Address correspondence and reprint requests to Dr Jerònia Lladó, Departament de Biologia, Ed. Guillem Colom, Universitat de les Illes Balears, Cra de Valldemossa, km 7.5, E-07122 Palma de Mallorca, Illes Balears, Spain. E-mail address: firstname.lastname@example.org
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by the selective death of motoneurons. Recently, vascular endothelial growth factor (VEGF) has been identified as a neurotrophic factor and has been implicated in the mechanisms of pathogenesis of ALS and other neurological diseases. The potential neuroprotective effects of VEGF in a rat spinal cord organotypic culture were studied in a model of chronic glutamate excitotoxicity in which glutamate transporters are inhibited by threohydroxyaspartate (THA). Particularly, we focused on the effects of VEGF in the survival and vulnerability to excitotoxicity of spinal cord motoneurons. VEGF receptor-2 was present on spinal cord neurons, including motoneurons. Chronic (3 weeks) treatment with THA induced a significant loss of motoneurons that was inhibited by co-exposure to VEGF (50 ng/mL). VEGF activated the phosphatidylinositol 3-kinase/Akt (PI3-K/Akt) signal transduction pathway in the spinal cord cultures, and the effect on motoneuron survival was fully reversed by the specific PI3-K inhibitor, LY294002. VEGF also prevented the down-regulation of Bcl-2 and survivin, two proteins implicated in anti-apoptotic and/or anti-excitotoxic effects, after THA exposure. Together, these findings indicate that VEGF has neuroprotective effects in rat spinal cord against chronic glutamate excitotoxicity by activating the PI3-K/Akt signal transduction pathway and also reinforce the hypothesis of the potential therapeutic effects of VEGF in the prevention of motoneuron degeneration in human ALS.
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by the selective death of motoneurons in the spinal cord, brainstem and motor cortex which causes progressive muscle weakness and paralysis. The causes are still unknown; however, several hypothesis to explain motoneuron degeneration have been proposed, including mitochondrial dysfunction, protein aggregate formation, excitotoxicity, deficient axonal transport, lack of growth factors, and neuroinflammation (Boillee et al. 2006a). The astrocytic glutamate transporter, excitatory amino acid transporter 2/glutamate transporter 1 is down-regulated in cerebral cortex and spinal cord of ALS patients (Rothstein et al. 1995). This down-regulation accounts for an excessive concentration of glutamate in the synaptic cleft and the overstimulation of glutamate receptors which causes excitotoxic neuronal death. Although in ALS there is a selective death of motoneurons, there is a compelling evidence suggesting that glial cells may be involved in the initiation and/or propagation of the disease (Clement et al. 2003; Boillee et al. 2006b). Besides their function in regulating extracellular glutamate levels, glial cells are also a significant source of neurotrophic factors which play an important role in the development of the nervous system and providing also trophic support to the adult populations of neurons. Moreover, neurotrophic factors protect neurons against different kinds of excitotoxic insults. In this sense, it is possible that glial cells affected by the disease (i.e., ALS) are unable to secrete the right amount of neurotrophic factors required to sustain motoneurons.
Vascular endothelial growth factor (VEGF) was originally described as a factor with a regulatory role in vascular growth and development; however, several studies have revealed that VEGF has direct effects, such as stimulation of neuritic outgrowth (Sondell et al. 2000; Rosenstein et al. 2003) and neuronal survival (Jin et al. 2000a; Ogunshola et al. 2002) on a variety of neuronal cell types. Thus, VEGF has also been known for their neurotrophic and neuroprotective properties in nervous tissue and possibly are independent from its vascular actions (Storkebaum et al. 2004). The discovery that a deletion in the hypoxia-response element in the promoter region of the VEGF gene causes motoneuron degeneration with features similar to ALS (Oosthuyse et al. 2001) suggested that VEGF may be involved in the pathogenesis of ALS. In this regard, reduction in the levels of VEGF in the superoxide dismutase (SOD1) mutant mice, a model of ALS, worsens the disease (Lambrechts et al. 2003); and the treatment of mutant SOD1 mice and rats with VEGF clearly ameliorates the illness (Azzouz et al. 2004; Storkebaum et al. 2005) suggesting that VEGF could be a neurotrophic factor for motoneurons. Moreover, VEGF expression is decreased in the spinal cord of ALS patients (Brockington et al. 2006).
