Microglial cells play a critical role in the neuroinflammatory response that accompanies various diseases of the central nervous system, such as ischemic stroke, and ATP is a major signaling molecule regulating the response of these cells to these pathophysiological conditions. Experiments were carried out to determine the effects of afobazole on microglial function and to identify the molecular mechanisms by which afobazole affects microglial cells. Afobazole was found to inhibit migration of microglial cells in response to ATP and UTP chemoattraction in a concentration-dependent manner. Inhibition of either σ-1 or σ-2 receptors decreased the effects of afobazole on microglia. In addition to inhibiting microglial cell migration, activation of σ receptors by afobazole decreased intracellular calcium elevation produced by focal application of ATP and UTP in isolated microglial cells. Furthermore, afobazole blocked membrane currents elicited by rapid application of ATP in microglial cells. Taken together, our data indicate that afobazole inhibits microglial response to P2Y and P2X purinergic receptor activation by functioning as a pan-selective σ-receptor agonist. In addition to modulating response to purinergic receptor activation, the effects of afobazole on microglial survival during in vitro ischemia were assessed. Application of afobazole during in vitro ischemia decreased microglial cell death during the ischemic episode and after a 24-h recovery period. Moreover, when afobazole was only applied after the ischemic episode, a significant enhancement in cell survival was still observed. Thus, afobazole acts via σ receptors to decrease microglial response to ATP and provides cytoprotection during and after ischemia.
Microglia are the resident macrophages of the brain and constitute nearly 10% of the total cell population of this organ. The function of these cells in normal and pathophysiological conditions has been an area of significant research, and considerable controversy surrounds the role of these cells in diseases of the central nervous system. It has been shown that microglia transform from a resting state to an activated state in response to various stimuli, such as neuronal injury or pathogen infiltration (Hanisch and Kettenmann, 2007). Upon stimulation, these cells migrate to the site of injury, phagocytose debris, and release proinflammatory compounds (Hanisch and Kettenmann, 2007). Such a microglial response has been reported to occur as a result of an ischemic insult to the brain, after which microglia are attracted to the site of injury and proliferate (Weinstein et al., 2010). It has been suggested that these microglia then contribute to the neuroinflammatory response that results in the demise of neurons (Rogove et al., 2002; Streit et al., 2004; Dheen et al., 2007). Thus, there is significant interest in finding agents that modulate the function of microglia and reduce their proinflammatory response.
Microglial cells are highly susceptible to ischemic injury and are more likely to undergo apoptosis than astrocytes after an ischemic insult (Lyons and Kettenmann, 1998). In contrast to the theory that microglia compound neuronal injury after stroke, it has been suggested that nonactivated microglia can provide neurotrophic factors that lessen the inflammatory response and can mitigate neuronal death (Lalancette-Hébert et al., 2007). It has also been shown that injection of microglia into the lateral ventricles of adult rats undergoing a middle cerebral artery occlusion results in migration of the transplanted cells into the infarcted striatum and reduction of infarct size (Kitamura et al., 2004). Thus, enhancing the survival of microglia may be of significant benefit following stroke injury.
Studies in our laboratory have shown that σ receptors in microglial cells can significantly decrease the conversion of these cells into the activated state and depress production of proinflammatory mediators (Hall et al., 2009). This suppression of microglial activation may in part explain why the σ-receptor agonist, DTG, can decrease neuronal injury in a rat model of ischemic stroke even when administered at delayed time points (Ajmo et al., 2006). However, the specific σ receptor involved in these effects has not been identified. Furthermore, how σ-receptor activation influences microglial survival after ischemia remains to be determined.
Afobazole [5-ethoxy-2-[2-(morpholino)-ethylthio]benzimidazole], an anxiolytic currently in clinical use in Russia, has shown potential both in vitro and in vivo as a neuroprotective agent (Galaeva et al., 2005; Zenina et al., 2005; Seredenin et al., 2008). To date, however, the effects of afobazole on microglial cell function and survival have not been studied. Given that afobazole may act as a σ-receptor ligand, it is of significant interest to determine how afobazole affects microglial responses and if the effects of this drug are linked to the activation of σ receptors.
