Abstract

Glutamate neurotoxicity, which can result in both neuronal damage via c-Jun N-terminal kinase (JNK) /c-Jun pathway and increased GluR6 SUMOylation, has been related to hypoxia or ischemia. Here we studied the effects of glutamate on GluR6 SUMOylation status and the impact that these modifications have on c-Jun activity. We showed that SUMOylation of GluR6 at lysine 886 (K886) could be induced by glutamate treatment and occurred in a rapid manner; however, induced mutation at this site would block SUMOylation in vitro. The activation of c-Jun and PC12 cell death significantly increased after inhibition of SUMO pathways in cells treated with glutamate. We further demonstrated that Ubc9 played a functional role in GluR6 SUMOylation and that the C93S mutant of Ubc9 abrogated SUMO-1 conjugation activity. In conclusion, SUMOylation of GluR6 has key roles in the regulation of glutamate neuronal excitotoxicity. These data reveal another role of SUMO in protection against cell death.

Keywords: GluR6, glutamate, SUMOylation, c-Jun phosphorylation, cell death, neurotoxicity
Introduction

Glutamate is a major excitatory neurotransmitter in the central nervous system [1] as well as a potent neurotoxin that may lead to the death of nerve cells. Under pathological conditions, such as ischemia and hypoxia, glutamate is elevated globally in the brain, and its excessive presence can induce neurotoxicity [2, 3]. Glutamate cytotoxicity has been attributed to excitatory actions through three types of excitatory amino acid receptors, including N-methyl-D-aspartate receptors (NMDAR), alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPAR), and kainate receptors (KAR) [3, 4]. KAR subunit GluR6 has been shown to play an important role in neuronal death induced by cerebral ischemia [5]. Previous studies also showed that kainite (KA) can influence the assembly of the GluR6-PSD95-MLK3 signaling module and can subsequently activate signaling pathways downstream of JNK, ultimately resulting in neuronal cell death [6-8].

The small ubiquitin related modifier (SUMO) is a novel post-translational protein [9] that contains 101 amino acid residues and modifies other proteins through the SUMOylation pathway. SUMOylation usually occurs on lysine residues within the consensus SUMOylation motif (ΨKXE/D), which consists of an aliphatic amino acid (Ψ) next to the lysine residue to be modified (K), followed by another amino acid (X) of any type and an acidic side chain (E/D) [10, 11]. This post-translational modification is catalyzed by an enzymatic cascade, which requires an activation enzyme (E1) [12], a conjugase (E2) [13] , and for most substrates, a ligase (E3) [14, 15]. It has been reported that KAR subunit GluR6 is a SUMO target and the SUMO-conjugating enzyme Ubc9 and the SUMO ligase PIAS3 selectively bind to the C-terminal domain of GluR6 in yeast two-hybrid assay [16]. Recently, an Ubc9 fusion-directed SUMOylation (UFDS) system is developed; it strongly increases the efficiency of SUMOylation of a specific substrate protein in vivo in SUMOylation sites fused to Ubc9. UFDS is independent of SUMO ligases [17]; hence, we applied UFDS for the identification of GluR6 SUMOylation.

Interestingly, GluR6 exhibits low levels of SUMOylation under resting condition but is rapidly SUMOylated in response to the agonist activation by glutamate. Wild type GluR6, but not the mutated K886R GluR6, is endocytosed in response to kainate in COS-7 cells. Furthermore, electrophysiological recordings in hippocampal slices demonstrate that KAR- mediated excitatory postsynaptic currents are decreased by SUMOylation and enhanced by deSUMOylation [16]. Based on these analyses, we hypothesized that SUMOylation might downregulate the surface expression of GluR6, which, in turn, would reduce excitotoxicity and subsequent cell death dependent on a JNK/c-Jun activation pathway. To confirm the hypothesis, we transfected both HEK293T cells and PC12 cells with either wild type GluR6 or mutated GluR6, and then investigated the changes in phosphorylation level of c-Jun in HEK293T cells and PC12 cells survival evoked by glutamate.

Materials and Methods

Plasmid Constructs. The plasmid PcDNA3.1-GluR6, which contains a 6×myc tag at the N-terminus, was kindly provided by Henley JM (University of Bristol, UK). Since SUMOylation of GluR6 is hardly detectable under resting conditions in vitro, we used UFDS to enhance GluR6 SUMOylation and constructed the plasmids PcDNA3.1- GluR6-myc-Ubc9, PcDNA3.1-GluR6 (K886)-myc-Ubc9, and PcDNA3.1-GluR6-myc-Ubc9 (C93S). To fuse the coding sequences of Ubc9 and its mutant Ubc9 (C93S) to the N-terminus of GluR6 or GluR6 (K886), Ubc9 and its mutant Ubc9(C93S) fragments were amplified by standard PCR methods with a forward primer containing a Kpn1 site before the start codon and a reverse primer  inserting a stop codon followed by a Sean site. Amplified fragments were digested and inserted into the PcDNA3.1 vector. The full-length GluR6 or GluR6 (K886) fragments amplified from PcDNA3.1-GluR6 with a forward primer containing an XhoI site before the start codon and a reverse primer inserting a KpnI site were digested and inserted into the PcDNA3.1-Ubc9. The plasmid PcDNA-GluR6 (K886)-myc and Ubc9 (C93S) were generated by site-directed mutagenesis with commercially synthesized mutagenic primers (Invitrogen Biotechnology, China). The predicted SUMOylation sites at the K886 residue of GluR6 was changed to arginine (R), which prevents SUMOylation [13] D.F. Van, E.L. Delvaux, d. Van and V.M.V. Chavez, Repression of the transactivating capacity of the oncoprotein PLAG1 by SUMOylation, J. Biol. Chem.  (2004), pp. 36121–36131.[16]. Mutation of the cysteine (C) residue to serine (S) at the active site of Ubc9 prevents the formation of a thioester bond, rendering the mutant protein incapable of SUMO conjugation [18]. All mutants were confirmed by DNA sequencing analysis (Invitrogen Biotechnology, China).

Cell Lines and Culture Conditions. Human embryonic kidney (HEK293T) cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco) with 2mM L-glutamine, 10% fetal bovine serum (Gibco) at 37°C with 5% CO₂ humidified atmosphere. Cell medium was replace every 3 days. For transfection, cells at a number of 2–6×10⁵ were seeded on the 6cm culture dish for 24 h. The rat pheochromocytoma cell line (PC12), provided by Shanghai Life Scientific Institute of Chinese Academy of Science, were grown in DMEM supplemented with 10% FBS at 37 °C, 5% CO₂. The cells were cultured at an inoculation density of 1–3×10⁵ cells on 3cm dishes.

Transfection. When cells entered the exponential period growth, they were transfected with various subunit combinations by lipofection using 8ug of cDNA and 10 ul of LipoFectamine 2000 (Invitrogen, USA) per 6cm culture dish.

Treatments and viability assay. 24 h after transfection, HEK293T cells were treated with 4mM glutamate, 12mM glutamate (Sigma), or vehicle alone (normal medium). Cells were incubated for 10 min, 30 min, or 12 h prior to protein extraction. PC12 cell viability was assessed by cell count kit-8 (CCK-8) (DOJINDO Molecular Technologies, INC), which is based on combining WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-

(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt] and 1-methoxy PMS [19]. PC12 cells were seeded onto 96-well plates at a density of 2×10⁴cells/well. After incubation for 12h, cells were treated with various concentrations of glutamate for 12 h. At the end of the treatment, the medium was removed and then 100ul serum-free medium and 10µl of CCK-8 were added into each well of the 96-well assay plate, followed by 1 h of incubation. The absorbance of each well was measured at 450 nm.

Immunoprecipitation and immunoblotting. 500-1000 ug of solubilized protein prepared as above was incubated with 2-4 mg of mouse monoclonal anti-myc antibody (from Santa Cruz) overnight at 4°C and then with 30 ml protein G-agarose beads (Sigma) for 1-3 h at 4°C. Immunoprecipitates were washed four times with lysis buffer and proteins were eluted from the beads by boiling in sample buffer. The supernatants were separated on 10% SDS-PAGE gels and then electrotransferred to PVDF membranes. After blocking for 2 h in 10% non-fat milk with 0.1% Tween 20 (TBST), membranes were incubated overnight at 4 °C with primary antibodies in TBST containing 5% bovine serum albumin. Membranes were then washed with TBS buffer and incubated with secondary antibodies in 5% non-fat milk/TBST for 1 h. The immunoreactivity was detected using enhanced chemiluminescence (ECL). The ECL-exposed films were digitized, and densitometric quantification of immunoreactive bands were performed using ImageJ (Wayne Rasband, National Institutes of Health, Bethesda, MD, USA).

Statistical analysis. Each data point presented in the studies was collected from at least three individual experiments. All statistical analyses were carried out with Graphpad Prism software. The data were summarized and plotted as mean±SEM. The significance of activity changes was determined by the Student’s t-test. A probability value of less than 0.05 (P < 0.05) was accepted as a significant difference from the control or between two samples in comparison as indicated.

Results

SUMO-1 modified GluR6 on lysine 886

We identified that the KAR subunit GluR6 was a SUMO substrate and that K886 was the site of SUMOylation. The C-terminus of wild type, but not K886R, GluR6 was robustly SUMOylated (Fig.1A and C). To determine the role of Ubc9 for GluR6 SUMOylation, we conjugated coding sequences of wild type Ubc9 or the mutant Ubc9 (C93S) to the N-terminus of GluR6. These fusion protein expression vectors were transfected independently or co-transfected with either GFP-SUMO-1 or SRa-SUMO-1 expression vectors into HEK293 cells. Proteins of the transfectants were analyzed by immunoblotting using antibodies for myc or GluR6. GluR6 SUMOylation was abolished by the C93S substitution in Ubc9 (Fig.1A and B), which disabled SUMO-1 conjugation.

Glutamate induced an increase in GluR6 SUMOylation

It was reported GluR6 was rapidly SUMOylated in response to glutamate [16]. We therefore tested the effect of this agonist on GluR6 SUMOylation by immunoprecipitation. Both glutamate (4mM or 12mM) treatments after 30 min (Fig.2A) and 12mM glutamate treatment after 10 min or 30 min (Fig.2B) increased the basal level of GluR6 SUMOylation.

SUMOylation of GluR6 inhibited c-Jun phosphorylation

Glutamate (4mM or 12mM) treatments after 12 h (Fig.3A) upregulated the low basal level of GluR6 SUMOylation but had no effect on GluR6 (K886R). Meanwhile, the level of c-Jun phosphorylation (Fig.3A and B) was also evaluated by western blot. It was shown that SUMOylation decreased the level of c-Jun phosphorylation.

Glutamate induced death of PC12 cells transfected with the mutant GluR6 (K886R)

The morphologies of PC12 cells cultured in different groups induced by glutamate for 12h are shown in Fig.4A. The PC12 cells transfected with wild type GluR6 and Glutamate-unstimulated group were in polygon or spindle-shaped that could connect with each other whereas in the mutant GluR6 (K886R) and 10mM glutamate treatment groups, most of the PC12 cells shrank and some even degenerated and detached from the substrate. The viability of PC12 cells as evaluated by a CCK-8 assay showed significant difference between wild type GluR6 and mutant GluR6 (K886R) groups (Fig.4B). At 10mM, the response of PC12 to glutamate exhibited significant difference in comparison to the unstimulated control group; however, there was no significant difference in PC12 cell survival among the control group and the 2mM, 4mM, or 8mM glutamate treatment groups.

Discussion

This study demonstrated that SUMOylation at K886 of GluR6 downregulated c-Jun phosphorylation, which was correlated to increased viability of PC12 cells after a glutamate challenge, indicating that SUMOylation of the kainate receptor GluR6 subunit modulates the response to excitotoxicity towards survival. To our knowledge, this is the first time to report that the SUMO-1 interaction with GluR6 is cytoprotective in glutamate-induced cell death.

It is known that cerebral ischemia has profound effects on multiple cellular processes, including posttranslational modification of proteins. Recently it has been reported that GluR6 is implicated in neuronal cell death following ischemic insult [20] via a JNK/c-Jun activation pathway [21, 22]. SUMO1 is known for having the capability of suppressing apoptosis signal-regulating kinase (ASK1)-mediated cell death and downregulating c-Jun activity in cultured BOSC23 cells [23]. Furthermore, in rats, transient middle cerebral artery occlusion (MCAO) results in a dramatic increase in SUMOylation by SUMO-1 while the level of KARs were decreased in the infarct regions [24]. Furthermore, glutamate could increase the low basal level of GluR6 SUMOylation and evoked GluR6 internalization, but had no effect on mutated GluR6 [16]. One possibility is that upregulation of SUMO levels leads to a deregulation in KAR thereby reducing neuronal excitability and increasing cell viability in response to glutamate released under conditions such as ischemia. Our results suggested that glutamate upregulates GluR6 SUMOylation, which in turn leads to an increase of cell viability via a c-Jun phosphorylation-dependent mechanism in vitro. Future studies are required to determine the significance of GluR6 SUMOylation in the nervous system.

However, in this study we found that, compared with each unstimulated control group, the level of c-Jun phosphorylation was increased in both the wild type GluR6 group induced by glutamate (12mM) and the mutant GluR6(K886R) groups treated with glutamate (4mM, 12mM). Therefore, the involvement of other mechanisms of glutamate affecting c-Jun phosphorylation cannot be excluded. A previous study shows that brain ischemia–reperfusion activates nNOS and may further enhance the release of nitric oxide (NO). Simultaneously, the increased production of NO facilitates S-nitrosylation of GluR6, which results in the upregulation of the activated GluR6, leading to the further assembling of GluR6 and MLK3 with PSD95 and the activation of the JNK downstream signaling pathway [8]. S-nitrosylation of GluR6 may play an important role in glutamate leading to the activation of c-Jun and the precise mechanism merits further investigation.

    In summary, the current study showed that the GluR6 SUMOylation could suppress  c-Jun phosphorylation in HEK293T cells and had a protective effect on PC12 cells against death. Thus, our results showed new evidence for the role of SUMO-1 in cytoprotective function. Elevation of GluR6 SUMOylation levels may be an appropriate target for drug discovery in the cerebrovascular disease field.

Acknowledgments

This work was supported by the National Natural Sciences Foundation of China (No. 30400421). We also thank Prof. Henley JM (University of Bristol, UK) for providing the expression vector for Myc-GluR6.

References

[1]    Fonnum F. (1984) Glutamate: a neurotransmitter in mammalian brain. J Neurochem 42, 1-11

[2]    Choi D.W., Rothman S.M. (1990) The role of Glutamate neurotoxicity in hypoxic-ischemic neuronal death. Annu Rev Neurosci 13，82-171

[3]    Choi D.W. (1988) Glutamate neurotoxicity and diseases of the nervous system. Neuron 1，34-623

[4]    Monaghan D.T., Bridges R.J., Cotman C.W. (1989) The excitatory amino acid receptors: their classes, pharmacology, and distinct properties in the function of the central nervous system. Annu Rev Pharmacol Toxicol 29，365-402

[5]    Pei D.S., Guan Q.H. , Sun Y.F., Zhang Q.X., Xu T.L., Zhang G.Y. (2005) Neuroprotective effects of GluR6 antisense oligodeoxynucleotides on transient brain ischemia/reperfusion-induced neuronal death in rat hippocampal CA1 region. J Neurosci Res 82，9-642

[6]    Pei D.S., Wang X.T., Liu Y. , Sun Y.F., Guan Q.H., Wang W., Yan J.Z., Zong Y.Y., Xu T.L., Zhang G.Y. (2006) Neuroprotection against ischemic brain injury by a GluR6-9c peptide containing the TAT protein transduction sequence. Brain129, 79-465

[7]    Zhang Q.G., Tian H. Li., Zhang G.Y. (2006) Antioxidant N-acetylcysteine inhibits the activation of JNK3 mediated by the GluR6-PSD95-MLK3 signaling module during cerebral ischemia in rat hippocampus. Neurosci Lett 408，64-159

[8]    Yu H.M., Xu J., Li C., Zhou C., Zhang F., Han D., Zhang G.Y. (2008) Coupling between neuronal nitric oxide synthase and Glutamate receptor 6-mediated c-Jun N-terminal kinase signaling pathway via S-nitrosylation contributes to ischemia neuronal death. Neuroscience 155，32-1120

[9]    Seeler J.S., Dejean A. (2001) SUMO: of branched proteins and nuclear bodies. Oncogene 20，9-7243

[10]  Scheschonka A., Tang Z., Betz H. (2007) SUMOylation in neurons: nuclear and synaptic roles?. Trends Neurosci 30，85-91

[11] RodriguezM.S., Dargemont C., Hay R.T. (2001) SUMO-1 conjugation in vivo requires both a consensus modification motif and nuclear targeting. J Biol Chem 276,  9-12654.

[12]  Okuma T., Honda R. , Ichikawa G.,  Tsumagari N., Yasuda H. (1999)  In vitro SUMO-1 modification requires two enzymatic steps, E1 and E2. Biochem Biophys Res Commun 254, 8-693

[13]  Schwarz S.E., Matuschewski K., Liakopoulos D, Scheffner M, Jentsch S. (1998) The ubiquitin-like proteins SMT3 and SUMO-1 are conjugated by the UBC9 E2 enzyme. Proc Natl Acad Sci USA 95, 4-560

[14]  Schwartz D.C., Hochstrasser M. (2003) A superfamily of protein tags: ubiquitin, SUMO and related modifiers. Trends Biochem Sci 28, 8-321

[15]  Johnson E.S., Gupta A.A. (2001) An E3-like factor that promotes SUMO conjugation to the yeast septins. Cell 106, 44-735

[16]  Martin S., Nishimune A., Mellor J.R., et al. (2007) SUMOylation regulates kainate-receptor- mediated synaptic transmission. Nature 447, 5-321

[17]  Jakobs A., Koehnke J.,  Himstedt F.,  Funk M, Korn B, Gaestel M, Niedenthal R. (2007) Ubc9 fusion-directed SUMOylation (UFDS): a method to analyze function of protein SUMOylation. Nat Methods 4, 50-245

[18] Chakrabarti S.R., Sood R., Ganguly S., Bohlander S., Shen Z., Nucifora G. (1999) Modulation of TEL transcription activity by interaction with the ubiquitin-conjugating enzyme UBC9. Proc Nat Acad Sci USA 96, 7467–7472

[19] Park H.H., Lee K.Y., Kim S.H., Lee Y.J., Koh S.H. (2009)  l-DOPA-induced neurotoxicity is reduced by the activation of the PI3K signaling pathway. Toxicology 3, 80-86

[20] Paschen W., Schmitt J., Uto A. (1996) RNA editing of Glutamate receptor subunits GluR2, GluR5 and GluR6 in transient cerebral ischemia in the rat. J Cereb Blood Flow Metab 16, 56-548

[21]  Pei D.S., Sun Y.F. , Guan Q.H. , Hao Z.B., Xu T.L., Zhang G.Y. (2004) Postsynaptic density protein 95 antisense oligodeoxynucleotides inhibits the activation of MLK3 and JNK3 via the GluR6.PSD-95.MLK3 signaling module after transient cerebral ischemia in rat hippocampus. Neurosci Lett 367, 5-71

[22] Tian H., Zhang Q.G. , Zhu G.X.,  Pei D.S., Guan Q.H., Zhang G.Y. (2005) Activation of c-Jun NH2-terminal kinase 3 is mediated by the GluR6.PSD-95.MLK3 signaling module following cerebral ischemia in rat hippocampus, Brain Res 1061, 57-66

[23]  Lee Y.S., Jang M.S. , Lee J.S., Choi E.J., Kim E. (2005) SUMO-1 represses apoptosis signal-regulating kinase 1 activation through physical interaction and not through covalent modification. EMBO Rep 10, 655-949

[24]  Cimarosti H., Lindberg C., Bomholt S.F., Rønn L.C., Henley J.M. (2008) Increased protein SUMOylation following focal cerebral ischemia. Neuropharmacology 54, 9-280