Vascular endothelial growth factor is a homodimeric growth factor whose biological activity is mediated through binding to the receptors VEGF receptor-1 (VEGFR1, Flt-1), VEGF receptor-2 (VEGFR2, Flk-1), VEGF receptor-3 (VEGFR3, Flt-4), and neuropilins (Nrp-1; Nrp-2). It is accepted that the major mediator of angiogenic permeability-enhancing effects as well as trophic effects on neurons is VEGFR2 [reviewed by Lambrechts and Carmeliet (2006)], which is expressed in rat spinal cord (Oosthuyse et al. 2001). VEGF expression in the normal spinal cord is low in the adult, however, it increases in response to injury (Tsao et al. 1999; Skold et al. 2000; Islamov et al. 2004; Fu et al. 2005). Motoneurons, astroglia (Oosthuyse et al. 2001), and microglia (Bartholdi et al. 1997) are sources of VEGF in the spinal cord.
Upon ligand binding, VEGFR2 undergoes phosphorylation (Meyer et al. 1999), activating in turn, intracellular signal transduction pathways, including phosphatidylinositol 3-kinase (PI3-K)/Akt and mitogen-activated protein kinase and ERK kinase (MEK)/extracellular signal-regulated kinase (ERK) pathways which are known to mediate cell survival in a variety of cells, including neurons. [Correction added after online publication (22/01/08): In the preceding sentence, mitogen-activated protein kinase 2nd ERK kinase was changed to mitogen-activated protein kinase and ERK kinase] In this sense, VEGF has neuroprotective effects in cultured cortical neurons after ischemia, serum deprivation, or hypoxia (Jin et al. 2000a,b). As hypoxic neuronal death also involves glutamate release and subsequent excitotoxic injury (Choi 1994), VEGF could be a good candidate in protecting neurons against glutamate-induced excitotoxicity; in this sense, VEGF protects cultured hippocampal neurons from glutamate (Matsuzaki et al. 2001) and NMDA-mediated (Svensson et al. 2002) toxicity. However, the potential neuroprotective effects of VEGF on spinal cord motoneurons after glutamate excitotoxicity have not been clearly established.
In this study, we investigated the potential role of VEGF in protecting spinal cord cultures against chronic glutamate-induced excitotoxicity, which is a cell death mechanism in ALS. VEGF exerts its protective effects on motoneurons against hypoxia-induced toxicity through the activation of PI3-K/Akt and the MEK/ERK signaling pathways (Shiote et al. 2005). Therefore, we examined if VEGF could activate these pro-survival intracellular signaling pathways and also if the activation of the PI3-K/Akt pathway would have a function protecting spinal cord; particularly motoneurons against glutamate-induced excitotoxicity.
Materials and methods
Human recombinant VEGF was purchased from Peprotech EC (London, UK), and was used at the indicated times and doses. The PI3-K inhibitor, LY294002 and the MEK inhibitor, U0126 were from Calbiochem (San Diego, CA, USA). Other reagents were from Sigma-Aldrich (St Louis, MO, USA) except when indicated.
Postnatal rat spinal cord organotypic culture
Organotypic spinal cord cultures were prepared from lumbar spinal cords of 8-day-old Sprague–Dawley rat pups (P8), as previously described (Llado et al. 2004). Lumbar spinal cords were collected under sterile conditions and transferred to sterile Gey’s balanced salt solution containing glucose (6.4 mg/mL). They were transversely sectioned into 350-μm slices with a McIIwain Tissue Chopper (Gomshall, Surrey, UK). Sections were carefully placed on Millipore Millicell-CM porous membranes (0.4 μm, Millipore; Billerica, MA, USA) in 35-mm culture wells containing 1 mL of incubation media [50% minimal essential medium, 25 mM HEPES, 25% heat-inactivated horse serum, 2 mM glutamine and 25% Hanks’ balanced salt solution (Invitrogen, Carlsbad, CA, USA) supplemented with d-glucose (25.6 mg/mL) at a final pH of 7.2]. Cultures were incubated at 37°C in a 5% CO2/95% air humidified environment. Medium was changed twice every week. Under these conditions, cultures can be maintained up to 3 months with steady motoneuron survival and preservation of synaptic morphology. Cultures were let to stabilize for 1 week, after this point the motoneuron population reaches a steady number and remains stable from 1 to 4 weeks (Rothstein et al. 1993). Thus, all treatments started 7 days after the explant procedure.
