Address correspondence and reprint requests to Gerd Bicker, Division of Cell Biology, Institute of Physiology, University of Veterinary Medicine Hannover, Bischofsholer Damm 15, D-30173 Hannover, Germany. E-mail: email@example.com
Developmental studies in both vertebrates and invertebrates implicate an involvement of nitric oxide (NO) signaling in cell proliferation, neuronal motility, and synaptic maturation. However, it is unknown whether NO plays a role in the development of the human nervous system. We used a model of human neuronal precursor cells from a well-characterized teratocarcinoma cell line (NT2). The precursor cells proliferate during retinoic acid treatment as spherical aggregate culture that stains for nestin and βIII-tubulin. Cells migrate out of the aggregates to acquire fully differentiated neuronal phenotypes. The cells express neuronal nitric oxide synthase and soluble guanylyl cyclase (sGC), an enzyme that synthesizes cGMP upon activation by NO. The migration of the neuronal precursor cell is blocked by the use of nNOS, sGC, and protein kinase G (PKG) inhibitors. Inhibition of sGC can be rescued by a membrane permeable analog of cGMP. In gain of function experiments the application of a NO donor and cGMP analog facilitate cell migration. Our results from the differentiating NT2 model neurons point towards a vital role of the NO/cGMP/PKG signaling cascade as positive regulator of cell migration in the developing human brain.
Migration of neuronal progenitors from defined proliferative zones to their final location is a key event that underlies the development, regeneration, and plasticity of the brain. During early development, newly born neurons undergo extensive migration to set up the ordered organization of the CNS and PNS. In the adult brain, the subventricular zone (SVZ) and the subgranular region of the hippocampal dentate gyrus maintain the capacity of generating new neurons that migrate out to reach their ultimate locations (Hatten 1999; Gage 2000; Lambert de Rouvroit and Goffinet 2001; Nadarajah et al. 2003; Zheng and Poo 2007). The molecular mechanisms of neuronal migration involve a rearrangement of cytoskeletal components in response to extracellular cues, mediated by numerous intra- and intercellular signals (reviewed by Ayala et al. 2007; Kawauchi and Hoshino 2008). The gaseous messenger nitric oxide (NO) and its main target soluble guanylyl cyclase (sGC), an enzyme that synthesizes cGMP have been implicated in neuronal development including neurogenesis and neuronal migrations (reviewed by Enikolopov et al. 1999; Jurado et al. 2004; Cárdenas et al. 2005; Bicker 2005, 2007). NO signal transduction has been shown to differentially regulate cell proliferation and motility in early developing Xenopus (Peunova et al. 2007). Several studies indicate that NO enhances neuronal cell migration (Tanaka et al. 1994; Zhang et al. 2001; Haase and Bicker 2003). The transient expression of NOS during development has provided additional evidence for an involvement of NO signaling in neuronal migration. For example, certain neurons in the fetal human spinal cord express a histochemical marker for NOS as they migrate to their final destination, suggesting a functional role of NO during this developmental process (Foster and Phelps 2000). The limited availability of fetal tissue will certainly restrict experimental approaches to unravel the function of NO signaling in human brain development.
In this study we used NT2 spherical aggregates as a model of human neuronal precursor cells to elucidate the involvement of NO-cGMP signaling in neuronal precursor cell migration. The NT2 spherical aggregates were characterized by immunocytochemical methods for the expression of neuronal markers, the neuronal isoform of the NO synthesizing enzyme, neuronal nitric oxide synthase (nNOS), and its target enzyme, sGC. Application of enzyme inhibitors and activators to the differentiating NT2 cell aggregate provides firm evidence that the NO/cGMP/ protein kinase G (PKG) signal transduction pathway positively regulates human neuronal precursor cell migration.
Materials and methods
The NO donor, NOC-18 (2,2-(Hydroxynitrosohydrazino) bis-ethanamine]), was purchased from Calbiochem (Darmstadt, Germany), and 8-bromo-cylic guanosine-monophosphate (8-Br-cGMP) and RP isomer of 8-Br-cGMP were purchased from Alexis Biochemicals (Lörrach, Germany). All the antibodies and other substances were obtained from Sigma (Taufkirchen, Germany) unless otherwise noted.
