OUP user menu

Effects of the Adenosine A2A Receptor Antagonist SCH 58621 on Cyclooxygenase-2 Expression, Glial Activation, and Brain-Derived Neurotrophic Factor Availability in a Rat Model of Striatal Neurodegeneration

Luisa Minghetti PhD, Anita Greco DSc, Rosa Luisa Potenza PhD, Antonella Pezzola PhD, David Blum PhD, Kadiombo Bantubungi PhD, Patrizia Popoli MD
DOI: http://dx.doi.org/10.1097/nen.0b013e3180517477 363-371 First published online: 1 May 2007

Abstract

Inhibition of adenosine A2A receptors (A2ARs) is neuroprotective in several experimental models of striatal diseases. However, the mechanisms elicited by A2AR blockade are only partially known, and critical aspects about the potential beneficial effects of A2AR antagonism in models of neurodegeneration still await elucidation. In the present study, we analyzed the influence of the selective A2AR antagonist SCH 58261 in a rat model of striatal excitotoxicity obtained by unilateral intrastriatal injection of quinolinic acid (QA). We found that SCH 58261 differently affected the expression of cyclooxygenase-2 (COX-2) induced by QA in cortex and striatum. The antagonist enhanced COX-2 expression in cortical neurons and prevented it in striatal microglia-like cells. Similarly, SCH 58261 differently regulated astrogliosis and microglial activation in the 2 brain regions. In addition, the A2AR antagonist prevented the QA-induced increase in striatal brain-derived neurotrophic factor levels. Because COX-2 activity has been linked to excitotoxic processes and because brain-derived neurotrophic factor depletion has been observed in mouse models as well as in patients with Huntington disease, we suggest that the final outcome of A2AR blockade (namely neuroprotection vs neurodegeneration) is likely to depend on the balance among its various and region-specific effects.

Key Words
  • Adenosine A2A receptor
  • Brain-derived neurotrophic factor
  • Cyclooxygenase-2
  • F2-isoprostanes
  • Neurodegeneration
  • Neuroprotection
  • Prostaglandin E2
  • Striatal degeneration.

Introduction

Adenosine 2A receptors (A2ARs) belong to a family of at least 4 types of G protein-coupled receptors (A1, A2A, A2B, and A3) (1) that mediate the multiple functions of the purine nucleoside adenosine. In the central nervous system, A2A receptors are localized mainly in the striatum, although lower levels of expression are detected in the cortex and in the hippocampus (2, 3). Besides their role in modulating dopamine-dependent activities in both physiologic and pathologic conditions (4), A2ARs are implicated in neuronal cell death associated with excitotoxicity (see Reference 5 for review). Given their preferential striatal localization, A2ARs are regarded as suitable targets for the development of neuroprotective strategies for treating disorders that are characterized by basal ganglia dysfunctions, such as Parkinson disease and Huntington disease (3, 6, 7). Indeed, in several experimental settings, A2AR blockade, either by administration of selective antagonists or genetic ablation, is neuroprotective and/or exerts antiparkinsonian activities (8-14).

Despite accumulating evidence, the neuroprotective mechanisms elicited by A2AR blockade are only partially known, and critical aspects about the potentially beneficial as well as detrimental effects of A2AR antagonism in models of neurodegeneration still await clarification (5). We have previously reported that the selective A2A antagonist SCH 58261 [5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-e]-1,2,4-triazolo[1.5-c]pyrimidine] protects from quinolinic acid (QA)-induced striatal degeneration (12), a phenotypic model of Huntington disease (15, 16). This protective effect was suggested to arise from a presynaptic action of the antagonist on the glutamatergic corticostriatal fibers resulting in the prevention of the QA-induced increase in striatal glutamate outflow. However, we also showed that SCH 58261 amplified the QA-induced intracellular calcium levels in striatal neurons (an effect potentially detrimental) at the postsynaptic level (12). An opposite outcome resulting from pre- versus postsynaptic inhibition of striatal A2ARs was also demonstrated in the 3-nitropropionic acid model of striatal degeneration (9).

