| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Australian Centre for Blood Diseases, Monash University, Alfred Medical Research and Education Precinct, Prahran, Victoria, Australia
1Correspondence: Australian Centre for Blood Diseases, Monash University, 89 Commercial Rd., Prahran 3181, Victoria, Australia. E-mail: robert.medcalf{at}med.monash.edu.au
ABSTRACT
Oncostatin M (OsM) is a member of the interleukin (IL)-6 family of cytokines and is well known for its role in inflammation, cell proliferation, and hematopoiesis. OsM, together with its glycoprotein 130 containing receptor complex, is expressed and regulated in most cells of the central nervous system (CNS), yet the function of OsM within this compartment is poorly understood. Here we have investigated the effect of OsM using in vitro and in vivo models of excitotoxic injury. Using primary cultures of mouse cortical neurons, OsM was shown to reduce N-methyl-D-aspartate (NMDA) -induced neuronal death by 50% when added simultaneously with NMDA while pretreatment of neurons with OsM fully prevented NMDA toxicity indicating a profound protective effect of this cytokine. OsM was also shown to inhibit NMDA-mediated increase in levels of free intracellular calcium and to selectively reduce neuronal expression of the NR2C subunit of the NMDA receptor. Finally, using an in vivo model of excitotoxic injury, OsM significantly reduced the NMDA-induced lesion volume when coinjected with NMDA into the mouse striatum. Taken together, these results identify OsM as a powerful neuroprotective cytokine and provide a rational foundation to explore the therapeutic potential for OsM in diseases of the CNS.—Weiss, T. W., Samson, A. L., Niego, B., Daniel, P. B., Medcalf, R. L. Oncostatin M is a neuroprotective cytokine that inhibits excitotoxic injury in vitro and in vivo.
Key Words: inflammation • glycoprotein 130 • glutamate • stroke
ONCOSTATIN M (OSM) BELONGS to the IL-6 (interleukin-6) family of neuropoietic factors that include the archetypical founder IL-6, as well as IL-11, leukemia inhibitory factor (LIF), ciliary neurotrophic factor, cardiotrophin-1, and cardiotrophin-like cytokine. This family is characterized by the pluripotency and redundancy of their biological responses and acts via heterodimeric receptor complexes. These receptors all contain the common signal transducer glycoprotein 130 (gp130) while selective cellular responses to individual cytokines are determined by the composition of the gp130 partnering molecule within the receptor complex (1)
.
In humans, two functional OsM receptors have been described: the type I receptor that consists of gp130 and LIF receptor subunits, and the type II receptor containing gp130 and the OsM receptor ß (OsMRß). Although the type I receptor is used by both OsM and LIF, the type II receptor is used exclusively by OsM. In the murine system, however, mouse OsM binds only to the type II receptor and does not engage the type I receptor (2)
.
The biological functions of OsM are complex and variable and depend on the cellular microenvironment. Indeed, OsM has been reported to confer both enhancing and suppressive effects on the inflammatory response and is often implicated in processes underlying cell proliferation and tumor development (3
, 4)
. Within the IL-6 family, OsM is most closely related to LIF by structural, functional, and genetic criteria (3)
. Although OsM and LIF display overlapping functions due to the common usage of the type I receptor, more recent findings have shown that OsM exhibits unique activities not shared with LIF, including its selective role during hematopoiesis and its effect on the regulation of certain target genes (5
, 6)
.