NT2/D1 precursor cells were purchased from American Type Culture Collection (ATTC, Manassas, VA, USA) and treated as indicated in the ATTC instruction manual. NT2 cells were cultured as free floating spherical aggregate as described by Paquet-Durand et al. (2003). Briefly, the NT2 precursor cells (passages 24–32) were seeded in 80 mm, bacteriological grade Petri dishes (Greiner, Hamburg, FRG) at a density of 5 × 106 cells/dish in 10 mL of Dulbecco’s modified eagle medium (DMEM/F-12; Gibco–Invitrogen, Karlsruhe, Germany) supplemented with 10% fetal bovine serum (Invitrogen, Karlsruhe, Germany) for 24 h. A differentiation medium, DMEM containing RA at a final concentration of 10 μM was used with medium change every 2–3 days. One group of spherical aggregates were cultured in a dish with RA-containing medium for 7 days (1 week) and the other group for 14 days (2 weeks).
The spherical aggregate culture generated in a Petri dish was seeded into poly-d-lysine (PDL) and Matrigel (Becton-Dickinson, Bedford, MA, USA) coated microdishes (Ibidi GmbH, Munich, Germany) or 12-mm cover glasses at a density of approximately 2–10 aggregates. The culture was allowed to attach for 90 min in an incubator at 37°C/5% CO2. NOC-18 was prepared in 10 mM NaOH as 100 mM stock solution, 1H-[1,2,4]-oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) dissolved in dimethylsulfoxide as 20 mM stock, 7-nitroindazole (7NI) dissolved in dimethylsulfoxide as 200 mM stock, and 8-Br-cGMP and RP-8-Br-cGMP were dissolved in the medium. DMEM-containing RA and chemical compounds were added into the culture and incubated at 37°C/5% CO2 for 48 h. To determine the migration of cells out of the spherical aggregate we acquired images at desired times as described in Moors et al. (2007).
For immunocytochemical stainings, we either used spherical aggregate cultures after 48 h of the migration assay or aggregates that were mechanically dispersed into single cells. The cultures were fixed in 4% paraformaldehyde dissolved in phosphate-buffered saline (PBS, 10 mM sodium phosphate, 150 mM NaCl, pH 7.4) for 30 min. They were permeabilized by washing three times for 5 min in PBS containing 0.2% Triton X-100 (PTX). Blocking solution containing PTX and 5% normal horse or rabbit serum was applied for 1 h. The primary anti-βIII-tubulin (1 : 10000; Sigma) and anti-nestin (1 : 400; Calbiochem) were diluted in PTX-containing blocking solution and applied overnight at 4°C. Wells were washed three times for 5 min in PTX and incubated with secondary biotinylated antibody (Vector, Burlingame, CA, USA) for 1 h at 20°C. After three more washing steps in PTX, the immunofluorescence was detected using streptavidin-CY3 (Sigma) or streptavidin-Alexa Fluor 488 (Mobitec, Göttingen, Germany). For nuclear counter-staining cells were incubated with 4′-6-diamidino-2-phenylindole at a concentration of 1 μg/mL for 5 min in PBS. After final washing steps in PBS, preparations were mounted in 90/10% glycerol/PBS with 4% sodium n-propyl-gallate as antifading agent and sealed with nail varnish.
For the detection of cGMP immunoreactivity (cGMP-IR), spherical aggregates were pre-incubated for 20 min at 20°C with 1 mM sodium nitroprusside (SNP) as a NO donor, 20 μM YC-1 [3-(50-Hydroxymethyl-20-furyl)-1-benzyl indazole] as an enhancer of NO-induced activity of sGC, and 1 mM 3-isobutyl-1-methylxanthine as phosphodiesterase inhibitor. Cultures were washed once with PBS and then we followed the same staining procedures as above with a blocking solution containing PTX and 5% normal rabbit serum. The polyclonal sheep cGMP antiserum (1 : 10000; a kind gift from Dr. J. de Vente, Maastricht University, Netherlands) was used as primary antibody to detect the level of cGMP. To test for the contribution of endogenous enzyme activity to cGMP levels, we used 50 μM ODQ as a specific inhibitor of sGC.