Besides the aforementioned complexity, the relative role of neuronal versus extraneuronal A2ARs in modulating striatal damage has been recently brought into question. In addition to neurons, A2ARs are expressed by microglial (17, 18) and astroglial cells (19). Blockade of A2ARs at such extraneuronal sites regulates potential neurotoxic activities associated with astrocyte and/or microglia activation, including release of glutamate and nitric oxide (5, 18, 20). In microglial cells, A2AR activation induces the expression of cyclooxygenase-2 (COX-2), the inducible isoform of the enzyme catalyzing the first committed step in prostaglandin (PG) synthesis. An early and transient increase in striatal levels of PGE2, one of the major PGs associated with COX-2, has been observed after striatal injection of QA (5), and COX-2 activity has been implicated in QA-induced neurotoxicity (21).

Finally, A2ARs could positively regulate the availability of brain-derived neurotrophic factor (BDNF) (22), a member of the neurotrophin family, which plays a significant role in the maintenance of function and survival of neurons and whose levels have been reported to change in QA-lesioned striata (23).

To go further in the comprehension of the neural mechanisms involving A2ARs, we have evaluated the influence of SCH 58261 on the following QA-induced effects: 1) COX-2 expression and activity; 2) free radical generation; 3) glial activation; and 4) changes in BDNF levels. We found that A2AR blockade exerted time- and region-dependent effects, suggesting that the beneficial or detrimental influence of A2AR blockade may partly depend on the balance of such varied and region-specific activities.

Materials and Methods

Animals and Surgery

Adult male Wistar rats (250-280 g) were used. Animals were kept under standardized temperature, humidity, and lighting conditions, with free access to water and food. Animal care and use followed the directives of the Council of the European Communities (86/609/EEC). Animals were anesthetized with Equithesin (3 ml/kg intraperitoneally). QA (180 nmol/1 μL) was unilaterally injected in the striatum (coordinates: anterior = +1.7; lateral = +2.7; ventral = −4.8 mm from bregma and dura), and vehicle (1 μL of PBS) was injected in the other side. SCH 58261 was administered at the dose of 0.01 mg/kg intraperitoneally, 20 minutes before the QA lesion. This dose of SCH 58261 has been consistently reported as neuroprotective in rats against QA-induced toxicity (12) and brain ischemia (24, 25).

Immunocytochemistry

Groups of at least 3 QA-lesioned rats each were killed at 1 or 14 days after the lesion. Animals were decapitated under ether anesthesia, and brains were removed and immediately frozen. For each brain, serial 20-μm coronal sections were cut on a cryostat microtome, mounted on slides, fixed in 4% paraformaldehyde in PBS containing 0.12 mol/L sucrose for 15 minutes at room temperature and washed in PBS.

COX-2 immunostaining was performed using a goat polyclonal anti-COX-2 antibody (M19; Santa Cruz Biotechnology, Santa Cruz, CA), as described previously (26). Briefly, after pretreatment with 0.3% hydrogen peroxide in absolute methanol, sections were blocked with 10% preimmune serum for 2 hours at room temperature and then incubated with the primary antibody (1:1000) overnight at room temperature.

Other primary antibodies used were the ED-1 monoclonal antibody (Serotec, Kidlington, UK), that recognizes a lysosomal glycoprotein shown to correlate well with immune cell activation and widely used to detect activated microglia (26, 27) and the rabbit polyclonal antibodies against the glial fibrillary acid protein (GFAP) (Dako, Cambridge, UK) as a marker for astrocytes.

Binding of the primary antibodies was detected using biotinylated secondary antibodies and the avidin-biotin-peroxidase method (ABC Elite; Vector Laboratories, Peterborough, UK) with diaminobenzidine as the substrate. All sections were counterstained with hematoxylin.

