Regulator of G protein signaling (RGS) family members, such as RGS2, interact with Gα subunits of heterotrimeric G proteins, accelerating the rate of GTP hydrolysis and attenuating the intracellular signaling triggered by the G protein-coupled receptor-ligand interaction. They are also reported to regulate G protein-effector interactions and form multiprotein signaling complexes. Ischemic stress-induced changes in RGS2 expression have been described in astrocytes, and these changes are associated with intracellular signaling cascades, suggesting that RGS2 upregulation may be an important mechanism by which astrocytes may regulate RGS2 function in response to physiological stress. However, information on the functional roles of stress-induced modulation of RGS2 protein expression in astrocyte function is limited. We report the role of ischemic stress in RGS2 protein expression in rat C6 astrocytoma cells and primary mouse astrocytes. A marked increase in RGS2 occurred after ischemic stress induced by chemicals (sodium azide and 2-deoxyglucose) or oxygen-glucose deprivation (OGD, real ischemia). RGS2 mRNA expression was markedly enhanced by 1 h of exposure to chemical ischemia or 6 h of OGD followed by 2 or 6 h of recovery, respectively. This enhanced expression in primary astrocytes and C6 cells was restored to baseline levels after 12 h of recovery from chemically induced ischemic stress or 4–6 h of recovery from OGD. RGS2 protein was also significantly expressed at 12–24 h of recovery from ischemic insult. Ischemia-induced RGS2 upregulation was associated with enhanced apoptosis. It significantly increased annexin V-positive cells, cleaved caspase-3, and enhanced DNA ladder formation and cell cycle arrest. However, a small interfering RNA (siRNA)-mediated RGS2 knockdown reversed the apoptotic cell death associated with ischemia-induced RGS2 upregulation. Upregulated RGS2 was significantly inhibited by SB-203580, a p38 MAPK inhibitor. Rottlerin, a potent inhibitor of PKCδ, completely abrogated the increased RGS2 expression. We also examine whether ischemia-induced RGS2-mediated apoptosis is affected by siRNA-targeted endogenous PKCδ downregulation or its phosphorylation. Although RGS2 upregulation was not affected, siRNA transfection significantly suppressed endogenous PKCδ mRNA and protein expressions. Ischemia-induced PKCδ phosphorylation and caspase-3 cleavage were dose dependently inhibited by PKCδ knockdown, and this endogenous PKCδ suppression reversed ischemia-induced annexin V-positive cells. This study suggests that ischemic stress increases RGS2 expression and that this condition contributes to enhanced apoptosis in C6 cells and primary astrocytes. The signaling it follows may involve PKCδ and p38 MAPK pathways.
- protein kinase C
- p38 MAPK
astrocytes, the most abundant glial cell type in the brain, provide metabolic and trophic support to neurons and modulate synaptic activity (52). It has been reported that astrocytes play an important role in the metabolism of neurotransmitters and in the underlying mechanisms of neurotransmitter uptake that protect neurons from injury due to glutamate toxicity (41). Accordingly, Takuma et al. (53) reported that impairments in these astrocyte functions can critically influence neuronal survival. Dorf et al. (9) further reported that astrocytes act as inflammatory response mediators and are implicated in the pathophysiology of neurodegenerative diseases. Recent studies show that astrocyte apoptosis may contribute to the pathogenesis of many acute and chronic degenerative disorders, such as cerebral ischemia, Alzheimer's disease, and Parkinson's disease (27, 52).
Because of the significance of G protein-coupled receptor (GPCR)-mediated signaling in astrocytes, which surround and protect neuronal cells, improved understanding of regulator of G protein signaling (RGS) regulatory mechanisms is likely to be important. Regulators of RGS proteins are important in regulation of GPCR-induced signaling by enhancing GTP hydrolysis, thereby inactivating the G protein activation cycle. A large body of evidence revealed >30 RGS and RGS-like mammalian family members, which have been classified into 6 distinct subfamilies on the basis of amino acid sequence, overall protein structure, and functions within subfamilies (14, 18, 40). RGS proteins are thus defined by a shared 120- to 130-amino acid domain that binds directly to activate Gα subunits and enhances its intrinsic GTPase activity (1, 14, 18). However, in their review, Abramow-Newerly et al. (1) indicate that RGSs are not simply GTPase-activating proteins, but are also involved in regulation of G protein-effector interactions and interactions with multiprotein signaling complexes. RGS proteins can also bind to proteins other than Gα, such as GPCRs (e.g., muscarinic, dopaminergic, adrenergic, angiotensin, interleukin, and opioid receptors) and effectors (e.g., adenylyl cyclase, G protein-activated inwardly rectifying K+ channels, phosphodiesterase-γ, PLC-β, and Ca2+ channels), as well as RGS binding partners (e.g., Gαi-interacting protein COOH terminus, spinophilin, and 14-3-3), that underlie the formation of signaling scaffolds or govern RGS protein availability and/or activity.
Grant et al. (12) reported that RGS mRNA expression is modulated by GPCR ligands and second messengers and is highly regulated in vivo and in vitro. In the brain, RGS expression is regulated by physiological and pharmacological agents. This seems especially pertinent for RGS2, inasmuch as its expression is rapidly and substantially increased in response to neuronal activity (21), second messengers (37, 50), and certain cellular stressors, such as heat shock and oxidative stress (51, 58).
Ingi et al. (21) and Zmijewski et al. (58) reported that stressors have been shown to transiently increase RGS2 mRNA. It was further reported that intracellular signaling mechanisms that mediate RGS2 upregulation are system-dependent and have been reported to include the cAMP-adenylyl cyclase-PKA (17, 19, 34, 54, 59), Ca2+ (36, 37), and PKC (12, 36, 50, 51) pathways.
Overall, the rapid and robust changes induced by a variety of agents associated with several intracellular signaling cascades suggest that modulation of RGS2 expression may be an important mechanism by which cells regulate RGS2 function. However, information on the functional roles of ischemia-induced modulation of RGS2 protein expression in astrocyte function is limited. This study, therefore, evaluates the effects of ischemic stress on the level of RGS2 protein expression observed, the cellular apoptosis involved, and the possible paths followed. We found that ischemic stress-upregulated RGS2 expression may be linked to enhanced apoptosis involving PKCδ and p38 MAPK pathways.
MATERIALS AND METHODS
RGS2 plasmid constructs were purchased from University of Missouri-Rolla cDNA Resource Center (Rolla, MO); antibodies to cleaved caspase-3, phosphorylated (Thr505) PKCδ, and β-actin from Cell Signaling (Beverly, MA); antibodies against RGS2 and PKCδ (C-17) from Santa Cruz Biotechnology (Santa Cruz, CA); annexin V antibody and propidium iodide (PI) from BD Biosciences Pharmingen (San Diego CA); sodium azide, 2-deoxy-d-glucose, HEPES, rottlerin, H89, and H7 from Sigma (St. Louis, MO); and PD-98059, SB-203580, and SP-600125 from Tocris (Ellisville, MO). The enzyme chemiluminescence kit was obtained from Supex (Pohang, Korea). All other reagents are of chemical grade.
Rat C6 glioma cells (American Type Culture Collection, Rockville, MD) were cultured at 37°C in 5% CO2-95% air in DMEM (GIBCO BRL) containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin.
Highly enriched astroglial primary cultures were prepared as previously described (45). Briefly, forebrain cortices of newborn (1-day-old) mouse pups were dissected, and the meninges and pia mater were carefully removed. The tissue was trypsinized, mechanically triturated, and seeded in tissue culture flasks. Cultures were grown at 37°C in a humidified atmosphere with 5% CO2 in DMEM supplemented with 10% FBS, 2 mM glutamine, penicillin (100 U/ml), and streptomycin (100 μg/ml). The culture medium was changed after 5 days and then every 3 days. When the cells reached confluence after 12–14 days, the flasks were shaken at 250 rpm for 24 h to remove nonadherent cells. The cells were dissociated using trypsin-EDTA (GIBCO BRL), centrifuged at 1,000 rpm for 10 min, and replated, and then experiments were performed. The purity of astrocytes (95%) was determined by immunocytochemical staining with glial fibrillary acidic protein, an astrocytic marker. All experimental procedures and protocols involving the use of laboratory animals were reviewed and approved by the Ethics Committee of our university.
RGS2 Plasmid (cDNA) and Small Interfering RNA Transfection
Astrocytes and C6 cells were transfected with hRGS2-pEGFP-C1 plasmid or mock transfected using WelFect-M GOLD transfection reagent (WelGENE) in accordance with the manufacturer's instructions. Transfection efficiency (32–36%) was monitored using fluorescein-activated cell sorting analysis for green fluorescent protein-tagged RGS2 plasmids.
Small interfering RNA (siRNA) duplex oligonucleotides directed against RGS2 and PKCδ were synthesized and obtained from Bioneer. For siRNA transfection, C6 cells and primary astrocytes were seeded at 1 × 106 cells/well using 60-mm cell culture plates (Falcon, BD Biosciences) in antibiotic-free medium. On the following day, cells were transfected with an si-control-siRNA (siCON) or an siRNA targeting RGS2 or PKCδ using the X-tremeGENE siRNA transfection kit (Roche) in serum- and antibiotic-free media, in accordance with the manufacturer's instructions. The response of the cells to treatments was measured 48 h later. Gene silencing was confirmed by real-time (quantitative) PCR (qPCR) analysis of RGS2 and PKCδ expression levels in siCON- vs. siRGS2-transfected wells. The following oligonucleotides were used: RGS2 [5′-CGGGAGAAAAUGAAGCGGA-3′ (sense) and 5′-UCCGCUUCAUUUUCUCCCG-3′ (antisense)], control [5′-CCUACGCCACCAAUUUCGU-3′ (sense) and 5′-ACGAAAUUGGUGGCGUAGG-3′ (antisense)], PKCδ [5′-CUCACAGUACUUCCUCUGU-3′ (sense) and 5′-ACAGAGGAAGUACUGUGAG-3′ (antisense)], and control [5′-GUCCCUUUCACCCGUAUUA-3′ (sense) and 5′-UAAUACGGGUGAAAGGGAC-3′ (antisense)].