Cells were homogenized in a ristocetin-induced platelet agglutination lysis buffer and HALT® protease inhibitor cocktail (Pierce, Rockford, IL, USA) for about 30 min. The cell lysates were centrifuged at speed of 6000 g for 10 min and the protein level of supernatant was estimated by the BCATM protein assay kit (Pierce). Subsequently, equal amount of protein (100 μg) for NT2 and NT2 + RA-treated samples were boiled for 3 min in 2× Laemmli buffer and subjected to electrophoresis on 8% sodium dodecyl sulfate acrylamide gel. The gel was blotted on polyvinylidene difloride membrane at 4°C. The membrane was then blocked with 3% dried milk powder in PBS for 2 h and incubated overnight at 4°C with a monoclonal antibody against nNOS (Sigma) at 1 : 1000 dilution in blocking solution. The biotinylated secondary antibody was applied for 1 h. Bound antiserum was detected by a standard peroxidase staining techniques using the Vectastain ABC Kit (Vector). After reactivation with methanol, membranes were additionally stained for anti-acetylated-α-tubulin diluted 1 : 10000 in blocking solution.
Cell proliferation was assessed by application of the thymidine analog 5-bromo-2′-deoxyuridine (BrdU), which was incorporated into the DNA of dividing cells. Spherical aggregates were exposed to chemicals for 48 h and then incubated with BrdU (50 μM) for 4 h. After fixation with 4% paraformaldehyde for 30 min, they were incubated with 2 M HCl solution in PBS for 20 min. Preparations were washed three times with PBS to completely remove the acid. They were permeabilized by washing three times for 5 min in 0.2% PTX and blocked for 1 h in 5% normal horse serum. Then, we followed the same immunocytochemical staining procedures as described in the Immunocytochemistry with the primary monoclonal anti-BrdU (clone Bu-33, Sigma) 1 : 200 in blocking solution.
Cell viability/cytotoxicity assay
To monitor cytotoxic effects of the drugs, we used the Alamar Blue viability assay (Trek Diagnostic Systems, East Grinstead, UK) and the Live/Dead viability/cytotoxicity assay (Molecular Probes, Eugene, OR, USA). For Alamar Blue viability assay the NT2 spherical aggregates were mechanically dispersed into single cells. About 10000 cell/well were seeded into PDL and Matrigel-coated 96-well plate and allowed to attach for 90 min. RA and chemical compounds containing medium were added and incubated for 48 h at 37°C/5% CO2. After 48 h the medium was completely changed into a cell culture medium containing 3% Alamar Blue. Fluorescence intensity of the Alamar Blue was detected at excitation/emission wavelength of 530/590 nm using a microplate reader (Infinite M200, Tecan, Männedorf, Switzerland) after 4 h. The actual number of living or dead cells was determined by incubating NT2 aggregates with 4 μM of ethidium homodimer-1 (EthD-1) and 2 μM calcein in PBS for 30 min at 37°C/5% CO2. After incubation the cultures were washed in PBS and viewed under microscope. The average numbers of living and dead cells were calculated from at least 10 images of two independent experiments and compared with live controls which was normalized to 100% for each experiment.
