objective per coverslip were analyzed, analyzed, n = 3 coverslips; P < 0.005, unpaired t-test). (E) Co-labeling of n = 4 coverslips). (H) At 3 DIV, stage-specific embryonic antigen 1 (SSEA-1) EdU and nestin showing that some of the NSPCs (4 DIV) were actively (green), a cell surface carbohydrate epitope found on uncommitted NSPCs, dividing. EdU was added to culture medium for only 6 h before fixation at appeared to localize to the plasma membrane of cells with undifferentiated 4 DIV. (i) Proliferating (EdU+) cells (green) during the 6 h pulse. (ii) Merged morphology. All scale bars: 20 μm. of EdU+ cells or the total cell number when compared to the cells compared to the vehicle control. Z-VAD-FMK, a general cas- vehicle control (Figures 3A,B). TUNEL analysis was performed to pase inhibitor, did not decrease the number of TUNEL-positive address whether the inhibitors affected cell survival. None of the cells in the presence of any of the inhibitors (Figure 3C), indicat- inhibitors significantly increased the number of TUNEL-positive ing that the decrease in EdU+ cell and total cell number as a result Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 3 Chan et al. Pathways regulating spinal cord NSPC FIGURE 2 | Neuronal differentiation in adult spinal cord NSPC cultures. Examples of newly generated MAP2+ cells (green EdU+ nuclei and red (A) Immunostaining for doublecortin (DCX) at 0 DIV (top left panel) and 1 DIV MAP2+ somata) by 6 DIV. As EdU was added to the culture medium for a 24-h (bottom left panel). Blue indicates Hoechst-stained nuclei. (Right panel) period at 5 DIV, only those cells that were dividing then were labeled green. Quantification of the percentage of DCX+ cells at both time points (20 random Blue indicates Hoechst-stained nuclei. (ii) Quantification of the percentage of fields with a 40× objective per coverslip were analyzed, n = 4 coverslips). (B) EdU+/MAP2+ cells at 6 DIV (10 random fields per coverslips were analyzed, i, Microtubule-associated protein 2 (MAP2) staining (green) of adult rat spinal n = 8 coverslips from three experiments). (E) Immunostaining for vesicular cord cultures at 0 DIV, arrowheads denote MAP2+ cells (left), and at 6 DIV in glutamate transporter 1 (VGLUT1) and MAP2 revealed that 42% of the culture medium containing FGF2 (right). Blue indicates Hoechst-stained nuclei. MAP2+ neurons had a glutamatergic phenotype (n = 539 cells). Cultures (ii) Fold change of the number of MAP2+ cells at 6 DIV (n = 13 coverslips from were counterstained with Hoechst (blue). (F) The majority (96%) of VGLUT1+ five experiments; P < 0.0001, unpaired t-test). The data was normalized to cells were negative for glutamic acid decarboxylase 67 (GAD-67) at 6 DIV. (i) A the mean total number of cells at 0 DIV. (C) As the population of NSPCs was scatter plot of the fluorescent intensity of each individual cell (n = 554 cells). enriched, many of the cells differentiated toward a neuronal lineage as the Red and green lines denote “cut-offs” for VGLUT1 and GAD-67 number of MAP2+ cells increased over time (10 random fields with a 40× immunoreactivity, respectively. Cells with intensity values above each line ∗∗ objective per coverslip were analyzed, n = 3 coverslips; P < 0.01, were considered positively stained. (ii) Example of VGLUT1+/GAD-67- cells at P < 0.001, One-Way ANOVA Dunnett’s multiple comparison post-test). (D) i, 6 DIV. Cells were counterstained with Hoechst (blue). Scale bars: 20 μm. of the treatment with U0126 or LY294002 was not due to apopto- NEURONAL DIFFERENTIATION IN ADULT SPINAL CORD NSPC sis. The specificity of each of the pharmacological inhibitors used CULTURES IS ERK1/2-DEPENDENT was verified in the adult spinal cord culture enriched in NSPCs To evaluate whether each of the inhibitors might influence the by Western blot analysis. By using antibodies specific for ERK1/2, neuronal differentiation potential of NSPCs, we examined the Akt, and PLCγ1 and their phosphorylated isoforms, the ratio of number of MAP2+ neurons in each treatment. We found that phosphorylated protein to total protein (pERK1/2 to ERK1/2, neuronal differentiation of the adult spinal cord-derived NSPCs pAkt to Akt, and pPLCγ1to PLCγ1) in thepresenceofeachof was only suppressed by treatment with 10 μM U0126, inhibitor the inhibitor was measured. We confirmed that, at the concentra- of MEK and therefore ERK1/2 activation, such that there was tions used in the cultures, each of the three inhibitors specifically 40% less MAP2+ neuronsascompared to the DMSO control and significantly blocked the phosphorylation of ERK1/2, Akt, by 6 DIV (Figures 4A i,ii, B). Inhibition of the PI3K/Akt and and PLCγ1(Figure 3D). PLCγ pathways had no significant effect on the number of cells Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 4 Chan et al. Pathways regulating spinal cord NSPC FIGURE 3 | Cell proliferation in adult spinal cord NSPC cultures requires each of the inhibitors. None of the treatment significantly changed cell activation of ERK1/2 and Akt. (A) Quantification of the normalized number survival as compared to DMSO control, even with the addition of Z-VAD-FMK of EdU+ cells showed that inhibition of ERK1/2 (with MEK inhibitor, U0126, (20 μM), a general caspase inhibitor (20 random fields with a 20× objective 10 μM) and Akt (with PI3K inhibitor, LY294002, 10 μM) activation suppressed per coverslip were analyzed, n = 3 coverslips for all except for the first bar proliferation of adult spinal cord cells at 4 DIV; inhibition of PLCγ activation (DNase as a positive control for the TUNEL assay): n = 2 coverslips). (D) i–iii, (with U73122, 2.5 μM) had no effect (10 random fields with a 40× objective Western blot analysis showed that the MEK/ERK1/2, PI3K/Akt, and PLCγ ∗∗ † per coverslip were analyzed, n = 3 coverslips; P < 0.01, P < 0.001, pathways were activated in adult spinal cord cultures (6 DIV). The right panel One-Way ANOVA Bonferroni’s multiple comparison post-test). EdU was shows the ratios of phosphorylated protein to total protein (pERK1/2 to added to culture medium 6 h before fixation. (B) Normalized total number of ERK1/2, pAkt to Akt, and pPLCγ1toPLCγ1) in the presence of each of the Hoechst-stained nuclei to the DMSO control grouped showed that inhibition inhibitors. Each of the pharmacological inhibitors specifically blocked of both ERK1/2 and Akt activation decreased total number of cells at 4–6 DIV phosphorylation of its respective target protein when exposed to FGF2. (n = 9 coverslips, except LY294002: n = 8 coverslips, and U73122: n = 6 Quantified results of the blots generated using the ImageJ software # ∗∗ † coverslips, from three experiments; P ≤ 0.0001, One-Way ANOVA (n = 3 experiments; *P < 0.05, P < 0.01, P < 0.001, One-Way ANOVA Bonferroni’s multiple comparison post-test). All values were normalized to the Bonferroni’s multiple comparison post-test). (iv) Total PLCγ protein was pulled DMSO control group in each experiment. (C) Quantification of the number of down and then the membrane was blotted for tyrosine residues. Upper right TUNEL-positive cells at 4 DIV to determine cell survival after treatment with corner shows that PLCγ phosphorylation was blocked by 2.5 μM of U73122. expressing MAP2 (Figures 4A i,iii,iv, B). Taken together, this After 6 days, the culture medium containing FGF2 was supple- indicates that the MEK/ERK1/2 pathway had a crucial role in mented instead with neurite outgrowth-inducing factors (BDNF, neuronal differentiation. GDNF, and cAMP) for 7 more days (except for the NB-A/B27 control condition in which no exogenous growth factor was NEURITE BRANCHING OF MAP2+ NEURONS DERIVED FROM ADULT added) (Figure 5A). By 13 DIV, added growth factors doubled SPINAL CORD NSPC CULTURES REQUIRES ACTIVATION OF ERK1/2 AND the number of neurite branches in DMSO (vehicle) control from PLCγ 4.0 to 7.9 branches per cell; whereas the number of branches in Previous studies on PC12, adult rat dorsal root ganglion neu- NB-A/B27 control with no added growth factor remained