Induction of Ischemia
Chemical ischemia was induced as described previously with slight modifications (20). Briefly, when primary astrocytes and C6 glioma cells reached 80% confluence, they were exposed to oxidative stress. Cells were washed with PBS once and then exposed to chemical ischemic buffer containing 10 mM sodium azide and 10 mM 2-deoxyglucose in HEPES-buffered saline (in mM: 120 NaCl, 5 KCl, 0.62 MgSO4, 1.8 CaCl2, and 10 HEPES, pH 7.4) for 1 h at 37°C in a 5% CO2-95% air incubator. Sodium azide, an inhibitor of oxidative phosphorylation, is used to induce chemical anoxia, and 2-deoxyglucose is used to inhibit glycolysis. To validate the result, cells were, in parallel, incubated with oxygen-glucose deprivation (OGD) medium (conventional; in mM: 116 NaCl, 5.4 KCl, 0.8 MgSO4, 1 NaH2PO4, 26.2 NaHCO3, 1.8 CaCl2, 0.01 glycine, and 0.6 l-arginine) for 6 h. The low (1–5%) oxygen systems of the incubator were established. After induction of ischemia, the medium was replaced with normal medium during recovery. Whenever necessary, pharmacological inhibitors were applied 30 min before induction of ischemia.
RNA Isolation and RT-PCR
Total cellular RNA was extracted using Easy Blue kits (iNtRON Biotechnology) according to the manufacturer's instructions and stored at −70°C until use. One microgram of RNA was annealed with poly(dT)18 for 10 min at 70°C and cooled for 5 min on ice, reverse transcribed using RT premix (Bioneer) in 20 μl of reaction mixture containing 5× buffer (10 mM dNTP, 0.1 mM dithiothreitol, and 2 U of murine leukemia virus reverse transcriptase), and run for 90 min at 42.5°C using a thermal cycler. The reactions were terminated at 95°C for 5 min to inactivate the reverse transcriptase. The semiquantitative PCR (RT-PCR) was performed using aliquots of cDNA obtained from RT reaction in a PCR premix (Bioneer) containing a 10× buffer [10 mM Tris·HCl (pH 8.3), 50 mM KCl, 0.1% Triton X-100, 0.25 mM dNTP, 25 mM MgCl2, and 1 U of Taq polymerase]. Amplification conditions were as follows: 5 min of predenaturation at 95°C followed by 30–35 cycles consisting of denaturation for 45 s at 95°C, annealing for 45 s at 55–60°C, and elongation for 45 s at 72°C. Final extension at the end of the cycle was done for 10 min at 72°C.
The PCR products were electrophoresed in 1.3% agarose gel stained with ethidium bromide and visualized using Eagle Eyes image analysis software (Stratagene, La Jolla, CA). The intensity of band densities for RGS mRNA expression levels was normalized for the corresponding GAPDH, a housekeeping gene used as an RNA internal standard, and ratios were compared. The sequence of primers used was as follows: GAPDH [5′-CACTCACGGCAAATTCAACGGC-3′ (forward) and 5′-CCTTGGCAGCACCAGTGGATGCAGG-3′ (reverse)], mouse RGS2 [5′-TGAAGCGGACACTCTTAAAG-3′ (forward) and 5′-TGCTGTTTGCTTTTCTTGCC-3′ (reverse)], rat RGS2 [5′-TCTGGTTGGCTTGCGAAGAC-3′ (forward) and 5′-TCTCTTTGGGAGCTTCCTTC-3′ (reverse)], RGS9 [5′-GAAGCCTGTGAGGACCTGAAGTACG-3′ (forward) and 5′-AGGAGGCAGCTCCTTTTTGAGTTG-3′ (reverse)], RGS16 [5′-TCAGAAGATGTACTGGGATGGAG-3′ (forward) and 5′-GCGATAAGCTGGTGATTGAGGAAG-3′ (reverse)], and RGS3 [5′-CGGACTCTCCTGCTGTCGGAGGA-3′ (forward) and 5′-TGGGTGGGAGGTCTTGTCCTACA-3′ (reverse)].
RGS2 and PKCδ mRNA expression was quantified by qPCR in a 25-μl total reaction mixture using Power SYBR Green PCR Master Mix (Applied Biosystems). DNA amplification and detection were carried out in a real-time PCR system (ABI 7500, Perkin Elmer Applied Biosystems, Foster City, CA). The cycle threshold (CT) values, corresponding to the PCR cycle number at which fluorescence emission in real time reaches a threshold above the baseline emission, were determined using ABI Prism 7700 software. The qPCR conditions consisted of initial denaturation at 95°C for 10 min followed by amplification as follows: 40 cycles of denaturation at 95°C for 15 s and annealing at 60°C for 1 min. Standard curves were generated using fivefold serial dilutions of genomic DNA, the linearity of the resulting CT values was confirmed for RGS2 and PKCδ, and values were normalized to β-actin primer set values. The sequence of primers was as follows: mouse RGS2 [5′-TGAAGCGGACACTCTTAAAG-3′ (sense) and 5′-TGCTGTTTGCTTTTCTTGCC-3′ (antisense)], rat RGS2 [5′-TCTGGTTGGCTTGCGAAGAC-3′ (sense) and 5′-TCTCTTTGGGAGCTTCCTTC-3′ (antisense)], rat PKCδ [5′-CCAAGGAGTCCAAGGACATC-3′ (sense) and 5′-GCAGGTTCCAGTTGATAGTC-3′ (antisense)], mouse β-actin [5′-GACTCCTATGTGGGTGACGA-3′ (sense) and 5′-CCAGATCTTCTCCATGTCGT-3′ (antisense)], and rat β-actin [5′-CCTCTGAACCCTAAGGCCAA-3′ (sense) and 5′-AGCCTGGATGGCTACGTACA-3′ (antisense)].
DNA Fragmentation Assay
To assess whether ischemia induces cell death, DNA fragmentation assay was performed. The cells were washed with PBS and lysed in a lysis buffer [20 mM EDTA, 50 mM Tris · HCl (pH 7.4), and 0.5% Triton X-100] for 20 min at 4°C. The lysates were centrifuged at 13,000 rpm for collection of supernatant containing low-molecular-weight DNA. The supernatants were extracted with phenol-chloroform, and nucleic acids were precipitated in ethanol-containing sodium acetate, centrifuged at 13,000 rpm for 10 min, and suspended in 30 μl of RNase. Each sample was then electrophoresed on a 1.4% agarose gel. The agarose gel was incubated for ≥3 h at 37°C with 20 mg/ml RNase A (Sigma), stained with ethidium bromide, and visualized.
Annexin V staining.
Apoptosis was analyzed by flow cytometry, which measures cells positively stained with annexin V and PI. Astrocytes and C6 cells were plated onto a 60-mm dish, trypsin was added to loosen the cells from the plate, and the cells were harvested after the appropriate treatment periods. Briefly, cells were washed twice with ice-cold PBS and pelleted by centrifugation at 2,000 rpm for 10 min, and the cell pellets were resuspended in 1× annexin V binding buffer. To a 100-μl aliquot of the cell suspension, 10 μl of PI (50 μg/ml) and then 10 μl of annexin V [FITC-conjugated annexin V (BD Pharmingen)] were added, and the cells were incubated in darkness for 15 min at room temperature. Flow cytometry was performed on a FACScan flow cytometer (Becton Dickinson, San Jose, CA). Data from a total of 10,000 events were analyzed using CellQuest software (Becton Dickinson Immunocytometry Systems, San Jose, CA). The percentage of annexin V-positive, but PI-negative, cells was calculated to determine cells in the early stage of apoptosis.
Cell cycle analysis.
Ischemia-induced cell death was identified as a sub-G1 peak in the cell cycle. Cells were stimulated to synchronously enter the cell cycle by trypsinization and replating in serum-containing medium. Briefly, cells were washed with cold PBS and fixed in 70% ethanol at −20°C for 10 min, stained with 40 μg/ml of PI and 100 μg/ml of RNase A at 37°C for 30 min, and analyzed for different cell cycle phases.
Treated cells were lysed in a lysis buffer (PRO-PREP, iNtRON Biotechnology) containing 1 mM PMSF, 2 μg/ml aprotinin, 1 μg/ml leupeptin, 1 μg/ml pepstatin A, 2 mM sodium fluoride, and 1 mM sodium orthovanadate. Protein was then measured using the PRO-MEASURE assay kit (iNtRON Biotechnology), with bovine serum albumin as a standard. Equal amounts of protein (40 μg) per lane were boiled with 1% SDS and 1% β-mercaptoethanol for 5 min, separated on 8–10% SDS-polyacrylamide gels, and then transferred onto nitrocellulose membranes. The membrane was washed three times with TBS-Tween 20 and blocked for 1 h in blocking solution [10 mM Tris·HCl (pH 7.5), 150 mM NaCl, 0.1% Tween 20, and 5% nonfat dry milk] and then incubated with primary antibodies diluted with blocking solution overnight at 4°C. After they were washed three times in PBS + Tween 20, the membranes were incubated with secondary horseradish peroxidase-linked antibodies for 1 h at room temperature. Bound antibody was visualized using enhanced chemiluminescence (Supex).
The data were analyzed by one-way analysis of variance followed by Dunnett's post hoc test. All experiments represent at least three independent replications performed in triplicate, and values are means ± SE. The difference between treatment groups was compared by t-test. P < 0.05 was considered statistically significant.
Ischemia-Induced RGS2 Protein Upregulation in C6 Cells and Primary Astrocytes
Since G protein signaling is involved in the pathophysiology of the nervous system, we evaluated whether oxidative stress could modulate expression of RGS proteins in C6 cells and primary astrocytes. We observed that chemically induced ischemic stress or OGD increased endogenous RGS2 levels above baseline. However, the expression levels of RGS3, RGS9, and RGS16, which were constitutively expressed in C6 cells, were not affected by the ischemic stress (Fig. 1A). The kinetics of RGS2 gene expression were further examined. As shown in Fig. 1, A–E, an ischemia-induced increase in endogenous RGS2 mRNA expression was clearly detected starting at 0–1 h of recovery after ischemic insult, peaked (>5-fold) at 2 h (Fig. 1, B and C), and was sustained (∼3-fold) for up to 6 h in chemical ischemia. Similarly, in C6 cells and astrocytes subjected to OGD, the expression peaked (>3-fold) at 1 h (Fig. 1, D and E) and was sustained (>2-fold) for up to 2 h. Enhanced RGS2 expression was restored to basal level at 12 h of recovery after 1 h of chemical insult (Fig. 1, B and C) or at 6 h of recovery after OGD (Fig. 1, D and E).
For more accurate quantitation, ischemia-induced RGS2 mRNA upregulations in C6 cells and primary astrocytes were further analyzed by qPCR. The result obtained from qPCR analysis followed a trend similar to that obtained from RT-PCR. RGS2 was upregulated at 0 h, peaked at 1 h, was sustained for 2 h, and then returned to baseline levels after 4 h of recovery (Fig. 1, F and G).
Western blot analysis showed that RGS2 protein expression was markedly enhanced after 12–24 h of recovery from ischemic insult in primary astrocytes (Fig. 2A). To estimate the observed RGS2 protein expression levels above baseline, we transfected C6 cells with RGS2 plasmid construct in a dose-dependent manner. RGS2 protein expression was much higher in cells transfected with a higher dose of plasmid than in hypoxia/ischemia-exposed cells (Fig. 2B).
siRNA-Mediated RGS2 Knockdown
To assess functional consequences of selective downregulation of endogenous RGS2 on ischemia/hypoxia-induced RGS2 expression, we used RT-PCR to examine transfected targeting sequences for effective RGS2 RNA interference (Fig. 3A). siRNA transfection significantly and dose dependently inhibited RGS2 mRNA expressions in ischemia-exposed and unexposed C6 cells. For more accurate quantitation, we examined RGS2 siRNA-transfected cells by qPCR and observed a pattern of siRNA-mediated endogenous RGS2 downregulation in C6 cells (Fig. 3B) and astrocytes (Fig. 3C) similar to that demonstrated by RT-PCR.
To determine whether RGS2 protein and mRNA expression patterns are similar and to examine cDNA or siRNA transfection effects on the protein expression levels, we used Western blot analysis. The results confirmed higher (>3-fold) RGS2 protein expression in RGS2 plasmid-transfected cells and cells subjected to ischemic insult than in control cells (Fig. 3D, lanes 2, 4, and 5). However, the enhanced protein expression was greatly downregulated by siRNA transfection in stressed and normal C6 cells (Fig. 3D).
Ischemia-Induced RGS2 Upregulation Enhances Apoptosis
Since previous reports indicate that ischemia/hypoxia or OGD induces apoptosis in astrocytes (11, 31, 55) and RGS2 expression is coupled to cellular stress and cell cycle arrest (58, 60), we next used flow cytometry (for annexin V-positive cells and cell cycle analysis), DNA fragmentation assay (for apoptotic cell death), and Western blotting (for caspase-3 cleavage) to examine whether ischemia-induced RGS2 upregulation was associated with apoptosis. As shown in Fig. 4, A and B, 1 h of chemical treatment followed by 12 and 24 h of recovery significantly increased annexin V labeling in C6 cells and primary astrocytes. We further examined the extent of apoptosis caused by stress-induced RGS2 expression on OGD-treated or untreated and/or RGS2 plasmid-transfected C6 cells. Plasmid transfection increased annexin V labeling in untreated cells. This transfection condition approximated the ischemic stress-induced increase in annexin V-positive cell levels. Transfection also augmented OGD-induced annexin V labeling (Fig. 4, C and D). These results suggest the involvement of RGS2 induction in enhanced apoptosis. To identify whether siRNA-mediated endogenous RGS2 knockdown inhibits the increase in annexin V labeling, we transfected C6 cells with RGS2 siRNA and measured OGD-induced annexin V labeling. siRNA transfection completely reversed the percentage of annexin V-positive cells to control levels (Fig. 4, C and D). These results indicate that RGS2 upregulation is involved in stress-induced astrocyte apoptosis.