Microscopy and data analysis
The morphological observation and migration assay were performed on a Zeiss Axiovert 200 microscope (Zeiss, Göttingen, Germany) equipped with a CoolSNAP digital camera. Immunofluorescence was detected on a Zeiss Axiovert 25 microscope equipped with an Axiocam 3900 digital camera. The acquired images were processed using Adobe Photoshop (Adobe Systems GmbH, München, Germany). The migration was quantified by measuring the distance from the edge of the sphere to the furthest migrated cell bodies at four different locations per spherical aggregate (Moors et al. 2007). The data were presented as mean ± SEM of at least eight spherical aggregate. Each experiment was repeated at least three times. The percentage of nestin, βIII-tubulin, and BrdU were determined dividing the number of positively stained cells by the total number of cells. The data were presented as mean ± SEM of two independent experiments. Statistical comparisons of controls versus treatment were performed with the unpaired two-tailed Student’s t-test. Levels of significance were indicated as *p <0.05, **p <0.01, and ***p <0.001.
Migration of cells out of the NT2 spherical aggregates depends on the length of RA treatment
After forming confluent monolayers NT2 precursor cells were split and seeded into bacteriological dishes. The cells formed a loose aggregate after 24 h of the proliferation step followed by subsequent RA treatment for 2 days. Upon further incubation with medium containing RA for 1 week the aggregates became spherical in shape, stable, and surrounded by fewer isolated cells. When the use of RA was prolonged for 2 weeks, the aggregates attained a more spherical shape and elaborated neuronal processes. Accordingly, we classified the aggregates into two separate groups which were based on age as 1 or 2 weeks cell spheres.
The aggregates were seeded into PDL and Matrigel-coated microdishes or cover glasses containing a marker grid, which allowed us to follow the migration of cells out of the same sphere for 48 h. Cells started to detach radially from the rim of 1-week-old NT2 sphere after about 3–6 h and the migrated distance increased with time (Fig. 1a–c). Cells migrated to a lesser extent from 2-week-old NT2 spheres but elaborated neuronal processes (Fig. 1d–f). Higher magnification of the migrated area of 1-week-old spheres showed mostly single cells devoid of neuronal processes (Fig. 1g). However, 2-week-old spheres elaborated long neuronal processes (Fig. 1h) and in this case length of neurites was excluded from the migration analysis. Accordingly, migratory cells derived from 1- and 2-week-old spheres cover, after 48 h, a distance of 171.0 ± 5.5 and 77.5 ± 2.1 μm, respectively (Fig. 1i). Thus, the migration of cells out of the NT2 spheres depended on the duration of the RA treatment in the Petri dishes.
NT2 spheres express both precursor and neuronal markers
To monitor the progress of neuronal differentiation within the developing spherical aggregate, we performed immunocytochemical staining against cytoskeletal markers of progenitor cells and neurons. At the end of a 48 h lasting migration experiment spheres were fixed and stained against the neuronal progenitor marker nestin and an early marker of neuronal differentiation, βIII-tubulin. After 1 week of RA treatment the NT2 spheres expressed nestin and βIII-tubulin (Fig. 2a and c). For quantification, the spheres were dispersed mechanically into single cells and stained against both markers. While 10.3% of the cells stained positively for βIII-tubulin (Fig. 2e and h), about 84.3% remained nestin-positive (Fig. 2g). Upon further incubation with RA for an additional week, the spheres displayed an increasing amount of βIII-tubulin-stained cells (Fig. 2d). The immunoreactivity was not only expressed in the cell bodies but also in the emerging neuronal processes (Fig. 2d and f). We counted about 46%βIII-tubulin- and 63% nestin-positive cells (Fig. 2g and h). An increase in the level of βIII-tubulin, elaboration of neuronal processes together with reduction in the level of nestin upon longer incubation with RA indicated the progress of neuronal differentiation.
Studies on gene expression of NT2 cell during and shortly after differentiation with RA showed down-regulations of several stem cell markers and up-regulation of nestin and neurofilament transcripts that support our present observations at the protein levels (Houldsworth et al. 2002; Przyborski et al. 2003). Because of their high cellular motility we further characterized the type of cells that migrated out of 1-week-old NT2 spheres. Only 3.5 ± 0.5% of the migratory cells were βIII-tubulin-positive and almost all of the furthest migrated cells lacked the βIII-tubulin staining (Fig. 2c). Hence, the cells at the front of the migration expressed mainly the precursor marker nestin. Costaining of BrdU together with nestin indicated that most of the proliferating cells were nestin-positive neuronal progenitor cells (Fig. S1a). Costaining of BrdU together with βIII-tubulin showed that none of the BrdU-incorporated cells were βIII-tubulin positive (Fig. S1b). Thus, migration as measured in our cell culture model was mainly based on the motility of nestin-positive neural precursor cells (Fig. 2a). Along these lines we used the NT2 cell spheres to study the role of NO signaling in neuronal precursor migration.