Chronic glutamate neurotoxicity was induced by exposure to threohydroxyaspartate (THA; 1–4 weeks), a potent glutamate transporter inhibitor (100 μM). This compound has been described to produce a dose-dependent sustained elevation of glutamate levels causing degeneration of motoneurons (Rothstein et al. 1993). To assess the potential neuroprotective effects of VEGF, it was added to the organotypic culture medium (50 ng/mL) at the same time as THA. In the PI3-K and MEK inhibition studies, LY294002 (10 μM) and U0126 (10 μM) were added at the same time as VEGF. All drugs were added to the organotypic culture medium starting from 7 days after culturing and were maintained in all subsequent medium changes.
After several days of growing in spinal cord medium or exposure to THA, VEGF or THA plus VEGF in the presence or absence of kinase inhibitors, organotypic cultures were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 for 30 min at 21 ± 1°C. After blocking with 5% normal horse serum in 0.2% Triton X-100 for 1 h, the cultures were incubated overnight at 4°C with the primary antibody. Different primary antibodies were used: anti-neurofilament heavy chain (NF-H, SMI-32; 1 : 1000) from Stenberger Monoclonals Inc. (Baltimore, MD, USA); anti-VEGFR2 (1 : 50) from Santa Cruz Biotechnology (Santa Cruz, CA, USA); anti-glial fibrillary acidic protein (anti-GFAP; 1 : 1000) from DAKO (Glostrup, Denmark); anti-phosphoERK 1/2 (p44/42 MAPK) (1 : 250) and anti-phosphoAkt (1 : 200) from Cell Signaling Technology (Danvers, MA, USA); anti-Bcl-2 (1 : 100) from BD Biosciences (Franklin Lakes, NJ, USA); anti-survivin (1 : 500) from Novus Biologicals (Littleton, CO, USA); and anti-microtubule associated protein 2 (1 : 1000) and anti-neurofilament of low molecular weight (1 : 100) from Chemicon (Temecula, CA, USA). Specificity controls were made by incubating without primary antibody. For immunofluorescence, sections were incubated for 1 h with the appropriate secondary antibody, Fluorescein anti-rabbit or Texas red anti-mouse (Vector Laboratories, Burlingame, CA, USA). Cultures were then washed and mounted using Gelmount solution. For motoneuron counting, the peroxidase-based detection system was used. The secondary antibody used was a biotinylated anti-mouse and avidin-biotin complex reagents (Vector Laboratories); the diaminobenzidine reaction was used for color development. Images were obtained using a LEICA DMR epifluorescence microscope (Leica Microsystems, Wetzlar, Germany) equipped with a Leica DC300 camera and software. A Leica TCS SP2 confocal laser scanning microscope was also used for co-localization studies.
Motoneurons in the organotypic spinal cord sections were identified by SMI-32 immunostaining and on the basis of their morphology and size (> 25 μm) and their localization in the ventral horn. All motoneurons meeting these criteria were blindly counted in each spinal cord section. A minimum of 30 sections were counted for each experimental condition.
Western blotting was performed in spinal cord cultures after several days of growing in spinal cord medium or exposure to THA, VEGF or THA plus VEGF in the presence or absence of kinase inhibitors for the selected periods and concentrations. After treatments, spinal cord cultures were rinsed in ice-cold phospate-buffered saline (PBS) and lysed in 50 mM Tris–HCl, pH 6.8, 2 mM EDTA and 0.5% Triton X-100 buffer. Lysates were sonicated and proteins quantified by means of the DC Protein Assay from Bio-Rad Laboratories (Hercules, CA, USA). Protein equivalents from each sample (15 μg/well) were resolved in sodium dodecyl sulfate–polyacrylamide gel electrophoresis and electro transferred to 0.45 micron nitrocellulose membranes (Amersham, Buckinghamshire, UK) using a Bio-Rad semidry Trans-Blot, according to the manufacturer’s instructions. Membranes were blocked at 21 ± 1°C for 1 h with PBS containing 5% non-fat dry milk, 0.5% bovine serum albumine and 0.2% Tween 20. Membranes were probed overnight at 4°C with the adequate primary antibodies: anti-VEGFR1 (1 : 500) and anti-VEGFR2 (1 : 500); SMI-32 (1 : 1000); anti-ERK 1/2 (p44/42 MAPK; 1 : 1000), anti-Akt (1 : 1000) from Cell Signaling Technology; anti-Bcl-2 (1 : 1000); and anti-survivin (1 : 1000). Membranes were then washed with PBS for 10 min at room temperature (three times) and incubated for 2 h with the appropriate peroxidase-conjugated secondary antibodies. All the antibodies were used at the dilutions recommended by the suppliers. Blots were finally developed with the chemiluminiscent peroxidase substrate (Sigma-Aldrich) and visualized in chemiluminiscence film (Amersham). The apparent molecular weight of proteins was determined by calibrating the blots with pre-stained molecular weight markers (Bio-Rad). Commercial lysates of A10 thoracic aorta myoblasts and EVC304 transformed endothelial cells (Santa Cruz Biotechnology) were used as a positive control for VEGFR1 and VEGFR2 expression, respectively.