Brain Dissection and Prostaglandin E2, 15-F2t-Isoprostane, and Brain-Derived Neurotrophic Factor Extractions

After decapitation, cortices and striata were immediately dissected out from both hemispheres, placed in plastic tubes, weighted, frozen on dry ice, and stored at −80°C until metabolite extraction. A detailed procedure for PGE2 and F2-isoprostane (15-F2t-IsoP) extraction has been described elsewhere (28). In brief, 200 μL of ice-cold Tris-HCl buffer, pH 7.5, containing 10 μg/mL of the COX inhibitor indomethacin (stock solution 100× in ethanol) to avoid ex vivo PGE2 synthesis and 10 μmol/L radical scavenger butylated hydroxytoluene (stock solution 100× in ethanol) to avoid auto-oxidation, were added to each frozen sample, which was quickly thawed, homogenized with a Teflon pestle (Sigma-Aldrich, Milan, Italy) (20 cycles in an ice bath), vigorously vortexed, and centrifuged at 14,000 rpm for 45 minutes at 4°C. Supernatants were collected and stored at −80°C until analysis. For BDNF extraction, tissue samples were rapidly thawed in ice-cold PBS containing a protease inhibitor cocktail (1:10 of a stock solution in H2O, prepared following the manufacturer's instructions; Sigma-Aldrich,), and processed as before.

Prostaglandin E2 and 15-F2t-Isoprostane Measurement

PGE2 and 15-F2t-IsoP were measured in tissue extracts by high-sensitivity colorimetric enzyme immunoassays (EIA kits, detection limit for PGE2 7.8 pg/mL; Assay Designs, Inc., Ann Arbor, MI; detection limit for 15-F2t-IsoP 2 pg/mL; Cayman Chemical, Ann Arbor, MI). According to the manufacturers, the cross-reactivity of the anti-PGE2 antibody with 8-15-F2t-IsoP was <0.25% and that of anti-15-F2t-IsoP antibody for other prostaglandins was <1% (0.02% for PGE2). All measurements were run at least in duplicate for each sample. Results are expressed as pg/mg of wet tissue.

Brain-Derived Neurotrophic Factor Measurement

The amounts of free mature BDNF in tissue extracts were measured (in duplicate or triplicate) by a high-sensitivity colorimetric enzyme immunoassay (EIA kit,detection limit 7.8 pg/mL; Promega, Madison, WI), following the manufacturer's instructions. According to the manufacturer's instructions, cross-reactivity of the anti-BDNF antibody with other related neurotrophic factors (NGF, NT-3, and NT-4) is <3%. Results are expressed as pg/mg of wet tissue.

Statistical Analysis

Results are expressed as means ± SEM. Comparisons among groups were made using Student t-test. Correlation between experimental variables was measured by Spearman rank correlation (rs). Analyses were performed using Stata 8.1 software (Stata Corp., College Station, TX).

Results

Effect of Quinolinic Acid Lesion and SCH 58261 on Cyclooxygenase-2 Expression, Prostaglandin Synthesis, and Free Radical Formation

Unilateral injection of QA (180 nmol, 1 μL) in the dorsal striatum of adult rats causes excitotoxic striatal degeneration and glial activation as previously characterized by Bantubungi et al (29).

First, we addressed the influence of the A2AR antagonist on the expression of COX-2 enzyme, the synthesis of its major metabolite PGE2, and the production of 15-F2t-IsoP (also known as 8-epi-PGF), a product of lipid peroxidation used as marker of free radical formation (30).

One day after QA injection, there was a clear induction of COX-2 expression in a few cells at the core of the lesion in the ipsilateral side (Fig. 1A), corresponding to a subset of ED-1-positive cells in adjacent sections (Fig. 2A) and thus identified as microglia/macrophage-like cells. COX-2 immunoreactivity was substantially decreased by SCH 58261 treatment. Interestingly, although QA induces a primary excitotoxicity within the ipsilateral striatum (29), we found that in the ipsilateral cortex, QA injection induced a strong COX-2 expression in pyramidal neurons (Fig. 1C). Pretreatment with SCH 58261 did not prevent the induction of COX-2 in pyramidal neurons but, conversely, enhanced it. QA-induced striatal and cortical COX-2 expression was transient because 14 days after QA lesion the levels of COX-2 expression in the ipsilateral striatum (Fig. 1B) and cortex (Fig. 1D) were back to the levels detected in the contralateral vehicle-injected side. No differences were observed in animals receiving SCH 58261 pretreatment.

FIGURE 1.

Influence of peripheral SCH 58261 on quinolinic acid (QA)-induced cyclooxygenase-2 (COX-2) expression in striatum and cortex. (A, B) COX-2 immunoreactivity in striatum at 1 day and 14 days after unilateral striatal injection of QA, respectively. (C, D) COX-2 immunoreactivity in the cortex at 1 day and 14 days after unilateral striatal injection of QA, respectively.

FIGURE 2.

Influence of peripheral SCH 58261 on quinolinic acid (QA)-induced microglial reactivity. Microglial reactivity was studied using ED-1 monoclonal antibodies. (A, B) ED-1 immunostaining in striatum at 1 day and 14 days after unilateral striatal injection of QA, respectively. (A) Inset: cc, corpus callosum; st, striatum. (C, D) ED-1 immunostaining in the cortex at 1 day and 14 days after unilateral striatal injection of QA, respectively.

As an index of COX-2 activity, we measured the production of one of its major metabolites, PGE2 (Table). One day after QA injection, the levels of PGE2 in striatal homogenates were found to be increased in the ipsilateral side compared with the contralateral vehicle-injected side (134.6 ± 12.8%; p < 0.05, n = 7). Such an increase was abolished in SCH 58261-treated animals, as PGE2 levels were comparable in contralateral and ipsilateral sides. In agreement with the strong COX-2 expression observed in cortical neurons, QA injection induced a remarkable increase in PGE2 cortical levels in the ipsilateral side (Table). At odds with COX-2 expression, the cortical levels of PGE2 were not further increased by SCH 58261 pretreatment. This apparent discrepancy could be explained by rapid consumption of the COX-2 substrate arachidonic acid, whose availability might become a limiting factor in PGE2 synthesis regardless of the overexpression of COX-2.

View this table:
TABLE.

QA injection as well as SCH 58261 pretreatment did not significantly affect radical formation in the striatum, as indicated by the striatal levels of 15-F2t-IsoP (Table). However, 15-F2t-IsoP levels were strongly increased in the cortex, reaching an 8-fold increase compared with the contralateral side. In animals receiving SCH 58261 pretreatment, the levels of 15-F2t-IsoP induced by QA injection were still elevated, showing a 5-fold increase over the contralateral side, but, as for PGE2 levels, they tended to decrease versus QA alone. Such a decrease, however, was not statistically significant. Because of the normalization of COX-2 expression at 14 days after the lesion, the determination of PGE2 and 15-F2t-IsoP levels was not performed at that time point.

Effect of Quinolinic Acid Lesion and SCH 58261 on Glial Activation

To investigate the effect of QA striatal injection on glial activation, we immunostained adjacent sections using ED-1 monoclonal antibodies, which specifically recognize macrophages and activated microglia, and polyclonal antibodies against GFAP, a marker for astrocytes, at 1 and 14 days after lesion.

At 1 day after lesion, ED-1-positive cells were detected in the striatum, broadly distributed around the injection site (Fig. 2A) and extending up to the cortex of the injected side (Fig. 2B). The SCH 58261 pretreatment moderately reduced the number of ED-1 immunoreactive cells induced by QA injection in the striatum and cortex (Fig. 2A, C). Interestingly, in the contralateral (vehicle-injected) side of SCH 58261 pretreated animals, there was an increase in ED-1-positive cells in the striatum and, more prominently, in the corpus callosum with respect to QA alone (Fig. 2A, inset).

At 14 days after lesion, ED-1 immunoreactivity was still very strong, particularly in the ipsilateral striatum and cortex, but not in the contralateral vehicle-injected side (Fig. 2B, D, upper panels). In SCH 58261 pretreated rats, no remarkable differences were observed in the lesioned side. However, a significant increase in the number of ED-1-positive cells was detected in the contralateral cortex (Fig. 2C, D, lower panels).

Intrastriatal QA injection induced widespread GFAP immunoreactivity at 1 day after lesion, which extended to the contralateral side. The degree of GFAP immunoreactivity was not influenced by SCH 58261 (not shown). At 14 days after lesion (Fig. 3), GFAP immunoreactivity was still evident, consistently with the observations previously reported in a bilateral model of QA toxicity (12). QA-stimulated GFAP immunoreactivity was decreased in rats that were pretreated with SCH 58261, particularly in the striatum (Fig. 3A).

FIGURE 3.

Influence of peripheral SCH 58261 on quinolinic acid (QA)-induced astrocytosis. Astrogliosis was detected 14 days after unilateral striatal injection of QA using polyclonal antibodies against glial fibrillary acidic protein (GFAP), a marker for astrocytes. (A) GFAP immunoreactivity in striatum. (B) GFAP immunoreactivity in cortex.

Effect of Quinolinic Acid Lesion and SCH 58261 on Striatal Brain-Derived Neurotrophic Factor Levels

BDNF plays a significant role in the maintenance of function and survival of neurons. The decreased levels of BDNF found in the striatum of A2AR knockout versus wild-type mice (23) suggest that adenosine A2ARs may positively regulate BDNF availability. Expression of COX-2 and BDNF mRNAs after systemic administration of kainic acid shows a very similar pattern (31) and a link between COX-2 activity and BDNF is suggested by recent studies showing that COX inhibitors blocked the increase of both PGE2 and BDNF after spatial learning and long-term potentiation (32).

Based on the above observations, we sought to investigate whether SCH 58261 influenced BDNF levels and whether a correlation existed between PGE2 and BDNF levels in QA-injected brains. We focused our analysis at 1 day after lesion. As shown in Figure 4A, the striatal levels of BDNF were significantly higher in the QA-injected side than in the contralateral vehicle-injected side. In addition, within the same hemisphere, striatal BDNF levels and cortical PGE2 levels (Fig. 4B) were correlated (Fig. 4C).

FIGURE 4.

Influence of intrastriatal quinolinic acid (QA) injection on striatal brain-derived neurotrophic factor (BDNF) levels and cortical prostaglandin (PG) E2 levels. One day after unilateral striatal QA injection, a significant increase in BDNF (A) was found in striatal homogenates compared with the contralateral vehicle-injected side. A significant increase in PGE2 was found in the corresponding cortical homogenates (B). Bars represent mean ± SEM values; *, p < 0.05 versus the contralateral side. Striatal BDNF levels and cortical PGE2 levels within the same hemisphere were correlated (C) (Spearman rank correlation, rs = 0.640, n = 10.

When rats were pretreated with SCH 58261, the levels of BDNF in striatal homogenates were decreased in both QA- and vehicle-injected sides compared with rats that did not receive the A2A antagonist. As shown in Figure 5, BDNF levels were around or below the limits of the assay detection in SCH 58261 pretreated rats.

FIGURE 5.

Influence of peripheral SCH 58261 on quinolinic acid (QA)-induced striatal brain-derived neutotrophic factor (BDNF) levels. One day after unilateral striatal injection of QA, BDNF levels were significantly increased in striatal homogenates compared with the contralateral vehicle-injected side. SCH 58261 pretreatment (SCH 0.01) reduced levels of BDNF on both ipsilateral and contralateral sides around or below the assay detection limits (dotted line). Circles represent individual values and lines represent the mean values. Data are expressed as means ± SEM, n = 5. *, p < 0.05 versus the contralateral side.

Discussion

In this study we show that the effects elicited by the A2AR antagonist SCH 58261 on COX-2 expression, PGE2 synthesis, lipid peroxidation, and glial activation in a model of QA-induced neurotoxicity are dependent on the specific brain region analyzed.

In the brain, COX-2 overexpression and activity are often associated with excitotoxic and inflammatory neurodegenerative processes in several acute and chronic diseases (33). Besides neurons, other brain cells, including endothelial cells, astrocytes, and microglia can express COX-2 under pathologic conditions. Despite the neuroprotective action of COX-2 inhibitors under varied experimental conditions, the mechanisms mediating COX-2-dependent neurotoxicity are only partially known. PGE2, one of the major PGs associated with COX-2 activity, has neuroprotective as well as neurotoxic activities, depending on the experimental model and the involvement of specific PGE2 receptor subtypes (34). Besides PGs, a bystander generation of free radicals during COX-2 enzymatic activity could lead to neuronal death. It has been recently demonstrated that COX-2 is not a major source of superoxide radicals (35, 36), but rather it generates carbon-centred radicals, capable of causing lipid peroxidation (37). In line with these findings, we have reported a transient COX-2-dependent release of PGE2 and 15-F2t-IsoP, after in vivo acute activation of NMDA glutamate receptors in the hippocampus of freely moving rats (38). A rapid and transient increase of both PGE2 and 15-F2t-IsoP was also reported in a model of bilateral intrastriatal injection of QA and was prevented by a neuroprotective concentration of SCH 58261 (5). In the present study, we report that striatal PGE2 but not 15-F2t-IsoP was elevated 1 day after QA injection compared with the contralateral side. At this time point, COX-2 immunoreactivity was also increased and localized mainly to a subset of ED-1-immunopositive cells at the core of the lesion. In the cortex, both PGE2 and 15-F2t-IsoP levels were remarkably increased and a strong COX-2 immunoreactivity was associated with cortical neurons. The expression of COX-2 in cortical glutamatergic neurons is dependent on normal synaptic activity and is rapidly upregulated after seizures, ischemia, or activation of N-methyl-d-aspartate (NMDA) or α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (39-41). The induction of COX-2 expression in cortical pyramidal neurons and the consequent synthesis of PGE2 can be due to the ability of QA to stimulate NMDA receptors and to increase the extracellular levels of glutamate (42). The formation of 15-F2t-IsoP could be related to the QA-induced COX-2 activity, as described after in vivo acute activation of hippocampal NMDA glutamate receptors (38) or to other enzymatic pathways evoked by glutamate receptor activation, such as nitric oxide synthase, protein kinase C, or NADPH oxidase (43). In the striatum, the increase in PGE2 levels in the absence of 15-F2t-IsoP accumulation may be related to the expression of COX-2 in activated (ED-1-positive) microglial cells, which are richer than neurons in endogenous antioxidant defenses that protect cells from oxidant injury, thus limiting or preventing the formation of 15-F2t-IsoP. Unexpectedly, the administration of the selective A2AR antagonist SCH 58261 before intrastriatal QA injection led to the enhancement of neuronal COX-2 expression in the cortex and, at the same time, to the abrogation of microglial COX-2 expression at the core of the striatal lesion. Experimental evidence indicates that A2AR blockade can potentiate NMDA-induced neurotoxicity (5, 44). Thus, SCH 58261 may enhance the QA-induced COX-2 expression in cortical neurons by increasing the state of activation of NMDA receptors on these cells. On the other hand, in a rat focal ischemia model, SCH 58261 has been reported to reduce p38 mitogen-activated protein kinase (MAPK) phosphorylation, hence its activation in microglia (45). Because p38 MAPK controls crucial steps associated with microglia activation including COX-2 induction (46), its inhibition by SCH 58261 could explain the abrogation of COX-2 immunoreactivity in striatal microglia. This hypothesis is also in line with the observation that adenosine A2A receptor stimulation upregulates COX-2 expression in rat microglial cultures (17).