Agents that modulate cell migration via NO signaling do not alter cell viability and differentiation
We initially examined the potential cytotoxicity of bioactive chemical compounds affecting cell migration via NO/cGMP signal transduction (Fig. 3a). Thus, we performed Alamar Blue cell viability and the live/dead assay under identical experimental conditions as the migration assay. Both assay showed that none of the pharmacological agents had any effect on cell viability at the concentration used to modulate cell migration (Fig. 3b and c). This was also further confirmed by morphological observation of the spheres including the migrating cells in the presence of the compounds. However, when the NOS inhibitor, 7NI; sGC inhibitor; ODQ; and NO donor, NOC-18 were used at the rather high concentration of 1 mM, they significantly affected the survival of cells (Fig. 3d–f). Next, we used the βIII-tubulin immunocytochemical staining to determine the effect of chemicals on neuronal differentiation. At the concentration used to modulate cell migration, only 7NI appeared to slightly decrease while 8-Br-cGMP increased the βIII-tubulin staining. However, the effect of both chemicals was not statistically significant (Fig. 3g).
NT2 cells express nNOS and functional sGC
Western blotting revealed the presence of the nNOS in both RA-treated and untreated precursor cells (Fig. 4a). Using a monoclonal antibody against nNOS, the blot resolved a major protein band at 155 kDa. This result is in line with studies showing that untreated NT2 cells express nNOS but not endothelial NOS and inducible NOS (Lee et al. 2001; Hyun et al. 2002). For comparison of the NOS content in precursor NT2 with NT2 + RA-treated cells, the ratio of nNOS signal to the loading control was calculated and averaged for all probed blots (n = 3). We found no significant differences between NT2 + RA and NT2 cells (0.76 ± 0.07 vs. 0.68 ± 0.11, respectively). To reveal potential cellular targets of a NO signal endogenous to the NT2 cell spheres, we used an antibody against cGMP (de Vente et al. 1987; Tanaka et al. 1997). In the absence of an exogenous source of NO, a low level of cGMP was detected which further confirmed the presence of an endogenous source of NO (Fig. 4b). The level of cGMP increased dramatically upon stimulation with the NO donor SNP (Fig. 4c). We could not observe the increment in cGMP-IR when SNP stimulation was accompanied by the sGC inhibitor, ODQ (Fig. 4d). When ODQ was used alone, hardly any cGMP-IR became detectable (Fig. 4e). All these experiments showed the expression of a NO-sensitive sGC in cells of the NT2 sphere culture.
NO is a positive regulator of cell migration via the cGMP and PKG pathway
As NO/cGMP signal transduction could be a positive regulator of cell motility during neural development, we examined the effect of nNOS inhibition on migratory behavior of the differentiating NT2 cells. In the presence of nNOS inhibitor, 7NI, the migration of cells out of the sphere was significantly reduced (Fig. 5a, b and d). Application of the sGC inhibitor, ODQ, at a concentration of 200 μM decreased the migration of the cells by 60%. In a concentration range of 50–200 μM, ODQ inhibited cell migration in a dose-dependent manner (Fig. 5c and e). Potential downstream effector proteins for the cGMP signaling pathway are PKG that exist in two isoforms, PKG-I and PKG-II (Madhusoodanan and Murad 2007). The irreversibly PKG-I binding inhibitor RP-8-Br-cGMP blocked the migration of cells in a dose-dependent manner (Fig. 5f).