Extracellular signal-regulated kinase 1/2 and Akt kinases activation was measured by their phosphorylation using primary antibodies against phosphorylated (Thr202/Tyr204) ERK1/2 (1 : 1000) and phosphorylated (Ser473) Akt (1 : 1000), respectively (Cell Signaling Technology) in cultures subjected to 24 h of serum deprivation and then exposed to VEGF (100 ng/mL) for different times in the presence or absence of the MEK inhibitor, U0126 (10 μM) or the PI3-K inhibitor, LY294002 (10 μM).
All data are expressed as mean ± SEM values. anova followed by Bonferroni test was used for the statistical evaluations. The level of significance was chosen as p ≤ 0.05.
VEGFR2 is expressed in spinal cord organotypic cultures
To investigate the effects of VEGF on spinal cord neurons, we first studied the expression of VEGF receptors in our culture model. Western blot analysis showed the presence of VEGFR2, but not VEGFR1 in the spinal cord organotypic cultures (Fig. 1b). Immunohistochemistry also revealed VEGFR2 but not VEGFR1 expression in these cultures. Double-immunolabeling with VEGFR2 and microtubule associated protein 2 or GFAP showed neuronal localization of VEGFR2, specifically in large neurons in the ventral horn of the spinal cord that were identified as motoneurons. VEGFR2 did not co-localize with the astrocytic marker GFAP (Fig. 1a).
VEGF protects motoneurons against chronic glutamate excitotoxicity
Excitotoxic degeneration of motoneurons has been proposed as a pathogenic mechanism in ALS. The potential neuroprotective effects of VEGF against chronic THA-induced cell death were studied by exposing organotypic spinal cord cultures to this glutamate-uptake inhibitor, as a model of excitotoxic motoneuron degeneration. Spinal cord organotypic cultures were exposed to THA (100 μM) for 1 to 4 weeks. A significant (p <0.001) loss of motoneurons at 3 (61%) and 4 weeks (83%; Fig. 2a) as well as a loss of integrity at the neuropil was observed (Fig. 2b).
We next investigated the ability of VEGF to protect spinal cord neurons against excitotoxicity by co-administration of THA along with VEGF (50 ng/mL). Western blot experiments revealed that VEGF was able to attenuate the loss of NF-H (assessed by SMI-32 antibody) induced by THA in the spinal cord cultures (Fig. 3a). To assess if this VEGF protective effect could also be reflected in the spinal cord motoneuron population, immunohistochemistry with SMI-32 antibody was performed. Cultures treated with THA alone showed a loss of motoneurons; however, in the VEGF plus THA condition, a relative preservation of neurites in the explant, as well as integrity of motoneuron somas was observed. In fact, motoneuron counts revealed that VEGF significantly protected spinal cord motoneurons from excitotoxic death (44% and 49% of protection after 3 and 4 weeks of THA exposure, respectively; Fig. 3b–e). Exposure to VEGF alone did not affect the survival of motoneurons (Fig. 3b, f).
Other studies have shown that pathological conditions could increase the expression of VEGF or its receptors (Ijichi et al. 1995). Thus, an increased expression of VEGF receptors because of glutamate-induced excitotoxicity could partly explain the increased survival of motoneurons induced by VEGF. However, there was not any modulation of VEGF or VEGFR2 by any of the experimental treatments (data not shown).