Experiments were carried out to examine the effects of afobazole on several key markers of microglial activation and to determine the molecular mechanisms and pathway(s) underlying these effects. Afobazole was found to inhibit microglial migration in response to ATP and UTP. These effects of afobazole were attributed to the activation of both σ-1 and σ-2 receptors. Afobazole acting via both σ-receptor subtypes decreased P2Y- and P2X-mediated responses, resulting in decreases in [Ca2+]i elevations and membrane currents produced by the purinergic receptors. Moreover, afobazole enhanced microglial survival after ischemia, even when the drug was applied only after the ischemic insult.
Materials and Methods
Primary Cultures of Microglia.
Primary cultures of microglia were prepared from Sprague-Dawley rat pups (postnatal days 3–4) as described previously by our laboratory (Hall et al., 2009). The mixed glial cultures were incubated for 7 to 12 days at 37°C before experiments were conducted. Microglia were separated from other cells, physically, by shaking the flasks containing the mixed cultures at high speed. After shaking, the supernatant was collected and spun at 2200 rpm. The cell pellet was resuspended in DMEM. Cells were used immediately for the migration and cytotoxicity assays or plated on poly-l-lysine-coated glass coverslips for 1 day before use in calcium imaging or electrophysiological experiments.
Migration assays were carried out using a 48-well microchemotaxis chamber (Neuro Probe, Inc., Gaithersburg, MD). The bottom wells of the chamber were filled with the chemoattractant (ATP or UTP) in DMEM in the absence and presence of the chemicals tested (i.e., afobazole ± σ ligands). The upper chamber contained 1 × 106 freshly isolated microglia suspended in DMEM, also with or without afobazole ± σ ligands to match the chemical milieu of the lower chamber. The two chambers were separated by a polycarbonate membrane containing 8-μm pores at a density of 103 pores/mm2 (exposed filter area was 8 mm2). Before the experiments, the membrane was coated with fibronectin at 10 μg/ml in phosphate buffer saline solution (PBS) containing 144 mM NaCl, 10 mM Na2HPO4, 3 mM KH2PO4, pH to 7.4, with NaOH for 1 h at room temperature. The microglia were permitted to migrate for 2 h while incubated at 37°C. The membranes were then removed, and nonmigrating microglia adhering to the top of the membrane were removed. The membrane was then incubated in 4% paraformaldehyde in PBS for 15 min at room temperature, washed two times in ice-cold PBS, and washed once in distilled water. The membrane was then cut to separate the wells and placed on microscope slides, with membrane bottom facing upward. Once the membrane had dried, Vectashield Hardset mounting medium containing 4′-6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, CA) was applied to the membrane, and a coverslip was affixed to the slide. Cells were illuminated at 359 nm and visualized at 461 nm using a Zeiss Axioskop 2 outfitted with a 20× objective. Images were collected using OPENLAB software (PerkinElmer Life and Analytical Sciences, Waltham, MA) and analyzed using ImageJ (http://rsbweb.nih.gov/nih-image/). Stained nuclei were isolated from the fainter visible pores by setting excluding size (30–300 μm2) and intensity thresholds. DAPI-positive cells were identified and counted in four random fields/well. The number of cells that migrated in each well was taken as the average number of cells in the four fields of view. For each experiment, a minimum of three wells were used, and the results of at least three experiments were averaged together.
Calcium Imaging Measurements.
The effects of afobazole on intracellular Ca2+ elevations produced by ATP and UTP were examined using fluorescent-imaging techniques with Fura-2 as the indicator as reported previously (Hall et al., 2009). Microglia plated on coverslips were incubated for 1 h at room temperature in DMEM containing 3 μM acetoxymethyl ester fura-2 and 0.3% dimethyl sulfoxide. The coverslips were washed in fura-2-free saline solution before the experiments. The control bath solution for calcium imaging and electrophysiological experiments was a physiological saline solution (PSS) containing 140 mM NaCl, 5.4 mM KCl, 1.3 mM CaCl2, 1.0 mM MgCl2, 20 mM glucose, and 25 mM HEPES, pH to 7.4 with NaOH. All drugs for the calcium imaging and electrophysiological studies were applied in PSS using a rapid application system as described previously (Cuevas and Berg, 1998).