Within the CNS, however, the role of OsM is less well understood. In humans, elevated levels of OsM have been described in several CNS pathologies, including multiple sclerosis, HIV-associated dementia, and epileptic seizure (7
8
9)
. Although OsM is expressed in neurons and astrocytes, microglia are considered to be major producers of OsM in the brain (4)
. In vivo, neurons are closely associated with microglia and astrocytes and are likely to be exposed to high concentrations of glial-derived OsM generated after inflammatory, ischemic, or excitotoxic injuries. The inflammatory response after brain injury may also significantly influence neuronal outcome. Although the consequences of elevated or enforced OsM expression in the brain are unknown, other inflammatory cytokines have been shown to confer either beneficial (e.g., TGF-ß) or detrimental outcomes (e.g., IL-1ß) in models of cerebral ischemia (10
, 11)
. Strong links have also been established between inflammatory processes, excitotoxicity, and some neuropathological conditions (12
, 13)
. Excitotoxic pathways initiated after excessive glutamate release have been implicated in acute disorders such as epileptic seizures, stroke, and traumatic brain and spinal cord injury, as well as in some chronic age-related neurodegenerative disorders, including Alzheimer’s disease and amyotrophic lateral sclerosis (14)
. In the context of the gp130/interleukin-6 family, expression levels of gp130 are increased in hippocampal neurons during ischemia, whereas levels of IL-6 itself are enhanced after ischemic injury (13
, 15)
. IL-6 has been reported to inhibit glutamate release and the spread of excitation in the cerebral cortex, and to partially block the neurotoxic effects of the glutamate analogue N-methyl-D-aspartate (NMDA) in vitro and in vivo; hence, a neuroprotective effect for this family member has been proposed (13
, 16
, 17)
. However, this is an area of controversy as other reports have indicated that IL-6 can promote neurodegeneration and enhance excitotoxic injury (18
, 19)
.
We have investigated the effect of OsM using in vitro and in vivo models of excitotoxic injury. Our data indicate that OsM significantly attenuates excitotoxic cell death in vitro and in vivo. The results of this study therefore identify OsM as a novel and powerful neuroprotective cytokine, and provide additional impetus to investigate further the biological role and potential of other gp130 ligands within the CNS.
MATERIALS AND METHODS
Animals
Wild-type (WT) (C57/Black 6) male mice between 8–10 wk of age were used for this study. Animal manipulations and stereotactic microinjection procedures were performed in accordance with the guidelines of the National Health and Medical Research Council of Australia for live animal use. The experiments were approved by Animal Ethics committees of Monash University and the Alfred Medical Research and Education Precinct (AMREP).
Preparation of cortical neurons from embryonic mice
Cortical neurons were prepared as described elsewhere with some modifications (20)
. In brief, cortices were removed and digested in HBSS containing 0.2 g/l trypsin (Sigma, St. Louis, MO, USA) and 80 U/ml DNase (Sigma) for 5 min at 37°C with agitation. The digested tissues were triturated through an 18-gauge blunt-ended needle to obtain a single cell suspension. The suspension was centrifuged and the pellet resuspended in Neurobasal media (NBM; Invitrogen, Carlsbad, CA, USA) containing 1 x B27 supplement (Invitrogen), 10% dialyzed fetal calf serum, 0.5 mM L-glutamine and 50U/ml penicillin/streptomycin). The cell suspension was seeded onto poly-D-lysine-coated, 24-well or 12-well plates (with or without glass coverslips) at a density of 125,000 cells/cm2. Cells were maintained in a humidified 37°C incubator containing 5% CO2 and 8% O2. Twenty-four hours after seeding, the serum-containing media was aspirated and replaced with NBM media supplemented with 1.25xB27, 0.5 mM L-glutamine and 50 U/ml penicillin/streptomycin (NBM+). On day 5, an equal volume of NBM+ media was added. All experiments were performed between days 12 and 13. Neuronal cells were characterized on day 12 by positive staining for NeuN. The presence of microglia, oligodendrocytes, and astrocytes was assessed by expression levels of CD11b, PLP-1, and GFAP, respectively. Microglia and oligodendrocytes are absent from our DIV12 cultures, but GFAP-positive astrocytes were present in our cultures. Although not directly quantitated, astrocyte contamination is estimated to be <20% in accordance with previous estimates using a similar culture technique (21)
.
Cell viability
Viability of the cortical neurons after treatment with NMDA in the presence or absence of OsM was assessed using the methyl-thiazole-tetrazolium (MTS) assay (Cell Titer 96®, Promega, Madison, WI, USA) according to the manufacturer’s instructions. Results were expressed relative to untreated cells after correction for values obtained from blank (cell-free) media. Percent viability was determined relative to untreated (control) cells and results presented are mean values ±SD of quadruplicate determinations (each well containing 250,000 cells) from three independent experiments.