VEGF activates PI3-K/Akt and MEK/ERK signal transduction pathways in spinal cord cultures
The PI3-K/Akt and MEK/ERK pathways are important regulators of cell survival and have been shown to mediate the anti-apoptotic effects of VEGF in endothelial cells (Gupta et al. 1999) and neurons (Shiote et al. 2005). To elucidate the capacity of VEGF to stimulate cell signaling pathways known to increase survival of motoneurons, spinal cord cultures were exposed to VEGF (100 ng/mL) for increasing periods (6–24 h), and homogenates analyzed by western blotting using anti-phosphoAkt and ERK as phosphorylation at specific sites is essential for their activity. These studies demonstrated that VEGF stimulation induces an increase in the phosphorylation of both Akt and ERK, with a maximal activation of both kinases after 12 h of exposure. No changes in the expression of the kinases above were detected after this short-term (6–24 h) VEGF exposure (Fig. 4a, b). To assess if the phosphorylation of these kinases could also be detected specifically in the spinal cord motoneurons after stimulation with VEGF (12 h), immunohistochemistry co-localization experiments were performed and showed the presence of the phosphorylated forms of both Akt and ERK in motoneurons (Fig. 4c).
The phosphorylation of Akt and ERK stimulated by VEGF was abrogated in the presence of the PI3-K inhibitor, LY294002 (10 μM) or the MEK inhibitor, U0126 (10 μM), respectively; indicating that VEGF activates these kinases via PI3-K and MEK, respectively (Fig. 4d). Chronic (1 week) treatments with THA and VEGF, alone or in combination did not induce a modulation in the total Akt or ERK proteins (Fig. 4e), suggesting that the long-term neuroprotective effects of VEGF are mediated by increased phosphorylation and not by increased expression of these kinases.
VEGF mediates neuroprotection against chronic glutamate-induced neurotoxicity and restores the levels of Bcl-2 and survivin via PI3-K/Akt
It is well documented that when Akt is phosphorylated in response to a wide variety of growth factor stimuli, it plays a key role in promoting neuronal survival by opposing apoptotic pathways (Dolcet et al. 1999; Soler et al. 1999; Namikawa et al. 2000). To assess whether activation of VEGF signaling through PI3-K/Akt is a component of VEGF-mediated neuroprotection, THA and VEGF were co-administered in the presence or absence of the PI3-K inhibitor, LY294002 (10 μM). Western blot analysis revealed that the protective effect of VEGF against the loss of the NF-H protein found after 2 weeks of THA treatment was abrogated in the presence of the PI3-K inhibitor, LY294002 (Fig. 5a). Furthermore, LY294002 also reversed the rescue of motoneurons observed with VEGF after 3 weeks of THA treatment (Fig. 5b). Administration of LY294002 alone did not affect motoneuron survival (Fig. 5b). The study of the potential implication of the MEK/ERK signal transduction pathway on the neuroprotective effects of VEGF could not be achieved as the MEK inhibitor, U0126 (1 to 10 μM) became toxic for the spinal cord cultures (data not shown).
Up-regulation of Bcl-2 expression has been identified as a critical mechanism by which growth factors promote cell survival in motoneurons (Newbern et al. 2005). Western blot experiments demonstrated that Bcl-2 expression was decreased after THA treatment; however, VEGF was able to attenuate the down-regulation of the molecule, restoring normal levels. The addition of the PI3-K inhibitor LY294002 reversed the effect of VEGF (Fig. 6a, b). Immunohistochemistry experiments showed that these protein changes also occur in motoneurons (Fig. 6c).
Survivin, a member of the family of inhibitor of apoptosis proteins, prevents apoptosis by blocking activation of both initiator and effector caspases (Deveraux and Reed 1999). Western blot experiments showed that survivin expression was reduced after THA treatment (Fig. 7a, b). However, exposure to THA in the presence of VEGF totally prevented the down-regulation of survivin, and the addition of the PI3-K inhibitor LY294002 reversed the effect of VEGF (Fig. 7a, b). Immunohistochemistry experiments showed expression of survivin in neurons, but not in motoneurons; thus, the previously observed changes by western blot in the expression of survivin may involve other neuronal cell types (Fig. 7c).
Together, these data suggest that the PI3-K/Akt signal transduction pathway plays a role in the neuroprotective effects of VEGF against chronic THA-induced cell death in rat spinal cord and that the restoration of the normal levels of the anti-apoptotic proteins Bcl-2 and survivin could be implicated in its neuroprotective effects.
Recent insights into the role of VEGF in neurological diseases suggest that VEGF could be a promising candidate in the therapy of ALS. Several lines of evidence indicate that VEGF acts on motoneurons as a neurotrophic factor (Oosthuyse et al. 2001; Shiote et al. 2005), being neuroprotective against oxidative stress (Li et al. 2003), and after hypoxia/hypoglycemia-induced cell death (Van Den Bosch et al. 2004). The present study provides evidence that VEGF protects motoneurons against glutamate-induced excitotoxicity in spinal cord organotypic cultures. Our results in agreement with others previously published (Matsuzaki et al. 2001; Li et al. 2003; Kilic et al. 2006), suggest that VEGF exerts its neuroprotective action on spinal cord motoneurons through a molecular mechanism involving Akt phosphorylation, and also the anti-apoptotic proteins Bcl-2 and survivin.
As our studies were carried out in a paradigm of chronic excitotoxicity/slow degeneration, the chances to observe motoneurons with apoptotic features are minimal in this model. However, if not apoptosis per se, molecular elements of the programmed cell death cascade may be implicated in this type of motoneuron degeneration as the excitotoxic treatments down-regulated Bcl-2 and survivin.
In contrast to observations by Silverman et al. (1999) demonstrating an increased survival of dopaminergic neurons in mesencephalic explant cultures in response to VEGF, basal motoneuron survival was not modulated by VEGF. This discrepancy may be because of several factors: first, intrinsic differences in the response to VEGF between dopaminergic neurons and motoneurons; second, embryonic mesencephalic explants may be more dependent on growth factors than the postnatal spinal cord cultures; and third, VEGF was added at the time of performing the mesencephalic explant cultures and after 1 week in our experimental paradigm; after this point, the population of motoneurons remains stable from 1 to 4 weeks (Fig. 2a), and thus, in this condition a pro-survival effect of VEGF cannot be expected.
Our findings suggesting a potential role of VEGF interfering with excitotoxic neuronal death pathways in rat spinal cord are in agreement with Svensson et al. (2002) and Matsuzaki et al. (2001), who also describe neuronal protection by VEGF against glutamate-induced excitotoxicity in primary rat hippocampal cultures. However, others have demonstrated that VEGF does not modify the vulnerability of motoneurons to excitotoxicity in co-cultures of motoneurons on a glial feeder layer (Van Den Bosch et al. 2004). Differences between these and the present results may be partly because of the models used: Van Den Bosch et al. 2004 induced acute kainic acid excitotoxic death while we have a model of chronic slow excitotoxicity. In addition, motoneuron response to drugs may be different if they grow on a glial feeder layer or in the organotypic culture which has an excellent preservation of the interactions between neurons and glia. Moreover, neuroprotection by VEGF in an in vivo model of progressive spinal motoneuron death induced by overactivation of α-amino-3-hydroxy-5-methylisoxazole-4-propionate receptors has been reported recently (Tovar et al. 2007).
As VEGFR2 but not VEGFR1 antisense oligonucleotides reverse the neuroprotective activity of VEGF (Matsuzaki et al. 2001) and neuropilin-1 ligands are unable to mimic these effects (Jin et al. 2000b), it seems that VEGFR2 is implicated in VEGF-mediated neuroprotection. In our spinal cord culture, VEGFR2 was detected by western blot and immunohistochemistry showed its expression in neurons but not in astrocytes (Fig. 1a,b). By contrast, VEGFR1 was undetectable neither by western blot studies nor by immunohistochemistry. Thus, in our model the neuroprotective effect of VEGF is probably mediated by the stimulation of neuronal VEGFR2, and an indirect effect through the astrocytes is unlikely. VEGFR2 immunoreactivity was only found in a subset of spinal cord motoneurons; in this regard, VEGF and VEGFR2 receptors are not generally expressed but present only in subsets of neurons in the CNS (Ogunshola et al. 2002). Thus, the fact that VEGF did not achieve a complete protection of the population of spinal cord motoneurons at all times studied may be because of VEGFR2 expression only in a subset of motoneurons or a low expression of these receptors on spinal cord motoneurons.