Microglia plated on glass coverslips were moved to a recording chamber mounted on a Zeiss Axiovert 200 and visualized at 400×. Methods for the electrical recordings of these cells were identical to those used previously in our laboratory to record from neurons (Herrera et al., 2008). Cells were electrically accessed using the amphotericin B-perforated patch method (Rae et al., 1991). An amphotericin B stock solution (60 mg/ml in DMSO) was made daily, kept on ice, light-protected, and diluted to 240 μg/ml (0.4% DMSO) in control pipette solution immediately before patch-clamp experiments. The pipette solution consisted of 75 mM K2SO4, 55 mM KCl, 5 mM MgSO4, and 10 mM HEPES (titrated to pH 7.2 with N-methyl-d-glucamine). Patch electrodes were made from thin-walled borosilicate glass (World Precision Instruments Inc., Sarasota, FL) and had resistances of 1.0 to 1.5 MΩ. Only cells with stable access resistance (Rs) ≤ 20 MΩ throughout the experiment were used, and this resistance was always compensated at 40%. All cells were voltage-clamped at −70 mV.
All chemicals used in this investigation were of analytic grade. The following drugs were used: amphotericin B, DTG, and metaphit [1-(1-(3-isothiocyanato)phenyl)cyclohexylpiperidine] (Sigma-Aldrich, St. Louis, MO; SM-21 [(±)-tropanyl 2-(4-chlorophenoxy)butanoate maleate], rimcazole [9-[3-(cis-3,5-dimethyl-1-piperazinyl)propyl]-9H-carbazole dihydrochloride], NE-100 [4-methoxy-3-(2-phenylethoxy)-N,N-dipropylbenzeneethanamine hydrochloride], BD 1063 [1-[2-(3,4-dichlorophenyl)ethyl]-4-methylpiperazine dihydrochloride], and BD 1047 [N-[2-(3,4-dichlorophenyl)ethyl]-N-methyl-2-(dimethylamino)ethylamine dihydrobromide] (Tocris Bioscience, Ellisville, MO); and fura-2 acetoxymethyl ester (Invitrogen, Carlsbad, CA). Afobazole was generously provided by Masterclone (Moscow, Russia).
Data analysis of Ca2+-imaging and electrophysiological studies were carried out using Clampfit 9 (Molecular Devices, Sunnyvale, CA). Imaging data files were collected using SlideBook 4.02 (Intelligent Imaging Innovations, Inc., Denver, CO), converted to text format, and imported into Clampfit for final analysis. Statistical analysis was conducted using SigmaPlot 9 and SigmaStat 3 software (Systat Software, Inc., San Jose, CA). For migration results, n value refers to the number of runs, with 3 wells per run for each condition. Statistical differences were considered significant if p < 0.05. Differences were determined using paired and unpaired t tests for within group and between group experiments, respectively. Likewise, either a one-way or a two-way analysis of variance (with or without repeat measures) was used for multiple group comparisons, as appropriate. Post hoc analysis of the analysis of variance data was conducted when necessary using a Tukey's test to determine differences between individual groups. For the generation of concentration-response curves, data were best fit using a single-site or two-site Langmuir-Hill equation.
The release of ATP from injured or dying neurons is one of the earliest signals involved in the activation of microglia following brain injury (Davalos et al., 2005). ATP, acting via ionotropic (P2X) and metabotropic (P2Y) purinergic receptors on microglia, causes these cells to migrate to the site of injury (Honda et al., 2001). Thus, experiments were carried out to determine how afobazole affects the migration of microglial cells in response to ATP, which activates both P2X and P2Y receptors, and UTP, which selectively activates P2Y receptors. Figure 1A shows photomicrographs of microglial nuclei stained using DAPI. The bright, visible nuclei represent microglial cells that have migrated across a polycarbonate filter in a chemotaxis chamber over a 2-h period in response to 100 μM ATP (left panels) or 100 μM UTP (right panels) in the absence (Control) and presence of afobazole at the indicated concentrations. Application of afobazole decreased the number of migrating microglia. This decrease in the number of cells was not the result of changes in survival rates of the cells (data not shown). In identical experiments, afobazole produced a concentration-dependent decrease in microglial migration. Figure 1B shows the concentration-response relationship for afobazole inhibition of microglial migration in response to ATP and UTP. For both compounds, the concentration-response relationship was best fit by a two-site Langmuir-Hill equation.