sim- rons, and corticospinal motor neurons have reported that neurite ilar to that measured at 6 DIV (4.6 vs. 4.0 branches per cell) outgrowth involves growth factor-mediated ERK1/2 activation (Figure 5C; dotted line indicates level at 6 DIV). Hence, the addi- tion of exogenous factors (i.e., BDNF, GDNF, and cAMP) indeed (Hollis et al., 2009; Hashimoto and Ishima, 2011). To investi- gate whether the MEK/ERK1/2 pathway, as well as the PI3K/Akt promoted further neuronal differentiation of MAP2+ neurons, as indicated by the further increase in neurite outgrowth (com- and PLCγ signaling pathways played a role in neurite outgrowth, both neurite branching and the average length of the longest neu- pare Figure 2B with Figure 5B). We found that neurite branching of MAP2+ neurons was significantly suppressed when activa- rite of MAP2+ neurons were quantified in cultures treated with the corresponding inhibitors and vehicle control (Figures 5A,B). tion of ERK1/2 and PLCγ was blocked (5.2 and 6.1 branches per Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 5 Chan et al. Pathways regulating spinal cord NSPC and were therefore more differentiated (Figure 5D). By adding EdU to the culture medium at 5 DIV for 24 h, followed by a switch to the medium with neurite-inducing factors for 7 days (as in Figure 5A), we found that some of the MAP2+ cells with pro- cesses that came in contact with neighboring MAP2+ cells were also EdU+, once again confirming that they were newly gener- ated neurons (Figure 5E). Co-labeling of GAD-67 and VGLUT1 revealed that 63.3% of the cells were only VGLUT1+ at 14 DIV, while 97.4% of the VGLUT1+ cells were negative for GAD-67 (n = 428 cells from two coverslips) (Figures 5F,G). This suggests that many of the cells in culture remained glutamatergic since 6DIV (Figure 2E). DISCUSSION Since the role of the MEK/ERK1/2, PI3K/Akt, and PLCγ signaling pathways in adult spinal cord NSPCs has not been established, we assessed the contribution of these pathways to the prolifera- tion and neuronal differentiation of NSPCs derived from adult rat spinal cord cultures. Proliferation was evaluated using the thymidine analogue, EdU; while neuronal differentiation was demonstrated through a time course of MAP2 staining and neurite outgrowth, as well as by the appearance of synapto- physin in the neurites. First, our data demonstrate that both the MEK/ERK1/2 and PI3K/Akt pathways are essential for pro- liferation of adult spinal cord NSPCs. Second, we show that the activation of ERK1/2 is necessary for directing adult spinal cord NSPCs toward a neuronal fate. Finally, our data indicate that neu- rite branching is dependent on the activation of the MEK/ERK1/2 and PLCγ pathways. Currently there are only few publications that describe the FIGURE 4 | Neuronal differentiation in adult spinal cord NSPC cultures stem cell niche in the adult spinal cord (Sabourin et al., 2009; is ERK1/2-dependent. (A) Immunostaining for MAP2 in adult spinal cord Hugnot and Franzen, 2011). Among those studies, cells that NSPC cultures after 6 DIV in vehicle control (DMSO) (i), U0126 (10 μM) (ii), express the immature neuronal marker, DCX were detected LY294002 (10 μM) (iii), and U73122 (2.5 μM) (iv). Compared to the vehicle control (i), the number of MAP2+ cells was noticeably less when the throughout the ependymal region in the adult spinal cord, where cells were treated with the MEK inhibitor, U0126 (ii). Blue represents occasional proliferating nestin+ cells were also found (Hamilton Hoechst-stained nuclei. Scale bar: 20 μm. (B) Quantification of the et al., 2009; Sabourin et al., 2009; Hugnot and Franzen, 2011). normalized number of MAP2+ cells in adult spinal cord NSPC cultures In addition, proliferating NG2+ cells found throughout the (6 DIV). By 6 days, there was a similar increase in MAP2+ cell number as parenchyma in the adult spinal cord have been reported to coex- shown in Figure 2. Only U0126 (which inhibited ERK1/2 activation) significantly inhibited the increase of the number of MAP2+ cells after 6 press DCX (Shechter et al., 2010), even though conventionally days. LY294002 (10 μM) and U73122 (2.5 μM) had no significant effect on NG2 is a marker for oligodendrocyte progenitor cells. Together neuronal differentiation (10 random fields with a 40× objective per coverslip with in vivo studies which describes coexpression of NG2 and were analyzed, n = 6 coverslips, except LY294002: n = 5 coverslips and † # DCX in the adult neocortex (Tamura et al., 2007)and in the U73122: n = 3 coverslips; P < 0.001, P ≤ 0.0001, One-Way ANOVA adult piriform cortex (Guo et al., 2010), these findings suggest Bonferroni’s multiple comparison post-test). All values are normalized to the DMSO control in each experiment. that at least a subpopulation of NG2 cells can be less committed to the oligodendrocyte lineage than previously thought and has the potential to give rise to neurons. cell, respectively) but not when activation of Akt was blocked Nestin, a NSPC marker (Lendahl et al., 1990; Johansson et al., (7.7 branches per cell). It is interesting to note that while neu- 1999; Namiki and Tator, 1999; Fu et al., 2005), was used in this rite branching was ERK1/2- and PLCγ-dependent, the inhibition study to characterize the expansion of adult spinal cord NSPCs in of these two signaling pathways in turn resulted in an increase culture medium containing FGF2. We showed that by 4 DIV 50% in neurite length of the MAP2+ neurons (from 91 μmat 6DIV of the cells were nestin+ and 30% were MAP2+.Based on the to 130.4 μm and 125.4, respectively) (Figure 5C). The ability of experiments done at early time points (at 0–3 DIV), the remain- these MAP2+ cells to extend neurites also indicates that neurite ing population of the cells are most likely NG2, DCX, and/or elongation of these cells is not ERK1/2- and PLCγ-dependent. SSEA-1 cells. The presence of cells expressing various progeni- The expression of synaptophysin in MAP2+ neurons indicates tor markers (nestin, NG2, and SSEA-1) indicate the presence of that those MAP2+ neurons in the control condition began to different uncommitted progenitor cells with a range of poten- establish synapses with neighboring cells after 2 weeks of culture tials, and hence, the term “neural stem/progenitor cells (NSPCs).” Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 6 Chan et al. Pathways regulating spinal cord NSPC FIGURE 5 | Neurites branching requires ERK1/2 and PLCγ activation; the neurite length of the MAP2+ neurons (U0126: 130.4 ± 5.6 μm, n = 83 some neurons form synapses and are glutamatergic. (A) Schematic of cells; U73122: 125.4 ± 4.4 μm, n = 96 cells). PI3K inhibitor, LY294002 experimental protocol used in the neurite studies. FGF2 was present in (10 μM), had no effect on neurite branching or elongation (n = 98 cells). culture to expand nestin+ NSPCs (0–6 DIV). Culture medium containing Inhibitor(s) or vehicle control (DMSO) was added to culture from 6 DIV, FGF2 was then replaced with BDNF, GDNF (GFs), and cAMP to promote except in the last condition (NB-A/B27) in which the culture was further differentiation of MAP2+ neurons (6–13 DIV). (B) A typical MAP2+ maintained in basal medium only with no growth factor treatment (For ∗∗ † neuron derived from an adult spinal cord NSPC culture (13 DIV). The each condition, three coverslips were analyzed; P < 0.01, P < 0.001, number of neurite branches of each MAP2+ neuron in each treatment was One-Way ANOVA Bonferroni’s multiple comparison post-test). (D) MAP2+ analyzed by scoring those branches that intersected the perimeter of a (red) neurons displayed synapse formation with one another, as evidenced 30 μm-circle. The mean neurite length was quantified by measuring the by co-labeling with anti-synaptophysin (syn, green) (13 DIV). Scale bar: length of the longest neurite of each MAP2+ neuron in each treatment. 