Our results indicate that the PI3-K/Akt pathway participates in the VEGF-mediated neuroprotection against chronic glutamate excitotoxicity. In fact, several studies have also demonstrated that VEGF-mediated neuronal survival is dependent on PI3-K but independent of MAPK activity. Thus, VEGF is able to protect a motoneuron-like neuroblastoma × spinal cord cell line from stress and death caused by mutant SOD1 via PI3-K activity (Li et al. 2003). Akt is the critical protein activated by PI3-K, regulating the balance between survival and apoptosis. Although Akt phosphorylation by VEGF is transient, it may be sufficient to transmit signals for survival to downstream targets. This observation is in agreement with Matsuzaki et al. (2001), who also showed that a transient phosphorylation of Akt induced by VEGF is able to rescue hippocampal neurons from glutamate-induced toxicity. Thus, the signal transmitted to downstream targets by transient kinase activation is enough to promote long-term survival of neurons. Recently, a loss of Akt phosphorylation both in human ALS and in the mouse model of the disease has been reported. Interestingly, VEGF counteracts the loss of phospho-Akt in the mouse model (Dewil et al. 2007). In PC12 cells, Akt phosphorylates and activates the transcription factor, cyclic AMP-response element binding protein, implicated in the transcription of the Bcl-2 gene (Pugazhenthi et al. 2000) and in this sense, VEGF has been shown to induce Bcl-2 expression in neuroblastoma cells, protecting them from apoptosis (Beierle et al. 2002). Moreover, inhibition of glutamate toxicity by Bcl-2 has also been demonstrated (Zhong et al. 1993; Lawrence et al. 1996; Howard et al. 2002). In this context, over-expression of Bcl-2 has been shown to attenuate motoneuron degeneration in the SOD1 model of ALS, in which excitotoxicity seems to play a role in the development of the disease (Azzouz et al. 2000). In addition to its actions in blocking apoptosis after cytochrome c release, Bcl-2 has been shown to increase the calcium uptake and buffering capacity of mitochondria (Zhong et al. 1993). As calcium is implicated in excitotoxic neuronal death, Bcl-2 would facilitate the adaptation to higher concentrations of calcium and this might protect neurons against glutamate toxicity. In this regard, we have shown that excitotoxic conditions decreased Bcl-2 expression in the spinal cord cultures, and that the VEGF-induced neuroprotection against glutamate toxicity could be related to the restoration of Bcl-2 levels in the spinal cord cultures, and specifically in motoneurons.
Survivin is a novel member of the inhibitor of apoptosis proteins that exerts its actions by physically associating with both initiator and effector caspases, preventing their activation (Deveraux and Reed 1999). Survivin has been implicated in the survival-promoting effects of VEGF in endothelial (Mesri et al. 2001), and also in neuroblastoma cells (Beierle et al. 2005). Survivin also plays a critical role in protecting neurons from deregulated apoptosis (Jiang et al. 2005) and is up-regulated in the brain after traumatic injury (Johnson et al. 2004); therefore, its actions could be relevant on neuronal repair and neurodegenerative diseases. Our results indicating that the modulation by VEGF of survivin expression is mediated by activation of the PI3-K/Akt pathway are in agreement with others showing survivin over-expression through the same pathway after VEGF exposure in neuroblastoma cells (Beierle et al. 2005). As motoneurons do not seem to express detectable amounts of survivin, the effect of VEGF restoring survivin expression may protect other spinal cord neurons having an indirect effect on the survival of motoneurons.
The present results suggest that VEGF may have therapeutic effects in the prevention of motoneuron degeneration in human ALS, and the molecular mechanisms presented here could be considered as therapeutic targets underlying the neuroprotection of motoneurons provided by VEGF in the animal ALS models, in which VEGF over-expression (Wang et al. 2007), intracerebroventricular delivery of VEGF (Storkebaum et al. 2005) or VEGF delivery with a lentivector (Azzouz et al. 2004) proved to be somewhat effective in ameliorating the disease.
This study was supported by the Spanish Ministry of Health through the ‘Fondo de Investigación Sanitaria’ Grants FIS PI030632 (JL), FIS 02/3008 (JL) and FIS PI041507 (GO), and by ‘Fundació La Marató de TV3’. LT was supported by a pre-doctoral fellowship from ‘Govern Balear, Conselleria d’Economia Hisenda i Innovació’. MM was supported by a pre-doctoral fellowship from the ‘Fondo de Investigación Sanitaria’. JL was supported by the Ramón y Cajal Program from the Spanish Ministry of Education and Science.