Activation of σ receptors has been linked to inhibition of microglial cell migration and afobazole has been suggested to interact with σ receptors (Seredenin et al., 2009). Experiments were carried out to determine whether afobazole acts through σ receptors in microglial cells to inhibit migration. Preincubation of microglial cells in the pan-selective, irreversible σ-receptor antagonist, metaphit, decreased the migration of microglia (Fig. 2A). However, a similar number of cells migrated in the presence of afobazole with and without metaphit preincubation (Fig. 2A). Thus, metaphit had effects on microglial migration on its own but effectively disrupted afobazole inhibition of migration. The inhibition of migration observed under control conditions (no metaphit preincubation, Media) was 46 ± 2% but decreased to 27 ± 2 after metaphit preincubation (Fig. 2B). Similar results were observed when the pan-selective σ-receptor agonist, DTG, was used (Fig. 2, A and B). When UTP was used as the chemoattractant, metaphit once again decreased migration on its own (Fig. 2C). However, the number of migrating cells was greater when afobazole was applied after metaphit preincubation than when afobazole was applied alone (Fig. 2C). The net result is a decrease in afobazole-induced inhibition of migration from 49 ± 1% (Media) to 25 ± 1% when cells were preincubated in metaphit (Fig. 2D). As with ATP, DTG inhibited UTP-evoked migration in a metaphit-sensitive manner (Fig. 2, C and D).
To date, the specific σ-receptor subtype(s) responsible for inhibiting microglial migration has not been identified. Thus, σ-receptor subtype-selective antagonists were used to determine the specific σ-receptor subtype responsible for the afobazole-evoked inhibition of microglial cell migration. Figure 3A shows the mean number of cells migrating (per field of view) when ATP was applied alone (ATP) or along with three different σ-1-receptor-selective antagonists (BD 1063, BD 1047, and NE-100). Afobazole was then applied along with the ATP (Afob) or with the ATP and the σ-1 antagonists (Afob + BD 1063, Afob + BD 1047, and Afob + NE-100). Similar to metaphit, all σ-1-receptor antagonists produced a small inhibition of migration on their own (Fig. 3A). However, a greater number of microglial cells migrated in the presence of afobazole when these σ-1 antagonists were applied than when afobazole was administered alone (Fig. 3A). The net result was that afobazole-mediated inhibition of ATP-evoked migration was decreased by ≥47% when the σ-1-receptor antagonists were coapplied (Fig. 3B). Thus, inhibition of σ-1 receptors decreases afobazole-evoked inhibition of microglial migration.
To determine whether σ-2 receptors are also involved in the regulation of microglial cell migration by afobazole, σ-2-receptor-selective antagonists were used in a series of experiments. Similar to σ-1 antagonists and metaphit, σ-2-selective antagonists produced a small decrease in microglial cell migration when applied alone (Fig. 4A). However, when applied with afobazole, both SM-21 and rimcazole significantly lessened the inhibition of migration produced by afobazole (Fig. 4, A and B). There was a 44 ± 0.3% decrease in migration in the afobazole-treated group but only a 22 ± 0.5 and 31 ± 0.4% decrease when afobazole was applied along with SM-21 and rimcazole, respectively (Fig. 4B).
Given the link between σ receptors, intracellular calcium and ATP-evoked microglial cell migration, experiments were carried out to determine whether afobazole can suppress elevations in [Ca2+]i evoked by ATP. Figure 5A shows representative traces of [Ca2+]i as a function of time recorded from two cells exposed to ATP in the absence (Control) and presence of 100 μM afobazole (Afobazole, top traces) and 100 μM DTG (DTG, bottom traces). Both afobazole and DTG reduced ATP-evoked increases in [Ca2+]i. In identical experiments, both peak (Fig. 5B, left) and net [Ca2+]i elevations (Fig. 5B, right) were decreased by afobazole and DTG. Afobazole and DTG decreased peak [Ca2+]i elevations produced by 100 μM ATP by 81 ± 2 and 58 ± 4%, respectively.
Microglia express both metabotropic (P2Y) and ionotropic (P2X) purinergic receptors, and both of these receptors can contribute to increases in [Ca2+]i. To determine whether σ receptors can regulate ionotropic receptors, whole-cell patch-clamp electrophysiological recordings were performed. Figure 6A shows traces of membrane currents recorded from a single microglia held at −70 mV in response to focal application of 300 μM ATP. ATP evoked a rapidly activating inward current that decayed over time. Application of 300 μM afobazole decreased the peak response elicited by ATP (Fig. 6A). Compiling the data from similar experiments demonstrates that 300 μm of afobazole significantly decreases ATP-evoked currents by approximately 50%. Thus, afobazole can affect P2X channels.