Induction of apoptosis was further examined by caspase-3 activity, DNA fragmentation, and flow cytometric cell cycle arrest analysis. Dose-dependent cleavage of caspase-3 was observed in RGS2 plasmid-transfected C6 cells and astrocytes (Fig. 5, A and B) and was correlated to the extent of ischemia-induced caspase-3 cleavage (Fig. 5C). Detection of caspase-3 cleavage was significant after 24 h of recovery from ischemic insult alone Fig. 5C, lane 2), weak after recovery from transfection alone (lane 3), and strong after recovery from transfection and ischemic insult (lane 4), suggesting that cell death potentiation may be caused by transfection and ischemic induction, which may be evidence for the association of ischemia-induced RGS2 upregulation with increased astrocyte apoptosis. Figure 6A shows that ischemia-induced RGS2 expression increases DNA fragmentation, further supporting the presence of apoptotic cell death (lane 2). In addition, flow cytometric cell cycle arrest analysis showed that the ratio of cells with apoptotic nuclei to cells with nuclei in the G1 phase (M1/M2) increased in stressed and/or transfected and stressed cells (Fig. 6, B and C).
Ischemia-Induced RGS2 Upregulation May Involve p38 MAPK and PKCδ Pathways
Since MAPKs, a specific class of serine/threonine protein kinases, are also involved in many cellular stressors in eukaryotes, we next assessed whether MAPKs participate in the regulation of ischemia-induced RGS2 overexpression in C6 cells. MAPK inhibitors such as PD-98059 (2′-amino-3′-methoxyflavone, 25 μM), an ERK1/2 inhibitor, SB-203580 (4-[5-(4-fluorophenyl)-2-[4-(methylsulfonyl) phenyl]-1H-imidazole-4-yl] pyridine, 25 μM), a p38 MAPK inhibitor, and SP-600125 (anthra1–9-cd]pyrazol-6(2H)-one, 25 μM), a JNK inhibitor, were examined. Only SB-203580 suppressed the ischemia-induced RGS2 mRNA overexpression by ∼40% (Fig. 7, B–D, lanes 4 and 5). However, PD-98059 and SP-600125 had no significant effects on stress-induced RGS2 mRNA overexpression (Fig. 7B, lanes 3 and 4). Furthermore, FACS analysis of cell cycle arrest showed that the apoptotic cell population increased (∼5-fold) under ischemic conditions compared with control (Fig. 7, E and F).
Since the serine/threonine protein kinases, such as PKA and PKC, are also involved in cellular stress-induced apoptosis (28, 33), we used inhibitors to examine whether the PKA and/or PKC pathways are involved in ischemia-induced RGS2 mRNA overexpression. Rottlerin, a specific inhibitor of PKCδ (5 μM), completely blocked the ischemia-induced RGS2 mRNA upregulation (Fig. 7A, lane 5, and Fig. 7, C and D). However, staurosporine (1 μM), a broad-spectrum PKC inhibitor, had a slight, but nonsignificant, effect on RGS2 mRNA (Fig. 7A, lane 6). On the other hand, the PKA inhibitors H89 (1 μM) and H7 (1 μM) had no effect on RGS2 expression (Fig. 7A, lanes 3 and 4). PKCδ has been reported to be a potent large-conductance Ca2+-activated K+ (BKCa) channel activator; thus the effect of NS-1619 (a BKCa channel opener) was examined. However, NS-1619 did not affect the enhanced RGS2 expression (data not shown). Since recent reports indicate that rottlerin blocks other kinase and nonkinase proteins and uncouples mitochondria (49), we examined whether ischemia-induced RGS2-linked apoptosis is affected by siRNA-mediated endogenous PKCδ downregulation or its phosphorylation. qPCR analysis (Fig. 8A) shows that although ischemia-induced RGS2 upregulation was not affected, siRNA transfection significantly suppressed (>50%) endogenous PKCδ expression. Similarly, its protein expression was also markedly inhibited (Fig. 8B). Furthermore, ischemia-induced PKCδ phosphorylation and caspase-3 cleavage were dose dependently inhibited by PKCδ knockdown (Fig. 8C), and this siRNA-mediated PKCδ downregulation reversed ischemia-induced annexin V labeling (Fig. 8E). These findings suggest that the apoptotic processes linked to ischemia-induced RGS2 upregulation may be mediated by PKCδ.
In primary astrocytes, upregulated RGS2 mRNA was also completely abolished by rottlerin and significantly attenuated by SB-203580 (data not shown). These results suggest that PKCδ and p38 MAPK may likely be involved in RGS2 mRNA upregulation following ischemia in C6 cells and primary astrocytes.
Astrocytes, the most abundant glial cell types in the brain, provide metabolic and trophic support to neurons, are involved in transmitter reuptake and release, modulate synaptic activity, regulate ion concentration, and repair nervous system injury. Accordingly, impairment of these astrocyte functions can critically influence neuronal survival. A large body of evidence suggests that astrocyte apoptosis caused by chemicals, glucose deprivation, or ischemic/hypoxic or oxidative stress may contribute to the pathogenesis of many acute and chronic neurodegenerative disorders, such as cerebral ischemia, Alzheimer's disease, and Parkinson's disease. On the other hand, cellular stresses, including apoptotic agents such as camptothecin, have been reported to increase the expression of RGS2 mRNA, suggesting that induction of RGS2 by cellular stress may be an important mechanism by which neurons and glial cells regulate RGS2 function. However, information on the roles of RGS2 expression levels in astrocytes in response to physiological stress is limited.
In a previous study employing inhibitors (25), we found that isoproterenol upregulates RGS2 mRNA expression in a protein tyrosine kinase-, PKC-, and p38 MAPK-dependent manner in C6 glioma cells and astrocytes. In this study, we determined that treatment of C6 glioma cells and primary astrocytes with chemical stress or OGD significantly increases RGS2 mRNA and protein levels above baseline. The kinetics of mRNA expression patterns show upregulated transcripts within 0 h of recovery from ischemic insult, with peak expressions at 1 or 2 h of recovery, depending on the type of stress (chemical or OGD), and return to basal levels at 6–12 h. These findings add further to the previous reports that RGS2 mRNA expressions are highly regulated in the brain by physiological and pharmacological agents that cause a rapid and transient increase in RGS2 mRNA levels. In vivo studies indicate that stimuli that evoke neuronal plasticity, including electroconvulsive seizures in neurons of the hippocampus, cortex, amygdala, and striatum (21), and drugs, such as cocaine, amphetamine, or methamphetamine, in the caudate putamen (5, 6) rapidly increase mRNA levels. In vitro reports also indicate that RGS2 mRNA is upregulated by stimuli such as prostaglandins E1 and E2 in U937 cells (3), carbachol or heat shock muscarinic receptor activation in neuroblastoma SH-SY5Y cells, or muscarinic and β-adrenergic receptor activation in astrocytoma 1321N1 cells (50, 51). In addition to mRNA expression, we showed that RGS2 protein expression is markedly increased after 12–24 h of recovery from ischemic stress in C6 glioma cells and astrocytes, suggesting that ischemia-induced RGS2 upregulation may be involved at the transcriptional and translational levels. This conclusion was further confirmed, in that whereas plasmid transfection approximated the level of ischemia-induced RGS2 upregulation, siRNA-mediated knockdown significantly suppressed RGS2 expression at mRNA and protein levels in C6 glioma cells and astrocytes subjected to ischemia. Similar enhanced RGS2 protein expressions have been reported in response to mechanical stress in periodontal ligament cells (44), angiotensin II stimulation in H295R human adrenocortical cells (39), mononuclear cells in Bartter's syndrome (8), and hypertrophic cardiomyocytes (57) and breast cancer cells (47), in which functional consequences of selective downregulation of endogenous RGS2 was assessed using siRNA.
Our findings, for the first time, show that increased RGS2 expression above baseline in C6 glioma cells and astrocytes in response to ischemic stress is linked to various steps of cell apoptosis. In cells subjected to ischemic insult, caspase-3 activity was enhanced, annexin V labeling was increased, DNA ladder formation was triggered, and the cell cycle was arrested. In addition, the extent of apoptotic cell death (caspase-3 activity and percent annexin V-positive cells) caused by ischemia-mediated RGS2 upregulation was dose dependently related to that caused by plasmid transfection. Furthermore, siRNA transfection targeted to RGS2 downregulation reversed all apoptotic effects to the basal level. Thus the study shows that an increase in RGS2 expression in C6 glioma cells and astrocytes is preceded by the development of apoptosis and may therefore contribute to the pathophysiological importance of the events triggering apoptosis. We, however, do not know, at this stage, whether the link of upregulated RGS2 activity to apoptotic events is related to or distinct from its established RGS domain function, i.e., the ability to bind to and inactivate Gα subunits of G-protein-mediated signals by enhancing GTPase activity. Beside RGS2 domain functions, recent reports indicate that RGS2 has a potential to integrate multiple signaling networks. For example, the amino-terminal domain of RGS2 was shown to interact with and regulate three different effector proteins: adenylate cyclase (42), tubulin (15), and the cation channel TRPV6 (46). The growing list of RGS2-interacting partners may be through the amino-terminal domain of RGS2, which comprises several short effector protein interaction motifs and adopts distinct structures to bind various targets (16). These studies further suggest that RGS2 is a key point of integration for multiple intracellular signaling pathways, and they highlight the role of RGS proteins as dynamic, multifunctional signaling centers that coordinate a diverse range of cellular functions. In this study, RGS2 induces apoptosis, only at a higher level of expression. Thus it is also possible that RGS2-induced apoptosis may be an indirect effect, i.e., due to effects of nonphysiological levels of RGS2 on processes whose proper functions are required to maintain cell survival.
Consistent with the present finding, Jin et al. (22) reported that overexpression of RGS5 in human umbilical vein endothelial cells induced apoptosis with increased activation of caspase-3 and Bax-to-Bcl-2 ratio. They concluded that RGS5 is a novel hypoxia-inducible factor-1α-dependent, hypoxia-induced gene that is involved in the induction of endothelial apoptosis. The link between nuclear RGS protein and apoptosis was first reported by Dulin et al. (10), who indicate that overexpression of the truncated form RGS3T, but not RGS3, induces apoptotic cell death. They suggest that nuclear localization of RGS3T is critical for RGS3T-induced cell death with an uninvestigated underlying mechanism.
Since several reports indicate that the intracellular signaling mechanisms that mediate RGS2 upregulation are system-dependent and include the adenylate cyclase-cAMP-PKA-Ca2+ (19–21, 34) or PKC (26–28, 30, 33, 35) pathway, this study examined the possible mechanisms that can be used to follow apoptotic events.
Rottlerin, a specific inhibitor of PKCδ, completely abolished the ischemia-induced RGS2 mRNA upregulation, whereas staurosporine, a broad-spectrum PKC inhibitor, exhibited a slight, but nonsignificant, effect. However, the PKA inhibitors H89 and H7 had no significant effects on ischemia-evoked RGS2 expression, suggesting that PKCδ, but not PKA, may be an effector molecule that can be involved in mediating the observed apoptotic cell death. Since recent reports indicate that rottlerin has several nonselective biological activities, such as blockade of other kinase and nonkinase proteins, activation of Ca2+-sensitive K+ channels, uncoupling of mitochondria, and effects associated with uncoupling events (49), we employed siRNA-targeted endogenous PKCδ downregulation and evaluated PKCδ mRNA and protein suppression. Our findings confirm that the events of apoptotic cell death, i.e., PKCδ phosphorylation, caspase-3 cleavage, and increased annexin V labeling, caused by ischemia-induced RGS upregulation are completely reversed by siRNA-mediated endogenous PKCδ knockdown. Thus these results suggest that the apoptotic processes linked to ischemia-induced RGS2 upregulation may be mediated by PKCδ activation. A number of in vitro and in vivo studies have shown that the activation of PKCδ is a critical mediator of apoptosis (2, 4, 13, 30, 35). Several lines of evidence support the present findings, in that the ability of PKCδ to mediate apoptosis is intimately linked to its ability to stimulate G1 cell cycle progression and subsequent arrest in the S phase, an event that triggers caspase-dependent apoptotic cell death (48), which is associated with the caspase-3-dependent cleavage of PKCδ (56). PKCδ is activated by numerous apoptotic stimuli, including genotoxins (38), oxidative stress (29), and death receptors (24). Conversely, inhibition of PKCδ with rottlerin (38) or expression of kinase dead PKCδ (PKCδKD) (32) inhibits apoptosis induced by a variety of stimuli. Inhibitors of PKCδ and expression of PKCδKD inhibit apoptotic events, including caspase activation and DNA fragmentation, as well as loss of mitochondrial membrane potential.
Yoshida et al. (55) reviewed signaling pathways that may participate directly or indirectly in the apoptotic effect of PKCδ and may involve members of the MAPK superfamily that are activated in response to various extracellular stimuli and stress. In the present study using inhibitors, p38 MAPK, but not JNK and ERK pathways, appears to be involved in ischemia-induced RGS2 upregulation and apoptosis. These findings suggest that ischemia-induced RGS2 expression and cell death may be linked to signaling pathways that involve p38 MAPK and PKCδ and may play significant roles in stress-induced apoptosis. Similar reports implicated MAPK pathways in cellular stress, such as oxidative stress and chemical toxicity, which lead to apoptosis signaling (7, 23, 43). RGS5-induced apoptosis in human umbilical vein endothelial cells was reported to activate p38 signaling (22), suggesting that RGS5 could be an important target for apoptotic therapy.
In conclusion, the main findings of this study provide evidence that 1) ischemic stress induces RGS2 protein upregulation, 2) increased RGS2 expression above the basal level is linked to apoptotic cell death in response to physiological stress, 3) the signaling pathways of apoptosis may involve PKCδ and p38 MAPK, and 4) RGS2 and PKCδ gene knockdown reverses events of apoptosis and the signaling pathways it follows. Finally, our data suggest a potential role for astrocytes in regulation of molecular mechanisms of RGS2 function in response to physiological stress that cause apoptosis and neurodegenerative disorders.
This work was supported by the National Research Foundation (ROI-2007-000-20609-0).
No conflicts of interest are declared by the author(s).
- Copyright © 2010 the American Physiological Society