To test whether cell migration could be enhanced by an exogenous stimulation of NO/cGMP signaling, we applied the NO donor NOC-18 which decays with long half-life (about 57 h at 22°C). After 24 h exposure to 50 μM of NOC-18, the migration of cells increased by more than twofold. Furthermore, NOC-18 significantly enhanced the migration in a dose-dependent manner in the concentration range of 1–100 μM (Fig. 6a–c). At relatively higher concentration (1 mM), NOC-18 appeared to significantly inhibit cell migration (Fig. 6c). A viability assay and morphological observation showed that 1 mM of NOC-18 significantly affected cell survival (Fig. 3e). The inhibitory effect on cell migration at this high concentration seemed to be because of cytotoxicity (Figs. 3e and 6c). As we determined cell migration based on the furthest migrated cells, an increased rate of cell proliferation might contribute to the facilitation of migration. Hence, we conducted a BrdU cell proliferation assay (Fig. 6d–h) in addition to viability testing. At the concentration range where we found an increasing trend of cell migration (1–50 μM), NOC-18 seemed to have a modest but not significant effect on cell proliferation (Fig. 6d–g). Intriguingly, 100 μM of NOC-18 appeared to facilitate cell migration while inhibiting proliferation (Fig. 6c,f and g). A BrdU proliferation assay was also performed in the presence of nNOS/sGC/PKG inhibitors. We could not observe any effect on proliferation of NT2 cell spheres at the concentration of chemical inhibitors that blocked cell motility (Fig. 6h). Additionally, migration was determined as the distance of the furthest migrated nestin-positive cells after NOC-18 treatment. We found out a significantly higher migration distance of the nestin-positive cells after 50 μM NOC-18 applications (Fig. 7a–c) suggesting that NOC-18 facilitated the migration of neuronal precursor cells. A BrdU proliferation assay was also performed for the nestin-positive cells which further confirmed that 50 μM NOC-18 did not alter the neuronal precursor cell proliferation (Fig. 7d).
To demonstrate that the enhanced migration by exogenous NO application was mediated via the cGMP pathway (Fig. 3a), we exposed the NT2 cell spheres to NOC-18 while blocking the downstream target enzymes sGC and PKG. We observed a significant reduction of NO-induced facilitation of cell migration when NOC-18 was used in combination with ODQ or RP-8-Br-cGMP (Fig. 8a). This provided direct evidence for the involvement of cGMP and PKG in NO-mediated neuronal precursor cell migration. Furthermore, a direct facilitation of cell migration was observed by a cell permeable analog of cGMP (8-Br-cGMP) (Fig. 8b–d). A rescue experiment showed that 8-Br-cGMP reversed the inhibitory effect of ODQ to the control level (Fig. 8e).
NT2 cell spheres as a model for early developing human neuron
The Ntera2 clone, D1 (NT2), is a well characterized cell line that has been derived from a human testicular cancer. NT2 cells terminally differentiate into post-mitotic neurons by exposure to micromolar doses of RA and mitotic inhibitors (Andrews 1984; Pleasure et al. 1992; Paquet-Durand et al. 2003). The differentiated neurons express a large number of neuronal markers and form functional synapses [for review see Paquet-Durand and Bicker (2007)].
Neurosphere-based cultures have been employed to study the involvement of signaling pathways that regulate neuronal precursor cell proliferation, differentiation, and migration (Leone et al. 2005; Mizuno et al. 2005; Moors et al. 2007). Here, we used 10 μM of RA to generate NT2 aggregate cultures that expresses both neuronal progenitor and early neuronal markers. When the cultures were seeded on PDL and Matrigel-coated substrate, cells could migrate out of the spheres. Most of the cells that migrated out after 1 week RA treatment were nestin-positive with only few cells displaying βIII-tubulin staining. This indicates that the migrating cell population is mainly composed of neuronal precursor cells (Fig. 2a and c). When the time of RA application was prolonged to 2 week, we observed an increase in the level of βIII-tubulin in cell somata and the neuronal processes. Expression of neuronal markers such as glutamate receptors, MAP2, Tau, and NeuN just after 14-day exposure with RA has been reported previously for NT2 cells (Megiorni et al. 2005). Furthermore, the migration assay showed that upon prolonged treatment with RA, the cells tended to remain inside the sphere and send out neuronal processes. This indicates that RA treatment induces neuronal differentiation while decreasing cell migration (Figs. 1 and 2). RA has been shown to reduce the migration of airway smooth muscle cells (Day et al. 2006) and neuroblastoma cell lines (Voigt and Zintl 2003; Messi et al. 2008). The early expression of cytoskeletal neuronal markers that resemble mammalian neurogenesis (Przyborski et al. 2000, 2003; Houldsworth et al. 2002; Paquet-Durand and Bicker 2007) suggests that the NT2 aggregate culture system is a valid model for developing human neuronal cells.
NO/cGMP/PKG pathway mediates the migration of human neuronal precursor cells
In this study we showed for the first time that NO signaling positively regulated cell migration in a model of human neuronal precursor cells. Both undifferentiated NT2 precursors and differentiating NT2 cell spheres expressed nNOS (Fig. 4a) that could serve as endogenous source of NO. The presence of functional sGC was directly demonstrated by exogenous stimulation of cells with a NO donor and subsequent immunocytochemical detection of cGMP. The NO-induced cGMP-IR was reduced by the specific sGC inhibitor ODQ (Fig. 4b–e). After showing the presence of the NO signal transduction pathway in the NT2 cultures we asked whether NO modulated the motility of cells out of the spheres. Chemical manipulation of the target enzymes involved in NO transduction demonstrated that inhibition of nNOS, sGC, and PKG reduced the motility of cells (Fig. 5a–f). These loss of function effects suggested that NO signal transduction positively regulated cell motility. Possible cytotoxic effects of the inhibitors could be excluded as the concentrations we used were well below the dose that caused significant cell death (Fig. 3b–f).
A second line of evidence derives from a gain of function after the exogenous application of NO and cGMP. The NO donor and membrane permeable analog of cGMP enhanced cell migration (Figs. 6a–c,7a–c and 8b–d) which directly confirmed the positive regulatory role of NO and cGMP. Interestingly, NO-induced facilitation of cell migration was abolished when NOC-18 was used together with the sGC inhibitor ODQ or PKG inhibitor RP-8-Br-cGMP (Fig. 8a). This inhibition of migration by sGC and PKG inhibitors infers that a certain level of cGMP is required for cell motility. As pharmacological inhibition of the sGC enzyme can be completely rescued by application of the cGMP analog (Fig. 8e), it is unlikely that unspecific side effects of the chemical blocker contribute to inhibition of cell migration. Our combined result implicate that NO facilitates cell migration by inducing sGC to increase cGMP levels which in turn may activate PKG.
NO signal transduction and neuronal motility
Nitric oxide and cGMP have been shown to mediate the migration of various cell types including neutrophils, epithelial, and endothelial cells (Ziche et al. 1994; Elferink and VanUffelen 1996; Noiri et al. 1996, 1998). Chemical manipulations of cultured nervous systems implicate NO/cGMP as regulator also for neuronal cell motility [reviewed by Bicker (2005, 2007)]. Neuroanatomical studies using markers against NOS and sGC suggest the migrating neuroblasts of the rostral migratory stream as potential targets for NO signaling in the adult brain (Moreno-López et al. 2000; Gutièrrez-Mecinas et al. 2007). NO signal transduction pathways regulate the migration of cerebellar neurons, insect enteric neurons, and early developing Xenopus neuronal cells (Tanaka et al. 1994; Haase and Bicker 2003; Peunova et al. 2007; Knipp and Bicker 2009). The application of a NO donor to young adult rats before and after stroke increased cell proliferation and migration in the SVZ and dentate gyrus (Zhang et al. 2001). Some studies report that NO has no effect on post-mitotic neuronal cell migration (Moreno-López et al. 2004) or decreased ependymal/SVZ cell migration after focal cerebral ischemia in rats (Zhang et al. 2007). Such inconsistency in the effects of NO on neuronal migration could be because of the use of different drug concentrations, species differences, pathological, and developmental stages of the experimental animals. Here, we provide further evidence for a significant role of NO in human neuronal precursor cell migration. The present study shows a concentration-dependent facilitation of human neuronal precursor cell migration by NO donor application. The inverted U-shaped dose-–response curve (Fig. 6c) might in part account for the inconsistency in the role of NO. In our cell culture model, NOC-18 increased the migration of cells out of spheres in a concentration range of 1–50 μM without affecting cell viability and proliferation (Figs. 3e and 6a–g). Analysis of migration distance and proliferation of the nestin-positive cells confirmed that NOC-18 facilitated the migration of the neuronal precursor cells (Fig. 7a–c) without affecting proliferation (Fig. 7d). At 100 μM NOC-18 cell migration was enhanced while proliferation was reduced (Fig. 6c, f and g). This result implicates that NO differentially coordinates cell migration and proliferation, as has also been reported by Peunova et al. (2007). At the relatively high concentration of 1 mM, the increment in cell migration was not observed. This reduction was most probably because of cytotoxicity (Figs. 3e and 6g). Low concentration of exogenous NO has been shown to enhance random migration of neutrophils by stimulating sGC, while a higher concentration inhibited migration (Elferink and VanUffelen 1996). Presumably, at low concentration NO activates only its main target enzyme sGC (Garthwaite 2008) whereas at sufficiently higher concentrations it rapidly reacts with proteins and may even cause cell death [for review see Blaise et al. (2005) and Madhusoodanan and Murad (2007)].
Presently, we do not know how the NO/cGMP/PKG pathway causes the cytoskeletal reorganization that leads to cell migration out of the NT2 spheres. In primary cultured smooth muscle cells, the migration stimulatory effect of NO donors and cGMP is associated with altered cell morphology and dissociation of actin filaments (Brown et al. 1999). In insect enteric neurons, where the NO/cGMP/PKG pathway is crucial for cell migration, a block of this pathway results in a relocalization of F-actin bundles from the neurites to the cell body (Haase and Bicker 2003). The stimulation of the NO/cGMP/PKG pathway in astrocytes caused a redistributation of glial fibrillary acidic protein and depolymerization of actin with a loss of stress fibers. These cytoskeletal rearrangements resulted in a facilitation of astrocyte motility (Borán and García 2007). Such changes in neurofilaments and actin are attributed to inhibition of RhoA GTPase by NO-stimulated PKG pathway (Sawada et al. 2001; Gudi et al. 2002; Borán and García 2007; Peunova et al. 2007). Thus, there is a link between the NO signal transduction pathway and actin cytoskeleton, possibly via RhoA GTPase. Phosphorylation of Enabled/vasodilator-stimulated phosphoprotein family proteins that regulate actin polymerization by PKG has also been reported to account for the action of cGMP in cell motility (Sporbert et al. 1999; Chen et al. 2007; Lindsay et al. 2007). Most likely changes in the cytosolic calcium concentration will contribute to the morphology of motile cell processes. Recent investigations have shown that NO regulates growth cone filopodial behavior involving sGC, PKG, and cyclic adenosine diphosphate ribose which causes the release of calcium from intracellular stores via the ryanodine receptor (Welshhans and Rehder 2005, 2007).
This study demonstrated that NO facilitated the migration of differentiating model neuronal cells through the cGMP and PKG signaling pathway. In our accessible NT2 sphere culture system we can now investigate the mechanisms by which NO regulates human neuronal precursor motility.
We thank Dr. J. de Vente for his kind gift of the cGMP antiserum, Dr M. Stern for the discussion and help during the preparation of the manuscript, and S. Tan for technical support. M.A. Tegenge was supported by a Georg–Christoph–Lichtenberg scholarship from the Ministry for Science and Culture of Lower Saxony and G. Bicker was supported by DFG grant BI 262/16-1.