The relationship between afobazole interaction with σ receptors and regulation of ATP-evoked [Ca2+]i elevations in microglia was examined pharmacologically. The pan-selective σ-receptor antagonist, metaphit, was used to inhibit σ-receptor function in microglia and the effects of afobazole on ATP-evoked [Ca2+]i elevations studied using Ca2+ fluorometry. Figure 7A shows [Ca2+]i traces as a function of time recorded from two isolated microglial cells. Both cells were stimulated with 100 μM ATP in the absence and presence of 100 μM afobazole, but the traces on the right are from a cell preincubated in metaphit. Although afobazole decreases the [Ca2+]i responses in both cells, afobazole was more effective in the cell preincubated in PSS alone. A comparison of results obtained from multiple cells shows that afobazole significantly decreases ATP-evoked [Ca2+]i increases in the absence and presence of metaphit preincubation, but that metaphit preincubation diminishes the effects of afobazole (Fig. 7B). Afobazole (100 μM) inhibition of the ATP-evoked rise in [Ca2+]i decreases from 48 ± 4 to 21 ± 3% following metaphit preincubation (Fig. 7C).
Although afobazole inhibition of P2X channels may account for the decrease in [Ca2+]i, the effects of afobazole on P2Y-mediated [Ca2+]i responses remain to be studied. Application of the P2Y agonist, UTP (100 μM), evoked increases in [Ca2+]i similar to those elicited by ATP (Fig. 7D). In two individual microglia, afobazole (100 μM) decreased [Ca2+]i responses to UTP, but the response to afobazole was reduced by preincubation in metaphit (Fig. 7D). In identical experiments, it was found that preincubation in metaphit produced a small but significant reduction in the UTP-evoked elevation in [Ca2+]i (Fig. 7E). It is noteworthy that the [Ca2+]i response observed when afobazole was applied after metaphit preincubation was greater than that observed in control cells (Fig. 7E, PSS), indicating that the effects of afobazole were diminished by metaphit preincubation. The inhibition UPT-evoked increases in [Ca2+]i by afobazole decreased from 59 ± 4% in the absence of metaphit preincubation (PSS) to 25 ± 2% in the presence of metaphit preincubation (Fig. 7F).
Experiments using the σ-1-receptor-selective antagonists, BD 1063 and BD 1047, were carried out to ascertain whether that σ-receptor subtype was specifically involved in the effects of afobazole on ATP- and UTP-evoked increases in microglial [Ca2+]i. Neither BD 1063 (100 μM) nor BD 1047 (10 μM) significantly affected ATP-evoked increases in [Ca2+]i (Fig. 8, A and B). However, in the presence of these σ-1-receptor antagonists, the effects of afobazole (100 μM) on the [Ca2+]i responses to ATP were significantly decreased (Fig. 8, A and B). Both BD 1063 (100 μM) and BD 1047 (10 μM) reduced the percentage inhibition produced by afobazole by approximately 30% (Fig. 8, C and D).
Given that σ-2 receptors are also involved in afobazole modulation of microglial migration, the role of these receptors in afobazole regulation of ATP-evoked increases in [Ca2+]i was also tested. The σ-2-receptor-selective antagonists SM-21 and rimcazole were used in an attempt to disrupt afobazole signaling via this receptor. Figure 9 shows the results obtained in these experiments. Application of either SM-21 or rimcazole blunted the inhibitory effects of afobazole on mean peak ATP-evoked [Ca2+]i elevations (Fig. 9, A and B). There was an approximately 30% reduction in the afobazole-induced percentage inhibition of the [Ca2+]i responses in the microglial cells when either of these σ-2 antagonists was applied (Fig. 9, C and D).
