Address correspondence and reprint requests to Carla I. Tasca, Departamento de Bioquímica, CCB, UFSC, Trindade, 88040-900 Florianópolis, SC, Brazil. E-mail: email@example.com
Guanosine (GUO) is an endogenous modulator of glutamatergic excitotoxicity and has been shown to promote neuroprotection in in vivo and in vitro models of neurotoxicity. This study was designed to understand the neuroprotective mechanism of GUO against oxidative damage promoted by oxygen/glucose deprivation and reoxygenation (OGD). GUO (100 μM) reduced reactive oxygen species production and prevented mitochondrial membrane depolarization induced by OGD. GUO also exhibited anti-inflammatory actions as inhibition of nuclear factor kappa B activation and reduction of inducible nitric oxide synthase induction induced by OGD. These GUO neuroprotective effects were mediated by adenosine A1 receptor, phosphatidylinositol-3 kinase and MAPK/ERK. Furthermore, GUO recovered the impairment of glutamate uptake caused by OGD, an effect that occurred via a Pertussis toxin-sensitive G-protein-coupled signaling, blockade of adenosine A2A receptors (A2AR), but not via A1 receptor. The modulation of glutamate uptake by GUO also involved MAPK/ERK activation. In conclusion, GUO, by modulating adenosine receptor function and activating MAPK/ERK, affords neuroprotection of hippocampal slices subjected to OGD by a mechanism that implicates the following: (i) prevention of mitochondrial membrane depolarization, (ii) reduction of oxidative stress, (iii) regulation of inflammation by inhibition of nuclear factor kappa B and inducible nitric oxide synthase, and (iv) promoting glutamate uptake.
Cerebral ischemic injury remains a leading cause of death and disability in industrialized countries. The lack of oxygen and glucose because of ischemic injury results in a cascade of events such as loss of mitochondrial membrane potential and reduction in ATP production that impairs ATP-dependent processes as Na+/K+ ATPase activity. These events result in disruption of membrane potential triggering a release of excitatory amino acids such as glutamate (Bonde et al. 2003). Excessive extracellular glutamate leads to overstimulation of glutamate receptors and consequent influx of Na+, Cl−, and Ca2+ ions through the channels gated by those receptors. The increase in Ca2+ levels inside the cell results in activation of Ca2+-dependent enzymes which leads to processes such as proteolysis, hydrolysis, lipid peroxidation, and reactive oxygen species (ROS) production (Choi 1988). Glutamate uptake is a crucial process to maintain extracellular glutamate concentrations below toxic levels. This effect is achieved through specific high-affinity sodium-dependent excitatory amino acid transporters that are mainly present in astrocytes. Glutamate transporters are modulated by the cell redox status; thus, increased ROS production may result in glutamate uptake impairment (Trotti et al. 1998).
Excessive ROS production during an ischemic event can trigger an inflammatory response through activation of the transcriptional nuclear factor kappa B (NF-κB) (Gloire et al. 2006). NF-κB dimer is retained in the cytosol by interacting with inhibitory IκB proteins. When active, NF-κB subunit translocates to the nucleus and promotes expression of a large number of genes related with pathology of cerebral ischemia, including those involved in inflammatory response, such as interleukin-1, tumor necrosis factor-alpha and enzymes like inducible nitric oxide synthase (iNOS) and cyclooxigenase-2 (Sethi et al. 2008). The iNOS isoform is a mediator of inflammatory reactions and may catalyze substantial synthesis of nitric oxide (NO) in the injured brain, thus contributing to glutamate excitotoxicity (Bal-Price and Brown 2001).
There are increasing evidences that point to the importance of purines during hypoxic/ischemic events (Thauerer et al. 2012). For example, guanine derivatives can be released from astrocytes when subjected to hypoxia or hypoglycemia (Ciccarelli et al. 1999). Furthermore, guanine nucleotides and nucleoside levels increase during and after hypoxic or hypoglycemic periods. In the particular case of guanosine (GUO), its levels increase progressively under hypoxic or hypoglycemic conditions as a result of extracellular hydrolysis of nucleotides like as GTP, GDP, and GMP (Ciccarelli et al. 2001).
Extracellular effects of GUO have been implicated in neuroprotection by counteracting glutamate excitotoxicity. It has been shown that GUO protects from seizures induced by quinolinic acid in vivo (Schmidt et al. 2000; de Oliveira et al. 2004), and it exerts neuroprotection in in vivo and in vitro models of ischemia (Oleskovicz et al. 2008; Rathbone et al. 2011). Moreover, GUO protects hippocampal slices from glutamate-induced cell death by a mechanism that involves reduction in iNOS expression (Molz et al. 2011).
Despite the evidence showing that GUO displays a relevant neuroprotective effect in different neurotoxicity models (for review, see Schmidt et al. 2007), its extracellular site of interaction and its intracellular signaling pathway have not yet been fully characterized. It is suggested that GUO modulates cell proliferation, neurite outgrowth and cellular protection by a mechanism that involves adenosine receptors activation (Ciccarelli et al. 2000; Thauerer et al. 2010; Dal-Cim et al. 2012). However, GUO and GTP bind to adenosine receptors with a very low affinity (Muller and Scior 1993). Interestingly, in a recent study, we have showed that GUO is able to afford protection to hippocampal slices against oxygen/glucose deprivation and reoxygenation (OGD) by stimulating glutamate uptake, which is dependent on large conductance Ca2+-activated K+ (BK) channels and phosphatidylinositol-3 kinase (PI3K) pathway activation, (Dal-Cim et al. 2011). Taken together, these findings suggest that GUO may have distinct cellular targets besides the already known purinergic receptors.
The results of this study show that GUO, by modulating adenosine receptors and activating the MAPK extracellular-signal regulated kinase (ERK) kinase (MEK), promotes neuroprotection of hippocampal slices subjected to OGD. The neuroprotective mechanism of action of GUO involves reduction ROS production, prevention of mitochondrial membrane potential disruption, inhibition of NF-κB and iNOS induction, and promotion of glutamate uptake. Some of the data present here have appeared in the form of an abstract (Tasca et al. 2011).
Materials and methods
Male adult Wistar rats (60–90 days post-natal) maintained on a 12 h light–12 h dark schedule at 25°C, with food and water ad libitum, have been obtained from our local breeding colony. Experiments have followed the ‘Principles of laboratory animal care’ (NIH publication No. 85-23, revised in 1985) and they have been approved by the local Ethical Committee for Animal Research (CEUA/UFSC).
Preparation and incubation of hippocampal slices
Animals were deeply anesthetized by i.p. injection of sodium pentobarbital (60 mg/kg), decapitated, and each brain was rapidly removed from the skull and placed into ice-cold Krebs-Ringer bicarbonate buffer (KRB); dissection buffer (pH 7.4) containing, in mM: NaCl 120, KCl 2, CaCl2, 0.5, NaHCO3 26, MgSO4 10, KH2PO4 1.18, glucose 11, and sucrose 200.The buffer was bubbled with 95% O2-5% CO2 up to pH 7.4. Slices (0.4 mm) were prepared using a McIlwain Tissue Chopper (The Mickle Laboratory Engineering Co. ltd., Gomshall, England), separated in KRB at 4°C. Immediately after sectioning, slices were transferred to a vial of sucrose-free dissection buffer, bubbled with 95% O2/5% CO2 for 30 min at 34°C to recover from slicing trauma, before starting the experiments (equilibration period). The slices were completely submerged and protected from the vigorous bubbling in the chamber by a semipermeable nylon mesh.
The slices corresponding to the basal group were incubated at the end of the experiment in a normal KRB. OGD was induced by incubating the slices for a 15 min period in an OGD buffer in Hank's balanced salt solution (HBSS), composition in mM: 1.3 CaCl2, 137 NaCl, 5 KCl, 0.65 MgSO4, 0.3 Na2HPO4, 1.1 KH2PO4 and, 5 HEPES, where 10 mM d-glucose was replaced by 10 mM 2-deoxy-glucose (Pocock and Nicholls, 1998) equilibrated with a 95% N2/5% CO2 gas mixture. After this OGD period, slices returned to an oxygenated regular KRB containing glucose (reoxygenation period) for 2 h. These experiments were performed at 37°C.
When present, guanosine (GUO, 100 μM, Sigma, St Louis, MO, USA) was added in the reoxygenation period. Charybdotoxin (100 nM, Sigma, a potassium channel blocker), 1,3-dipropyl-8-cyclopentylxanthine (DPCPX) (250 nM, Sigma, an A1R antagonist), CGS21680 (200 nM, Sigma, an A2AR agonist), ZM241385 (50 nM, Sigma, an A2AR antagonist), PD98059 (25 μM, Sigma, a MEK-ERK pathway inhibitor), LY294002(10 μM, Alomone, a PI3K pathway inhibitor), or Pertussis toxin (500 μg/mL, Sigma, an inhibitor of Gi/Go-proteins) were incubated in the reoxygenation period. All compounds cited above were pre-incubated for 15 min before adding GUO and they were kept together during the 2 h of incubation.
Evaluation of cell viability
Cell viability was determined through the ability of cells to reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (from Sigma) as previously described (Liu et al. 1997). Hippocampal slices were incubated with MTT (0.5 mg/mL) in KRB for 30 min at 37°C. The tetrazolium ring of MTT can be cleaved by active dehydrogenases to produce a precipitated formazan. The medium was withdrawn, the precipitated formazan was solubilized with 200 μL of dimethyl sulfoxide and cellular viability was quantified spectrophotometrically at a wavelength of 550 nm.
Measurement of ROS production and mitochondrial membrane potential
To measure cellular ROS production, the molecular probe 2′,7′-dichlorofluorescin diacetate (H2DCFDA) was used. Hippocampal slices cells were loaded with 80 μM of H2DCFDA for 45 min at 37°C. Slices were then washed twice with KRB and maintained for 15 min, subsequent to the experiments of OGD. H2DCFDA diffuses through the cell membrane and it is hydrolyzed by intracellular esterases to the non-fluorescent form 2′,7′-dichlorodihydrofluorescein (DCFH). DCFH reacts with intracellular H2O2 to form 2′,7′-dichlorofluorescein (DCF), a green fluorescent dye. A separate set of slices were loaded with the mitochondrion selective dye fluorescent, tetramethylrhodamine ethyl ester (TMRE, 3 μM) for 15 min at 37°C (Egea et al. 2007), for analyzing the mitochondrial membrane potential after hippocampal slices being subjected to OGD. At the end of both experiments, slices were washed three times with KRB. Finally, fluorescence was measured in a fluorescence inverted NIKON eclipse T2000-U microscope (Nikon Instruments Europe, Badhoevedorp, Netherlands). Wavelengths of excitation and emission for DCFDA (ROS production) were 450 and 490 nm, and for TMRE (mitochondrial membrane potential) were 550 and 590 nm, respectively. Images were taken at CA1 region of the hippocampus at magnifications of 100×. Fluorescence analysis was performed using the Metamorph program version 7.0 (Molecular Devices, LLC, Sunnyvale, CA, USA). Fluorescence was measured in three different areas within the same hippocampal region to obtain the average value. Average fluorescence in basal conditions was then taken as 1, and experimental variables were normalized with respect to this value (Espada et al. 2010).
Isolation of nuclear extracts
Cytosolic and nuclear fractions were prepared as previously described (Rojo et al. 2006) with slight modifications. Briefly, hippocampal slices were mechanically disaggregated in three volumes of cold buffer A (20 mM HEPES, pH 7.0; 0.15 mM EDTA, 0.015 mM EGTA, 10 mM KCl, 1% Nonidet P-40, 1 mg/mL leupeptin, and 1 mM phenylmethylsulfonyl fluoride). The homogenates were kept in ice during 10 min and were subsequently centrifuged at 600 g for 5 min. The supernatant, corresponding to the cytosolic fraction, was resolved in sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and blotted. The nuclear pellet was re-suspended in five pellet volumes of cold buffer B (10 mM HEPES, pH 8.0; 0.1 mmol/L EDTA, 25% glycerol, 0.1 mol/L NaCl, 1 mg/mL leupeptin, and 1 mM phenylmethylsulfonyl fluoride). After centrifugation in the same conditions indicated above, the nuclei were re-suspended in two pellet volumes of hypotonic cold buffer A. Nuclear debris were removed by centrifugation at 600 g for 5 min at 4°C. The supernatant corresponding to the nuclear fraction was resolved in SDS–PAGE and blotted.
After treatment, hippocampal slices were lysed in 100 μL ice-cold lysis buffer (1% Nonidet P-40, 10% glycerol, 137 mM NaCl, 20 mM Tris–HCl, pH 7.5, 1 μg/mL leupeptin, 1 mM phenylmethylsulfonylfluoride, 20 mM NaF, 1 mM sodium pyrophosphate, and 1 mM Na3VO4). Protein (60 μg) from this cell lysate was resolved by SDS– PAGE and transferred to Immobilon-P membranes (Millipore Corp., Billerica, MA, USA). Membranes were blocked with 2% bovine serum albumin in Tris-buffered saline-T (10 mM Tris, pH 7.4; 150 mM NaCl; 0.2% Tween-20) during 2 h. After that, membranes were incubated with primary antibodies in Tris-buffered saline-T, rabbit polyclonal anti-iNOS (1 : 1000) (Cell Signalling) and rabbit polyclonal anti-p65 (1 : 1000) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Appropriate peroxidase-conjugated secondary antibodies (1 :10 000) were used to detect proteins. After incubation with primary and secondary antibodies, membranes were extensive washed and were incubated with enhanced chemiluminescence-Plus (GE Healthcare, Buckinghamshire, UK). Membranes were then exposed to the camera of a ChemicDoC MP System (Bio-Rad Laboratories, Hercules, CA, USA), and pixel intensities of the immunoreactive bands were quantified using Scion Image program (Scion Corporation, Frederick, MD, USA).
L-[3H] glutamate uptake
L-[3H]glutamate uptake was evaluated as previously described (Molz et al. 2009). After OGD and reoxygenation, hippocampal slices were washed for 15 min at 37°C in a HBSS, composition in mM: 1.3 CaCl2, 137 NaCl, 5 KCl, 0.65 MgSO4, 0.3 Na2HPO4, 1.1 KH2PO4, 2 glucose, and 5 HEPES. Uptake was assessed by adding 0.33 μCi/mL L-[3H] glutamate with 100 μM unlabeled glutamate in a final volume of 300 μL. Incubation was stopped immediately after 7 min by discarding the incubation medium, and slices were subjected to two ice-cold washes with 1 mL HBSS. Slices were solubilized by adding a solution with 0.1% NaOH/0.01% SDS and incubated overnight. Aliquots of slice lysates were taken for determination of the intracellular content of L-[3H] glutamate by scintillation counting. Sodium-independent uptake was determined using choline chloride, instead of sodium chloride in the HBSS buffer. Unspecific sodium-independent uptake was subtracted from total uptake to obtain the specific sodium-dependent glutamate uptake. Results were obtained as nmol of L-[3H]glutamate taken up per milligram of protein per minute and were expressed in percentage related to control levels.
Comparison among groups was performed by one-way anova followed by Duncan's test if necessary.
Implication of BK channels and A1R in the protective effect of guanosine against oxidative damage induced by OGD
Mitochondrial activity is essential for energy production into cells; however, Ca2+ overload because of excessive activation of glutamate receptors and reduced activity of the mitochondria is a crucial event in the excitotoxic cascade that follows an ischemic stroke (Murphy et al. 1999). Considering that mitochondrial activity is dependent on maintenance of its membrane potential, we evaluated the mitochondrial membrane potential in hippocampal slices subjected to oxygen/glucose deprivation and reoxygenation (OGD) and to guanosine treatment.
Hippocampal slices subjected to 15 min of oxygen/glucose deprivation followed by 2 h of reoxygenation (OGD) increased the fluorescence emission of TMRE in CA1 as an indication of mitochondria membrane depolarization (Fig. 1).
Previous studies have shown that 100 μM of GUO was neuroprotective against in vitro model of ischemia (Oleskovicz et al. 2008; Dal-Cim et al. 2011); therefore, this was the concentration used in this study. The presence of GUO (100 μM) during the reoxygenation period, significantly reduced mitochondrial depolarization induced by OGD (Fig. 1).
Blockade of BK channels with charybdotoxin at 100 nM does not alter cellular viability of hippocampal cells, but it can prevent the protective effect of GUO on cellular viability and glutamate uptake under OGD conditions (Dal-Cim et al. 2011). In this study, we have used charybdotoxin to evaluate a putative extracellular interaction site for GUO. Interestingly, we have observed that the concentration of BK inhibitor (charybdotoxin, 100 nM) used to prevent the effects of GUO on cellular viability and glutamate uptake in previous studies (Dal-Cim et al. 2011), did not prevent GUO action on mitochondria membrane potential (Fig. 1a).
Since GUO may also interact with adenosine A1 receptors (A1R), and considering that activation of A1R is involved in neuroprotective effects against oxidative damage (Dal-Cim et al. 2012), we verified the role of A1R in protective effects triggered by GUO in hippocampal slices. The presence of the A1R antagonist, DPCPX (250 nM; Sperlágh et al. 2007) abolished the protective effect of GUO on mitochondrial membrane potential (Fig. 1b).
The main source of ROS is the mitochondria, where they are generated as by-products of the electron transport chain and enzymes of the tricarboxylic acid cycle. (Addabbo et al. 2009). Since OGD induces changes in the mitochondrial membrane potential we also analyzed ROS production in rat hippocampal slices subjected to OGD. Hippocampal slices subjected to OGD show an excessive ROS production in CA1, when compared to basal situation (Fig. 1c and d). However, when the hippocampal slices were incubated with GUO, the amount of ROS produced by cells was reduced to basal levels (Fig. 1c and d). Evaluation of the mechanism of action triggered by GUO demonstrated that the BK channel blocker, charybdotoxin (Fig. 1c), and the A1R antagonist, DPCPX (Fig. 1d) prevented the reduction in ROS elicited by GUO in hippocampal slices subjected to OGD.
Implication of PI3K and MAPK/ERK in the protective effect of guanosine against oxidative damage induced by OGD
The intracellular signaling pathways PI3K and MAPK/ERK have been shown to be involved in the neuroprotective effect of GUO against OGD (Oleskovicz et al. 2008). Therefore, the participation of these signaling pathways on the effects of GUO on mitochondrial membrane potential was analyzed. The protective effect of GUO on mitochondrial depolarization exerted on hippocampal slices exposed to OGD was not affected by the PI3K inhibitor LY294002 (10 μM) (Fig. 2a) but was blocked by the MEK inhibitor PD98059 (25 μM) (Fig. 2b). When we analyzed ROS production under the same experimental conditions as above, a similar pattern was observed (Fig. 2c and d).
Guanosine inhibits NF-κB activation and iNOS induction
Excessive ROS production leads to activation of the transcription factor NF-κB and this event is associated with cell death during cerebral ischemia (Gloire et al. 2006; (Zhang et al. 2005). We analyzed the role of NF-κB in our experimental model through the evaluation of expression of the active subunit of NF-κB, p65, in the cytoplasm and the nucleus of hippocampal slices subjected to OGD. An increased immunodetection of the p65 subunit in the nuclear fraction of hippocampal slices subjected to OGD, as indication of NF-κB activation, was observed. GUO treatment during reoxygenation period reduced p65 expression in the nucleus (Fig. 3). Evaluation of GUO mechanism of action showed that charybdotoxin and LY294002 did not block the translocation of the transcription factor NF-κB promoted by GUO. Moreover, inhibition of MEK by PD98059, or the presence of the A1R antagonist, DPCPX, was effective in blocking the translocation of NF-κB promoted by GUO (Fig. 3b).
OGD, excessive ROS production or glutamate toxicity induces iNOS expression (Martin-de-Saavedra et al. 2011; Molz et al. 2011). Furthermore, the promoter sequence of iNOS contains a binding site for the transcription factor NF-κB (Mattson et al. 2000). In this study, we showed that hippocampal slices subjected to OGD show augmented ROS production. To evaluated the participation of ROS production and iNOS induction in the mechanism of OGD-induced cell damage, we have used NAC (5 mM; Yuan et al. 2012) a ROS scavenger, and 1400W (20 or 50 μM; Cárdenas et al. 1998) a selective iNOS inhibitor, to assess the hippocampal cellular viability after OGD. NAC (5 mM) was able to prevent the decrease in cellular viability induced by OGD (Fig. 4c), confirming that ROS generation was involved in the mechanism of cell damage evoked by OGD. The iNOS inhibitor, 1400W, was ineffective at 20 μM concentration; however, it partially prevented the cellular damage evoked by OGD when tested at 50 μM concentration.
The evaluation of iNOS expression, as an indicative of iNOs induction, showed that OGD increased iNOS levels when compared to control hippocampal slices (Fig. 4b), demonstrating that iNOS induction participates in the cellular damage evoked by OGD. In GUO-treated slices, iNOS expression returned to basal levels and this effect was blocked in the presence of MAPK/ERK and PI3K inhibitors (Fig. 4b and c). Furthermore, DPCPX, the A1R antagonist, also reversed GUO effect on OGD-induced iNOS levels (Fig. 4c),
Guanosine increases glutamate uptake: dependence on Gi/o-protein-coupled signaling
Previous reports have shown that the effects triggered by GUO are mediated by cellular signaling coupled to G-proteins, mainly from Gi/Go-proteins family (Di Iorio et al. 2004; D'Alimonte et al. 2007; Volpini et al. 2011). To assess the involvement of G-proteins in the neuroprotection promoted by GUO, a potent and specific inhibitor of Gi/Go-proteins, Pertussis toxin (PTX), was used. PTX (500 ng/mL) blocked the neuroprotective effect of GUO in hippocampal slices subject to OGD (Fig. 5a). As previously described by our group, the protective effect of GUO in vitro is directly related to its ability to modulate glutamate uptake (Dal-Cim et al. 2011; Molz et al. 2011). On the basis of these studies, we evaluated glutamate uptake in hippocampal slices subjected to OGD. As expected, OGD reduced glutamate uptake and GUO recovered this glutamate uptake impairment. When hippocampal slices were pre-incubated with PTX, increased glutamate uptake favored by GUO was no longer observed (Fig. 5b). Thus, modulation of glutamate uptake promoted by GUO seems to be mediated by activation of a Gi/Go-protein coupled signaling pathway.
Role of MAPK/ERK and adenosine receptors on GUO-induced enhancement of glutamate uptake
Considering that the protective actions of GUO against oxidative damage were mediated by A1R and its effects on glutamate uptake depended on activation of a Gi/Go-protein coupled signaling, we evaluated the putative participation of A1R in the modulation of glutamate uptake promoted by GUO. In hippocampal slices incubated with the A1R antagonist DPCPX, GUO did not present its neuroprotective effect on cellular viability (Fig. 6a). Surprisingly, antagonism of A1R by DPCPX did not prevent the increase in glutamate uptake afforded by GUO under OGD (Fig. 6b). The intracellular signaling triggered by GUO to counteract oxidative damage was the same that promoted cellular protection and increased glutamate uptake against OGD-induced damage, since the presence of PD98059, the MEK inhibitor, prevented GUO action on this parameter (Fig. 6a and b).
Considering that A1R and A2AR exert antagonistic functions on glutamate release in the striatum and hippocampus (Quarta et al. 2004) and that in this study A1R mediated some of the protective effects of GUO, but not the increase in glutamate uptake, we aimed to evaluate the role of A2AR in the modulation of glutamate uptake in our experimental protocol. The presence of the A2AR agonist CGS21680 (200 nM), or the A2AR antagonist ZM241385 (50 nM) did not alter cellular viability. When hippocampal slices were incubated with ZM241385, no alteration in the neuroprotective actions of GUO were observed. However, CGS21680 prevented GUO-induced neuroprotection (Fig. 7a). Similar effects were observed when glutamate uptake was analyzed. The A2AR activation with its agonist CGS21680 blocked the increased glutamate uptake promoted by GUO in hippocampal slices subject to OGD (Fig. 7b), suggesting a modulatory role of GUO also on adenosine receptors to exert its neuroprotective actions.
This study shows for the first time that guanosine protects hippocampal slices against oxidative and inflammatory processes induced by OGD via a mechanism that involves MAPK/ERK and A1R activation. Moreover, guanosine attenuates OGD-induced glutamate uptake impairment by activating MAPK/ERK, Gi/Go-proteins-coupled signaling and blockade of A2A adenosine receptors.
Disruption of mitochondrial membrane potential is a marker of mitochondrial dysfunction that contributes to cell death by reducing ATP production, increasing ROS production and releasing signaling molecules that regulate apoptotic cell death (Christophe and Nicolas 2006). The regulation of mitochondrial function via therapeutic approaches may facilitate neuroprotective cell responses. Recently, we demonstrated that GUO protects SH-SY5Y neuroblastoma cells from oxidative damage induced by mitochondrial activity impairment (Dal-Cim et al. 2012). Here, we show that GUO prevents disruption of the mitochondria potential induced by OGD in hippocampal slices. These effects point to the fact that GUO plays an important role in mitochondrial homeostasis in events that cause energetic unbalance. Reinforcing this notion, we show that GUO reduced ROS production induced by OGD. The precise mechanism by which GUO exerts these effects is not yet completely understood, but evidence suggests that it involves heme oxygenase-1 expression, via PI3K pathway, A1R and A2AR modulation and also BK channel activation (Dal-Cim et al. 2011, 2012).
In this study, we have found that the PI3K pathway does not directly mediate GUO′s protective effect on oxidative damage induced in an in vitro model of brain ischemia. We have also observed that BK channels mediated GUO-induced protective effect only against ROS production in the CA1 region of hippocampus, which usually is the most susceptible hippocampal region in an ischemic insult (Kirino and Sano 1984; Stanika et al. 2010). However, GUO maintained the mitochondrial potential and decreased ROS levels via activation of MAPK/ERK and A1R in the CA1 region of the hippocampus, indicating that GUO may have distinct cellular targets and mechanisms of action depending on the intensity of the insult and the seriousness of the cellular consequence observed.
The A1R is traditionally coupled to members of Pertussis toxin-sensitive family of G-proteins (Gi/Go) and their activation inhibits the adenylyl cyclase activity or modulates calcium channels activity. Activation of A1R at nerve endings inhibits transient calcium channels, whereas the same receptor in nerve cell bodies and dendrites may preferentially regulate potassium channels conductance (Boison et al. 2010). Thus, it is feasible that GUO could modulate A1R activity promoting a restoration of ionic gradient that was disrupted after OGD in ischemic events. Furthermore, intracellular signaling activated by A1R could contribute to GUO-induced neuroprotection.
Previous reports have shown that A1R activation is linked to modulation of PI3K and MAPK/ERK pathways (Jacobson and Gao 2006; Thauerer et al. 2012). Notwithstanding, activation of MEK in ischemic events is controversial. Studies have reported the involvement of MEK activation with pro-inflamatory processes and cell death following ischemia (Alessandrini et al. 1999; Maddahi et al. 2011). However, it was also demonstrated that MAPK/ERK pathway was involved in purine nucleoside-mediated protection of hippocampal slices and astrocytic or neuronal cells following hypoxic insults (Ciccarelli et al. 2007; Oleskovicz et al. 2008; Tomaselli et al. 2008). The results obtained in this study reinforce the importance of MAPK/ERK intracellular signaling activation by purine nucleosides, specifically GUO, since we demonstrate that blockade of MEK prevented GUO-induced neuroprotection against oxidative damage, impairment of glutamate uptake and inflammatory process via NF-κB and iNOS, which were induced by ischemia.
The transcriptional factor NF-κB is considered the major inflammatory mediator in neuronal tissue. Its activation during brain ischemia has been related to excessive ROS production. Activation of NF-κB leads to nuclear translocation of the p65 and/or p50 subunits to modulate the transcription of NF-κB responsive genes such as IL-6 and iNOS (Madrigal et al. 2006). The participation of GUO in inflammatory processes was previously reported by D′Alimonte that showed that GUO treatment (300 μM) inhibits TNF-α and amyloid-β peptide-induced p65 phosphorylation in mouse microglial cells (D'Alimonte et al. 2007). In this study, we show that lower concentrations of GUO (100 μM) are also capable of decreasing nuclear p65 expression induced by an oxidative insult because of OGD in hippocampal cells. Furthermore, analysis of the expression of iNOS, a NF-κB responsive gene, showed that GUO treatment could reduce iNOS levels under OGD conditions. Induction of iNOS results in excessive amounts of NO that changes its physiological neuromodulatory actions to neurotoxic effects such as inhibition of the complex-I of the electron transport chain. In addition, NO may react with superoxide anions to form nitrite peroxide, a strong pro-oxidant, that ultimately leads to irreversible cellular damage (Brown 2010). Thus, inhibition of iNOS has been proposed as an important therapeutic strategy against ischemic insults (Iadecola and Ross 1997; Moro et al. 2004). The mechanism elicited by GUO to counteract NF-κB activation and iNOS induction was mediated by the transduction signaling A1R and MAPK/ERK.
Several evidence support the notion that the neuroprotective effect of GUO against excitotoxicity is related to the modulation of glutamate uptake (Oliveira et al. 2004; Moretto et al. 2005; Schmidt et al. 2007), thus facilitating the clearance of glutamate from the synaptic cleft. Recently, we demonstrated that GUO treatment prevents glutamate release induced by glutamate excitotoxicity and it also increases glutamate uptake into hippocampal slices subject to OGD, effects that are mediated by activation of PI3K pathway (Dal-Cim et al. 2011; Molz et al. 2011). Here, we observed that GUO-induced stimulation of glutamate uptake was totally prevented by MEK inhibition. Thus, it seems that the effect of GUO on glutamate uptake may be related to its ability to reduce ROS production, since MAPK/ERK pathway also mediates GUO′s action on oxidative damage induced by OGD. In addition, it has been reported that signaling pathways such as PI3K and MAPK/ERK are involved in regulation of glutamate transporters activity, expression and insertion in the cell membrane (Li et al. 2006; Frizzo et al. 2007; Dal-Cim et al. 2011) thus, MEK activation by GUO could result in modulation of glutamate transporters activity, contributing in this way to its neuroprotective profile.
Notably, we show that the action of GUO on glutamate uptake is not related to its putative interaction with A1R. On the other hand, we previously demonstrated that increased glutamate uptake induced by GUO depends on BK channel activation (Dal-Cim et al. 2011), suggesting that GUO may display distinct cellular effects regarding its extracellular interaction site. Modulation of BK channels by GUO can result in a decline of intracellular Ca2+ levels, thus avoiding the loss of cell membrane potential, a critical event for maintaining the activity of glutamate transporters (Danbolt 2001). Moreover, A1R activation is associated with decreased vesicular glutamate release in pre-synaptic terminals (Sperlagh and Vizi 2011), an effect that does not depend on modulation of cell membrane glutamate transporters. It has been recently demonstrated that A2AR activation in cultured astrocytes, decreases glutamate uptake while activation of A1R has no effect on the modulation of glutamate transport (Matos et al. 2012). These evidences reinforce the idea that activation of A1R does not modulate the activity of glutamate transporters. Surprisingly, we have observed that A2AR stimulation blocks the effect of GUO on glutamate uptake in hippocampal slices subjected to OGD, suggesting that cellular signaling triggered by the activation of A2AR combined with the energy imbalance and oxidative damage caused by OGD dramatically affect the ability of GUO to recover the impaired glutamate transporters activity.
A1R and A2AR may closely interact through the formation of oligomers and in this heteromeric organization, they can modulate glutamate release from pre-synaptic terminals in the striatum (Ciruela et al. 2006), or GABA uptake into cultured cortical astrocytes (Cristóvão-Ferreira et al. 2011). It has been proposed that when A1R and A2AR are forming an oligomer, a hierarchical activation of A2AR over A1R will result in the functional tuning of purinergic transmission (Ciruela et al. 2012). In this scenario, GUO may act as an antagonic A2AR modulator, although it is surprising that GUO effect on glutamate uptake was dependent on Gi-protein coupled signaling, whereas A1R blockade did not alter GUO effect on glutamate uptake. On the other hand, it seems plausible that the blockade of A2AR activation would be involved on GUO effect, since it has been shown that A2AR antagonists may exert neuroprotective actions (Cunha 2005; Gomes et al. 2011).
Although our data strongly indicate an interaction among GUO and adenosine receptors (and in earlier studies, with the BK potassium channel), we cannot exclude the possibility that GUO acts through a selective receptor and secondarily modulates the activity of adenosine receptors and BK channels. Studies supporting this hypothesis have demonstrated specific putative binding sites for GUO (Tasca et al. 1999; Gysbers et al. 2000; Traversa et al. 2002; Volpini et al. 2011). Supporting this idea, we have seen that inhibition of Gi/Go-protein by Pertussis toxin reverses the effect of GUO on cell viability and glutamate uptake in hippocampal slices subject to OGD, indicating that the interaction site of GUO couples to a Gi/Go-protein. Still, the exact mechanism of GUO action on BK channels and/or adenosine receptors remains to be determined.
Concluding, GUO is neuroprotective against oxidative and inflammatory processes induced by ischemia via activation of A1R and MAPK/ERK pathway. GUO also prevented glutamate uptake impairment by modulation of A2A adenosine receptors and a pathway involving Gi/o-proteins-coupling and MAPK/ERK signaling (Fig. 8). The molecular targets and signaling pathways recruited by GUO to display its neuroprotective effects are still under evaluation in our laboratory.
Research supported by grants from the Brazilian funding agencies, CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) – Projects IBN-Net # 01.06.0842-00 and INCT (Instituto Nacional de Ciência e Tecnologia) for Excitotoxicity and Neuroprotection and FAPESC (Fundação de Amparo à Pesquisa e Inovação do Estado de Santa Catarina) – Project NENASC/PRONEX to C.I.T. and grants from the Spanish Ministry of Science and Innovation SAF2009-12150, Ministry of Education PBH2007-0004-PC and the Spanish Ministry of Health (Instituto de Salud Carlos III) RETICS-RD06/0026 to M.G.L. C.I.T. is recipient of CNPq productivity fellowship and T. D.C. was recipient of CAPES/DGU (Project No 173/2008) and CNPq pre-doctoral scholarship. The authors state no conflicts of interest. All authors have materially participated in the research and/or article preparation.