INTRODUCTION

ATP-sensitive potassium (KATP) channels originally were discovered in heart cardiac tissue. Uptill now, they have been found to be widely expressed in metabolically active tissues, and in many excitable cells including cardiac myocytes, hormone-secreting cells, skeletal and smooth muscle, and neurons (Ashcroft, 1988; Seino, 1999; Zawar et al, 1999). KATP channels play a critical role in coupling cellular metabolism to electrical activity, and have been shown to participate in some important biological processes such as insulin secretion, synaptic transmission, and excitability of cardiac, vascular, and nonvascular smooth muscle. It is generally agreed that KATP channels in cardiac muscle and neurons have a protective effect during cellular suffering from injury, such as ischemia and anoxia (Zini et al, 1993; Prensa and Parent, 2001). Within the brain, KATP channels exist in both pre- and post-synaptic membrane, as well as in glial cells. Activation of post-synaptic KATP channels causes membrane hyperpolarization, which limits neuronal excess excitation; activation of presynaptic KATP channels can directly modulate neurotransmitter release from nerve terminals. Therefore, KATP channel openers (KCOs) serve as efficient tools that can adjust cell excitability, and exhibit beneficial effects under pathological conditions such as ischemia, stroke, and neurodegenerative diseases (Yao and Gross, 1994; Sargent et al, 1993; Liss et al, 2005). Emerging evidence including from our laboratory has revealed that the opening of KATP channels play dominant protective roles in various models of neurodegenerative diseases such as Parkinson's disease (PD) (Tai et al, 2003; Wang et al, 2006; Yang et al, 2004).

It is well-known that astrocytic glutamate transporters (GluTs) perform the majority of glutamate uptake from the extracellular space in the brain, and play an important role in maintaining extracellular glutamate concentrations below neurotoxic levels (Matute et al, 2006; Schousboe and Waagepetersen, 2005). The dysfunction of astrocytic GluTs leads to an excessive extracellular glutamate concentration, thereby inducing excitotoxicity and neurodegeneration (Ferrarese et al, 2000). Thus, it has been proposed that improvement of glutamate uptake might be a potential strategy for neuroprotection (Wadiche and Gersdorff, 2006; Genoud et al, 2006; Matsugami et al, 2006; Takayasu et al, 2006). Our previous studies have revealed that KCOs could decrease extracellular glutamate concentrations and reduce excitotoxicity in PD animal and cellular models (Tai et al, 2003; Yang et al, 2004; Wang et al, 2004; Busija et al, 2004; Liss et al, 1999). We also demonstrated that KCOs could regulate glutamate uptake in PC12 cells and in synaptosomes from normal and PD rats (Inagaki et al, 1995). Since KATP channels are expressed in astrocytes and astrocytic GluTs play key roles in maintaining extracellular glutamate levels, whether activation of KATP channels plays a neuroprotective role via regulating the functions of GluTs is essential to be elucidated and is hitherto unknown. In the present study, we examined the effects of KCOs on glutamate uptake in primary cultured astrocytes to determine (1) whether activation of KATP channels would regulate GluTs function, (2) whether KCOs could protect astrocytes against neurotoxin-induced dysfunction of GluTs, and (3) the possible mechanisms by which KATP channels modulate glutamate uptake.

MATERIALS AND METHODS

Primary Astrocyte Cultures

Confluent primary astrocyte cultures were prepared from cerebral cortex of newborn Sprague–Dawley rats as described previously (Ivanova and Beyer, 2003), with minor modifications as listed below. All animal procedures were performed according to the NIH Guide for Animal Care and approved by the institutional animal care and use committee. Briefly, neonatal rats were killed by rapid decapitation, the cerebral cortices were removed and separated from meninges and basal ganglia, and tissue was dissociated with 0.25% trypase (Amresco) at 37°C and terminated by Dulbecco's modified Eagle's medium (DMEM, Gibco-BRL Life Technologies) supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin. After centrifugation at 1500 r.p.m. for 5 min, the cell pellets were resuspended and seeded on poly-lysine-coated (Sigma) flask. The cultures were maintained at 37°C in a humidified 5% CO2–95% air atmosphere. Culture medium was replaced 24 h later and then changed every 2–3 days. Before experiments, astrocytes were replated on poly-lysine-coated 6- or 24-well plates. Immunocytochemistry showed that 98% of the cells stained positively for the astrocytic marker glial fibrillary acid protein (GFAP).

Assays of Glutamate Uptake

Glutamate uptake determination was performed according to the procedure as described (Yao et al, 2005). Following pinacidil (Pina), diazoxide (Dia), P1075 or iptakalim (Ipt) administration for 48 h, uptake assays were initiated by adding [3H]D, L-glutamate (190 nM, final concentration) to reaction flasks covered with astrocytes, and incubated for 15 min at 37°C in 24-well plates. Nonspecific uptake was determined from samples incubated in the presence of the Na+-free buffer, and these values were subtracted from the total. All the reactions were terminated by rinsing thrice with ice-cold 0.9% NaCl. Then, the cells were lysed immediately with 0.3 M NaOH. After centrifugation at 10 000 r.p.m. for 15 min, supernatant was abstracted and radioactivity was determined by liquid scintillation counter. Protein level of each sample was detected using dye coomassie brilliant blue G-250.

Measurement of ROS generation

Formation of reactive oxygen species (ROS) was evaluated using 2′,7′-dichlorofluorescin diacetate (DCFH-DA), a membrane-permeable probe de-esterified intracellularly. The non-fluorescent dye freely penetrates cells, is then hydrolyzed by intracellular esterases to 2′,7′-dichlorodihydrofluororescein (DCFH), and trapped inside the cells. Upon oxidation by ROS, DCFH yields the highly fluorescent product dichlorofluorescein (DCF).

Treated with drugs for 48 h, cells were loaded with DCFH-DA (50 μM final concentration) in DMEM for 60 min in the dark and fixed by 4% formaldehyde. After rinsing cells twice with PBS solution (137 mM NaCl, 2.7 mM KCl, 9.5 mM Na2HPO4·12H2O, 1.5 mM KH2PO4, pH 7.20), fluorescence was read at the excitation wavelength (Ex) of 488 nm and the emission wavelength (Em) of 530±20 nm. DCFH-DA was initially dissolved to 40–50 mM in ethanol and stored at −20°C. A final concentration of ethanol (0.6% or less) did not alter the fluorescence measurement.

PepTag Assay for Non-Radioactive Detection of PKC Activity

Following KCOs treatment for 48 h, the cells were rinsed with ice-cold PBS solution and harvested by scraping on ice into 0.2 ml of cold protein kinases C (PKC) extraction buffer. The lysates were subjected to mild sonication and followed by centrifugation at 4°C, 14 000 × g in a microcentrifuge for 5 min. Then the supernatant was used for the PepTag assay, which is generally performed within 24 h.

According to the manufacturer's (Promega) protocol, the cell samples were incubated with PKC reaction mixture (25 μl) at 30°C for 30 min and the reactions were stopped by placing the tubes in a boiling water bath. After adding 80% glycerol (1 μl), the samples were loaded on a 0.8% agarose gel at 100 V for 20 min. The bands were visualized and photographed under UV light. The units of kinase activity in each slice of agarose were calculated according to the manufacturer's instructions.

Statistical Analysis

All data were presented as the mean±SEM. Statistical analysis for multiple comparisons was performed by one-way ANOVA test with Bonferroni's corrections using SPSS 10.0 for Windows (SPSS Inc., Chicago, IL, USA). The level of statistical significance is defined as p<0.05.

RESULTS

KCOs Facilitated Glutamate Uptake in Rat Primary Astrocyte Cultures in a Concentration-Dependent Manner

Astrocytes in culture exhibited a high-affinity glutamate uptake after 48 h of incubation in DMEM with Vmax calculated to 10.66±0.06 pM mg−1 min−1 (Figure 1a). As shown in Figure 1a, the uptake was significantly increased (117, 126, and 129% of the control) when exposed to pinacidil (Pina, 1, 10, and 100 μM respectively). Preincubation with 100, 200, or 500 μM diazoxide (Dia), a mitochondrial KATP (mitoKATP) channel opener, increased glutamate uptake to 120, 136, and 149%, respectively, of the control in a concentration-dependent manner (Figure 1b). However, compared with control group, no significant effect on astroglial glutamate uptake was detected in the presence of P1075 (p>0.05, Figure 1c), a selective sarcolemmal KATP (sarcKATP) channel opener. As a novel KCO, iptakalim (Ipt) increased glutamate uptake to 118 and 114% at the concentration of 1 and 10 μM, respectively (Figure 1d). These findings indicated that the KCOs could facilitate glutamate uptake by GluTs in astrocytes.

Figure 1
figure 1

Kinetics of glutamate uptake in differentially pretreated astrocyte populations. Values represented as mean±SEM of at least four independent experiments performed in triplicate. C, control; Pina, pinacidil; Dia, diazoxide; Ipt, iptakalim. *p<0.05, **p<0.01 vs control group.

KCOs Alleviated MPP+-Induced Glutamate Uptake Inhibition in Rat Primary Astrocyte Cultures

After incubation with 1-methyl-4-phenylpyridinium (MPP+, 150 μM) for 48 h, astrocytic glutamate uptake was remarkably decreased to 5.13±0.46 pmol mg−1 min−1 compared with control group (7.08±0.22 pmol mg−1 min−1, Figure 2a). As shown in Figure 2a, preincubation with Pina concentration-dependently (0.1–100 μM) alleviated glutamate uptake impairment in MPP+ cotreated astrocytes, which was completely blocked by glibenclamide (Glib, 10 μM) and mitoKATP channel blocker 5-hydroxydecanoate (5-HD, 250 μM) but not HMR1098 (20 μM, a sarcKATP channel blocker, Figure 2b). These results indicated that the prevention of the uptake improvement with Glib or 5-HD was due to the blockade of the mitoKATP channel, because Glib or 5-HD alone had no effects on glutamate uptake (Figure 2h). To examine further the involvement of mitoKATP channel in the glutamate uptake, Dia or P1075 was preincubated with MPP+. Dia (10, 100, 200 or 500 μM) and the novel KCO Ipt (0.1, 1, 10 or 100 μM) could concentration-dependently attenuated uptake inhibition induced by MPP+ (Figure 2c and e), P1075 (1, 10, 20, 50 or 100 μM) had no effect on astrocytic glutamate uptake (p>0.05, Figure 2g). The facilitating effects of Dia (100 μM) and Ipt (10 μM) on glutamate uptake were completely abolished by 5-HD (250 μM, Figure 2d and f), indicating that the protective role of Dia and Ipt in astrocitic glutamate clearance was actually due to the opening of mitoKATP channels.

Figure 2
figure 2

Protective effects of KCOs on glutamate uptake inhibition induced by MPP+ in primary cultured astrocytes. Results are mean±SEM performed at least four independent experiments in triplicate. C, control; MPP+, 1-methyl-4-phenylpyridinium; Pina, pinacidil; Dia, diazoxide; Ipt, iptakalim; Glib, glibenclamide; 5-HD, 5-hydroxydecanoate. *p<0.05, **p<0.01 vs unteated group; #p<0.05, ##p<0.01 vs MPP+-treated group.

KCOs Inhibited MPP+-Triggered ROS Elevation in Rat Primary Astrocyte Cultures

After, preincubation with different KCOs for 30 min, then followed by incubation with 150 μM MPP+ for 48 h, the production of ROS was measured with the DCFH-DA probe. MPP+ (Figure 3b) increased the ROS output to 472% of control group (Figure 3a). Either Dia (100 μM, Figure 3c) or Ipt (10 μM) (Figure 3e) could inhibit ROS production significantly. Furthermore, the effects of Dia and Ipt were reversed by 5-HD (250 μM) (Figure 3d and f). P1075 failed to reverse MPP+-induced ROS accumulation (Figure 3g).

Figure 3
figure 3

Effects of KCOs on ROS production in astrocytes incubated with MPP+. Values represented as mean±SEM of four independent experiments performed in triplicate. C, control; M&MPP+, 1-methyl-4-phenylpyridinium; Dia, diazoxide; Ipt, iptakalim; 5-HD, 5-hydroxydecanoate. **p<0.01 vs control group; ##p<0.01 vs MPP+-treated group.

KCOs Suppressed PKC Activity in MPP+-Treated Astrocyte Cultures

After preincubation with Dia (100 μM), Ipt (10 μM), P1075 (20 μM), and 5-HD (250 μM) for 30 min followed by exposure to MPP+ for 48 h, astroglial protein kinases C (PKC) activity was detected with PepTag non-radioactive protein kinase assays (Promega). As shown in Figure 4, MPP+ (150 μM) increased the astroglial PKC activity to 23 nM min−1 gPro−1. Pretreatment with Dia and Ipt significantly inhibited PKC activity to 13 nM min−1 gPro−1 and 15 nM min−1 gPro−1 repectively. However, P1075 failed to influence the increase of cellular PKC activity induced by MPP+.

Figure 4
figure 4

Effects of various KCOs on MPP+-induced PKC activation of primary cultured astrocytes. Compared with non-preconditioned cells, MPP+ remarkably increased intracellular PKC activity, which could be restrained by Dia and Ipt. There was no significant change detected in P1075-treated group. Results are mean±SEM of four independent experiments performed in triplicate. C, control; M&MPP+, 1-methyl-4-phenylpyridinium; Dia, diazoxide; Ipt, iptakalim. *p<0.05 vs control group; #p<0.05 vs MPP+-treated group.

DISCUSSION

In the brain, KATP channels do not open under physiological conditions, but they can be activated by either a specific opener or under many pathological conditions, such as acute ischemia/hypoxia, ischemic preconditioning or metabolic inhibition, thereby hyperpolarizing the membrane potential (Farkas et al, 2004; Thomzig et al, 2001; Nicholls and Attwell, 1990). KATP channels are regulated by intracellular concentrations of ATP ([ATP]i), being opened as [ATP]i decreases, and thus by linking neuronal metabolism to excitability, they are thought to play an important role in protecting cells from neuronal death caused by hypoxia, ischemia or metabolic inhibition (Tai et al, 2003; Yang et al, 2004; Busija et al, 2004; Liss et al, 1999).

KATP channels are hetero-octameric proteins composed of pore-forming Kir6.x subunits (Kir6.1 and Kir6.2) and regulatory sulfonylurea receptor (SUR) subunits (SUR1, SUR2A, and SUR2B) (Seino, 1999; Inagaki et al, 1995), and located in various cellular compartments, including the surface of the plasmalemmal membrane (sarcKATP channel) and the inner mitochondrial membrane (mitoKATP channel) (Farkas et al, 2004). In the brain, KATP channels are enriched in neurons as well as in glial cells, but their subunit combinations of KATP channels appear not to be homogeneous. For example, the subunits of KATP channels in neurons are Kir6.2 and SUR1/SUR2B (Zawar et al, 1999; Thomzig et al, 2001), and KATP channels in astrocytes are composed of Kir6.1 and SUR1 (Thomzig et al, 2001). These studies have raised the questions: what is the role of KATP channels expressed in discrete cellular populations and how do they work in particular physiological/pathophysiological episodes?

It has been well-known that activation of KATP channels can regulate release of neurotransmitters from neurons, and reduce neuronal excitability thereby protecting against excitotoxicity. How do KATP channels regulate the functions of astrocytes? Traditionally, astrocytes were thought to play minor roles in neuronal function and in directing overall activities in the brain, serving only a maintenance role in regulating brain homeostasis (Nicholls and Attwell, 1990; Anderson and Swanson, 2000). Notably, recent studies challenge these assumptions and suggest that rather than being an innocuous bystander, astrocytes may play crucial roles in regulating neuronal activity and signal transmission; deficiencies in these functions may contribute to neurodegeneration (Haydon, 2001). One way astrocytes wield their effects on neuronal function is through GluTs to maintain stimulatory, but nontoxic low levels of free intrasynaptic L-glutamate in the area adjacent to neurons (Nicholls and Attwell, 1990; Robinson, 1998; Lipton and Rosenberg, 1994). Abnormalities in this process result in the accumulation of extracellular glutamate in synaptic clefts, and subsequently induce overexcitation, and death of neurons (Nicholls and Attwell, 1990; Robinson, 1998; Lipton and Rosenberg, 1994). Therefore, the present study is to show the effects of KATP channel opening on glutamate uptake by GluTs in astrocytes.

We demonstrated that administration with Pina, Dia and Ipt increased glutamate uptake in rat primary cultured astrocytes, but sarcKATP channel opener P1075 failed to affect astrocytic glutamate uptake. Furthermore, Pina, Dia and Ipt could also reverse the inhibition of glutamate uptake induced by MPP+, and mitoKATP blocker 5-HD inhibited these effects. These findings suggest that activation of KATP channels, especially mitoKATP channel, facilitated the function of GluTs in astrocytes. Ipt is a lipophilic para-amino compound with low molecular weight, which can freely cross the blood–brain barrier and has been shown to be a novel KCO by pharmacological, electrophysiological and biochemical studies, and receptor binding test (Wang, 2003; Xie et al, 2005). Ipt was initially designed and synthesized as a novel antihypertensive drug (Wang, 1998). In the view of the technical requirement for novel antihypertensive drug approval, the preclinical investigation of Ipt has been completed and clinical trials are currently underway (Wang, 2003). Our previous studies provided compelling supports for the neuroprotective effects of Ipt at the dosages not affecting blood pressure, via preventing the excitotoxic insults and anti-neuroinflammation (Wang et al, 2006, 2004; Yang et al, 2004). The present study revealed a new mechanism involved in the neuroprotective effects of the novel KCO.

It has been reported that mitoKATP channel opening could reduce the rate of ATP loss, and decrease the rate of adenine nucleotide degradation so that ADP is available for phosphorylation. The results from our previous studies also showed that activation of mitoKATP channel inhibited the decrease of ADP-dependent state 3/4 oxygen consumption and respiratory control index injured by MPP+ in astrocytes. Consequently, activation of mitoKATP channel by KCOs preserves an adequate supply of ATP (Garlid et al, 2003). Glutamate uptake into astrocytes is driven by the electrochemical gradients of Na+ and K+, with a stoichiometry of 3 Na+, and 1 H+ in and 1 K+ out with the uptake of each glutamate anion (Sung et al, 2003). The resulting increase in [Na+]i might be corrected by a cycle of the Na+ pump and ATP consumption. Therefore, mitoKATP channel activation by KCOs might increase the supplementation of ATP, and thereby enhancing glutamate uptake by GluTs.

Activation of mitoKATP channel induces K+ influx, which partially restores the potential back toward basal level and enhances mitochondrial respiration by mitochondria matrix swelling, and subsequently yields a mild uncoupling (Starkov, 1997). Matrix swelling triggered by mitoKATP channel activation changes the geometry between the inner and outer mitochondrial membranes, and thus decreases intermembrane space, inducing faster electron transport and decreases in ROS release (Starkov, 1997; Jezek et al, 2004). Thus, mitoKATP channel acts as a potent governor to mitochondrial ROS release (Brookes et al, 2004). It has been demonstrated that oxidative stress can reduce glutamate transporter activities, and antioxidants can reverse the impairment of glutamate uptake (Jayakumar et al, 2006). The data from our laboratory also showed that antioxidants glutathione and N-acetylcysteine could improve glutamate uptake in lipopolysaccharide-treated astrocytes (Zhou et al, 2006). In the present study, treatment with MPP+, an inhibitor of mitochondrial complex I, that can induce ATP depletion, increased ROS in astrocytes, whereas diazoxide and iptakalim could inhibit ROS production. Taken together, inhibition of ROS production might be responsible for improvement of glutamate uptake by KCOs.

Many studies have suggested that glutamate uptake may be regulated by a variety of signaling pathways (Chaudhry et al, 1995; Rothstein et al, 1994), and that protein phosphorylation plays important roles in glutamate uptake and the expression and redistribution of GluTs (Rothstein et al, 1994; Gegelashvili and Schousboe, 1997). However, the effects of protein phosphorylation on the function of GluTs remain controversial because different transporters could be regulated in opposite directions by the same signaling pathway (Chaudhry et al, 1995; Danbolt, 2001). In the present study, the primary astrocyte cultures were prepared from cerebral cortex of rats where GLT-1 represents the predominant glial transporter (O’Shea et al, 2006; Marco and Michael, 2004). It has been reported that activation of PKC caused internalization of GLT-1 and decreased in cell-surface expression of GLT-1 (Kalandadze et al, 2002); PKC activation also inhibited the traffic process of GLT-1 whereby PKC phosphorylation reduced the function of GLT-1 (Ganel and Crosson, 1998). The present study showed that treatment with diazoxide and iptakalim inhibited PKC phosphorylation, thus it is reasonable to presume that inhibiting PKC phosphorylation by KCOs facilitates glutamate uptake. However, the detailed pathways involved in KCO inhibiting PKC activity deserve to be defined.

In summary, it is the first study to demonstrate that activation of KATP channel, especially mitoKATP channel, improves the function of GluTs in astrocytes. Reducing ROS production and downregulating PKC phosphorylation contribute to the facilitative effects of KCOs on glutamate uptake. Therefore, this study not only reveals a novel pharmacological profile of KCOs as regulators of GluTs, but also provides a new strategy for neuroprotection. Notably, Ipt, a novel and blood–brain barrier permeable KCO, may be a promising therapeutic agent for neurodegenerative diseases such as PD.