Studies have shown that microglia are highly susceptible to ischemic injury (Lyons and Kettenmann, 1998), and numerous studies have suggested that these cells are important for limiting injury after ischemic stroke (Weinstein et al., 2010). Thus, experiments were carried out to determine whether afobazole can provide cytoprotection in vitro during ischemia. Cultured microglia were exposed to a combination of oxygen-glucose deprivation and 4 mM azide for 2 h to stimulate ischemia-induced cell death. Microglial cell death was quantified using an lactate dehydrogenase (LDH) cytotoxicity assay. Figure 10A shows the relative degree of cell death observed under control conditions (Control, normoxia, and normal media) and when microglial cells were exposed to the ischemic conditions (Ischemia). Under both conditions, cells were treated with either no drug (DMEM) or with DTG and afobazole at the indicated concentrations (in micromolars). Both afobazole and DTG significantly decreased neuronal death immediately after ischemia in a concentration-dependent manner. When microglial cells were removed from the ischemic environment and placed in normal media during a 24-h recovery period, cell death continued at a faster rate in cells exposed to ischemia than in control cells (Fig. 10B). Both afobazole and DTG decreased the amount of microglial cell death that occurred after this 24-h recovery period (Fig. 10B). Combining the data for cell death during the ischemic episode and during the recovery period shows that microglial cell death during ischemia is nearly 3-fold that of control and that both afobazole and DTG enhance cell survival (Fig. 10C). Moreover, afobazole at 100 μM was more cytoprotective than DTG at that same concentration (Fig. 10C). Specifically, examining the ischemia-induced cell death shows that afobazole and DTG decrease cell death after the ischemic episode in a concentration-dependent manner, and both inhibit this cell death by approximately 40% (Fig. 10D). Likewise, the ischemia-induced cell death that occurs during the 24-h recovery period was reduced by these drugs, but the effects ranged from 64 ± 3 (30 μM DTG) to 78 ± 4% (100 μM afobazole) (Fig. 10E). The effect of these drugs on net microglial cell death was concentration- and drug-dependent (Fig. 10F). The most pronounced cytoprotection observed was provided by 100 μM afobazole, which reduced net cell death by over 50% (Fig. 10F).
There is significant interest in identifying agents that can provide neuroprotection and cytoprotection after an ischemic event has occurred. Thus, experiments were conducted to determine whether application of afobazole after in vitro ischemia can affect microglial survival. Figure 11A shows the result of multiple experiments in which microglial cells were exposed to ischemia for 2 h and later treated with either media (DMEM) as a control or with afobazole at 30 and 100 μM concentrations. As expected, there was no difference in the cell death observed for all three groups during the ischemic event, because all groups were treated with DMEM at this stage (Fig. 11A). Introduction of afobazole after the ischemic event (Afob 30, Afob 100) decreased the cell death that occurs during a 24-h recovery period after the ischemic episode (Fig. 11B). The reduction in microglial cell death observed with this delayed application of afobazole (Afob 30, Afob 100) was comparable with that seen when afobazole was applied during the ischemic event (compare Fig. 10B with Fig. 11B). When net cell death (during ischemia + recovery) is compared for these groups, delayed application of afobazole produced a statistically significant reduction in cell death (Fig. 11C). However, administration of afobazole throughout was the most effective treatment paradigm for reducing cell death (Fig. 10F). Upon a close examination of the ischemia-evoked component of cell death, there was no protection in the DMEM followed by afobazole groups, as expected (Fig. 11D). However, delayed application of afobazole decreased ischemia-induced microglial cell death by >60% (Fig. 11E). Figure 11F shows the percentage inhibition of net ischemia-evoked cell death. Delayed application of afobazole was able to reduce microglial cell death by approximately 20% at both concentrations tested (Fig. 11F).
The major discovery reported here is that afobazole acts through both σ-1 and σ-2 receptors to influence activation and survival in microglial cells. To the best of our knowledge, this is the first report of afobazole affecting the function of glial cells. Migration of microglial cells in response to ATP and UTP were blocked by afobazole, and this inhibition could be disrupted via the use of pan-selective (metaphit) and subtype selective (σ-1, NE-100, BD 1063, and BD 1047; σ-2, SM-21, and rimcazole) σ-receptor antagonists. In addition to disrupting microglial migration, afobazole blocked membrane currents evoked by ATP and calcium elevations elicited by ATP and UTP. The effects on Ca2+ increases could be blocked via the use of σ-receptor antagonists, indicating that afobazole acting on σ-1 and σ-2 receptors can decrease P2X and P2Y receptor function. Finally, data presented here show that afobazole can also protect microglial cells from ischemic injury, and application of this drug, even at delayed time points, can enhance survival of these cells in an in vitro ischemia model.