Apoptosis
Apoptosis was assessed by analysis of activation of caspase -3 and -7 using the substrate DEVD-aminoluciferin from Caspase-GloTM 3/7 assay kit (Promega, Mannheim, Germany), according to the manufacturer’s instruction.
RT-polymerase chain reaction (RT-PCR)
Cortical neurons grown in 12-well plates were used for the analysis of mRNA expression patterns. During the treatment period, neurons were incubated under serum-free NBM conditions and stimulated for 24 h in the presence or absence of 50 ng/ml mouse OsM (R&D Systems, Minneapolis, MN, USA). Total RNA was isolated using 300 µl Trizol (Invitrogen) per well in accordance with the manufacturer’s instructions. For cDNA synthesis, 1 µg of RNA (in 10 µl of water) was combined with 20 pmol of oligo(dT)18 and heated to 65°C for 5 min, then cooled on ice. The following reaction components (1 x First Strand Synthesis buffer, dNTPs (50 µM each), DTT (5 mM), 25 U of SuperScript III polymerase (Invitrogen) were then added to a final volume of 20 µl and each sample was incubated for 1 h at 50°C. All polymerase chain reaction (PCR) reactions were performed in a 30 µl volume containing 1.5 µl of cDNA template and the appropriate forward and reverse primers (0.4 µM each, see Table 1
), 0.2 mM each of dNTPs, and 20 U/ml of recombinant Taq polymerase (Invitrogen). Each PCR reaction was preincubated at 94°C for 3 min, denaturated at 94°C for 5 s, followed by annealing/extension at temperatures indicated in Table 1
for 30 s. After cycling, the reactions went through a final postamplification step at 72°C for 3 min. PCR products were resolved on 2% agarose gels containing ethidium bromide and visualized under UV light. Images were subsequently captured using the CHEMIgenius Bioimaging System (Syngene, Frederick, MD, USA).
|
Primers, as shown in Table 1
, were designed using the Primer3 Software (http://frodo.wi.mit.edu/).
Western blot analysis
Western blot analysis was performed as described previously (22)
. Briefly, 5 x 105 primary murine cortical neurons were incubated in serum-free NBM in the presence or absence of 50 ng/ml mouse OsM for 24 h. The media were removed and cells were lysed with 200 µl of ice-cold radio-immunoprecipitation assay (RIPA) buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 1 mM PMSF, 2 mM imidazole, 1 mM NaF, 1 mM Na3VO4) and lysates were collected and quantified using the Bio-Rad Protein Assay (Bio-Rad, Hercules. CA, USA), snap-frozen, and stored at –80°C for later use. For analysis of the expression of NR2C in the brain after intrastriatal injection with vehicle or 5 µg/ml OsM, lysates were prepared from a 2 mm-thick coronal plane (dissected as a complete block 1 mm rostral and 1 mm caudal from the injection site). The ipsilateral (IL) and contralateral (CL) regions were then separated. The volume of lysis buffer used was adjusted to the respective wt (between 59 and 88 mg) of the sample.
Samples containing 30–40 µg protein were boiled in the presence of DTT, subjected to SDS-PAGE, and transferred onto PVDF membranes. The membranes were probed overnight with primary antibodies: goat anti-NR2C C-terminal antibody (Ab) (Santa Cruz Biotechnology, Inc., CA, USA) at a 1:100 dilution or mouse antitubulin Ab (Chemicon International, Temecula, CA, USA) at a 1:5000 dilution. Membranes were washed by standard techniques, then incubated for 1 h with the appropriate secondary Ab [sheep anti-mouse horseradish peroxidase (HRP)-conjugated Ab (Chemicon International, 1:5000) or rabbit anti-goat HRP-conjugated Ab (Sigma, 1:5000)]. Signals were revealed using Supersignal® West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, USA). Quantitative analysis of the bands was performed with Image Quant 5.2 Software.
Measurement of free intracellular calcium
Calcium current experiments were performed as described elsewhere with some modifications (23
, 24)
. Neurons were cultured on poly-D-lysine coated coverslips and incubated in phenol red-free NBM+ media containing 1 µM Oregon Green® 488 BAPTA-1 acetoxymethyl ester (AM) (Invitrogen), for 45 min at 37°C in a humidified incubator containing 5% CO2 and 20% O2. The fluorophore-containing media was replaced with fresh fluorophore-free NBM+ media and incubated for a further 45 min. The coverslips were then assembled into a closed perfusion chamber that was maintained at 37°C. A representative field of neurons was selected and Flow Buffer (phenol red-free HBSS containing 2 mM CaCl2 and 0.6 mM MgCl2) perfused over the cells at a rate of 0.5 ml/min. To assess the effect on the NMDA-induced rise in levels of free intracellular calcium [Ca2+]i, a perfusion protocol which involves three stimulations was employed. Oregon Green® 488 BAPTA-1 Fluorescence Units (FU) were monitored in 1.8 s intervals before, during and after each agonist exposure. The first and second stimulations involved two identical 45 s exposures to NMDA (50 µM), while the third stimulation using 75 mM KCl was used to verify neuronal responses. By including the KCl controls, we could confirm that all neurons selected for analysis were indeed intact for the entire procedure. OsM (100 ng/ml) or HBSS buffer was perfused for 5 min over the cells during an intermediate period between the two transient NMDA exposures. NMDA and KCl exposures are known to elicit a sharp rise in free intracellular calcium and only those cells that displayed a sharp and definitive rise in fluorescence from all three stimulations were analyzed (on average, 86% of cells in a given field responded to all three treatments). The images were analyzed off-line using Leica physiology software with regions of interest (ROI) (corresponding to the cell body and axon hillock) being selected. The FU for each ROI was adjusted to the median FU for each unstimulated baseline period, and the integral of the curve resulting from the second NMDA stimulation was expressed relative to the integral of the curve from the first NMDA stimulation. The relative second curve integral for each ROI was then averaged across all ROI to obtain an n = 1 value. Each data set is the average of three independent cultures. In total, 
60 neurons were analyzed per treatment group. Quantitative data is presented as "relative area under the second curve".
NMDA-Induced Excitotoxicity in vivo
This method was performed as described previously (20
, 24)
. Briefly, male C57BL/6 mice were anesthetized with Avertin (2,2,2 tribromoethanol in tertiary amyl alcohol) (Sigma) at a concentration of 0.5g/kg body wt. Anesthetized mice were placed in a stereotactic frame. The left striatum was injected with 1 µl of a 50 mM NMDA solution either alone or in combination with 5 µg/ml murine OsM. To control for the vehicle of OsM (PBS containing 0.025% BSA), a third group of animals was injected with NMDA in combination with 0.025% of BSA. The injection coordinates were as follows: bregma + 0.4 mm, mediolateral 2.0 mm, and dorsoventral 3.0 mm. The 1 µl solutions were delivered over a 4 min period at 0.2 µl/min, and the needle remained in place for an additional 2 min after injection to minimize reflux of fluid. After 24 h mice were transcardially perfused with ice-cold PBS followed by fixation with 4% paraformaldehyde. The brains were removed, postfixed at 4°C in 4% paraformaldehyde for 24 h, then incubated in 30% sucrose for another 24 h. Thereafter, brains were frozen on dry ice and kept at –80°C.
Quantification of NMDA-induced lesion area
Coronal sections of the brain were cut at 40 µm intervals throughout the entire lesion. Using the ballistic light approach to identify the lesion (25)
, sections were scanned and the lesion area in each section determined using Image J software (NIH, Bethesda, MD, USA; http://rsb.info.nih.gov/ij) and the volume (mm3) of the lesion then calculated using the formula: V = 
mm2 * n * 0.04 mm, where x represents the average of the measured area of the lesion per slide, and n represents the number of sections used.
Statistical analyses
Data were compared statistically by 2-sided Student’s t test and ANOVA, where appropriate. To adjust for multiple comparisons a post hoc analysis according to Bonferroni and Holmes was performed. Values of P < 0.05 were considered significant.
RESULTS
OsM attenuates NMDA-induced neuronal cell death in vitro
To determine whether treatment with OsM modulated the degree of NMDA-induced injury, pure neuronal cultures were treated with 50 µM NMDA in the presence or absence of murine OsM. As shown in Fig. 1
A, 24 h treatment with OsM alone had no effect on neuronal viability whereas treatment with NMDA caused 40% cell death. However, cotreatment with NMDA together with 50 ng/ml OsM reduced the degree of NMDA-induced cell death to 20% (i.e., a 50% relative reduction; n=3, P<0.01), indicating a neuroprotective capability for OsM. Treatment of neurons with 100 ng/ml OsM produced no further inhibition of NMDA toxicity (data not shown).
|
Since OsM conferred a protective effect when added simultaneously with NMDA, we determined whether pretreatment of cells with OsM prior to challenge with NMDA would confer more effective protection. To address this, cells were first treated with OsM (50 ng/ml) for 24 h. Culture media was then replaced with media containing fresh OsM (50 ng/ml) or NMDA (50 µM) (added alone or in combination) and the degree of cell death assessed 24 h later. Prolonged exposure to OsM (2x24 h) had no effect on cell morphology or number (data not shown). As shown in Fig. 1B
, pretreatment of cells with 50 ng/ml OsM essentially reversed the toxic effect of NMDA. Indeed the toxic effect of NMDA was reduced to less than 5% indicating a profound neuroprotective effect of this cytokine. To determine the dose dependency of the effect, the same experiment was performed using 10 ng/ml OsM. Pretreatment of neurons with this lower concentration of OsM was still substantially protective, attenuating the degree of NMDA-induced injury by 25% (P<0.01) (Fig. 1B
). Taken together, these results indicate that OsM dose-dependently inhibits NMDA-induced excitotoxic injury.
OsM attenuates NMDA-induced cleavage of caspase-3/-7
To explore a possible effect of OsM on the apoptotic component of NMDA-mediated excitotoxic injury, neurons were treated with OsM (50 ng/ml) for 1 h, followed by exposure to NMDA (50 µM) for 6 h. As shown in Fig. 2
, NMDA-induced cleavage of caspase-3/-7 by 25% compared with untreated controls. OsM alone conferred a slightly antiapoptotic effect. However, pretreatment of cells with OsM reversed the effect of NMDA on caspase-3/-7 activation. This data suggests that OsM counteracts NMDA-induced apoptosis.
|
The effect of OsM on excitotoxic injury induced by other glutamate analogues
In addition to NMDA, other glutamate analogues commonly used to explore excitotoxic responses are kainic acid (KA) and alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA). All three analogues bind to distinct glutamate receptor subtypes. We next determined whether the protective effect of OsM on NMDA-induced cell death was specific to NMDA, or whether OsM could protect cells from excitotoxic injury initiated by these other glutamate analogues. To this end, the effect of OsM was determined on neurons treated with KA or AMPA, both used at a concentration of 100 µM (26
, 27)
. As shown in Fig. 3
A, treatment of neurons with either KA or AMPA alone caused significant cell death after 24 h. However, akin to the effect of OsM seen on NMDA-induced injury, cotreatment with OsM also conferred a protective effect, albeit smaller (20 to 30%), against both KA- and AMPA-induced cell death. Since OsM was profoundly protective against NMDA toxicity when cells were pretreated with this cytokine, neurons were pretreated with 100 ng/ml OsM for 24 h, then treated with OsM again in the presence or absence of either AMPA or KA (as above). Cell viability was assessed 24 h later. In contrast to the results observed with NMDA, OsM pretreatment produced no additional protective effect against either AMPA or KA induced cell death (Fig. 3B
). Since AMPA and KA-mediated excitotoxic injury are both known to produce secondary NMDA-dependent effects (28
, 29)
it is possible that the protective effect of OsM against AMPA and KA toxicity is via inhibition of this secondary NMDA signal. Based on the comparison between the three glutamate analogues, these data indicate that the protective effect of OsM on excitotoxic cell death is more specific to NMDA-induced injury.