5 μm. (E) Evidence of newly generated neurons (cells with green EdU+ Scale bar: 20 μm. (C) Quantification of neurite branching (solid bars) and nuclei and white MAP2+ labeling) at 13 DIV with processes that came in neurite length (patterned bars) of MAP2+ cells (13 DIV). Dotted gray line contact with neighboring cells. EdU was added to culture medium at 5 DIV represents average measurements (left Y-axis: number of neurite branches; for 24hr, followed by a switch of culture medium as described in (A). Scale right Y-axis: mean neurite length) at 6 DIV before growth factor treatment. bar: 20 μm. (F) i, Co-labeling of GAD-67 (green) and VGLUT1 (red) of adult By 13 DIV, added growth factors increased neurite branches in DMSO spinal cord NSPC cultures at 14 DIV showed that the majority of the cells control from 4.0 ± 0.4(at 6DIV) to7.9 ± 0.3 (at 13 DIV) (n = 25 and 87 were only VGLUT1+. (ii) Immunostaining negative control in the same cells, respectively). The number of branches in NB-A/B27 control with no culture condition as (i) at 14 DIV. Blue represents Hoechst-stained nuclei. added growth factor (4.6 ± 0.3 at 13 DIV; n = 94 cells) remained similar to Insets show phase contrast pictures of the same field. Scale bar: 20 μm. that at 6 DIV (4.0 ± 0.4; n = 25 cells). Neurite branching of MAP2+ (G) Scatter plot of the fluorescent intensity of each individual cell analyzed, neurons was significantly suppressed when the activation of ERK1/2 and which showed that 63.3% of the cells were VGLUT1+ at 14 DIV; while PLCγ was respectively blocked by 10 μM U0126 (5.2 ± 0.3at13DIV; 97.4% of the VGLUT1+ cells were negative for GAD-67 (n = 428 cells from n = 83 cells) and 2.5 μM U73122 (6.1 ± 0.2 at 13 DIV; n = 96 cells). The two coverslips were analyzed). Red and green lines denote the “cut-off” blockage of the activation of ERK1/2 and PLCγ resulted in an increase in for VGLUT1 and GAD-67 immunoreactivity, respectively. Moreover, we can infer from our data that by 14 DIV essentially With the use of proliferative markers, studies in blastocyst- all of the NSPCs in our study had undergone neuronal differ- derived embryonic stem cells (ESCs) (Li et al., 2007)and in entiation as suggested by the expression of the mature neuronal primary cultures of embyronic (E12.5-E13.5) cortical progen- marker, MAP2, as well as synaptophysin and VGLUT1. itor cells (Barnabe-Heider and Miller, 2003) indicate that the Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 7 Chan et al. Pathways regulating spinal cord NSPC activation of Akt, but not ERK1/2, is required for proliferation. In interact. Our cultures were enriched for NSPCs, but we cannot contrast, the activation of either Akt or ERK1/2 has been shown to rule out the presence of other adult spinal cord cells particularly promote proliferation in adult hippocampal (Peltier et al., 2007; at early DIV. However, we found that by 14 DIV all the cells were Ma et al., 2009) and in adult subventricular zone (Torroglosa MAP2+, indicating that all of the adult spinal cord NSPCs present et al., 2007; Lao et al., 2013) neural stem cell cultures. Similarly, in in the culture underwent neuronal differentiation. adult spinal cord NSPC cultures, we have shown that the inhibi- Collectively, these findings suggest that while the PI3K/Akt tion of either the PI3K/Akt pathway or the MEK/ERK1/2 pathway, pathway may be distinctly pivotal for proliferation of ESCs and but not the PLCγ pathway, markedly reduced cell division (pro- embryonic cortical progenitor cells, proliferation of adult neu- liferation). Past studies using different cell lines and primary ral stem and progenitor cells involves both the PI3K/Akt and neurons have suggested that cell survival involves the activation MEK/ERK1/2 pathways in a non-overlapping manner (Table 1). of these two pathways (Xia et al., 1995; Kennedy et al., 1997; Xue The downstream targets of these two pathways regulating pro- et al., 2000; Yamaguchi and Wang, 2001). Nonetheless, we have liferation in adult spinal cord NSPCs remain to be determined. shown that the inhibition of either the PI3K/Akt pathway or the However, recent studies of proliferation of adult neural stem cells MEK/ERK1/2 pathway did not induce additional cell death in the from other regions have identified additional pathways as well as adult spinal cord cultures when compared to the vehicle control. crucial downstream targets of the PI3K/Akt and MEK/ERK1/2 Hence, the reduction in cell proliferation when each of these two pathways, including the Wnt signaling pathway (Gage, 2010), pathways was inhibited was not due to apoptosis. In other words, the interleukin/Janus kinase/signal transducers and activators of both the PI3K/Akt and MEK/ERK1/2 pathways are crucial for transcription pathway (Gomez-Nicola et al., 2011), mammalian inducing cell proliferation of adult spinal cord NSPCs. target of rapamycin (Paliouras et al., 2012), and cyclin-dependent In the adult brain, the hippocampal subgranular zone (SGZ) kinase inhibitor (Marques-Torrejon et al., 2013). and the subventricular zone (SVZ) contain at least two popula- With respect to neural lineage differentiation, we have shown tions of stem/progenitor cells that are nestin+ (Doetsch et al., in cultures of adult spinal cord NSPCs activation of the 1997; Fukuda et al., 2003). Type 1 and B cells (in the SGZ and MEK/ERK1/2 (but not PI3K/Akt or PLCγ) pathway is required SVZ, respectively), which are also GFAP+ (glial fibrillary acidic for neuronal differentiation. On the other hand, it has been protein); and type 2 and C cells (in the SGZ and SVZ, respec- reported that, in adult hippocampal neural stem cell cultures, tively) which are GFAP- and occasionally DCX+.Type 1 and neuronal differentiation is blocked by ERK1/2 activation and B cells, respectively, give rise to Type 2 and C cells (Suh et al., is induced through activation of PLCγ (Ma et al., 2009). Our 2009). The proliferation of both of these cell types involves Akt findings provide evidence that, unlike in the adult hippocam- and ERK1/2 pathways (Lao et al., 2013). However, in the adult pal neural stem cells, but similar to mouse embryonic stem cells spinal cord, the populations of stem/progenitor cells are much (Li et al., 2006), ERK1/2 signaling promotes neuronal differentia- less well-characterized; they appear to be located at the dorsal area tion in adult spinal cord NSPC cultures (Table 1). The inhibition of the ependymal zone in which proliferating cells, nestin+ cells, of ERK1/2 activation likely prevents the existing/remaining adult GFAP+ cells, and BLBP+ (brain-lipid-binding protein) cells have spinal cord NSPCs from exiting cell cycle and from proceeding been detected (Shechter et al., 2007; Meletis et al., 2008; Hamilton to differentiate into MAP2+ neurons. Although the activation et al., 2009; Sabourin et al., 2009). In this area there are also of the MEK/ERK1/2 pathway has been implicated in promoting DCX+ cells (Shechter et al., 2007). Additional characterization neuronal survival (Xue et al., 2000), as mentioned earlier the inhi- is required to determine the number and type of progenitor pop- bition of the MEK/ERK1/2 pathway did not result in an increase ulations in the adult spinal cord, as well as how these populations of cell death. This implies that the decrease in the number of Table 1 | Contribution of ERK1/2, Akt, PLCγ to proliferation and neuronal differentiation of stem/progenitor cells. Stem or progenitor cell origin Proliferation Neuronal differentiation Publications MEK/ERK1/2 PI3K/Akt PLCγ MEK/ERK1/2 PI3K/Akt PLCγ Embryonic stem cells (human) No Yes n/d n/d n/d n/d Li et al., 2007 Embryonic stem cells (mouse) n/d n/d n/d Yes n/d n/d Li et al., 2006 E12.5 – E13.5 cortical progenitors (mouse) No Yes n/d Yes n/d n/d Barnabe-Heider and Miller, 2003 Adult hippocampal neural progenitor cells (rat) No Yes n/d n/d Inhibitory n/d Peltier et al., 2007 Adult hippocampal neural progenitor cells (rat) Yes n/d n/d Inhibitory n/d Yes Ma et al., 2009 Adult SVZ neurospheres (mouse) n/d Yes n/d n/d n/d n/d Torroglosa et al., 2007 Adult SVZ neurospheres (mouse) Yes Yes n/d n/d n/d n/d Lao et al., 2013 Adult spinal cord progenitor cells (rat) Yes Yes No Yes No No Present study No: activation of the pathway is not involved. Yes: activation of the pathway is involved. Inhibitory: activation of the pathway is inhibitory. n/d: not determined in the study. Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 8 Chan et al. Pathways regulating spinal cord NSPC MAP2+ neurons we observed after ERK1/2 inhibition was not as to understand the signaling mechanism that is unique to the due to an increase in neuronal death but rather a suppression cell type of interest. of neuronal differentiation. We have also shown in our NSPC As mentioned in the introduction, adult spinal cord NSPCs are cultures that the inhibition of ERK1/2 and Akt activation signif- stimulated to proliferate in response to injury (Johansson et al., icantly, and to the same extent, suppressed active proliferation, 1999; Horner et al., 2000; Yamamoto et al., 2001; Danilov et al., yet only the inhibition of ERK1/2 activation suppressed neuronal 2006). This surge of cell proliferation, however, produces mostly differentiation (Table 1; Figures 3A,B). Therefore, it is unlikely reactive astrocytes (Johansson et al., 1999; Shihabuddin et al., that the suppression in neuronal differentiation as a result of 2000; Horky et al., 2006; Yang et al., 2006). There is also support- the ERK1/2 pathway inhibition was simply due to a decrease in ing evidence that there is an upregulation of FGF2 expression by proliferation. the reactive astrocytes after spinal cord or spinal nerve injuries In addition to studying the role of the signaling pathways in (Madiai et al., 2003; Qi et al., 2003), despite the lack of neuronal neuronal differentiation, we have also investigated their role in regeneration. Intriguingly, Miura et al. have reported partial func- neurite outgrowth in the newly generated MAP2+ cells. Our tional recovery in adult rats infected with a constitutively active findings indicate that both the MEK/ERK1/2 and PLCγ path- form of MEK1 (activator of ERK) following spinal cord injury ways act to modulate cytoskeletal dynamics of the MAP2+ cells, (Miura et al., 2000). While they attributed the functional recov- as both MEK and PLCγ inhibitors, respectively, reduced neurite ery to axonal regeneration, neurogenesis of the endogenous spinal branching of these cells. Reminiscent of previous neurite studies cord NSPCs was not examined. Based on our study, it is likely (Mattson et al., 1988; Costantini and Isacson, 2000), the length of that the constitutively active MEK/ERK signaling pathway may the longest neurite increases in conditions where neurite branch- have also promoted neuronal differentiation of those endogenous ing is compromised. These findings suggest that the MEK/ERK1/2 NSPCs, resulting in the replenishment of lost neurons following and PLCγ pathways may have a specific effect on neurite complex- injury to the spinal cord. As we gain more knowledge regarding ity rather than neurite elongation. During neuronal development, the regulation of NSPCs in the adult spinal cord, more research to part of neurite outgrowth includes neurite elongation. Elongation better understand such a complex system is necessary to facilitate is dependent on the dynamics of cytoskeleton, which consists of future attempts to manipulate these NSPCs in pathological con- microfilaments, intermediate filaments, and microtubules (MTs). ditions. To conclude, we found in our study that the activation of As it turns out, several microtubule-binding proteins, includ- the MEK/ERK1/2 pathway and the PI3K/Akt pathway (but not the ing MAP2, can affect the polymerization and stability of MTs PLCγ pathway) are essential for proliferation of adult rat spinal (Sanchez et al., 2000). Based on previous studies a model for neu- cord NSPCs; and that the activation of the MEK/ERK1/2 path- rite elongation and branching has been already proposed (Hely way(butnot the PI3K/Aktor the PLCγ pathway) is required for et al., 2001; Kiddie et al., 2005). When it is phosphorylated, MAP2 neuronal differentiation of the adult spinal cord NSPCs. Together has low affinity for MTs. This decrease in MAP2 association with with other studies, our findings reinforce the idea that the con- MTs in turn decreases MT stability which results in a decrease tribution of signaling pathway activation in proliferation and in neurite elongation while facilitates neurite branching (Hely neuronal differentiation is unique in different cell types and is et al., 2001; Kiddie et al., 2005). On the other hand, when MAP2 dependent on the source of NSPCs. is de-phosphorylated, this favors the binding of MAP2 to MTs, MATERIALS AND METHODS which leads to MTs bundling and polymerization and therefore promotes neurite elongation while decreases neurite branching PRIMARY CULTURES (Hely et al., 2001; Kiddie et al., 2005). Our results, therefore, can All animals were used in accordance with guidelines approved by be explainedasfollows. AsMAP2is a known substrate formany the New York University Langone Medical Center Institutional protein kinases, including ERKs and protein kinase C (which is Animal Care and Use Committee. Adult male sprague-Dawley activated by PLCγ), it is likely that the inhibition of ERK1/2 and rats weighing 260–390 g were used. Animals were anesthetized PLCγ activation in fact promoted MAP2 dephosphorylation and and perfused with artificial cerebrospinal fluid (CSF) as described hence its binding to MTs, resulting in an increase in neurite length previously (Montoya et al., 2009). After perfusion and decapi- (elongation) and a decrease in neurite branching. tation of the animal, a 20 gauge needle and syringe filled with By comparing the findings in our study with those in oth- 10 mL of cold CSF was inserted into the sacral canal to extrude ers, there is a recurring theme that the effect of signaling the spinal cord. Microdissection of the spinal cord into approx- pathway activation is cellular context-dependent (Schlessinger, imately 1 mm pieces was carried out in cold hibernate A (HA) 2000). Along these lines, it is worth mentioning that while it (Brain Bits, Springfield, IL) supplemented with 2% B27 [self- has been suggested that the PI3K/Akt pathway can antagonize prepared; (Brewer et al., 1993)], GlutaMAX (0.5 mM), penicillin the MEK/ERK1/2 pathway (Mason, 2007), and vice versa (in (100 U/mL) and streptomycin (100 μg/mL) (all purchased from NIH3T3 cells) (Hayashi et al., 2008), we do not have any evi- GIBCO, Carlsbad, CA). The dissected spinal cord tissue was dence that supports such crosstalk between this two pathways in digested in papain (36 U/mL, Worthington, Lakewood, NJ) and our enriched adult spinal cord NSPC culture. For instance, from DNase I (0.02% w/v, Worthington) in a 15-mL conical tube for our Western blot results, we did not detect the potentiation of 5 min at room temperature (RT), then at 30 Cfor 25min. The Akt phosphorylation (vs. DMSO control) in response to ERK1/2 digested tissue was kept on ice and allowed to settle by gravity inactivation, or vice versa. Collectively, these data indicate that (∼1 min). Supernatant was discarded and the pellet was resus- even well-established pathways warrant further investigation so pended with 2 mL of HA containing 0.02% DNase I. Using three Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 9 Chan et al. Pathways regulating spinal cord NSPC fire-polished glass pipettes of decreasing diameter, the cell sus- IMMUNOCYTOCHEMISTRY pension was triturated with each no more than ten times. Before Adult rat spinal cord cells were fixed in 4% paraformalde- switching to another pipette, the suspension was allowed to settle hyde (15710, Electron Microscopy Sciences, Hatfield, PA) for (as above), and the supernatant was kept in another conical tube, 15 min and permeabilized in 0.3% Triton X-100 (X100RS, Sigma- while 2 mL of fresh HA (containing 0.02% DNase I) was added to Aldrich) [or 0.1% Tween 20 (P5927, Sigma-Aldrich) for mouse the suspension. At the end of the trituration procedure when all anti-stage-specific embryonic antigen 1 (SSEA-1)] for 10 min three glass pipettes were used, a cell suspension with a total vol- at RT. To label proliferating cells, 5-ethynyl-2 -deoxyuridine ume of 6 mL was layered on top of 5 mL of 6% OptiPrep (Accurate (EdU) (1.5–3 μM, from the Click-iT EdU Imaging Kit, C10337, Chemical and Scientific Corporation, Westbury, NY) gradient in Invitrogen), a modified thymidine analogue, was applied to a 50-mL conical tube. The gradient was centrifuged at 822 × gfor the culture (for 24 h, unless otherwise stated) before fixation. 13 min at 4 C. After centrifugation, the white myelin layer and the To detect EdU-labeled cells, a click-iT reaction between the top half of the supernatant were discarded. After gentle mixing, Alexa Fluor 488 azide and the modified EdU alkyne was car- the suspension was brought up to 20 mL with HA and was allowed ried out according to the instructions of the company. This to slowly pass through a 70 μm nylon mesh. The cell suspension reaction was carried out prior to immunostaining procedures was spun one last time at 480 × g for 5 min. The isolation protocol with various primary antibodies. Mouse anti-nestin (∼1 μg/mL, yielded ∼1% of the total cell number (∼110,000 cells isolated) in Rat-401-concentrate, Developmental Studies Hybridoma Bank, the spinal cord, which was calculated based on the DNA content Iowa City, IA; 2.5 μg/mL, MAB353, Millipore) and mouse anti- (PureLink Genomic DNA kits, K1820–01, Invitrogen, Carlsbad, SSEA-1 (∼1 μg/mL, MC-480, Developmental Studies Hybridoma CA) of the final pellet and previous studies of amount of DNA Bank) were used as markers for undifferentiated adult neu- per cell (Santen and Agranoff, 1963) and totalcellnumberin ral stem/progenitor cells. Rabbit anti-NG2 chondroitin sulfate the adult rat spinal cord (Bjugn, 1993). The final pellet was proteoglycan (0.67 μg/mL, AB5320, Millipore) to label neu- resuspended with 400 μL of NB-A/B27 [Neurobasal A (GIBCO) ral progenitor cells. Primary antibodies used to label neu- supplemented with 2% B27, GlutaMAX (0.5 mM), penicillin rons were mouse anti-microtubule associated protein-2a and (100 U/mL) and streptomycin (100 μg/mL)]. Ten microliter of -2b (0.4–0.67 μg/mL, MAP2, clone AP20, MAB3418, Millipore, cell suspension (∼3000 cells) was plated on poly-D-lysine- Temecular, CA), goat anti-doublecortin (0.8 μg/mL, DCX, sc- coated glass coverslips (50 μg/mL, P6407, Sigma-Aldrich, St. 8066, Santa Cruz Biotechnology, Santa Cruz, CA), mouse anti- Louis, MO) in 4-well plates. After 1 h in 5% CO ,37 C, glutamic acid decarboxylase (0.67 μg/mL, GAD-67, MAB5406, 500 μL of pre-warmed NB-A/B27 containing 20 ng/mL of FGF2 Millipore), and rabbit anti-vesicular glutamate transporter (13256-029, Invitrogen) was added to each well. Adult spinal 1 (1:2000 polyclonal antiserum, VGLUT1, 135302, Synaptic cord cultures were treated with 10 μM U0126 (Cat#662005), Systems, Germany). Rabbit anti-synaptophysin 1 (1:150 poly- 10 μM LY294002 (Cat#440202), 2.5 μM U73122 (Cat#662035; clonal antiserum, 101002, Synaptic Sytems) was used to iden- all from Calbiochem, San Diego, CA), or vehicle control DMSO tify synapse formation. Ten percent normal goat serum (G9023, (dimethyl sulfoxide, D2650, Sigma-Aldrich). For longer culture Sigma-Aldrich) was used to block non-specific binding. Cultures period, NB-A/B27 from 6 to 13 days in vitro were supple- were incubated overnight at 4 C with primary antibodies diluted mented with recombinant human brain-derived neurotrophic in 2% goat (or, forDCX,donkey) seruminPBS. Species- factor (BNDF, 1 ng/mL, 10909010, Invitrogen), recombinant appropriate IgG, IgG1, and IgG2a conjugated with Alexa Fluor human glial-derived neurotrophic factor (GDNF, 0.1 ng/mL, 488, 546, or 555 (1:800–1:2000, Molecular Probes, Eugene, OR) 10907012, Invitrogen), and membrane permeable cAMP ana- were used as secondary antibodies. After discarding the buffer log [8-(4-chlorophylthio)adenosine 3 ,5 -cyclic monophosphate containing the primary antibodies and extensive rinsing (at least sodium salt, 125 μM, C3912, Sigma-Aldrich]. Culture medium 5 rinses, 5 min each), the secondary antibodies were added and was changed every other day in all experiments. incubated for 1 h at RT. Cells were counterstained with Hoechst In Western blots, the PLC inhibitor U73122 inhibited PLCγ1 33342 for 15 min (6 μg/mL). Coverslips were mounted with activation at 10 μM in our cultures. Due to its cytotoxic effect Aqua-Poly/Mount (Polysciences, Warrington, PA). Negative con- in longer exposure (6 DIV), the concentration of U73122 was trols were prepared by omitting the primary antibodies. Rabbit eventually lowered to 2.5 μM. At 2.5 μM, U73122 still signifi- and mouse whole molecule IgG (Jackson Immuno Research, West cantly and specifically inhibited PLCγ1 activation while it did Grove, PA) at comparable concentrations to primary antibodies not induce a significant increase in cell death as compared to the used for immunostaining were also used as alternative negative vehicle control (see results). This U73122 concentration is similar controls. Appropriate cell types or tissues were used as immunos- to its reported IC value of 1–2.1 μM (product information, taining controls to test the specificity of the primary antibodies Cat#662035 Calbiochem). The MEK inhibitor U0126 was tested used for immunocytochemistry in this study (Figure 6). at various concentrations in our cultures, resulting in a decrease For TUNEL labeling, adult spinal cord cells were cultured for in the number EdU+,MAP2+,and EdU+/MAP2+ cells as 4 days in FGF2-containing NB-A/B27, treated with each of the the concentrations of U0126 was increased (not shown). At inhibitors (or DMSO), and fixed as described above. To prevent the highest dosage we tested, U0126 (10 μM) did not cause an apoptosis, 20 μM Z-VAD-FMK (Cat#219007, EMD Biosciences, increase in cell death when compared to vehicle control and San Diego, CA), a general caspase inhibitor, was used. Click-iT its action was specific (see Western blot results), hence this TUNEL Alexa Fluor 488 Imaging Assay (C10245, Invitrogen) was concentration was used for subsequent experiments. used to detect apoptotic cells. The TUNEL reaction, based on a Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 10 Chan et al. Pathways regulating spinal cord NSPC FIGURE 6 | Immunostaining controls of primary antibodies used in the the primary antibody omitted. (iii) Composite of an adult spinal cord section study. As positive controls, primary antibodies used in the study were tested taken at a lower magnification. Prominent MAP2 staining (white) was in different cell types or tissues where they have been detected in previous observed only in the gray matter [similar to a previous study studies. (A), Immunostaining of mouse monoclonal nestin (green) in (Suzuki-Yamamoto et al., 2000)]. Scale bar: 50 μm. (D), Co-labeling of rabbit hippocampal progenitor cell culture from embryonic day (E) 18 rat embryos polyclonal synaptophysin 1 (green) and mouse monoclonal MAP2 (red) in rat (1 DIV), where the majority of the cells were nestin+ (i) [previous reported in postnatal day 7 hippocampal culture (37 DIV). (i) MAP2+ neurons formed (Mistry et al., 2002)]. (ii) Negative control where the primary antibody was synapses (expressed synaptophysin) with neighboring neurons. (ii) Negative omitted. Scale bar: 20 μm. (B) i,ii, Immunostaining of mouse monoclonal control where the primary antibody was omitted. Whole molecule rabbit and GAD-67 (green) in an adult rat spinal cord section. (i) Fluorescent image mouse IgG at concentrations identical to those of the two primary antibodies showing that GAD-67 immunoreactivity was localized in lamina II of the were used to show specificity of the primary antibodies. Blue represents dorsal horn, which matches the typical staining pattern of GAD-67 in the Hoechst-stained nuclei. Scale bar: 20 μm. (E), Immunostaining of rabbit spinal cord (Edgerton et al., 2001). (ii) Phase contrast image of the same polyclonal VGLUT1 (red) in rat E18 hippocampal culture (22 DIV). (i) Negative dorsal horn field. Scale bar: 20 μm. (B) iii,iv, Immunostaining of rat postnatal control where the primary antibody was omitted. (ii) Only those cells with day 7 hippocampal culture (16 DIV) using mouse monoclonal GAD-67 neuronal morphology were immunoreactive for VGLUT1. (F) i, antibody. (iii) GAD-67 staining (green). (iv) Negative control in which the Immunostaining of goat polyclonal DCX (magenta) in an adult mouse brain primary antibody was omitted. Scale bar: 20 μm. (C), Immunostaining of section. DCX+ immature neurons were detected in the dentate gyrus of the mouse monoclonal MAP2 in adult rat spinal cord sections. (i) MAP2 staining hippocampus. (ii) Negative control with the primary antibody omitted. Blue (red) in ventral horn neurons. Scale bar: 20 μm. (ii) Negative control where represents Hoechst-stained nuclei. Scale bar: 20 μm. copper catalyzed reaction between the Alexa Fluor azide and a then exposed toFGF2 (20ng/mL) for30min afterwhich the modified dUTP alkyne, was carried out according to the company medium was discarded and the cells were rinsed once with ice- manual. DNase I was provided in the company kit as a positive cold PBS. The following steps were carried out on ice or at control of the TUNEL assay. 4 C: cells were lysed directly in Petri dishes by adding 50μlof lysis buffer: RIPA (20188, Millipore) supplemented with 0.1% IMMUNOBLOTTING AND IMMUNOPRECIPITATION (v/v) SDS and 1X phosphatase inhibitor cocktail (78420, Thermo Adult spinal cord cells were cultured with FGF2 for 6 days in Scientific, Rockford, IL). Solubilized cell lysate was collected to one side of the dish by scraping. Lysate was then transferred 35 mm Petri dishes. Cells were rinsed with NB-A/B27 four times, once every hour for 4 h to remove traces of FGF2, followed by to Eppendorf tubes and rocked for 5 min. Supernatant from each sample was collected after a 13-min spin at 13k rpm. Total a 1 h-pre-treatment with one of the following inhibitors, 10 μM protein concentrations were determined using the Lowry assay. U0126, 10 μM LY294002, 2.5 μM U73122, or DMSO. Cells were Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 11 Chan et al. Pathways regulating spinal cord NSPC Various volumes of samples containing 50 μg of total protein were IMAGE ACQUISITION AND ANALYSIS lyophilized and resuspended with 20 μL of sample buffer. After Fluorescent images were captured using Carl Zeiss Axiovert 200 boiling the samples for 5 min, proteins were separated by SDS- M. For each experiment, ten (using a 40× objective) to twenty PAGE and transferred to PVDF membranes (Amersham, UK) (using a 20× or 40× objective) random fields of each of the cov- for 3 h (100 V). Membranes were blocked in 5% BSA in Tris- erslips in each treatment were acquired. For total cell numbers, buffered saline, 0.1% Tween 20 (TBST, all from Sigma-Aldrich) Hoechst-stained nuclei were counted using a 10× objective over for 1 h at RT, and then they were incubated overnight at 4 C the entire coverslip. The exposure times of the fluorescent chan- with primary antibodies diluted in 4% BSA in TBST. Membranes nels (DAPI, GFP, and Rhodamine) was determined using negative were first probed with antibodies against phospho-proteins. Then controls where either the primary antibody was omitted or the the membranes were stripped (Restore Plus, 46430, Thermo whole molecule IgG of matching species was added in place of the Scientific) and reprobed with antibodies against total pro- primary antibody. The fluorescent intensities of individual cells teins. Primary antibodies used were mouse anti-phospho-p44/42 were analyzed using the AxioVision software. Cells were scored MAPK (ERK1/2) (1:2000, Thr202/Tyr204, 9106, Cell Signaling as immune-positive when their intensities were at least three Technology, Danvers, MA), rabbit anti-phospho-Akt (1:1000, standard deviations above the average values of their correspond- Ser-473, 9271, Cell Signaling), rabbit anti-phospho-PLCγ1 ing negative controls. Alternatively, the background fluorescence (1:700, Thy-783, 2821, Cell Signaling Technology), rabbit anti- from each negative control was subtracted from the cells and flu- p44/42 MAPK (1:1000, 9102, Cell Signaling Technology), mouse orescence above the background was considered as positive based anti-Akt (1:1000, 610860, BD Biosciences), and mouse anti- on the RGB histogram generated by AxioVision. All results are PLCγ1 (1:300, sc-7290, Santa Cruz Biotechnology). Membranes represented as mean + SEM. Statistical significance was deter- were incubated with HRP-conjugated goat anti-rabbit (1:5000, mined by unpaired t-test when comparing two groups. When sc-2004, Santa Cruz Biotechnology) or anti-mouse (1:5000, sc- comparing three or more groups One-Way ANOVA was used. 2005, Santa Cruz Biotechnology) secondary antibodies diluted in Bonferroni’s multiple comparison post-test was used when com- 2.5% BSA in TBST. Membranes were developed with SuperSignal paring three to five groups and Dunnett’s for five or more groups. ∗ ∗∗ § † # chemiluminescent substrates (West Pico, 34077, West Dura, P < 0.05, P < 0.01, P < 0.005, P < 0.001, P ≤ 0.0001. 37071, Thermo Scientific). Blots were quantified using the gel analyzer function in the ImageJ software. ACKNOWLEDGMENTS For immunoprecipitation, anti-PLCγ-coated Dynabeads The authors wish to thank Dr. Jin Zhang and Lourdes A. Martin (Dynabeads Protein G, 100.03D, Invitrogen) was incubated with Hernandez for their technical assistance and Joey Ting for his crit- solubilized cell lysate containing 40 μg of total protein overnight ical comments on the manuscript. The Rat-401 antibody devel- at 4 C. The lysate-antibody-Dynabeads mixture was washed three oped by Susan Hockfield and the MC-480 (SSEA-1) antibody times with coldPBS andthen resuspended with 20 μLof sample developed by David Solter were obtained from the Developmental buffer for subsequent Western blot analysis as described previ- Studies Hybridoma Bank developed under the auspices of the ously. Monoclonal phosphotyrosine antibody (1:1500 pTyr, PY99, NICHD and maintained by The University of Iowa, Department sc-7020) was purchased from Santa Cruz Biotechnology. of Biology, Iowa City, IA 52242. Bjugn, R. (1993). The use of the opti- Exp. Neurol. 164, 60–70. doi: Region-specific growth proper- REFERENCES Alvarez-Buylla, A., and Garcia- cal disector to estimate the num- 10.1006/exnr.2000.7417 ties and trophic requirements of Verdugo, J. M. (2002). Neurogenesis ber of neurons, glial and endothe- Danilov, A. I.,Covacu, R.,Moe, M. brain- and spinal cord-derived rat in adult subventricular zone. lial cells in the spinal cord of C., Langmoen, I. A., Johansson, embryonic neural precursor cells. J. Neurosci. 22, 629–634. the mouse–with a comparative note C. B., Olsson, T., et al. (2006). Neuroscience 135, 851–862. doi: Barnabe-Heider, F., Goritz, C., on the rat spinal cord. Brain Res. Neurogenesis in the adult spinal 10.1016/j.neuroscience.2005.05.049 Sabelstrom, H., Takebayashi, 627, 25–33. doi: 10.1016/0006-8993 cord in an experimental model of Fukuda, S., Kato, F., Tozuka, Y., H., Pfrieger, F. W., Meletis, K., et al. (93)90744-8 multiple sclerosis. Eur. J. Neurosci. Yamaguchi, M., Miyamoto, Y., and (2010). Origin of new glial cells Brewer,G.J., Torricelli, J. R., Evege,E. 23, 394–400. doi: 10.1111/j.1460- Hisatsune, T. (2003). Two distinct in intact and injured adult spinal K., and Price, P. J. (1993). Optimized 9568.2005.04563.x subpopulations of nestin-positive cord. Cell Stem Cell 7, 470–482. doi: survival of hippocampal neurons Doetsch, F., Garcia-Verdugo, J. M., and cells in adult mouse dentate gyrus. 10.1016/j.stem.2010.07.014 in B27-supplemented Neurobasal, a Alvarez-Buylla, A. (1997). Cellular J. Neurosci. 23, 9357–9366. Barnabe-Heider, F., and Miller, F. D. new serum-free medium combina- composition and three-dimensional Gage, F. H. (2010). Molecular and cellu- (2003). Endogenously produced tion. J. Neurosci. Res. 35, 567–576. organization of the subventricu- lar mechanisms contributing to the neurotrophins regulate survival and doi: 10.1002/jnr.490350513 lar germinal zone in the adult regulation, proliferation and differ- differentiation of cortical progeni- Capela, A., and Temple, S. (2002). mammalian brain. J. Neurosci. 17, entiation of neural stem cells in the tors via distinct signaling pathways. LeX/ssea-1 is expressed by adult 5046–5061. adult dentate gyrus. Keio J. Med. 59, J. Neurosci. 23, 5149–5160. mouse CNS stem cells, identify- Edgerton, V. R., de Leon, R. D., 79–83. doi: 10.2302/kjm.59.79 Belachew, S., Chittajallu, R., Aguirre, A. ing them as nonependymal. Neuron Harkema, S. J., Hodgson, J. A., Gomez-Nicola, D., Valle-Argos, A.,Yuan, X.,Kirby,M., Anderson, 35, 865–875. doi: 10.1016/S0896- London, N., Reinkensmeyer, B., Pallas-Bazarra, N., and S., et al. (2003). Postnatal NG2 6273(02)00835-8 D. J., et al. (2001). Retraining Nieto-Sampedro, M. (2011). proteoglycan-expressing progenitor Costantini, L. C., and Isacson, O. the injured spinal cord. J. Interleukin-15 regulates prolif- cells are intrinsically multipotent (2000). Immunophilin ligands and Physiol. 533, 15–22. doi: 10.1111/ eration and self-renewal of adult and generate functional neurons. GDNF enhance neurite branching j.1469-7793.2001.0015b.x neural stem cells. Mol. Biol. Cell 22, J. Cell Biol. 161, 169–186. doi: or elongation from developing Fu,S. L., Ma, Z.W., Yin, L.,Iannotti, 1960–1970. doi: 10.1091/mbc.E11- 10.1083/jcb.200210110 dopamine neurons in culture. C., Lu, P. H., and Xu, X. M. (2005). 01-0053 Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 12 Chan et al. Pathways regulating spinal cord NSPC Guo, F., Maeda, Y., Ma, J., Xu, Rev. Biochem. 69, 373–398. doi: Madiai,F., Hussain,S.R., Goettl, neurite outgrowth of adult J., Horiuchi, M., Miers, L., 10.1146/annurev.biochem.69.1.373 V. M.,Burry, R.W., Stephens, spinal cord motor neurons. et al. (2010). Pyramidal neu- Hugnot, J. P., and Franzen, R. (2011). R. L. Jr., and Hackshaw, K. V. Exp. Neurol. 220, 303–315. doi: rons are generated from The spinal cord ependymal region: (2003). Upregulation of FGF-2 10.1016/j.expneurol.2009.09.003 oligodendroglial progenitor a stem cell niche in the caudal cen- in reactive spinal cord astrocytes Namiki, J., and Tator, C. H. (1999). cells in adult piriform cortex. tral nervous system. Front. Biosci. following unilateral lumbar spinal Cell proliferation and nestin J. Neurosci. 30, 12036–12049. doi: 16, 1044–1059. doi: 10.2741/3734 nerve ligation. Exp. Brain Res. 148, expression in the ependyma of the 10.1523/JNEUROSCI.1360-10.2010 Johansson, C. B., Momma, S., Clarke, 366–376. adult rat spinal cord after injury. Hamilton, L.K., Truong,M. K., D. L., Risling, M., Lendahl, U., and Marques-Torrejon,M.A., Porlan, E., J. Neuropathol. Exp. Neurol. 58, Bednarczyk, M. R., Aumont, Frisen, J. (1999). Identification of Banito, A., Gomez-Ibarlucea, 489–498. doi: 10.1097/00005072- A., and Fernandes, K. J. (2009). a neural stem cell in the adult E., Lopez-Contreras, A. J., 199905000-00008 Cellular organization of the central mammalian central nervous system. Fernandez-Capetillo, O., et al. Paliouras, G. N., Hamilton, L. K., canal ependymal zone, a niche Cell 96, 25–34. doi: 10.1016/S0092- (2013). Cyclin-dependent kinase Aumont, A., Joppe, S. E., Barnabe- of latent neural stem cells in the 8674(00)80956-3 inhibitor p21 controls adult Heider, F., and Fernandes, K. adult mammalian spinal cord. Kennedy, S. G., Wagner, A. J., Conzen, neural stem cell expansion by J. (2012). Mammalian target Neuroscience 164, 1044–1056. doi: S. D.,Jordan, J.,Bellacosa, A., regulating Sox2 gene expression. of rapamycin signaling is a 10.1016/j.neuroscience.2009.09.006 Tsichlis, P. N., et al. (1997). The Cell Stem Cell 12, 88–100. doi: key regulator of the transit- Hashimoto, K., and Ishima, T. PI 3-kinase/Akt signaling pathway 10.1016/j.stem.2012.12.001 amplifying progenitor pool in (2011). Neurite outgrowth delivers an anti-apoptotic sig- Mason, I. (2007). Initiation to end the adult and aging forebrain. mediated by translation elon- nal. Genes Dev. 11, 701–713. doi: point: the multiple roles of fibrob- J. Neurosci. 32, 15012–15026. doi: gation factor eEF1A1: a target 10.1101/gad.11.6.701 last growth factors in neural 10.1523/JNEUROSCI.2248-12.2012 for antiplatelet agent cilosta- Kiddie, G., Mclean, D., Van Ooyen, A., development. Nat. Rev. Neurosci. 8, Palmer,T.D., Willhoite,A. R., and zol. PLoS ONE 6:e17431. doi: and Graham, B. (2005). Biologically 583–596. doi: 10.1038/nrn2189 Gage, F. H. (2000). Vascular niche 10.1371/journal.pone.0017431 plausible models of neurite out- Mattson, M. P., Guthrie, P. B., and for adult hippocampal neurogene- Hayashi, H., Tsuchiya, Y., Nakayama, growth. Prog. Brain Res. 147, Kater, S. B. (1988). Components of sis. J. Comp. Neurol. 425, 479–494. K.,Satoh, T.,and Nishida, E. 67–80. doi: 10.1016/S0079-6123 neurite outgrowth that determine Peltier, J., O’neill, A., and Schaffer, (2008). Down-regulation of the PI3- (04)47006-X neuronal cytoarchitecture: influence D. V. (2007). PI3K/Akt and kinase/Akt pathway by ERK MAP Lao, C. L., Lu, C. S., and Chen, J. C. of calcium and the growth substrate. CREB regulate adult neural kinase in growth factor signal- (2013). Dopamine D(3) recep- J. Neurosci. Res. 20, 331–345. doi: hippocampal progenitor prolif- ing. Genes Cells 13, 941–947. doi: tor activation promotes neural 10.1002/jnr.490200307 eration and differentiation. Dev. 10.1111/j.1365-2443.2008.01218.x stem/progenitor cell proliferation Meletis, K., Barnabe-Heider, F., Carlen, Neurobiol. 67, 1348–1361. doi: Hely, T. A., Graham, B., and Ooyen, through AKT and ERK1/2 path- M.,Evergren, E.,Tomilin,N., 10.1002/dneu.20506 A. V. (2001). A computational ways and expands type-B and -C Shupliakov, O., et al. (2008). Spinal Qi,M.L., Wakabayashi, Y., Haro, H., model of dendrite elongation cells in adult subventricular zone. cord injury reveals multilineage and Shinomiya, K. (2003). Changes and branching based on MAP2 Glia 61, 475–489. doi: 10.1002/ differentiation of ependymal in FGF-2 expression in the distal phosphorylation. J. Theor. Biol. glia.22449 cells. PLoS Biol. 6:e182. doi: spinal cord stump after complete 210, 375–384. doi: 10.1006/jtbi. Lendahl, U., Zimmerman, L. B., and 10.1371/journal.pbio.0060182 cord transection: a comparison 2001.2314 Mckay, R. D. (1990). CNS stem Ming, G. L., and Song, H. (2005). Adult between infant and adult rats. Spine Hollis, E. R. 2nd., Jamshidi, P., Low, cells express a new class of inter- neurogenesis in the mammalian (Phila Pa 1976) 28, 1934–1940. K., Blesch, A., and Tuszynski, M. H. mediate filament protein. Cell 60, central nervous system. Annu. Rev. doi: 10.1097/01.BRS.0000083323. (2009). Induction of corticospinal 585–595. doi: 10.1016/0092-8674 Neurosci. 28, 223–250. doi: 10.1146/ 38962.2A regeneration by lentiviral trkB- (90)90662-X annurev.neuro.28.051804.101459 Sabourin,J. C., Ackema,K.B., induced Erk activation. Proc. Natl. Li, J., Wang, G., Wang, C., Zhao, Y., Ming, G. L., and Song, H. (2011). Ohayon, D., Guichet, P. O., Perrin, Acad.Sci.U.S.A. 106, 7215–7220. Zhang, H., Tan, Z., et al. (2007). Adult neurogenesis in the F. E., Garces, A., et al. (2009). doi: 10.1073/pnas.0810624106 MEK/ERK signaling contributes mammalian brain: significant A mesenchymal-like ZEB1(+) Horky, L. L., Galimi, F., Gage, F. H., to the maintenance of human answers and significant ques- niche harbors dorsal radial glial and Horner, P. J. (2006). Fate embryonic stem cell self-renewal. tions. Neuron 70, 687–702. doi: fibrillary acidic protein-positive of endogenous stem/progenitor Differentiation 75, 299–307. doi: 10.1016/j.neuron.2011.05.001 stem cells in the spinal cord. Stem cells following spinal cord injury. 10.1111/j.1432-0436.2006.00143.x Mistry, S. K., Keefer, E. W., Cells 27, 2722–2733. doi: 10.1002/ J. Comp. Neurol. 498, 525–538. doi: Li,Z., Theus, M. H.,and Wei, L. Cunningham, B. A., Edelman, stem.226 10.1002/cne.21065 (2006). Role of ERK 1/2 signaling G. M., and Crossin, K. L. (2002). Sanchez, C., Diaz-Nido, J., and Avila, Horner, P. J., Power, A. E., in neuronal differentiation of cul- Cultured rat hippocampal neural J. (2000). Phosphorylation of Kempermann, G., Kuhn, H. G., tured embryonic stem cells. Dev. progenitors generate spontaneously microtubule-associated protein Palmer,T.D., Winkler, J., etal. Growth Differ. 48, 513–523. doi: active neural networks. Proc. Natl. 2 (MAP2) and its relevance for (2000). Proliferation and dif- 10.1111/j.1440-169X.2006.00889.x Acad.Sci.U.S.A. 99, 1621–1626. the regulation of the neuronal ferentiation of progenitor cells Lie, D. C., Colamarino, S. A., Song, H. doi: 10.1073/pnas.022646599 cytoskeleton function. Prog. throughout the intact adult J., Desire, L., Mira, H., Consiglio, Miura, T., Tanaka, S., Seichi, A., Neurobiol. 61, 133–168. doi: rat spinal cord. J. Neurosci. 20, A., et al. (2005). Wnt signalling reg- Arai, M., Goto, T., Katagiri, H., 10.1016/S0301-0082(99)00046-5 2218–2228. ulates adult hippocampal neuroge- et al. (2000). Partial functional Santen, R. J., and Agranoff, B. W. Huang, C. C., Lo, S. W., and Hsu, K. nesis. Nature 437, 1370–1375. doi: recovery of paraplegic rat by (1963). Studies on the estimation of S. (2001). Presynaptic mechanisms 10.1038/nature04108 adenovirus-mediated gene delivery deoxyribonucleic acid in rat brain. underlying cannabinoid inhibition Ma,D.K., Ponnusamy,K., Song, M. of constitutively active MEK1. Biochim. Biophys. Acta 72, 251–262. of excitatory synaptic transmission R., Ming, G. L., and Song, H. Exp. Neurol. 166, 115–126. doi: doi: 10.1016/0926-6550(63)90339-6 in rat striatal neurons. J. Physiol. (2009). Molecular genetic analysis 10.1006/exnr.2000.7493 Schlessinger, J. (2000). Cell signaling 532, 731–748. doi: 10.1111/j.1469- of FGFR1 signalling reveals distinct Montoya, G. J., Sutachan, J. J., Chan, by receptor tyrosine kinases. Cell 7793.2001.0731e.x roles of MAPK and PLCgamma1 W. S., Sideris, A., Blanck, T. 103, 211–225. doi: 10.1016/S0092- Hubbard,S.R., andTill, J. H. activation for self-renewal of adult J., and Recio-Pinto, E. (2009). 8674(00)00114-8 (2000). Protein tyrosine kinase neural stem cells. Mol. Brain 2, 16. Muscle-conditioned media and Shechter, R., Baruch, K., Schwartz, M., structure and function. Annu. doi: 10.1186/1756-6606-2-16 cAMP promote survival and and Rolls, A. (2010). Touch gives Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 13 Chan et al. Pathways regulating spinal cord NSPC new life: mechanosensation modu- stem cells. Nature 417, 39–44. doi: Eur. J. Neurosci. 26, 2777–2794. doi: (2006). Endogenous neurogenesis lates spinal cord adult neurogene- 10.1038/417039a 10.1111/j.1460-9568.2007.05871.x replaces oligodendrocytes and sis. Mol. Psychiatry 16, 342–352. doi: Suh, H.,Deng, W.,and Gage, F.H. Weiss, D. J.,Fried, G.W., Chancellor, astrocytes after primate spinal cord 10.1038/mp.2010.116 (2009). Signaling in adult neuro- M. B., Herbison, G. J., Ditunno, J. injury. J. Neurosci. 26, 2157–2166. Shechter, R., Ziv, Y., and Schwartz, genesis. Annu. Rev. Cell Dev. Biol. F. Jr., and Staas, W. E. Jr. (1996). doi: 10.1523/JNEUROSCI.4070- M. (2007). New GABAergic 25, 253–275. doi: 10.1146/annurev. Spinal cord injury and bladder 05.2005 interneurons supported by myelin- cellbio.042308.113256 recovery. Arch. Phys. Med. Rehabil. specific T cells are formed Suzuki-Yamamoto, T., Toida, K., 77, 1133–1135. doi: 10.1016/S0003- Conflict of Interest Statement: The in intact adult spinal cord. Tsuruo, Y., Watanabe, K., 9993(96)90135-5 authors declare that the research Stem Cells 25, 2277–2282. doi: and Ishimura, K. (2000). Xia, Z., Dickens, M., Raingeaud, J., was conducted in the absence of any 10.1634/stemcells.2006-0705 Immunocytochemical localiza- Davis, R. J., and Greenberg, M. E. commercial or financial relationships Shen, Q., Goderie, S. K., Jin, L., tion of lung-type prostaglandin (1995). Opposing effects of ERK that could be construed as a potential Karanth, N., Sun, Y., Abramova, F synthase in the rat spinal cord. and JNK-p38 MAP kinases on conflict of interest. N., et al. (2004). Endothelial Brain Res. 877, 391–395. doi: apoptosis. Science 270, 1326–1331. cells stimulate self-renewal and 10.1016/S0006-8993(00)02709-8 doi: 10.1126/science.270.5240.1326 Received: 17 April 2013; accepted: 01 expand neurogenesis of neural stem Tamura,Y., Kataoka, Y., Cui,Y., Xue, L., Murray, J. H., and August 2013; published online: 27 August cells. Science 304, 1338–1340. doi: Takamori, Y., Watanabe, Y., Tolkovsky, A. M. (2000). The 2013. 10.1126/science.1095505 and Yamada, H. (2007). Multi- Ras/phosphatidylinositol 3-kinase Citation: Chan WS, Sideris A, Sutachan Shen, Q., Wang, Y., Kokovay, E., Lin, directional differentiation and Ras/ERK pathways function JJ, Montoya G JV, Blanck TJJ and Recio- G.,Chuang, S. M.,Goderie,S. of doublecortin- and NG2- as independent survival modules Pinto E (2013) Differential regulation K., et al. (2008). Adult SVZ stem immunopositive progenitor cells each of which inhibits a distinct of proliferation and neuronal differen- cells lie in a vascular niche: a in the adult rat neocortex in vivo. apoptotic signaling pathway in tiation in adult rat spinal cord neural quantitative analysis of niche cell- Eur. J. Neurosci. 25, 3489–3498. doi: sympathetic neurons. J. Biol. stem/progenitors by ERK1/2, Akt, and cell interactions. Cell Stem Cell 10.1111/j.1460-9568.2007.05617.x Chem. 275, 8817–8824. doi: PLCγ. Front. Mol. Neurosci. 6:23. doi: 3, 289–300. doi: 10.1016/j.stem. Torroglosa, A., Murillo-Carretero, 10.1074/jbc.275.12.8817 10.3389/fnmol.2013.00023 2008.07.026 M., Romero-Grimaldi, C., Yamaguchi, H., and Wang, H. G. This article was submitted to the journal Shihabuddin, L. S., Horner, P. J., Ray, Matarredona, E. R., Campos-Caro, (2001). The protein kinase PKB/Akt Frontiers in Molecular Neuroscience. J., and Gage, F. H. (2000). Adult A., and Estrada, C. (2007). Nitric regulates cell survival and apopto- Copyright © 2013 Chan, Sideris, spinal cord stem cells generate neu- oxide decreases subventricular zone sis by inhibiting Bax conformational Sutachan, Montoya G, Blanck and rons after transplantation in the stem cell proliferation by inhibi- change. Oncogene 20, 7779–7786. Recio-Pinto. This is an open-access adult dentate gyrus. J. Neurosci. 20, tion of epidermal growth factor doi: 10.1038/sj.onc.1204984 article distributed under the terms of 8727–8735. receptor and phosphoinositide- Yamamoto, S., Yamamoto, N., the Creative Commons Attribution Shihabuddin, L. S., Ray, J., and Gage, 3-kinase/Akt pathway. Stem Cells Kitamura, T., Nakamura, K., and License (CC BY). The use, distribution F. H. (1997). FGF-2 is sufficient 25, 88–97. doi: 10.1634/stemcells. Nakafuku, M. (2001). Proliferation or reproduction in other forums is per- to isolate progenitors found in 2006-0131 of parenchymal neural pro- mitted, provided the original author(s) the adult mammalian spinal cord. Vessal, M.,Aycock, A.,Garton, M. T., genitors in response to injury or licensor are credited and that the Exp. Neurol. 148, 577–586. doi: Ciferri, M., and Darian-Smith, C. in the adult rat spinal cord. original publication in this journal 10.1006/exnr.1997.6697 (2007). Adult neurogenesis in pri- Exp. Neurol. 172, 115–127. doi: is cited, in accordance with accepted Song, H., Stevens, C. F., and Gage, mate and rodent spinal cord: com- 10.1006/exnr.2001.7798 academic practice. No use, distribution F. H. (2002). Astroglia induce paring a cervical dorsal rhizotomy Yang, H., Lu, P., Mckay, H. M., Bernot, or reproduction is permitted which does neurogenesis from adult neural with a dorsal column transection. T., Keirstead, H., Steward, O., et al. not comply with these terms. Frontiers in Molecular Neuroscience www.frontiersin.org August 2013 | Volume 6 | Article 23 | 14

Journal

Frontiers in Molecular Neuroscience – Unpaywall

Published: Jan 1, 2013

There are no references for this article.