Previous studies in our laboratory had shown that stimulation of σ receptors can decrease microglial cell activation in response to various stimuli, including ATP, monocyte chemoattractant protein-1, and lipopolysaccharide (Hall et al., 2009). However, the specific σ receptor involved remained to be identified. Furthermore, the possibility that afobazole could act as an agonist at one or both of these receptors to influence microglial activation remained unexplored. Here, we show that application of afobazole can completely abolish the migration of microglia in response to both ATP and UTP. The effects of afobazole were blocked by preincubating the cells in the pan-selective, irreversible inhibitor of σ receptor metaphit. The concentration required and extent of the block are consistent with the effects of afobazole being mediated by actions of this drug on σ receptors (Zhang and Cuevas, 2002; Wu, 2003; Katnik et al., 2006; Herrera et al., 2008).
The role of σ receptors in the actions of afobazole on migration were further supported via the use of subtype-selective antagonists. Three σ-1 receptor-selective antagonists NE-100, BD 1063, and BD 1047 decreased the ability of afobazole to inhibit microglial migration. At the same concentration used here (100 μM), NE-100 has been shown to block σ-1 receptor-mediated protection against oxidative stress in human retinal cells (Bucolo et al., 2006). BD 1063 and BD 1047 were first shown to be effective inhibitors of σ-1 receptors by Matsumoto et al. (1995). The responses seen with the concentrations of BD 1063 (1 μM) and BD 1047 (10 μM) used here are indicative of a σ-1 receptor-mediated effect (Matsumoto et al., 1995). Moreover, these concentrations are identical to those used previously in our laboratory to block σ-1 receptor-mediated effects in neurons (Katnik et al., 2006; Herrera et al., 2008).
The fact that a two-site Langmuir-Hill equation was required to fit the concentration-response relationship for afobazole inhibition of microglial migration suggests that these effects may be mediated by two receptors. The interaction with a second receptor was confirmed via the use of σ-2-receptor subtype-selective antagonists. The afobazole block of microglial migration was lessened by both rimcazole and SM-21. The fact that these molecules are effective at the nanomolar range confirms that afobazole acts as an agonist at σ-2 receptors to block microglial migration (Husbands et al., 1999; Ghelardini et al., 2000).
The effects of afobazole on microglial function may in part explain the clinical benefits of afobazole. Data presented here clearly demonstrate that afobazole is an agonist for σ receptors in microglial cells. Our laboratory has shown that activation of σ receptors on microglial cells decreases the secretion of proinflammatory cytokines and, in particular, TNF-α, by these cells (Hall et al., 2009). TNF-α has been linked to both anxiety and depression (Connor et al., 1998; Tuglu et al., 2003), and treatments that reduce TNF-α or prevent binding of this cytokine to its receptors is associated with reduction of affective disorder symptoms (Tuglu et al., 2003). Thus, afobazole may reduce the incidence of affective disorders by reducing the production of TNF-α in microglia. In animals, afobazole has been shown to be effective at decreasing the incidence of learned helplessness in a rat model of depression (Uyanaev and Fisenko, 2006). Data presented here demonstrate that application of afobazole can decrease the activity of P2X receptors in microglial cells. A recent study using mice deficient in P2X7 receptors, which are primarily expressed in microglia and peripheral macrophages, showed that these mice exhibited an antidepressant-like profile (Basso et al., 2009). Thus, by blocking P2X7 receptors, afobazole may decrease the incidence of depression.
In vivo, resting microglia exhibit a ramified morphology characterized by fine processes, or filopodia, that are highly dynamic and capable of rapidly moving toward a site of neuronal injury in response to a chemoattractant, such as ATP (Davalos et al., 2005). Prolonged activation of the microglia results in membrane ruffling, which involves withdrawal of the filopodia and conversion of the microglia into an amoeboid phenotype that is associated with an inflammatory state (Kreutzberg, 1996; Lyons et al., 2000). Our laboratory has shown that σ-receptor activation depresses the intracellular calcium signaling necessary for membrane ruffling to take place (Hall et al., 2009). Here we show that afobazole can prevent these increases in [Ca2+]i by acting on both σ-1 and σ-2 receptors. When ATP is used as the activating molecule, metabotropic P2Y12 receptors are responsible for increases in [Ca2+]i, which are critical for triggering this phenomenon (Honda et al., 2001). The fact that afobazole prevents UTP-evoked increases in [Ca2+]i indicates that afobazole is blocking P2Y [Ca2+]i signaling. The activation of P2X4 receptors expressed in microglia also contributes to ATP-evoked increases in [Ca2+]i. Given that afobazole also inhibits P2X channel function, [Ca2+]i elevations that may be attributed to these channels will also be suppressed by afobazole.