|
OsM selectively down-regulates NMDA receptor subunit gene expression
The NMDA receptor is a multisubunit complex that regulates ion fluxes and transmits intracellular signals after engagement with NMDA or glutamate. Changes in NMDA receptor subunit composition have also been shown to influence the response to ischemic and excitotoxic injury (30
31
32)
. To investigate whether OsM affects the expression of NMDA receptor subunits, primary cortical neurons were incubated in presence or absence of OsM for 24h. RNA was collected and changes in mRNA expression levels for NR1, NR2A, NR2B, and NR2C determined by RT-PCR. As shown in Fig. 4
A, mRNA levels for NR1, NR2A, and NR2B were unchanged after OsM treatment. However, expression of NR2C receptor subunit was reduced by OsM. This suppressive effect of OsM on NR2C expression was also evident at the protein level as assessed by Western blot analysis (Fig. 4A
, insert). To assess whether this effect of OsM seen in vitro is also operative in vivo, animals were injected with OsM or vehicle. Changes in NR2C protein level was determined in brain lysates by Western blotting. As can be seen from Fig. 4B
, OsM reduces expression of NR2C in vivo thereby supporting the in vitro data.
|
OsM inhibits NMDA-induced increase in free intracellular calcium levels
Our previous experiments were designed to determine the effects of OsM after chronic exposure to NMDA. It is well known that NMDA-induced cell death is preceded by changes at the level of free intracellular calcium ([Ca2+]i) that can be observed on NMDA treatment. To determine whether OsM could influence the acute effect of NMDA at this level, we used real-time confocal imaging to visualize changes in [Ca2+]i in individual neurons on NMDA treatment. Using this approach, treatment of primary neurons with OsM for 5 min prior to NMDA addition significantly attenuated NMDA-induced increase in [Ca2+]i (Fig. 5
; Table 2
). Taken together with the data presented in Figs. 1
, 4
, these findings suggest that OsM has both short-term (i.e., calcium transients) and long-term (i.e., modulation of gene expression) protective effects against NMDA-induced neuronal stimulation.
|
|
OsM is neuroprotective in vivo
An important question was to determine whether the neuroprotective effect of OsM on neurotoxic injury seen in vitro was also operative in vivo. To address this, the striatal region of mice was stereotaxically injected with NMDA (1 µl of 50 mM NMDA) in the presence or absence of OsM (5 µg/ml) or vehicle (PBS+0.025% BSA) and the lesion volume determined 24 h later. As shown in Fig. 6
, codelivery of OsM reduced the NMDA-induced lesion volume by 40% (n=10 for NMDA group; n=4 for NMDA+vehicle group; n=8 for NMDA+OsM group). The difference between NMDA+vehicle and NMDA+OsM was statistically significant; P<0.05). Hence, OsM confers neuroprotection against NMDA-induced toxicity in vivo.
|
DISCUSSION
OsM is well known for its pro- and anti-inflammatory capabilities, its effect on hematopoiesis, and as a modulator of the proliferative process (3
, 4)
. Although OsM is expressed by microglia, astrocytes, and neurons, and expression levels are elevated during some acute and chronic neuronal injuries, detailed studies addressing the role of OsM in the CNS are scant (1
, 33)
. This cytokine has also been described to regulate downstream molecules relevant in inflammatory brain diseases (e.g., IL-6 and monocyte chemoattractant protein-1) (33)
. A closer investigation into the role of OsM in the brain is therefore warranted. In this study we have used in vitro and in vivo models of excitotoxic injury and identified a novel and potentially important neuroprotective role for OsM. Results of our in vitro studies revealed that OsM reduced NMDA induced cell death by at least 50% when added simultaneously with NMDA. However OsM essentially reversed NMDA-induced toxicity if cells were pretreated with this cytokine for 24 h prior to NMDA challenge. With the exception of specific NMDA receptor antagonists (e.g., MK801), the complete reversal of NMDA-induced excitotoxic injury is a rare occurrence and, to the best of our knowledge, unprecedented for any other cytokine. In our study, only a single time period for pretreatment was used, and the minimum OsM pretreatment period required for full protection against NMDA injury remains to be determined. Although our primary cultures are predominantly neuronal, astrocytes are also present. However, since astrocytes are generally resistant to excitotoxic insults and NMDA toxicity is selective to neurons, we are confident that our viability assays are indeed reflecting neuronal cells (34)
. However, it remains an open question as to whether astrocytes contribute to the neuroprotective effect of OsM.
The NMDA receptor belongs to a group of ionotropic glutamate receptors that also includes receptors for kainic acid and AMPA (14)
. Our study revealed that the protective effect of OsM was more selective to NMDA-induced toxicity. OsM was found to confer some protection against AMPA- and KA-induced toxicity, but pretreatment of cells with OsM produced no additional protection against KA- or AMPA-induced cell death. AMPA- and KA-induced excitotoxic injury are known to produce secondary damaging effects via NMDA receptor activation (28
, 29)
, and it is plausible that the protective effect of OsM seen against AMPA and KA toxicity was by blocking this secondary NMDA-dependent effect. If this was indeed the case, it would be expected that OsM would display no protection against either AMPA or KA in the presence of the specific NMDA receptor antagonist MK801, a result that has already been shown for IL-6 (13)
.
Although we cannot exclude the possibility that OsM directly antagonizes the NMDA receptor on the cell surface, a more likely scenario is that OsM was influencing expression of protective genes or regulating NMDA receptor expression. For the latter, our RT-PCR and Western blot experiments revealed that OsM down-regulated expression of the NR2C receptor subunit while having no effect on mRNA expression levels of NR1, NR2A, or NR2B. Down-regulation of both NR2A and NR2C subunit expression by cytokines (e.g., FGF) and by brain-derived neurotrophic factor (BDNF) has been reported (32)
, but this is the first report showing that OsM selectively down-regulates expression of the NR2C gene. Although our study does not allow us to formally conclude that the decrease in NR2C is part of the mechanism underlying the protective effect of OsM, this is a reasonable possibility. Changes in NMDA receptor expression can influence the degree of neuronal damage after stroke and traumatic brain injury. Indeed, targeted disruption of the gene for the NR2C subunit confers protection from ischemic damage subsequent to permanent middle cerebral artery occlusion (30)
. Also, down-regulation of NR2A and NR2C expression by FGF and BDNF has been proposed as a mechanism underlying the protective effect of these factors on cerebellar granule cells (32)
. It is also possible that OsM may, in part, be having its protective effects by the modulation of these downstream effectors. It would be interesting to determine whether the complete neuroprotective effects produced by OsM when cells are pretreated with this cytokine would be observed using primary neurons from NR2C-deficient mice.
OsM has recently been shown to inhibit NMDA-mediated increase in GnRH from cultured neurons without having any effect itself on GnRH expression (35)
. In this report, the authors excluded a direct effect of OsM on the NMDA receptor and instead implicated a causal role for the OsM initiated MAPK Erk1/2 intracellular signaling pathway, although the molecular processes underlying this effect are still largely unknown.
OsM is a pleiotropic cytokine that has been shown to influence the expression pattern of many genes via the janus-activated kinase (JAK)-STAT or MAPK pathways (2)
. Although we did not directly assess the mechanism by which OsM exerts its neuroprotective effects, it is likely that engagement of ERK1/2 and JAK-STAT pathways are involved. It would be interesting to determine whether inhibition of ERK1/2 phosphorylation would rescue cells from the neuroprotective effect of OsM. Apart from the possible effect on NR2C, other OsM targets in neurons that are engaged to protect these cells from NMDA toxicity remain to be identified. However, we have performed RT-PCR experiments on primary neurons treated for 24 h with OsM and observed marked increase in mRNA expression levels of TIMP-1, neuroserpin, and the OsM receptor (OsM-Rß) (data not shown). TIMP-1 has been described to be induced by OsM in other cell systems and also to produce a neuroprotective effect by influencing the activity of matrix metalloproteases (MMP-2 and MMP-9) (4
, 6
, 36)
. Neuroserpin, on the other hand, inhibits the activity of tissue-type plasminogen activator (t-PA), a serine protease that is a major modulator of the excitotoxic cascade (37)
. Other studies have shown that OsM can increase expression of the other major t-PA inhibitor, plasminogen activator inhibitor type 1 (PAI-1), in astrocytes (38)
. Hence, OsM could indirectly attenuate NMDA toxicity by neuroserpin or PAI-1-mediated inhibition of t-PA. Our observation that OsM could increase the expression of its own receptor could, in principle, facilitate an amplication loop for OsM stimulation. A role for one or more of these OsM regulated genes in the neuroprotective arm of OsM is plausible, but direct experiments to address this are yet to be performed.
NMDA toxicity is invariably coupled with acute increases in the level of free intracellular calcium (14)
. Using a real-time approach to visualize changes in free intracellular calcium in individual neurons after NMDA treatment, OsM was shown to reduce the NMDA-mediated rise in levels of free intracellular calcium by at least 30%. The fact that OsM was capable of exerting an acute inhibitory effect on NMDA-induced changes in calcium currents suggests that this cytokine can directly influence signaling downstream of the NMDA receptor, in addition to its longer term effects on altering patterns of gene expression. One could speculate that the OsM receptor (gp130/OsMRß) is coupled to the NMDA receptor and/or that the JAK-STAT/MAPK pathways initiated by OsMRß engagement have a profound and rapid negative effect on the downstream effector arm of NMDA signaling (2)
. With this in mind, it could be envisaged that the OsMRß and the NMDA receptor are colocalized on neurons.
Neuroprotective effects of molecules observed in vitro are not always reciprocated when applied to in vivo models. If OsM was to exert neuroprotective effects in vivo, it would also have to overcome the additional effects of activated microglia or other infiltrating inflammatory cells that are recruited after injury. Therefore, we considered it important to determine whether OsM could confer neuroprotective capabilities in vivo. To this end, we used the intrastriatal NMDA lesion model of excitotoxic injury (20
, 24)
. In our hands, codelivery of OsM reduced the NMDA-induced lesion volume by 40%. This finding substantiates our in vitro results and further strengthens the basis for our conclusion that OsM is genuinely a neuroprotective cytokine. Since pretreatment with OsM prior to NMDA challenge completely reversed NMDA toxicity in vitro, it would be interesting to determine whether predelivery of OsM into the mouse striatum would produce more substantial neuroprotection.
There has been much interest in the possible neuroprotective effects of gp130 ligands, although this is an area with considerable controversy. Most studies have addressed the role of IL-6 in models of excitotoxic injury, but these in vitro investigations have yielded conflicting results (16
17
18
19)
possibly as a consequence of different neuronal cell types used in these studies.
In conclusion, we report a profound and unprecedented neuroprotective role for OsM. There has been a long-standing desire to protect neurons against glutamate-mediated excitotoxicity, as glutamate is generally considered to play an important role in neuronal cell death in a variety of acute and chronic neurodegenerative diseases (14
, 39
40
41)
. Our observation that OsM is neuroprotective in vivo raises the possibility of therapeutic options for this cytokine and provides impetus to explore in further detail the effects of other members of the gp130 ligand family in the CNS.
ACKNOWLEDGMENTS
T.W.W. was a recipient of an Erwin Schroedinger postdoctoral scholarship of the Austrian Science Fund. This study was also funded with grants obtained by R.L.M. from the National Health and Medical Research Council of Australia.
Received for publication January 26, 2006. Accepted for publication June 12, 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