Cortical neurons play a crucial role in the QA-mediated neurotoxicity as striatal QA injection is no longer toxic after the removal of the corticostriatal glutamatergic projection (47). Pretreatment with selective COX-2 inhibitors has been reported to preserve motor performance and to reduce volume lesion induced by QA intrastriatal injection (21), suggesting that the overexpression of COX-2 in cortical neurons could contribute to neurotoxicity. Our data also suggest that A2AR blockade, by influencing in opposite ways QA-induced COX-2 overexpression in striatal and cortical neurons, may exert region-specific effects on neurodegenerative processes.

BDNF is synthesized in the pyramidal neurons of cerebral cortex and released in the striatum by corticostriatal anterograde transport and is an important trophic and survival factor for striatal neurons (48). A reduced availability of BDNF appears to be a crucial event in the pathogenesis of striatal degenerative diseases such as Huntington disease as suggested by the decreased huntingtin-mediated BDNF gene transcription observed in patients with Huntington disease and in animal models (49). Huntingtin and BDNF immunoreactivities were found colocalized in cortical neurons in normal rat brain and were reduced 6 weeks after QA lesion (23). In the present study, the striatal levels of BDNF were increased 1 day after QA, in agreement with the augmented striatal levels of BDNF mRNA observed within 24 hours in a similar model of striatal excitotoxicity (50). On the other hand, this finding is not necessarily at odds with the reduction in cortical BDNF immunoreactivity found at later time points by Fusco et al (23), because it may represent an early response to the toxic insult. The expression of BDNF gene is controlled by the second messenger cyclic AMP, levels of which can be elevated by PGE2 through its EP2 receptor subtype, expressed by cortical and hippocampal neurons, but also by microglia and astrocytes (51, 52). Because striatal levels of BDNF and cortical levels of PGE2 were positively correlated in both vehicle- and QA-injected hemispheres, it is temping to speculate that the early rise in the cortical PGE2 may result in an enhanced synthesis of BDNF in cortical pyramidal neurons. Activated microglia also represent a possible source of striatal BDNF, as demonstrated in different conditions of striatal injury (53, 54). As for the ability of SCH 58261 to reduce BDNF levels, this is in line with findings indicating that the state of activation of A2ARs regulate BDNF availability (22). The early depletion of neurotrophic factor induced by SCH 58261 could be critical for neuronal survival. On the other hand, BDNF has been recently reported to exert protoxic effects on motor neurons, whereas antagonists of both A2A and Trk receptors were neuroprotective (55).

Finally, we observed a composite and regional-specific effect of SCH 58261 pretreatment on glial cells. In the striatum, the A2AR antagonist moderately reduced the high number of activated (ED-1-positive) microglial cells as the site of injection, whereas it caused an early and transient increase in the number of such cells in the contralateral hemisphere, particularly in the corpus callosum, and later in the cortex. Thus, although the reduction of ED1-positive cells in the lesioned hemisphere may reflect a neuroprotective effect of A2AR antagonists, in the intact striatum SCH 58261 seems to facilitate microglial reactivity. The diverse responses of microglia to SCH 58261 could be related to the different state of activation and to the activity of the agonist on the signaling pathways triggered by A2AR activation in immune cells (56). On one hand, by preventing cAMP accumulation, the A2AR antagonist might lower the threshold for microglial reactivity; on the other hand, by inhibiting p38 MAPK, it may limit microglial activation.

In conclusion, our findings suggest that the final outcome of A2AR blockade (i.e. neuroprotection vs. neurodegeneration) is likely to depend on the balance among region- specific activities of A2AR antagonists and underline the importance of considering the effects of A2AR antagonists on other brain regions that are functionally linked to the striatum to fully appreciate their neuroprotective potential.

Footnotes

  • This work was supported by Istituto Superiore di Sanità and the Italian Ministry of Health research projects 4AN/F3 and F7 and 533F/A/3 and 4 (LM and PP).

  • All figures can be viewed in color online at http://jnen.oxfordjournals.org/.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
View Abstract