Although some studies have indicated that microglia may play a role in neurodegeneration in various central nervous system pathophysiological states (Streit et al., 2004), other studies have suggested that microglia may be beneficial in certain diseases. Resting microglia have been shown to harbor brain-derived neurotrophic factor and provide neuroprotection after traumatic brain injury in heat-acclimated mice (Shein et al., 2008). Likewise, intravenous transplantation of the human microglia cell line HM06 into rats after middle cerebral artery occlusion decreases infarct volume and improves behavioral recovery (Narantuya et al., 2010). Two common results of microglial responses in heat-acclimated mice and the HM06 transplantation model are as follows: 1) a decrease in proinflammatory cytokines and 2) an increase in trophic factors in the brain (Shein et al., 2007, 2008; Narantuya et al., 2010). Thus, preventing the conversion of microglia into proinflammatory mediators and enhancing the survival of resting microglia is likely to provide the greatest benefit to stroke patients from a therapeutic standpoint. Data presented here indicate that afobazole can decrease microglial activation. Furthermore, afobazole functions as a cytoprotective agent and decreases microglial death caused by ischemia. When afobazole was applied to microglia during an ischemic episode, microglial cell death was significantly reduced. It was noted that cell death continued even when microglia were removed from the ischemic conditions and permitted to recover for 24 h. Incubating the microglia in afobazole also prevented this delayed cell death caused by ischemia. It is noteworthy that microglial cell death could be significantly reduced even if afobazole was applied after the ischemic episode. Thus, afobazole may be used to mitigate injury to microglia after an ischemic stroke, raising the possibility that afobazole can be cytoprotective at delayed time points.
In conclusion, it has been shown previously that activated microglia are involved in the inflammatory response accompanying neuronal injury during various disease states, including Alzheimer's, Parkinson's, multiple sclerosis, traumatic brain injury, and stroke. However, previous studies have also concluded that resting microglia can provide trophic factors that lessen the loss of neurons in response to some, if not all, of these diseases. Here, we show that afobazole acts as an agonist at both σ-1 and σ-2 receptors to prevent microglial activation, as reflected by inhibition of cell migration and decreasing [Ca2+]i elevations produced by ATP and UTP. In addition, afobazole is an effective glia-protective drug that significantly reduces microglial cell death during and after an ischemic injury. Thus, afobazole is a potent modulator of microglia. The effects of afobazole on microglia may in part explain the clinical benefits of this drug in affective disorders and raise the possibility that afobazole can be a useful therapeutic agent for the treatment of diseases involving microglia, such as ischemic stroke.
Participated in research design: Cuevas and Katnik.
Conducted experiments: Cuevas, Rodriguez, Behensky, and Katnik.
Performed data analysis: Cuevas, Rodriguez, Behensky, and Katnik.
Wrote or contributed to the writing of the manuscript: Cuevas and Katnik.
We thank Drs. Andrei Petrov and Ilia Yasny and Nivia Cuevas for comments on a draft of this article.
This study was supported by grants from RF Pharmaceuticals Sarl and MDR Pharmaceutical Limited (to J.C.).
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
- acetoxymethyl ester
- acid-sensing ion channels
- (±)-tropanyl 2-(4-chlorophenoxy)butanoate maleate], rimcazole [9-[3-(cis-3,5-dimethyl-1-piperazinyl)propyl]-9H-carbazole dihydrochloride
- BD 1063
- 1-[2-(3,4-dichlorophenyl)ethyl]-4-methylpiperazine dihydrochloride
- BD 1047
- N-[2-(3,4-dichlorophenyl)ethyl]-N-methyl-2-(dimethylamino)ethylamine dihydrobromide
- Dulbecco's modified Eagle's medium
- dimethyl sulfoxide
- phosphate-buffered saline
- physiological saline solution
- intracellular calcium concentration
- change in intracellular calcium concentration
- lactate dehydrogenase
- tissue necrosis factor-α
- 4-methoxy-3-(2-phenylethoxy)-N,N-dipropylbenzeneethanamine hydrochloride.
- Received April 8, 2011.
- Accepted June 28, 2011.
- Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics