- Email: [email protected]
The ganglioside GQ1b regulates BDNF expression via the NMDA receptor signaling pathway
Neuropharmacology 77 (2014) 414e421
Contents lists available at ScienceDirect
Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm
The ganglioside GQ1b regulates BDNF expression via the NMDA receptor signaling pathway Min Kyoo Shin a,1, Woo Ram Jung a,1, Hong Gi Kim a, Seung Eon Roh b, Choong Hwan Kwak a, Cheorl Ho Kim a, Sang Jeong Kim b, Kil Lyong Kim a, * a b
Department of Biological Science, Sungkyunkwan University, Suwon, Gyeonggi-Do 440-746, Republic of Korea Department of Physiology, Seoul National University College of Medicine, Seoul, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 May 2013 Received in revised form 17 September 2013 Accepted 18 October 2013
Gangliosides are sialic acid-containing glycosphingolipids which play a role in neuronal functions. Among the gangliosides, tetrasialoganglioside GQ1b shows neurotrophic factor-like actions, such as increasing neurite outgrowth, cell proliferation, and long-term potentiation. In addition, we recently reported that GQ1b improves spatial learning and memory performance in naïve rats. However, it is still unknown how GQ1b exerts its diverse neuronal functions. Thus, we hypothesized that GQ1b might influence synaptic activity by regulating brain-derived neurotrophic factor (BDNF) expression, which is an important protein for synaptic plasticity and cognition. Interestingly, GQ1b treatment increased BDNF expression in GQ1b-null SH-SY5Y cell lines and rat primary cortical neurons. Additionally, we confirmed whether the observed effects were due to GQ1b or due to a ganglioside with fewer sialic acid molecules (GT1b and GD1b) created by the sialidases present on the plasma membranes, by directly applying GT1b and GD1b or GQ1b co-treated with a sialidase inhibitor. Treatment with GT1b or GD1b had no effect on BDNF expression, whereas co-treatment with a sialidase inhibitor and GQ1b significantly increased BDNF levels. Moreover, GQ1b restored the decreased BDNF expression induced by the ganglioside synthesis inhibitor, D-PDMP, in rat primary cortical neurons. GQ1b treatment significantly increased BDNF levels, whereas pretreatment with the N-methyl-D-aspartate (NMDA) receptor antagonist D-AP5 blocked the effects of GQ1b on BDNF expression, suggesting that GQ1b regulates BDNF expression via the NMDA receptor signaling. Finally, we performed an intracerebroventricular GQ1b injection, which resulted in increased prefrontal and hippocampal BDNF expression in vivo. These findings demonstrate, for the first time, that tetrasialoganglioside GQ1b regulates BDNF expression in vitro and in vivo. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Ganglioside GQ1b Brain-derived neurotrophic factor N-methyl-D-aspartate receptor
1. Introduction Gangliosides are sialic acid-containing glycosphingolipids which are primarily expressed in the outer leaflet of the plasma membrane, and are particularly enriched in the cell membranes of the central nervous system (Schengrund, 1990; Zeller and
Abbreviations: BDNF, Brain-derived neurotrophic factor; NMDA, N-methyl-Daspartate; LTP, Long term potentiation; D-AP5, D-2-amino-5phosphonopentanoate; CREB, cAMP response element binding protein; ERK, extracellular signal-regulated kinase; D-PDMP, D-threo-1-phenyl-2decanoylamino-3-morpholino-1-propranol. * Corresponding author. Tel.: þ82 31 290 7017; fax: þ82 31 290 7015. E-mail addresses: [email protected] (M.K. Shin), [email protected] (W. R. Jung), [email protected] (H.G. Kim), doinfi[email protected] (S.E. Roh), [email protected] (C.H. Kwak), [email protected] (C.H. Kim), sangjkim@snu. ac.kr (S.J. Kim), [email protected] (K.L. Kim). 1 These authors contributed equally to this study. 0028-3908/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuropharm.2013.10.022
Marchase, 1992). Gangliosides have many physiological roles in the nervous system, including neuronal survival, differentiation, and neuritogenesis (Facci et al., 1984; Ferrari and Greene, 1998; Ledeen, 1984). In addition, they are also involved in neural development and neurological diseases, such as brain ischemia, seizures, and Alzheimer’s disease (Fighera et al., 2006; Lombardi and Moroni, 1992; Louis et al., 1983; Okada et al., 2007). GQ1b is a tetrasialoganglioside with four sialic acid residues, and which exhibits neurotrophic factor-like activity in vitro and in vivo. GQ1b treatment increases cell proliferation and neurite outgrowth in human neuroblastoma cell lines (Tsuji et al., 1983). In cultured cortical neurons, D-threo-1-phenyl-2-decanoylamino-3morpholino-1-propranol (D-PDMP), an inhibitor of ganglioside synthesis, inhibits synaptic activity, whereas GQ1b was capable of restoring the decreased synaptic activity induced by D-PDMP (Mizutani et al., 1996). Exogenously added GQ1b induces long-term potentiation (LTP), the cellular model of learning and memory, in
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
CA1 hippocampal neurons of guinea pig brain slices (Fujii et al., 2002; Furuse et al., 1998). Furthermore, we have recently shown that intracerebroventricular (ICV) GQ1b administration improves spatial learning and memory performance in naïve rats, as measured by the Y-maze and Morris water maze tests (Jung et al., 2008). Also, GQ1b activates N-methyl-D-aspartate (NMDA) receptor signaling pathway in H19-7 cell lines and rat hippocampus (Jung et al., 2010). NMDA receptors are glutamate-gated cation channels which play important roles in learning and memory formation (Bliss and Collingridge, 1993). Calcium influx through NMDA receptors can activate diverse signaling pathways, including the extracellular signal-regulated kinase (ERK) signaling pathway. Activation of ERK phosphorylates ser-133 of cAMP response element binding protein (CREB) (Huang et al., 2008; Johannessen et al., 2004). CREB is a transcription factor which modulates the transcription of genes containing cAMP response elements (CRE) in their promoters (Impey et al., 1998; Silva et al., 1998). It is known that CREB regulates synaptic plasticity related gene expression, such as that of brain-derived neurotrophic factor (BDNF). BDNF is a member of the neurotrophin family, including nerve growth factor, neurotrophin-3, and neurotrophin-4/5, and plays an important role in regulating neuronal survival, growth, and differentiation (Lewin and Barde, 1996; Lu et al., 2005). It is also involved with induction and maintenance of LTP and memory formation (Figurov et al., 1996). BDNF gene knockout or knockdown mice have impaired spatial learning and memory performance on hippocampus-dependent learning tasks (Heldt et al., 2007; Ma et al., 1998). Although GQ1b exhibits neurotrophic factor-like activity in vitro and in vivo, it is still unknown how GQ1b affects diverse neuronal functions. Thus, we hypothesized that GQ1b might regulate BDNF expression for synaptic plasticity in vitro and in vivo. In the current study, we investigated whether GQ1b regulates BDNF expression in SH-SY5Y cell lines and rat primary cortical neurons, and additionally, the effect of GQ1b on BDNF expression was confirmed by two independent experiments using direct comparison with other gangliosides or co-treatment with sialidase inhibitor. Furthermore, we examined that GQ1b regulates BDNF expression via the NMDA receptor signaling pathway in cultured rat primary cortical neurons. Finally, in vivo prefrontal and hippocampal BDNF levels were determined by administration of GQ1b through ICV injection in naïve rats. 2. Materials and methods 2.1. Cell culture and reagents The SH-SY5Y human neuroblastoma cell line was cultured at 37 C in 5% CO2, and complete DMEM (Life Technologies, Grand Island, NY, USA) supplemented with 2% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA, USA) and antibioticeantimycotic solution (Life Technologies). Ceramide, GM1, GD1b, GT1b, GQ1b, and Dthreo-1-phenyl-2-decanoylamino-3-morpholino-1-propranol (D-PDMP) were purchased from Alexis Biochemicals (Exeter, UK). D-AP5 and N-acetyl-2,3-dehydro-2deoxyneuraminic acid were purchased from SigmaeAldrich (St. Louis, MO, USA). 2.2. Primary cortical neurons Primary cortical neuron cultures were prepared from SpragueeDawley rats, purchased from Orient Bio (Seoul, Korea). Cerebral cortices were removed from the brains of embryonic day 17 rats, and dissociated cells were plated at a density of 2 106 cells on 6well plates which were precoated with poly-D-lysine (Sigma) and laminin (Invitrogen). The plating medium consisted of DMEM supplemented with 5% FBS. Cultures were maintained at 37 C in a humidified 5% CO2 atmosphere. Once astroglia became confluent underneath neurons at 7 days in vitro (DIV), the media was exchanged with neurobasal medium supplemented with B27 (Invitrogen) to stop non-neuronal cell overgrowth. Almost pure neuronal cultures were obtained at 10 DIV (Supplementary Fig. 1). 2.3. Quantitative real-time PCR Total RNA was isolated from SH-SY5Y cells and rat primary cortical neurons using the QIAzolÔ Lysis Reagent kit (Qiagen, Seoul, Korea), according to the manufacturer’s instructions. RT of the RNA was performed using AccuPower RT PreMix
415
(Bioneer, Daejeon, Korea). Total RNA (1 mg) and 100 pmol of oligo dT18 were preincubated at 70 C for 5 min, and were then transferred to a reaction tube. The reaction volume was 20 mL cDNA synthesis was performed at 42 C for 60 min, followed by RT inactivation at 94 C for 5 min. For quantitative real-time RT-PCR, iQ SYBR Green supermix (Bio-Rad) and an iCycler real-time PCR machine were used to measure the expression of genes under the following conditions; 45 cycles of 95 C for 30 s, 60 C (rat BDNF), 65 C (human BDNF), 60 C (NR2A, NR2B) for 30 s, and 72 C for 30 s. Primer pairs used to amplify the target genes were: rat BDNF forward primer sequence: 50 -GGACATATCCATGACCAGAAAGAAA-30 , reverse primer sequence: 50 -GCAACAAACCACAACATTATCGAG-30 ; human BDNF forward primer sequence: 50 -AAACATCCGAGGACAAGGTG-30 , reverse primer sequence: 50 -AGAAGAGGAGGCTCCAAAGG-30 ; NR2A forward primer sequence: 50 -ATACCGGCAGAACTCCACAC-30 , reverse primer sequence: 50 -CCTCCAGTAGCCGTTCTCTG-30 ; NR2B forward primer sequence: 50 -ACCCTCAAAGCCCGACTAAT-30 , reverse primer sequence: 50 -TGAAGCAAGCACTGGTCATC-30 . For normalization of the cycling threshold values obtained with the experimental samples, GAPDH was amplified under the same conditions. All reactions were repeated at least three times independently to ensure the reproducibility of the results. 2.4. Enzyme-linked immunosorbent assay (ELISA) The BDNF Emax ImmunoAssay System was purchased from Promega (Madison, WI, USA), and BDNF levels were measured as previously described (Park et al., 2011). This ELISA kit showed <3% cross-reactivity with other related neurotrophic factors (NGF, NT-3, and NT-4) at 100 ng/ml. Each well of the 96-well polystyrene plate was incubated overnight at 4 C with 100 mL anti-BDNF monoclonal antibody (mAb), which was diluted 1:1000 in carbonate coating buffer (25 mM sodium bicarbonate and 25 mM sodium carbonate, pH 9.7). Unabsorbed mAb was removed, and the plates were washed once with TBST wash buffer (20 mM TriseHCl (pH 7.6), 150 mM NaCl, and 0.05% (v/v) Tween 20). Immediately prior to blocking, cell lysates were removed from the freezer and were allowed to warm to room temperature. Plates were blocked using 200 mL Promega 1 Block and Sample buffers, followed by 1 h incubation at room temperature. Plates were then washed using TBST wash buffer. A 30 mg (100 mL total volume) aliquot of each sample or standard (0, 7.8, 15.6, 31.3, 62.5, 125, 250, 500 pg/ml) was added, in triplicate, to the plates. Plates were incubated for 2 h with shaking (500 rpm) at room temperature. Plates were then washed five times with TBST wash buffer. Anti-human BDNF polyclonal antibody (100 mL diluted 1:500 in 1 Block and Sample buffers) was added to each well, and the plates were incubated for 2 h with shaking (500 rpm) at room temperature. The plates were again washed five times using TBST wash buffer. Anti-IgY horseradish peroxidase conjugate (100 mL diluted 1:200 in 1 Block and Sample buffers) was then added to each well, and the plates were incubated for 1 h with shaking (500 rpm) at room temperature. The plates were emptied again and were washed using TBST wash buffer. Finally, the plates were developed using 100 mL Promega TMB One Solution, and the reaction was stopped by adding 100 mL 1 N HCl. The plates were read at 450 nm on a Versa Max ELISA reader (Molecular Devices, Sunnyvale, CA, USA). 2.5. Western blotting Proteins were extracted from cells and tissues with lysis buffer containing 50 mM TriseCl, 150 mM NaCl, 0.5% Triton X-100, 0.5% NP-40, 1 mM EDTA, a protease inhibitor cocktail solution (Amersham Bioscience, Piscataway, NJ, USA; Roche, Mannheim, Germany), and a phosphatase inhibitor cocktail solution (Roche). The extracts were centrifuged at 10,000 g for 30 min at 4 C. The protein concentrations in the supernatants were determined using the Bradford assay (Bio-Rad, Hercules, CA, USA). Western blotting was performed as described previously (Shin et al., 2011). Briefly, 15e20 mg samples were boiled in Laemmli sample buffer containing 10% b-mercaptoethanol, and were separated on 10e15% sodium dodecyl sulfate-polyacrylamide gel by electrophoresis, for analysis of BDNF, GAPDH, NR2A, NR2B, p-NR2B, extracellular signal-regulated kinase (ERK)1/2, phosphorylated ERK1/2 (p-ERK1/2), cAMP response element-binding (CREB) protein, and phosphorylated CREB (p-CREB). Proteins were transferred onto nitrocellulose membranes (Whatman, Milford, MA, USA) at 100 V, and the membranes were blocked for 1 h in TBST containing 7% skim milk. Antibodies recognizing NR2A (1:1000, Santa Cruz Biotechnology, Santa Cruz, CA, USA, Catalogue number: sc-1468, antibody name: NMDAε1 (C-17)), NR2B (1:1000, Millipore, Bedford, MA, USA, Catalogue number: 06-600, antibody name: Anti-NR2B rabbit polyclonal IgG), p-NR2B (1:1000, Sigma, Saint Louis, USA, Catalogue number: M2442, antibody name: AntiPhospho-N-Methyl D-Aspartate (NMDA) NR2B Subunit (pTyr1472)), ERK1/2 (1:2000, Cell Signaling Technology, Danvers, MA, USA, Catalogue number: 9102, antibody name: p44/42 MAP kinase Antibody), p-ERK1/2 (1:2000, Cell Signaling Technology, Catalogue number: 9101, antibody name: Phospho-p44/42 MAP kinase (Thr202/ Tyr204) antibody), CREB (1:2000, Cell signaling Technology, Catalogue number: 9197S, antibody name: CREB (48H2) Rabbit mAb), p-CREB (1:1000, Santa Cruz Biotechnology, Catalogue number: sc-7978, antibody name: p-CREB-1 (Ser 133)), BDNF (1:1000, Alomone labs, Jerusalem, Israel, Catalogue number: ANT-010, antibody name: Anti-BDNF), and GAPDH (1:5000, Millipore, Catalogue number: MAB374, antibody name: Mouse Anti-Glyceraldehyde-3-phosphate dehydrogenase monoclonal antibody) were used for immunodetection. All blots were developed using a chemiluminescent horseradish peroxidase substrate kit (Millipore).
416
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
Fig. 1. GQ1b increases BDNF mRNA and protein expression in SH-SY5Y cells. BDNF mRNA expression was analyzed by quantitative real-time RT-PCR. GQ1b 1 mM treatment for 12 h most significantly increased BDNF mRNA expression (A). Time course of expression levels of BDNF mRNA following 1 mM GQ1b treatment (B). An enzyme-linked immunosorbent assay was performed to measure BDNF protein levels in SH-SY5Y cells treated with 1 mM GQ1b for 12, 24 h. BDNF protein expression increased significantly following GQ1b treatment in SH-SY5Y cells (C). Western blotting and densitometry analysis showing that the gangliosides GD1b, GT1b, and ceramide had no effects on BDNF regulation, whereas GM1 and GQ1b increased BDNF levels. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (D). SH-SY5Y cells were co-treated with 50 mM of the sialidase inhibitor (NeuAc2en) and 1 mM GQ1b for 12 h GQ1b increased BDNF protein expression, and the expression level was not changed by co-treatment with sialidase inhibitor, as determined by western blotting and quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (E). These data represent typical results from three independent experiments and are expressed as mean values SEM. *p < 0.05, **p < 0.01, ***p < 0.001 vs. vehicle-treated group, #p < 0.05 vs. NeuAc2en-treated group.
2.6. RNAi Pre-designed BDNF siRNA duplexes sets from Genolution Pharmaceuticals Inc. (Seoul, Korea) were transfected to knockdown BDNF expression. Targeting siRNA and non-targeting control siRNA sequences were as follows: BDNF-1 siRNA (sense, 50 GUAUAGAUUAGAUUAUAUUUU-30 ; antisense, 50 -AAUAUAAUCUAAUCUAUACUU-30 ), BDNF-2 siRNA (sense, 50 -CUUUAAUUGUGAAUUGAUAUU-30 ; antisense, 50 -UAUCAAUUCACAAUUAAAGUU-30 ), BDNF-3 siRNA (sense, 50 -GAAUUCCUACAAUAUAUAUUU-30 ; antisense, 50 -AUAUAUAUUGUAGGAAUUCUU-30 ), BDNF-4 siRNA (sense, 50 -CAAAGAAUUUCAAUUUGUUUU-30 ; antisense, 50 -AACAAAUUGAAAUUCUUUGUU-30 ), BDNF-5 siRNA (sense, 50 -CUAAUUAACAUAUAAUAUAUU-30 ; antisense, 50 -UAUAUUAUAUGUUAAUUAGUU-30 ), and control siRNA (sense, 50 -CCUCGUGCCGUUCCAUCAGGUAGUU30 ; antisense, 50 -CUACCUGAUGGAACGGCACGAGGUU-30 ). SH-SY5Y cells were transfected with a final concentration of 30 nM of each siRNA duplexes set, using RNAiMaxTM reagent (Invitrogen, Carlsbad, CA, USA). The cells were then incubated for 24 h and were harvested for Western blot analysis to confirm knockdown of BDNF expression. 2.7. Thin-layer chromatography Cultured SH-SY5Y cells were washed twice with ice-cold PBS and were scraped from the dishes. Gangliosides were extracted according to the method of Svennerholm and Fredman (Svennerholm and Fredman, 1980). Briefly, the collected gangliosides were separated on plates of silica gel with chloroform/methanol/0.2% aqueous calcium chloride [55:45:10 (by vol.)] as a developing solution using TLC. The plates were sprayed with a resorcinol/HCl reagent and were then heated at 110 C for 30 min. 2.8. Animals and ICV injections of GQ1b Male SpragueeDawley rats (7-weeks-old) were purchased from Orient Bio (Seoul, Korea) and were acclimated for 1 week prior to the initiation of experiments. The rats were housed three per cage, and had free access to water and food prior to surgery. Rats were maintained under standard conditions: 12 h lightedark cycle, 22 2 C. Animal studies received prior approval from the animal research committee of Sungkyunkwan University. ICV injections to the left lateral ventricle were performed as described previously (Jung et al., 2008). Briefly, rats were surgically implanted with a CMA/12 guide cannula (CMA, Helsingborg, Sweden). Following
insertion of a guide cannula into the brain of each rat, they were housed one per cage, and were maintained and stabilized under standard conditions for 1 week. GQ1b (1 mg) was injected once into the lateral ventricle of the rat cerebrum, and the cannula was left in place for 30 s after infusion to allow for diffusion. Rats were sacrificed 12 h after injection. The whole brain was removed and was immediately frozen in liquid nitrogen and stored at 80 C. 2.9. Statistical analysis The presented data represent typical results from three independent experiments, and are expressed as mean standard error of the mean (SEM). Statistical analysis was conducted using SPSS v. 10.0.7 (SPSS Inc., Chicago, IL, USA). One-way analysis of variance followed by Tukey’s test was used as a post-hoc test. Statistical differences between two groups were determined using the Student’s t-test. A p value <0.05 was considered to be statistically significant.
3. Results 3.1. GQ1b increases BDNF mRNA and protein expression in SH-SY5Y cells To examine whether GQ1b regulates BDNF expression, we used the SH-SY5Y cell line, which is known not to express the ganglioside GQ1b (Hettmer et al., 2005; Sadeghlar et al., 2000). We confirmed the absence of GQ1b expression in SH-SY5Y cells using immunocytochemistry and thin-layer chromatography (Supplementary Fig. 2). SH-SY5Y cells were treated with various concentrations (0.01, 0.1, 1, and 10 mM) of GQ1b for 12 h. Quantitative real-time RT-PCR analysis revealed that BDNF mRNA expression was most effectively increased in SH-SY5Y cells following treatment with 1 mM of GQ1b (Fig. 1A). Next, we examined BDNF mRNA expression at various time points (1, 3, 6, 9, 12, 15,
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
417
Fig. 2. GQ1b regulates BDNF levels in rat primary cortical neurons. Primary cortical neurons were incubated with 1 mM GQ1b for 0, 1, 3, 6, 12, and 24 h. Western blotting and densitometry analysis showing that BDNF protein expression increased significantly in primary cortical neurons treated with 1 mM GQ1b after 12 and 24 h. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (A). An enzyme-linked immunosorbent assay revealed that 1 mM GQ1b treatment for 24 h significantly increased BDNF expression, as compared to that in a vehicle-treated group (B). Western blotting analysis showing that GQ1b increased BDNF levels in the presence or absence of sialidase inhibitor NeuAc2en. Band intensity was quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (C). Monosialoganglioside GM1 (1 mM) treatment for 24 h also increased BDNF protein expression, which was less than that of the GQ1b treatment. Western blotting band intensity was quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (D). Data represent typical results from three independent experiments, and are expressed as mean values SEM. *p < 0.05, **p < 0.01 vs. vehicle-treated group, #p < 0.01 vs. NeuAc2en-treated group.
and 24 h) after 1 mM of GQ1b treatment. GQ1b significantly increased the expression of BDNF mRNA after 1 h treatment, and maximal effects were observed between 9 and 15 h post-treatment (Fig. 1B). We also investigated BDNF protein levels by enzymelinked immunosorbent assay (ELISA). Treatment with 1 mM GQ1b for 12 and 24 h significantly increased BDNF protein expression, as compared to that of vehicle-treated control group, although levels decreased over time (Fig. 1C). However, these results could not exclude the possibility that increased BDNF expression might be caused by other gangliosides, such as GT1b and GD1b, which were generated from GQ1b by sialidases which are known to be present in the plasma membranes (Monti et al., 2000). Thus, we compared the effects of each ganglioside (GD1b, GT1b, and GQ1b) including ceramide backbone on BDNF expression in SH-SY5Y cells. In addition, monosialoganglioside GM1 was also treated as positive control, because GM1 is known to interact with BDNF receptor, TrkB, which results in the induction of its ligand, BDNF (Cheng et al., 2011; Pitto et al., 1998). Treatment with 1 mM of ceramide, GD1b, or GT1b for 12 h did not affect BDNF protein levels. Although both GM1 and GQ1b significantly increased BDNF expression, GQ1b was more effective in increasing BDNF in SH-SY5Y cells than GM1 (Fig. 1D). Next, we inhibited the sialidase function in SH-SY5Y cells by co-treatment with 1 mM GQ1b and the sialidase inhibitor N-
acetyl-2,3-dehydro-2-deoxyneuraminic acid (NeuAc2en) for 12 h GQ1b treatment significantly increased BDNF protein expression in the presence or absence of NeuAc2en (Fig. 1E), suggesting that tetrasialoganglioside GQ1b, but not GD1b and GT1b, could regulate BDNF expression. Furthermore, we validated BDNF antibody specificity using siRNA experiment. BDNF protein expression was significantly decreased by transfection with BDNF siRNA duplexes, as compared to control siRNA transfected SH-SY5Y cells (Supplementary Fig. 3). 3.2. GQ1b regulates BDNF expression in rat primary cortical neurons Next, we investigated the effects of GQ1b on BDNF expression in a physiologically relevant neuronal cell culture system, such as rat primary cortical neurons. BDNF protein levels were examined at various time points (1, 3, 6, 12, and 24 h) after 1 mM GQ1b treatment. We found that GQ1b significantly increased BDNF expression after 12 and 24 h (Fig. 2A). In addition, the ELISA data also showed increased BDNF levels following 1 mM GQ1b treatment for 24 h, as compared to those of the vehicle-treated control group (vehicletreated group ¼ 15.22 2.72 pg/ml, GQ1b-treated group ¼ 29.80 5.45 pg/ml; p < 0.01) (Fig. 2B). Next, to determine
418
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
Fig. 3. GQ1b regulates BDNF expression via the NMDA receptor signaling pathway in rat primary cortical neurons. Western blotting analysis showing that treatment with 1 mM GQ1b for 24 h increased protein levels of NMDA receptor signaling molecules, such as pERK1/2 and pCREB. Pretreatment with 20 mM of D-AP5 blocked GQ1b-induced activation of the NMDA receptor signaling pathway. Band intensity was quantified by densitometry analysis. pERK/ERK and pCREB/CREB levels in vehicle-treated cells were set as 1 (A). BDNF mRNA expression was analyzed by quantitative real-time RT-PCR. Treatment of 1 mM GQ1b significantly increased BDNF mRNA expression, but the effects of GQ1b on BDNF mRNA expression were abolished by pretreatment with 20 mM of D-AP5 (B). Western blotting analysis showing that BDNF protein levels were increased by 1 mM of GQ1b treatment, and its effect is blocked by D-AP5 pretreatment (C). The increase in BDNF protein expression following 1 mM GQ1b treatment was inhibited by D-AP5 pretreatment, as determined by an enzyme-linked immunosorbent assay (D). NMDA receptor subunit 2A and 2B mRNA expression was analyzed by quantitative real-time RT-PCR. Pretreatment of D-AP5 blocked the increase of NR2A and NR2B mRNA expression induced by GQ1b treatment (E). Western blotting analysis showing that 1 mM GQ1b treatment for 24 h increased NR2A, NR2B expression and pNR2B protein levels, whereas the increase in NR2A, NR2B expressions and pNR2B protein levels following GQ1b treatment was inhibited by D-AP5 pretreatment. Band intensity was quantified by densitometry analysis. NR2A, NR2B, and pNR2B levels in vehicle-treated cells were set as 1 (F). Data represent typical results from three independent experiments, and are expressed as mean values SEM. *p < 0.05, **p < 0.01 vs. vehicle-treated control group.
whether the enhanced BDNF expression was due to GQ1b or other sialic acid cleaved forms, including GT1b and GD1b, we co-treated sialidase inhibitor, NeuAc2en, with GQ1b for 24 h in rat primary cortical neurons. BDNF protein expression increased significantly following treatment with GQ1b with or without NeuAc2en cotreatment, suggesting that GQ1b regulates BDNF expression in rat primary cortical neurons (Fig. 2C). The monosialoganglioside GM1 increased BDNF expression in SH-SY5Y cell lines (Fig. 1D). Thus, to examine whether GM1 also regulates BDNF expression in rat primary cortical neurons, we treated them with 1 mM GM1 or GQ1b for 24 h. Similarly, GM1 increased BDNF protein expression in rat primary cortical neurons, but to a lesser degree than GQ1b treatment (Fig. 2D). Given that ganglioside biosynthesis is a complicated mechanism, it is difficult to specifically block ganglioside synthesis in cells. Therefore, we used D-PDMP (a glucosylceramide synthase inhibitor), which has been commonly used to inhibit total
ganglioside biosynthesis (Mizutani et al., 1996). To investigate whether BDNF expression increases following GQ1b treatment in ganglioside-depleted cells, primary cortical neurons were incubated in 20 mM D-PDMP with or without 1 mM GQ1b for 24 h. Treatment with 1 mM GQ1b significantly increased BDNF expression, whereas D-PDMP decreased its levels, as compared to the vehicle-treated group. When cells were co-treated with GQ1b and D-PDMP, exogenously added GQ1b restored decreased BDNF expression in DPDMP-treated cells (Supplementary Fig. 4). 3.3. GQ1b increases BDNF expression through NMDA receptor signaling in rat primary cortical neurons In our previous study, we demonstrated that GQ1b treatment activates the NMDA receptor signaling pathway, and its effect is blocked by D-AP5, an NMDA receptor antagonist, in the H19-7 cell
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
419
Fig. 4. Intracerebroventricular GQ1b injection increased BDNF expression in the rat prefrontal cortex and hippocampus. Intracerebroventricular GQ1b injection was performed (1 mg GQ1b in 2 mL vehicle), and the rat prefrontal cortex and hippocampus were collected individually (vehicle, lane 1e5, n ¼ 5; GQ1b, lane 6e10, n ¼ 5) (A, B). Western blotting analysis showing that GQ1b injection significantly increased BDNF expression in the rat prefrontal cortex (A) and hippocampus (B). Each number represents an individual rat. Band intensity was quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-injected group were set as 1. Data represent typical results from three independent experiments and are expressed as mean values SEM. *p < 0.01 vs. vehicle-injected group.
line (Jung et al., 2010). To investigate whether GQ1b regulates BDNF expression via the NMDA receptor signaling pathway, we first examined the protein levels of NMDA receptor signaling molecules, such as pERK1/2 and pCREB, in rat primary cortical neurons. GQ1b 1 mM treatment for 24 h significantly increased pERK1/2 and pCREB protein levels, as compared to non-treated control group, but pretreatment with 20 mM D-AP5 for 20 min prior to adding 1 mM GQ1b inhibited GQ1b-induced activation of NMDA receptor signaling (Fig. 3A). Next, we measured BDNF mRNA and protein expression by quantitative real-time RT-PCR, western blotting, and ELISA assay. BDNF mRNA and protein expression increased following treatment with GQ1b, whereas pretreatment of D-AP5 completely blocked the effects of GQ1b on BDNF expression (Fig. 3BeD). Although these results suggest that GQ1b regulates BDNF expression via the NMDA receptor signaling pathway, it is still unknown whether the enhancement of NMDA receptor signaling by GQ1b was due to the activation of NMDA receptor or the up-regulation of NMDA receptor expression. Thus, we measured NMDA receptor subunit 2A (NR2A) and 2B (NR2B) mRNA, as well as protein expression and pNR2B (Tyr-1472) protein levels by quantitative real-time RT-PCR and western blotting analysis, respectively. NR2A and NR2B mRNA and protein expression, as well as pNR2B protein levels, were significantly increased by GQ1b treatment, but its effect was blocked by D-AP5 pretreatment (Fig. 3EeF), suggesting that GQ1b might increase the NMDA receptor signaling pathway by activation of NMDA receptor and the up-regulation of NMDA receptor expression. 3.4. ICV GQ1b injection increases BDNF expression in rat prefrontal cortex and hippocampus GQ1b increased BDNF expression in vitro. To examine whether GQ1b regulates BDNF expression in vivo, 1 mg of GQ1b was injected into the ICV region of the brain, and western blotting analysis was performed to measure BDNF protein expression. This dose of GQ1b has been previously demonstrated to improve spatial learning and memory performance and to activate the NMDA receptor signaling pathways in rats (Jung et al., 2008, 2010). ICV injection of GQ1b significantly increased BDNF expression in the prefrontal cortex
(2.63 0.77-fold, p < 0.01) (Fig. 4A). In addition, hippocampal BDNF expression was also increased following GQ1b injection (3.65 0.53-fold, p < 0.01), as compared to the vehicle-injected group (Fig. 4B). These data indicate that GQ1b regulates BDNF expression in the prefrontal cortex and hippocampus. 4. Discussion Here we report that GQ1b treatment increased BDNF expression in both GQ1b-null SH-SY5Y cell lines and rat primary cortical neurons. We also confirmed that increased BDNF expression was due to GQ1b, but not GT1b or GD1b, which was generated from GQ1b by plasma membrane-associated sialidases. Direct application of GT1b or GD1b did not increase BDNF levels, whereas GQ1b treatment with or without a sialidase inhibitor significantly increased BDNF expression. In addition, GQ1b restored the decreased BDNF expression in ganglioside synthesis inhibitor (DPDMP) treated primary cortical neurons. Furthermore, we investigated the underlying mechanism of GQ1b-induced BDNF regulation. Treatment with GQ1b significantly increased BDNF levels, whereas its effect was completely blocked by pretreatment with the NMDA receptor antagonist D-AP5, suggesting that GQ1b regulates BDNF levels via the NMDA receptor signaling. Finally, we found that an ICV GQ1b injection significantly increased BDNF levels in the prefrontal cortex and hippocampus. These findings demonstrate, for the first time, that the tetrasialoganglioside GQ1b regulates BDNF expression in vitro and in vivo. GQ1b treatment increased BDNF expression in SH-SY5Y cell lines which did not express endogenous GQ1b, suggesting that upregulation of BDNF expression by GQ1b does not require the synthesis of endogenous GQ1b. Monosialoganglioside GM1 also increased BDNF expression in the SH-SY5Y cell line and rat primary cortical neurons, which was less than that of the GQ1b treatment. This finding seems to supports previous research showing that GM1 has neurotrophic functions and the ability to improve spatial learning and memory in aged rats (Duchemin et al., 2002; Fong et al., 1997; Mutoh et al., 1995). In addition, GQ1b-treated brain slices dominantly enhanced LTP compared with the GM1-treated slices (Furuse et al., 1998).
420
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
Inhibiting ganglioside biosynthesis with D-PDMP reduced BDNF expression, but GQ1b co-treatment with D-PDMP restored the decreased BDNF levels in rat primary cortical neurons. Mizutani et al. have shown that D-PDMP treatment significantly decreases intracellular Ca2þ levels, as measured by fura-2 calcium imaging. A series of gangliosides (GM3, GM1, GD3, GD1b, and GT1b) did not normalize the decreased intracellular Ca2þ levels, but that only GQ1b restored Ca2þ levels in D-PDMP-treated rat primary cortical neurons (Mizutani et al., 1996), suggesting that GQ1b might recover BDNF levels by normalizing intracellular Ca2þ concentration through the NMDA receptor in D-PDMP-treated cells. Although GQ1b-induced upregulation of BDNF levels does not require the synthesis of endogenous GQ1b expression, co-treatment of GQ1b with D-PDMP showed reduced effects of GQ1b on BDNF protein levels. These results may be explained by the fact that D-PDMP treatment alters membrane fluidity. Barbour et al. demonstrated that D-PDMP treatment changes in membrane lipid organization contribute to the disruption of the cellular function of proteins at the cell surface (Barbour et al., 1992). Thus, further studies are required to examine whether D-PDMP treatment affects cellular function of NMDA receptors in rat primary cortical neurons. We have previously reported that GQ1b activates NMDA receptor signaling pathways (Jung et al., 2010). Thus, to investigate whether GQ1b increases BDNF expression through the NMDA receptor signaling pathway, D-AP5 was pretreated prior to adding GQ1b to rat primary cortical neurons. GQ1b treatment significantly increased BDNF and NMDA receptor signaling pathways, but pretreatment with D-AP5 blocked the effects of GQ1b. Moreover, DAP5 also inhibited increased NR2A, NR2B, and pNR2B levels induced by GQ1b. These results indicate that GQ1b not only increased BDNF but also upregulated NMDA receptor expression and activation through the NMDA receptor signaling pathways. Increased NR2A and NR2B expression and phosphorylation of the NR2B tyrosine 1472 residue, which is associated with NMDA receptor activation and enrichment of synaptic NMDA receptors (Grosshans et al., 2002; Lavezzari et al., 2003; Roche et al., 2001), may affect the positive response, as GQ1b regulated BDNF expression through the NMDA receptor. These findings support previous reports demonstrating that the enhancing effect of GQ1b on ATPinduced LTP is mediated by modulation of NMDA receptors (Fujii et al., 2002; Furuse et al., 1998). However, the specific mechanism of how GQ1b affects NMDA receptor functions is still unknown. One possible explanation is that GQ1b modulates NMDA receptor functions by interacting with ecto-protein kinases, as a previous study has show that the ecto-protein kinase inhibitor K252b blocks the effect of GQ1b on ATP-induced LTP in brain slices (Fujii et al., 2002). Also, it has been suggested that GQ1b might act as a Ca2þ donor with a sialic acid residue, allowing for an increase in Ca2þ influx through the NMDA receptor (Fujii et al., 2002; Furuse et al., 1998; Mizutani et al., 1996). Our data also suggest that different mechanisms might be involved in BDNF regulation between GQ1b and GM1, because monosialoganglioside GM1 activates the BDNF receptor, TrkB, which results in the induction of its ligand, BDNF (Pitto et al., 1998; Cheng et al., 2011). Intracerebroventricular GQ1b injection (one microgram) increased BDNF protein expression in the hippocampus and prefrontal cortex. Our previous reports have shown that hippocampal NR2A, NR2B expression and protein levels of pNR2B, pERK and pCREB were increased by the same concentrations of GQ1b ICV injection (Jung et al., 2010). In addition, GQ1b-injected rats exhibited improved spatial learning and memory performance in Ymaze and Morris water maze tests (Jung et al., 2008). Taken together, our previous reports and current study suggest that GQ1b might enhance cognitive function through activation of the NMDA receptor signaling pathway, which increases BDNF levels in rats.
Acknowledgments We thank Professor Gwag Byoung Joo (Ajou University, Korea) for help with the rat primary neuron cultures and Dr. Chung Tae Wook (Sungkyunkwan University) for helpful discussion relating to the subject. This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF2011-0011630) and Samsung Research Fund, Sungkyunkwan University, 2010. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.neuropharm.2013.10.022. References Barbour, S., Edidin, M., Felding-Habermann, B., Taylor-Norton, J., Radin, N.S., Fenderson, B.A., 1992. Glycolipid depletion using a ceramide analogue (PDMP) alters growth, adhesion, and membrane lipid organization in human A431 cells. J. Cell. Physiol. 150, 610e619. Bliss, T.V., Collingridge, G.L., 1993. A synaptic model of memory: long term potentiation in the hippocampus. Nature 361, 31e39. Cheng, P.L., Song, A.H., Wong, Y.H., Wang, S., Zhang, X., Poo, M.M., 2011. Selfamplifying autocrine actions of BDNF in axon development. Proc. Natl. Acad. Sci. U. S. A. 108, 18430e18435. Duchemin, A.M., Ren, Q., Mo, L., Neff, N.H., Hadjiconstantinou, M., 2002. GM1 ganglioside induces phosphorylation and activation of Trk and Erk in brain. J. Neurochem. 81, 696e707. Facci, L., Leon, A., Toffano, G., Sonnino, S., Ghidoni, R., Tettamanti, G., 1984. Promotion of neuritogenesis in mouse neuroblastoma cells by exogenous gangliosides. Relationship between the effect and the cell association of ganglioside GM1. J. Neurochem. 42, 299e305. Ferrari, G., Greene, L.A., 1998. Promotion of neuronal survival by GM1 ganglioside. Phenomenology and mechanism of action. Ann. N. Y. Acad. Sci. 19, 263e273. Fighera, M.R., Royes, L.F., Furian, A.F., Oliveira, M.S., Fiorenza, N.G., Frussa-Filho, R., Petry, J.C., Coelho, R.C., Mello, C.F., 2006. GM1 ganglioside prevents seizures, Naþ, Kþ-ATPase activity inhibition and oxidative stress induced by glutaric acid and pentylenetetrazole. Neurobiol. Dis. 22, 611e623. Figurov, A., Pozzo-Miller, L.D., Olafsson, P., Wang, T., Lu, B., 1996. Regulation of synaptic responses to high-frequency stimulation and LTP by neurotrophins in the hippocampus. Nature 381, 706e709. Fong, T.G., Neff, N.H., Hadjiconstantinou, M., 1997. GM1 ganglioside improves spatial learning and memory of aged rats. Behav. Brain Res. 85, 203e211. Fujii, S., Igarashi, K., Sasaki, H., Furuse, H., Ito, K., Kaneko, K., Kato, H., Inokuchi, J., Waki, H., Ando, S., 2002. Effects of the mono and tetrasialogangliosides GM1 and GQ1b on ATP-induced long-term potentiation in hippocampal CA1 neurons. Glycobiology 12, 339e344. Furuse, H., Waki, H., Kaneko, K., Fujii, S., Miura, M., Sasaki, H., Ito, K.I., Kato, H., Ando, S., 1998. Effect of the mono and tetra-sialogangliosides, GM1 and GQ1b, on long-term potentiation in the CA1 hippocampal neurons of the guinea pig. Exp. Brain Res. 123, 307e314. Grosshans, D.R., Clayton, D.A., Coultrap, S.J., Browning, M.D., 2002. LTP leads to rapid surface expression of NMDA but not AMPA receptors in adult rat CA1. Nat. Neurosci. 5, 27e33. Heldt, S.A., Stanek, L., Chhatwal, J.P., Ressler, K.J., 2007. Hippocampus-specific deletion of BDNF in adult mice impairs spatial memory and extinction of aversive memories. Mol. Psychiatry 12, 656e670. Hettmer, S., Ladisch, S., Kaucic, K., 2005. Low complex ganglioside expression characterizes human neuroblastoma cell lines. Cancer Lett. 225, 141e149. Huang, Y.H., Lin, Y., Brown, T.E., Han, M.-H., Saal, D.B., Neve, R.L., Zukin, R.S., Sorg, B.A., Nestler, E.J., Malenka, R.C., Dong, Y., 2008. CREB modulates the functional output of nucleus accumbens neurons: a critical role of N-methyl-Daspartate glutamate receptor (NMDAR) receptors. J. Biol. Chem. 283, 2751e 2760. Impey, S., Obrietan, K., Wong, S.T., Poser, S., Yano, S., Wayman, G., Deloulme, J.C., Chan, G., Storm, D.R., 1998. Cross talk between ERK and PKA is required for Ca2þ stimulation of CREB-dependent transcription and ERK nuclear translocation. Neuron 21, 869e883. Johannessen, M., Delghandi, M.P., Moens, U., 2004. What turns CREB on? Cell. Signal 16, 1211e1227. Jung, W.R., Kim, H.G., Kim, K.L., 2008. Ganglioside GQ1b improves spatial learning and memory of rats as measured by the Y-maze and the Morris water maze tests. Neurosci. Lett. 439, 220e225. Jung, W.R., Kim, H.G., Shin, M.K., Park, D.I., Kim, K.L., 2010. The effect of ganglioside GQ1b on the NMDA receptor signaling pathway in H19-7 cells and rat hippocampus. Neuroscience 165, 159e167.
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421 Lavezzari, G., McCallum, J., Lee, R., Roche, K.W., 2003. Differential binding of the AP2 adaptor complex and PSD-95 to the C-terminus of the NMDA receptor subunit NR2B regulates surface expression. Neuropharmacology 45, 729e737. Ledeen, R.W., 1984. Biology of gangliosides: neuritogenic and neuronotrophic properties. J. Neurosci. Res. 12, 147e159. Lewin, G.R., Barde, Y.A., 1996. Physiology of the neurotrophins. Annu. Rev. Neurosci. 19, 289e317. Lombardi, G., Moroni, F., 1992. GM1 ganglioside reduces ischemia-induced excitatory amino acid output: a microdialysis study in the gerbil hippocampus. Neurosci. Lett. 134, 171e174. Louis, J.C., Gorio, A., Massarelli, R., Harth, S., Dreyfus, H., 1983. Effect of gangliosides on the development of neurons in cell cultures. Birth Defects Orig. Artic. Ser. 19, 505e510. Lu, B., Pang, P.T., Woo, N.H., 2005. The yin and yang of neurotrophin action. Nat. Rev. Neurosci. 6, 603e614. Ma, Y.L., Wang, H.L., Wu, H.C., Wei, C.L., Lee, E.H., 1998. Brain derived neurotrophic factor antisense oligonucleotide impairs memory retention and inhibits longterm potentiation in rats. Neuroscience 82, 957e967. Mizutani, A., Kuroda, Y., Muramoto, K., Kobayashi, K., Yamagishi, K., Inokuchi, J., 1996. Effects of glucosylceramide synthase inhibitor and ganglioside GQ1b on synchronous oscillations of intracellular Ca2þ in cultured cortical neurons. Biochem. Biophys. Res. Commun. 222, 494e498. Monti, E., Bassi, M.T., Papini, N., Riboni, M., Manzoni, M., Venerando, B., Croci, G., Preti, A., Ballabio, A., Tettamanti, G., Borsani, G., 2000. Identification and expression of NEU3, a novel human sialidase associated to the plasma membrane. Biochem. J. 349, 343e351. Mutoh, T., Tokuda, A., Miyadai, T., Hamaguchi, M., Fujiki, N., 1995. Ganglioside GM1 binds to the Trk protein and regulates receptor function. Proc. Natl. Acad. Sci. U. S. A. 92, 5087e5091.
421
Okada, T., Wakabayashi, M., Ikeda, K., Matsuzaki, K., 2007. Formation of toxic fibrils of Alzheimer’s amyloid beta-protein (1e40) by monosialoganglioside GM1, a neuronal membrane component. J. Mol. Biol. 371, 481e489. Park, D.I., Kim, H.G., Jung, W.R., Shin, M.K., Kim, K.L., 2011. Mecamylamine attenuates dexamethasone-induced anxiety-like behavior in association with brain derived neurotrophic factor upregulation in rat brains. Neuropharmacology 61, 276e282. Pitto, M., Mutoh, T., Kuriyama, M., Ferraretto, A., Palestini, P., Masserini, M., 1998. Influence of endogenous GM1 ganglioside on TrkB activity, in cultured neurons. FEBS Lett. 439, 93e96. Roche, K.W., Standley, S., McCallum, J., Dune, Ly C., Ehlers, M.D., Wenthold, R.J., 2001. Molecular determinants of NMDA receptor internalization. Nat. Neurosci. 4, 794e802. Sadeghlar, F., Sandhoff, K., van Echten-Deckert, G., 2000. Cell type specific localization of sphingomyelin biosynthesis. FEBS Lett. 478, 9e12. Schengrund, C.L., 1990. The role(s) of gangliosides in neural differentiation and repair: a perspective. Brain Res. Bull. 24, 131e141. Shin, M.K., Kim, H.G., Kim, K.L., 2011. A novel trimeric peptide, Neuropep-1stimulating brain-derived neurotrophic factor expression in rat brain improves spatial learning and memory as measured by the Y-maze and Morris water maze. J. Neurochem. 116, 205e216. Silva, A.J., Kogan, J.H., Frankland, P.W., Kida, S., 1998. CREB and memory. Annu. Rev. Neurosci. 21, 127e148. Svennerholm, L., Fredman, P., 1980. A procedure for the quantitative isolation of brain gangliosides. Biochim. Biophys. Acta. 617, 97e109. Tsuji, S., Arita, M., Nagai, Y., 1983. GQ1b, a bioactive ganglioside that exhibits novel nerve growth factor (NGF)-like activities in the two neuroblastoma cell lines. J. Biochem. 94, 303e306. Zeller, C.B., Marchase, R.B., 1992. Gangliosides as modulators of cell function. Am. J. Physiol. 262, 21341e21355.
Contents lists available at ScienceDirect
Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm
The ganglioside GQ1b regulates BDNF expression via the NMDA receptor signaling pathway Min Kyoo Shin a,1, Woo Ram Jung a,1, Hong Gi Kim a, Seung Eon Roh b, Choong Hwan Kwak a, Cheorl Ho Kim a, Sang Jeong Kim b, Kil Lyong Kim a, * a b
Department of Biological Science, Sungkyunkwan University, Suwon, Gyeonggi-Do 440-746, Republic of Korea Department of Physiology, Seoul National University College of Medicine, Seoul, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 May 2013 Received in revised form 17 September 2013 Accepted 18 October 2013
Gangliosides are sialic acid-containing glycosphingolipids which play a role in neuronal functions. Among the gangliosides, tetrasialoganglioside GQ1b shows neurotrophic factor-like actions, such as increasing neurite outgrowth, cell proliferation, and long-term potentiation. In addition, we recently reported that GQ1b improves spatial learning and memory performance in naïve rats. However, it is still unknown how GQ1b exerts its diverse neuronal functions. Thus, we hypothesized that GQ1b might influence synaptic activity by regulating brain-derived neurotrophic factor (BDNF) expression, which is an important protein for synaptic plasticity and cognition. Interestingly, GQ1b treatment increased BDNF expression in GQ1b-null SH-SY5Y cell lines and rat primary cortical neurons. Additionally, we confirmed whether the observed effects were due to GQ1b or due to a ganglioside with fewer sialic acid molecules (GT1b and GD1b) created by the sialidases present on the plasma membranes, by directly applying GT1b and GD1b or GQ1b co-treated with a sialidase inhibitor. Treatment with GT1b or GD1b had no effect on BDNF expression, whereas co-treatment with a sialidase inhibitor and GQ1b significantly increased BDNF levels. Moreover, GQ1b restored the decreased BDNF expression induced by the ganglioside synthesis inhibitor, D-PDMP, in rat primary cortical neurons. GQ1b treatment significantly increased BDNF levels, whereas pretreatment with the N-methyl-D-aspartate (NMDA) receptor antagonist D-AP5 blocked the effects of GQ1b on BDNF expression, suggesting that GQ1b regulates BDNF expression via the NMDA receptor signaling. Finally, we performed an intracerebroventricular GQ1b injection, which resulted in increased prefrontal and hippocampal BDNF expression in vivo. These findings demonstrate, for the first time, that tetrasialoganglioside GQ1b regulates BDNF expression in vitro and in vivo. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Ganglioside GQ1b Brain-derived neurotrophic factor N-methyl-D-aspartate receptor
1. Introduction Gangliosides are sialic acid-containing glycosphingolipids which are primarily expressed in the outer leaflet of the plasma membrane, and are particularly enriched in the cell membranes of the central nervous system (Schengrund, 1990; Zeller and
Abbreviations: BDNF, Brain-derived neurotrophic factor; NMDA, N-methyl-Daspartate; LTP, Long term potentiation; D-AP5, D-2-amino-5phosphonopentanoate; CREB, cAMP response element binding protein; ERK, extracellular signal-regulated kinase; D-PDMP, D-threo-1-phenyl-2decanoylamino-3-morpholino-1-propranol. * Corresponding author. Tel.: þ82 31 290 7017; fax: þ82 31 290 7015. E-mail addresses: [email protected] (M.K. Shin), [email protected] (W. R. Jung), [email protected] (H.G. Kim), doinfi[email protected] (S.E. Roh), [email protected] (C.H. Kwak), [email protected] (C.H. Kim), sangjkim@snu. ac.kr (S.J. Kim), [email protected] (K.L. Kim). 1 These authors contributed equally to this study. 0028-3908/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuropharm.2013.10.022
Marchase, 1992). Gangliosides have many physiological roles in the nervous system, including neuronal survival, differentiation, and neuritogenesis (Facci et al., 1984; Ferrari and Greene, 1998; Ledeen, 1984). In addition, they are also involved in neural development and neurological diseases, such as brain ischemia, seizures, and Alzheimer’s disease (Fighera et al., 2006; Lombardi and Moroni, 1992; Louis et al., 1983; Okada et al., 2007). GQ1b is a tetrasialoganglioside with four sialic acid residues, and which exhibits neurotrophic factor-like activity in vitro and in vivo. GQ1b treatment increases cell proliferation and neurite outgrowth in human neuroblastoma cell lines (Tsuji et al., 1983). In cultured cortical neurons, D-threo-1-phenyl-2-decanoylamino-3morpholino-1-propranol (D-PDMP), an inhibitor of ganglioside synthesis, inhibits synaptic activity, whereas GQ1b was capable of restoring the decreased synaptic activity induced by D-PDMP (Mizutani et al., 1996). Exogenously added GQ1b induces long-term potentiation (LTP), the cellular model of learning and memory, in
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
CA1 hippocampal neurons of guinea pig brain slices (Fujii et al., 2002; Furuse et al., 1998). Furthermore, we have recently shown that intracerebroventricular (ICV) GQ1b administration improves spatial learning and memory performance in naïve rats, as measured by the Y-maze and Morris water maze tests (Jung et al., 2008). Also, GQ1b activates N-methyl-D-aspartate (NMDA) receptor signaling pathway in H19-7 cell lines and rat hippocampus (Jung et al., 2010). NMDA receptors are glutamate-gated cation channels which play important roles in learning and memory formation (Bliss and Collingridge, 1993). Calcium influx through NMDA receptors can activate diverse signaling pathways, including the extracellular signal-regulated kinase (ERK) signaling pathway. Activation of ERK phosphorylates ser-133 of cAMP response element binding protein (CREB) (Huang et al., 2008; Johannessen et al., 2004). CREB is a transcription factor which modulates the transcription of genes containing cAMP response elements (CRE) in their promoters (Impey et al., 1998; Silva et al., 1998). It is known that CREB regulates synaptic plasticity related gene expression, such as that of brain-derived neurotrophic factor (BDNF). BDNF is a member of the neurotrophin family, including nerve growth factor, neurotrophin-3, and neurotrophin-4/5, and plays an important role in regulating neuronal survival, growth, and differentiation (Lewin and Barde, 1996; Lu et al., 2005). It is also involved with induction and maintenance of LTP and memory formation (Figurov et al., 1996). BDNF gene knockout or knockdown mice have impaired spatial learning and memory performance on hippocampus-dependent learning tasks (Heldt et al., 2007; Ma et al., 1998). Although GQ1b exhibits neurotrophic factor-like activity in vitro and in vivo, it is still unknown how GQ1b affects diverse neuronal functions. Thus, we hypothesized that GQ1b might regulate BDNF expression for synaptic plasticity in vitro and in vivo. In the current study, we investigated whether GQ1b regulates BDNF expression in SH-SY5Y cell lines and rat primary cortical neurons, and additionally, the effect of GQ1b on BDNF expression was confirmed by two independent experiments using direct comparison with other gangliosides or co-treatment with sialidase inhibitor. Furthermore, we examined that GQ1b regulates BDNF expression via the NMDA receptor signaling pathway in cultured rat primary cortical neurons. Finally, in vivo prefrontal and hippocampal BDNF levels were determined by administration of GQ1b through ICV injection in naïve rats. 2. Materials and methods 2.1. Cell culture and reagents The SH-SY5Y human neuroblastoma cell line was cultured at 37 C in 5% CO2, and complete DMEM (Life Technologies, Grand Island, NY, USA) supplemented with 2% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA, USA) and antibioticeantimycotic solution (Life Technologies). Ceramide, GM1, GD1b, GT1b, GQ1b, and Dthreo-1-phenyl-2-decanoylamino-3-morpholino-1-propranol (D-PDMP) were purchased from Alexis Biochemicals (Exeter, UK). D-AP5 and N-acetyl-2,3-dehydro-2deoxyneuraminic acid were purchased from SigmaeAldrich (St. Louis, MO, USA). 2.2. Primary cortical neurons Primary cortical neuron cultures were prepared from SpragueeDawley rats, purchased from Orient Bio (Seoul, Korea). Cerebral cortices were removed from the brains of embryonic day 17 rats, and dissociated cells were plated at a density of 2 106 cells on 6well plates which were precoated with poly-D-lysine (Sigma) and laminin (Invitrogen). The plating medium consisted of DMEM supplemented with 5% FBS. Cultures were maintained at 37 C in a humidified 5% CO2 atmosphere. Once astroglia became confluent underneath neurons at 7 days in vitro (DIV), the media was exchanged with neurobasal medium supplemented with B27 (Invitrogen) to stop non-neuronal cell overgrowth. Almost pure neuronal cultures were obtained at 10 DIV (Supplementary Fig. 1). 2.3. Quantitative real-time PCR Total RNA was isolated from SH-SY5Y cells and rat primary cortical neurons using the QIAzolÔ Lysis Reagent kit (Qiagen, Seoul, Korea), according to the manufacturer’s instructions. RT of the RNA was performed using AccuPower RT PreMix
415
(Bioneer, Daejeon, Korea). Total RNA (1 mg) and 100 pmol of oligo dT18 were preincubated at 70 C for 5 min, and were then transferred to a reaction tube. The reaction volume was 20 mL cDNA synthesis was performed at 42 C for 60 min, followed by RT inactivation at 94 C for 5 min. For quantitative real-time RT-PCR, iQ SYBR Green supermix (Bio-Rad) and an iCycler real-time PCR machine were used to measure the expression of genes under the following conditions; 45 cycles of 95 C for 30 s, 60 C (rat BDNF), 65 C (human BDNF), 60 C (NR2A, NR2B) for 30 s, and 72 C for 30 s. Primer pairs used to amplify the target genes were: rat BDNF forward primer sequence: 50 -GGACATATCCATGACCAGAAAGAAA-30 , reverse primer sequence: 50 -GCAACAAACCACAACATTATCGAG-30 ; human BDNF forward primer sequence: 50 -AAACATCCGAGGACAAGGTG-30 , reverse primer sequence: 50 -AGAAGAGGAGGCTCCAAAGG-30 ; NR2A forward primer sequence: 50 -ATACCGGCAGAACTCCACAC-30 , reverse primer sequence: 50 -CCTCCAGTAGCCGTTCTCTG-30 ; NR2B forward primer sequence: 50 -ACCCTCAAAGCCCGACTAAT-30 , reverse primer sequence: 50 -TGAAGCAAGCACTGGTCATC-30 . For normalization of the cycling threshold values obtained with the experimental samples, GAPDH was amplified under the same conditions. All reactions were repeated at least three times independently to ensure the reproducibility of the results. 2.4. Enzyme-linked immunosorbent assay (ELISA) The BDNF Emax ImmunoAssay System was purchased from Promega (Madison, WI, USA), and BDNF levels were measured as previously described (Park et al., 2011). This ELISA kit showed <3% cross-reactivity with other related neurotrophic factors (NGF, NT-3, and NT-4) at 100 ng/ml. Each well of the 96-well polystyrene plate was incubated overnight at 4 C with 100 mL anti-BDNF monoclonal antibody (mAb), which was diluted 1:1000 in carbonate coating buffer (25 mM sodium bicarbonate and 25 mM sodium carbonate, pH 9.7). Unabsorbed mAb was removed, and the plates were washed once with TBST wash buffer (20 mM TriseHCl (pH 7.6), 150 mM NaCl, and 0.05% (v/v) Tween 20). Immediately prior to blocking, cell lysates were removed from the freezer and were allowed to warm to room temperature. Plates were blocked using 200 mL Promega 1 Block and Sample buffers, followed by 1 h incubation at room temperature. Plates were then washed using TBST wash buffer. A 30 mg (100 mL total volume) aliquot of each sample or standard (0, 7.8, 15.6, 31.3, 62.5, 125, 250, 500 pg/ml) was added, in triplicate, to the plates. Plates were incubated for 2 h with shaking (500 rpm) at room temperature. Plates were then washed five times with TBST wash buffer. Anti-human BDNF polyclonal antibody (100 mL diluted 1:500 in 1 Block and Sample buffers) was added to each well, and the plates were incubated for 2 h with shaking (500 rpm) at room temperature. The plates were again washed five times using TBST wash buffer. Anti-IgY horseradish peroxidase conjugate (100 mL diluted 1:200 in 1 Block and Sample buffers) was then added to each well, and the plates were incubated for 1 h with shaking (500 rpm) at room temperature. The plates were emptied again and were washed using TBST wash buffer. Finally, the plates were developed using 100 mL Promega TMB One Solution, and the reaction was stopped by adding 100 mL 1 N HCl. The plates were read at 450 nm on a Versa Max ELISA reader (Molecular Devices, Sunnyvale, CA, USA). 2.5. Western blotting Proteins were extracted from cells and tissues with lysis buffer containing 50 mM TriseCl, 150 mM NaCl, 0.5% Triton X-100, 0.5% NP-40, 1 mM EDTA, a protease inhibitor cocktail solution (Amersham Bioscience, Piscataway, NJ, USA; Roche, Mannheim, Germany), and a phosphatase inhibitor cocktail solution (Roche). The extracts were centrifuged at 10,000 g for 30 min at 4 C. The protein concentrations in the supernatants were determined using the Bradford assay (Bio-Rad, Hercules, CA, USA). Western blotting was performed as described previously (Shin et al., 2011). Briefly, 15e20 mg samples were boiled in Laemmli sample buffer containing 10% b-mercaptoethanol, and were separated on 10e15% sodium dodecyl sulfate-polyacrylamide gel by electrophoresis, for analysis of BDNF, GAPDH, NR2A, NR2B, p-NR2B, extracellular signal-regulated kinase (ERK)1/2, phosphorylated ERK1/2 (p-ERK1/2), cAMP response element-binding (CREB) protein, and phosphorylated CREB (p-CREB). Proteins were transferred onto nitrocellulose membranes (Whatman, Milford, MA, USA) at 100 V, and the membranes were blocked for 1 h in TBST containing 7% skim milk. Antibodies recognizing NR2A (1:1000, Santa Cruz Biotechnology, Santa Cruz, CA, USA, Catalogue number: sc-1468, antibody name: NMDAε1 (C-17)), NR2B (1:1000, Millipore, Bedford, MA, USA, Catalogue number: 06-600, antibody name: Anti-NR2B rabbit polyclonal IgG), p-NR2B (1:1000, Sigma, Saint Louis, USA, Catalogue number: M2442, antibody name: AntiPhospho-N-Methyl D-Aspartate (NMDA) NR2B Subunit (pTyr1472)), ERK1/2 (1:2000, Cell Signaling Technology, Danvers, MA, USA, Catalogue number: 9102, antibody name: p44/42 MAP kinase Antibody), p-ERK1/2 (1:2000, Cell Signaling Technology, Catalogue number: 9101, antibody name: Phospho-p44/42 MAP kinase (Thr202/ Tyr204) antibody), CREB (1:2000, Cell signaling Technology, Catalogue number: 9197S, antibody name: CREB (48H2) Rabbit mAb), p-CREB (1:1000, Santa Cruz Biotechnology, Catalogue number: sc-7978, antibody name: p-CREB-1 (Ser 133)), BDNF (1:1000, Alomone labs, Jerusalem, Israel, Catalogue number: ANT-010, antibody name: Anti-BDNF), and GAPDH (1:5000, Millipore, Catalogue number: MAB374, antibody name: Mouse Anti-Glyceraldehyde-3-phosphate dehydrogenase monoclonal antibody) were used for immunodetection. All blots were developed using a chemiluminescent horseradish peroxidase substrate kit (Millipore).
416
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
Fig. 1. GQ1b increases BDNF mRNA and protein expression in SH-SY5Y cells. BDNF mRNA expression was analyzed by quantitative real-time RT-PCR. GQ1b 1 mM treatment for 12 h most significantly increased BDNF mRNA expression (A). Time course of expression levels of BDNF mRNA following 1 mM GQ1b treatment (B). An enzyme-linked immunosorbent assay was performed to measure BDNF protein levels in SH-SY5Y cells treated with 1 mM GQ1b for 12, 24 h. BDNF protein expression increased significantly following GQ1b treatment in SH-SY5Y cells (C). Western blotting and densitometry analysis showing that the gangliosides GD1b, GT1b, and ceramide had no effects on BDNF regulation, whereas GM1 and GQ1b increased BDNF levels. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (D). SH-SY5Y cells were co-treated with 50 mM of the sialidase inhibitor (NeuAc2en) and 1 mM GQ1b for 12 h GQ1b increased BDNF protein expression, and the expression level was not changed by co-treatment with sialidase inhibitor, as determined by western blotting and quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (E). These data represent typical results from three independent experiments and are expressed as mean values SEM. *p < 0.05, **p < 0.01, ***p < 0.001 vs. vehicle-treated group, #p < 0.05 vs. NeuAc2en-treated group.
2.6. RNAi Pre-designed BDNF siRNA duplexes sets from Genolution Pharmaceuticals Inc. (Seoul, Korea) were transfected to knockdown BDNF expression. Targeting siRNA and non-targeting control siRNA sequences were as follows: BDNF-1 siRNA (sense, 50 GUAUAGAUUAGAUUAUAUUUU-30 ; antisense, 50 -AAUAUAAUCUAAUCUAUACUU-30 ), BDNF-2 siRNA (sense, 50 -CUUUAAUUGUGAAUUGAUAUU-30 ; antisense, 50 -UAUCAAUUCACAAUUAAAGUU-30 ), BDNF-3 siRNA (sense, 50 -GAAUUCCUACAAUAUAUAUUU-30 ; antisense, 50 -AUAUAUAUUGUAGGAAUUCUU-30 ), BDNF-4 siRNA (sense, 50 -CAAAGAAUUUCAAUUUGUUUU-30 ; antisense, 50 -AACAAAUUGAAAUUCUUUGUU-30 ), BDNF-5 siRNA (sense, 50 -CUAAUUAACAUAUAAUAUAUU-30 ; antisense, 50 -UAUAUUAUAUGUUAAUUAGUU-30 ), and control siRNA (sense, 50 -CCUCGUGCCGUUCCAUCAGGUAGUU30 ; antisense, 50 -CUACCUGAUGGAACGGCACGAGGUU-30 ). SH-SY5Y cells were transfected with a final concentration of 30 nM of each siRNA duplexes set, using RNAiMaxTM reagent (Invitrogen, Carlsbad, CA, USA). The cells were then incubated for 24 h and were harvested for Western blot analysis to confirm knockdown of BDNF expression. 2.7. Thin-layer chromatography Cultured SH-SY5Y cells were washed twice with ice-cold PBS and were scraped from the dishes. Gangliosides were extracted according to the method of Svennerholm and Fredman (Svennerholm and Fredman, 1980). Briefly, the collected gangliosides were separated on plates of silica gel with chloroform/methanol/0.2% aqueous calcium chloride [55:45:10 (by vol.)] as a developing solution using TLC. The plates were sprayed with a resorcinol/HCl reagent and were then heated at 110 C for 30 min. 2.8. Animals and ICV injections of GQ1b Male SpragueeDawley rats (7-weeks-old) were purchased from Orient Bio (Seoul, Korea) and were acclimated for 1 week prior to the initiation of experiments. The rats were housed three per cage, and had free access to water and food prior to surgery. Rats were maintained under standard conditions: 12 h lightedark cycle, 22 2 C. Animal studies received prior approval from the animal research committee of Sungkyunkwan University. ICV injections to the left lateral ventricle were performed as described previously (Jung et al., 2008). Briefly, rats were surgically implanted with a CMA/12 guide cannula (CMA, Helsingborg, Sweden). Following
insertion of a guide cannula into the brain of each rat, they were housed one per cage, and were maintained and stabilized under standard conditions for 1 week. GQ1b (1 mg) was injected once into the lateral ventricle of the rat cerebrum, and the cannula was left in place for 30 s after infusion to allow for diffusion. Rats were sacrificed 12 h after injection. The whole brain was removed and was immediately frozen in liquid nitrogen and stored at 80 C. 2.9. Statistical analysis The presented data represent typical results from three independent experiments, and are expressed as mean standard error of the mean (SEM). Statistical analysis was conducted using SPSS v. 10.0.7 (SPSS Inc., Chicago, IL, USA). One-way analysis of variance followed by Tukey’s test was used as a post-hoc test. Statistical differences between two groups were determined using the Student’s t-test. A p value <0.05 was considered to be statistically significant.
3. Results 3.1. GQ1b increases BDNF mRNA and protein expression in SH-SY5Y cells To examine whether GQ1b regulates BDNF expression, we used the SH-SY5Y cell line, which is known not to express the ganglioside GQ1b (Hettmer et al., 2005; Sadeghlar et al., 2000). We confirmed the absence of GQ1b expression in SH-SY5Y cells using immunocytochemistry and thin-layer chromatography (Supplementary Fig. 2). SH-SY5Y cells were treated with various concentrations (0.01, 0.1, 1, and 10 mM) of GQ1b for 12 h. Quantitative real-time RT-PCR analysis revealed that BDNF mRNA expression was most effectively increased in SH-SY5Y cells following treatment with 1 mM of GQ1b (Fig. 1A). Next, we examined BDNF mRNA expression at various time points (1, 3, 6, 9, 12, 15,
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
417
Fig. 2. GQ1b regulates BDNF levels in rat primary cortical neurons. Primary cortical neurons were incubated with 1 mM GQ1b for 0, 1, 3, 6, 12, and 24 h. Western blotting and densitometry analysis showing that BDNF protein expression increased significantly in primary cortical neurons treated with 1 mM GQ1b after 12 and 24 h. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (A). An enzyme-linked immunosorbent assay revealed that 1 mM GQ1b treatment for 24 h significantly increased BDNF expression, as compared to that in a vehicle-treated group (B). Western blotting analysis showing that GQ1b increased BDNF levels in the presence or absence of sialidase inhibitor NeuAc2en. Band intensity was quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (C). Monosialoganglioside GM1 (1 mM) treatment for 24 h also increased BDNF protein expression, which was less than that of the GQ1b treatment. Western blotting band intensity was quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-treated cells were set as 1 (D). Data represent typical results from three independent experiments, and are expressed as mean values SEM. *p < 0.05, **p < 0.01 vs. vehicle-treated group, #p < 0.01 vs. NeuAc2en-treated group.
and 24 h) after 1 mM of GQ1b treatment. GQ1b significantly increased the expression of BDNF mRNA after 1 h treatment, and maximal effects were observed between 9 and 15 h post-treatment (Fig. 1B). We also investigated BDNF protein levels by enzymelinked immunosorbent assay (ELISA). Treatment with 1 mM GQ1b for 12 and 24 h significantly increased BDNF protein expression, as compared to that of vehicle-treated control group, although levels decreased over time (Fig. 1C). However, these results could not exclude the possibility that increased BDNF expression might be caused by other gangliosides, such as GT1b and GD1b, which were generated from GQ1b by sialidases which are known to be present in the plasma membranes (Monti et al., 2000). Thus, we compared the effects of each ganglioside (GD1b, GT1b, and GQ1b) including ceramide backbone on BDNF expression in SH-SY5Y cells. In addition, monosialoganglioside GM1 was also treated as positive control, because GM1 is known to interact with BDNF receptor, TrkB, which results in the induction of its ligand, BDNF (Cheng et al., 2011; Pitto et al., 1998). Treatment with 1 mM of ceramide, GD1b, or GT1b for 12 h did not affect BDNF protein levels. Although both GM1 and GQ1b significantly increased BDNF expression, GQ1b was more effective in increasing BDNF in SH-SY5Y cells than GM1 (Fig. 1D). Next, we inhibited the sialidase function in SH-SY5Y cells by co-treatment with 1 mM GQ1b and the sialidase inhibitor N-
acetyl-2,3-dehydro-2-deoxyneuraminic acid (NeuAc2en) for 12 h GQ1b treatment significantly increased BDNF protein expression in the presence or absence of NeuAc2en (Fig. 1E), suggesting that tetrasialoganglioside GQ1b, but not GD1b and GT1b, could regulate BDNF expression. Furthermore, we validated BDNF antibody specificity using siRNA experiment. BDNF protein expression was significantly decreased by transfection with BDNF siRNA duplexes, as compared to control siRNA transfected SH-SY5Y cells (Supplementary Fig. 3). 3.2. GQ1b regulates BDNF expression in rat primary cortical neurons Next, we investigated the effects of GQ1b on BDNF expression in a physiologically relevant neuronal cell culture system, such as rat primary cortical neurons. BDNF protein levels were examined at various time points (1, 3, 6, 12, and 24 h) after 1 mM GQ1b treatment. We found that GQ1b significantly increased BDNF expression after 12 and 24 h (Fig. 2A). In addition, the ELISA data also showed increased BDNF levels following 1 mM GQ1b treatment for 24 h, as compared to those of the vehicle-treated control group (vehicletreated group ¼ 15.22 2.72 pg/ml, GQ1b-treated group ¼ 29.80 5.45 pg/ml; p < 0.01) (Fig. 2B). Next, to determine
418
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
Fig. 3. GQ1b regulates BDNF expression via the NMDA receptor signaling pathway in rat primary cortical neurons. Western blotting analysis showing that treatment with 1 mM GQ1b for 24 h increased protein levels of NMDA receptor signaling molecules, such as pERK1/2 and pCREB. Pretreatment with 20 mM of D-AP5 blocked GQ1b-induced activation of the NMDA receptor signaling pathway. Band intensity was quantified by densitometry analysis. pERK/ERK and pCREB/CREB levels in vehicle-treated cells were set as 1 (A). BDNF mRNA expression was analyzed by quantitative real-time RT-PCR. Treatment of 1 mM GQ1b significantly increased BDNF mRNA expression, but the effects of GQ1b on BDNF mRNA expression were abolished by pretreatment with 20 mM of D-AP5 (B). Western blotting analysis showing that BDNF protein levels were increased by 1 mM of GQ1b treatment, and its effect is blocked by D-AP5 pretreatment (C). The increase in BDNF protein expression following 1 mM GQ1b treatment was inhibited by D-AP5 pretreatment, as determined by an enzyme-linked immunosorbent assay (D). NMDA receptor subunit 2A and 2B mRNA expression was analyzed by quantitative real-time RT-PCR. Pretreatment of D-AP5 blocked the increase of NR2A and NR2B mRNA expression induced by GQ1b treatment (E). Western blotting analysis showing that 1 mM GQ1b treatment for 24 h increased NR2A, NR2B expression and pNR2B protein levels, whereas the increase in NR2A, NR2B expressions and pNR2B protein levels following GQ1b treatment was inhibited by D-AP5 pretreatment. Band intensity was quantified by densitometry analysis. NR2A, NR2B, and pNR2B levels in vehicle-treated cells were set as 1 (F). Data represent typical results from three independent experiments, and are expressed as mean values SEM. *p < 0.05, **p < 0.01 vs. vehicle-treated control group.
whether the enhanced BDNF expression was due to GQ1b or other sialic acid cleaved forms, including GT1b and GD1b, we co-treated sialidase inhibitor, NeuAc2en, with GQ1b for 24 h in rat primary cortical neurons. BDNF protein expression increased significantly following treatment with GQ1b with or without NeuAc2en cotreatment, suggesting that GQ1b regulates BDNF expression in rat primary cortical neurons (Fig. 2C). The monosialoganglioside GM1 increased BDNF expression in SH-SY5Y cell lines (Fig. 1D). Thus, to examine whether GM1 also regulates BDNF expression in rat primary cortical neurons, we treated them with 1 mM GM1 or GQ1b for 24 h. Similarly, GM1 increased BDNF protein expression in rat primary cortical neurons, but to a lesser degree than GQ1b treatment (Fig. 2D). Given that ganglioside biosynthesis is a complicated mechanism, it is difficult to specifically block ganglioside synthesis in cells. Therefore, we used D-PDMP (a glucosylceramide synthase inhibitor), which has been commonly used to inhibit total
ganglioside biosynthesis (Mizutani et al., 1996). To investigate whether BDNF expression increases following GQ1b treatment in ganglioside-depleted cells, primary cortical neurons were incubated in 20 mM D-PDMP with or without 1 mM GQ1b for 24 h. Treatment with 1 mM GQ1b significantly increased BDNF expression, whereas D-PDMP decreased its levels, as compared to the vehicle-treated group. When cells were co-treated with GQ1b and D-PDMP, exogenously added GQ1b restored decreased BDNF expression in DPDMP-treated cells (Supplementary Fig. 4). 3.3. GQ1b increases BDNF expression through NMDA receptor signaling in rat primary cortical neurons In our previous study, we demonstrated that GQ1b treatment activates the NMDA receptor signaling pathway, and its effect is blocked by D-AP5, an NMDA receptor antagonist, in the H19-7 cell
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
419
Fig. 4. Intracerebroventricular GQ1b injection increased BDNF expression in the rat prefrontal cortex and hippocampus. Intracerebroventricular GQ1b injection was performed (1 mg GQ1b in 2 mL vehicle), and the rat prefrontal cortex and hippocampus were collected individually (vehicle, lane 1e5, n ¼ 5; GQ1b, lane 6e10, n ¼ 5) (A, B). Western blotting analysis showing that GQ1b injection significantly increased BDNF expression in the rat prefrontal cortex (A) and hippocampus (B). Each number represents an individual rat. Band intensity was quantified by densitometry analysis. BDNF/GAPDH levels in vehicle-injected group were set as 1. Data represent typical results from three independent experiments and are expressed as mean values SEM. *p < 0.01 vs. vehicle-injected group.
line (Jung et al., 2010). To investigate whether GQ1b regulates BDNF expression via the NMDA receptor signaling pathway, we first examined the protein levels of NMDA receptor signaling molecules, such as pERK1/2 and pCREB, in rat primary cortical neurons. GQ1b 1 mM treatment for 24 h significantly increased pERK1/2 and pCREB protein levels, as compared to non-treated control group, but pretreatment with 20 mM D-AP5 for 20 min prior to adding 1 mM GQ1b inhibited GQ1b-induced activation of NMDA receptor signaling (Fig. 3A). Next, we measured BDNF mRNA and protein expression by quantitative real-time RT-PCR, western blotting, and ELISA assay. BDNF mRNA and protein expression increased following treatment with GQ1b, whereas pretreatment of D-AP5 completely blocked the effects of GQ1b on BDNF expression (Fig. 3BeD). Although these results suggest that GQ1b regulates BDNF expression via the NMDA receptor signaling pathway, it is still unknown whether the enhancement of NMDA receptor signaling by GQ1b was due to the activation of NMDA receptor or the up-regulation of NMDA receptor expression. Thus, we measured NMDA receptor subunit 2A (NR2A) and 2B (NR2B) mRNA, as well as protein expression and pNR2B (Tyr-1472) protein levels by quantitative real-time RT-PCR and western blotting analysis, respectively. NR2A and NR2B mRNA and protein expression, as well as pNR2B protein levels, were significantly increased by GQ1b treatment, but its effect was blocked by D-AP5 pretreatment (Fig. 3EeF), suggesting that GQ1b might increase the NMDA receptor signaling pathway by activation of NMDA receptor and the up-regulation of NMDA receptor expression. 3.4. ICV GQ1b injection increases BDNF expression in rat prefrontal cortex and hippocampus GQ1b increased BDNF expression in vitro. To examine whether GQ1b regulates BDNF expression in vivo, 1 mg of GQ1b was injected into the ICV region of the brain, and western blotting analysis was performed to measure BDNF protein expression. This dose of GQ1b has been previously demonstrated to improve spatial learning and memory performance and to activate the NMDA receptor signaling pathways in rats (Jung et al., 2008, 2010). ICV injection of GQ1b significantly increased BDNF expression in the prefrontal cortex
(2.63 0.77-fold, p < 0.01) (Fig. 4A). In addition, hippocampal BDNF expression was also increased following GQ1b injection (3.65 0.53-fold, p < 0.01), as compared to the vehicle-injected group (Fig. 4B). These data indicate that GQ1b regulates BDNF expression in the prefrontal cortex and hippocampus. 4. Discussion Here we report that GQ1b treatment increased BDNF expression in both GQ1b-null SH-SY5Y cell lines and rat primary cortical neurons. We also confirmed that increased BDNF expression was due to GQ1b, but not GT1b or GD1b, which was generated from GQ1b by plasma membrane-associated sialidases. Direct application of GT1b or GD1b did not increase BDNF levels, whereas GQ1b treatment with or without a sialidase inhibitor significantly increased BDNF expression. In addition, GQ1b restored the decreased BDNF expression in ganglioside synthesis inhibitor (DPDMP) treated primary cortical neurons. Furthermore, we investigated the underlying mechanism of GQ1b-induced BDNF regulation. Treatment with GQ1b significantly increased BDNF levels, whereas its effect was completely blocked by pretreatment with the NMDA receptor antagonist D-AP5, suggesting that GQ1b regulates BDNF levels via the NMDA receptor signaling. Finally, we found that an ICV GQ1b injection significantly increased BDNF levels in the prefrontal cortex and hippocampus. These findings demonstrate, for the first time, that the tetrasialoganglioside GQ1b regulates BDNF expression in vitro and in vivo. GQ1b treatment increased BDNF expression in SH-SY5Y cell lines which did not express endogenous GQ1b, suggesting that upregulation of BDNF expression by GQ1b does not require the synthesis of endogenous GQ1b. Monosialoganglioside GM1 also increased BDNF expression in the SH-SY5Y cell line and rat primary cortical neurons, which was less than that of the GQ1b treatment. This finding seems to supports previous research showing that GM1 has neurotrophic functions and the ability to improve spatial learning and memory in aged rats (Duchemin et al., 2002; Fong et al., 1997; Mutoh et al., 1995). In addition, GQ1b-treated brain slices dominantly enhanced LTP compared with the GM1-treated slices (Furuse et al., 1998).
420
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421
Inhibiting ganglioside biosynthesis with D-PDMP reduced BDNF expression, but GQ1b co-treatment with D-PDMP restored the decreased BDNF levels in rat primary cortical neurons. Mizutani et al. have shown that D-PDMP treatment significantly decreases intracellular Ca2þ levels, as measured by fura-2 calcium imaging. A series of gangliosides (GM3, GM1, GD3, GD1b, and GT1b) did not normalize the decreased intracellular Ca2þ levels, but that only GQ1b restored Ca2þ levels in D-PDMP-treated rat primary cortical neurons (Mizutani et al., 1996), suggesting that GQ1b might recover BDNF levels by normalizing intracellular Ca2þ concentration through the NMDA receptor in D-PDMP-treated cells. Although GQ1b-induced upregulation of BDNF levels does not require the synthesis of endogenous GQ1b expression, co-treatment of GQ1b with D-PDMP showed reduced effects of GQ1b on BDNF protein levels. These results may be explained by the fact that D-PDMP treatment alters membrane fluidity. Barbour et al. demonstrated that D-PDMP treatment changes in membrane lipid organization contribute to the disruption of the cellular function of proteins at the cell surface (Barbour et al., 1992). Thus, further studies are required to examine whether D-PDMP treatment affects cellular function of NMDA receptors in rat primary cortical neurons. We have previously reported that GQ1b activates NMDA receptor signaling pathways (Jung et al., 2010). Thus, to investigate whether GQ1b increases BDNF expression through the NMDA receptor signaling pathway, D-AP5 was pretreated prior to adding GQ1b to rat primary cortical neurons. GQ1b treatment significantly increased BDNF and NMDA receptor signaling pathways, but pretreatment with D-AP5 blocked the effects of GQ1b. Moreover, DAP5 also inhibited increased NR2A, NR2B, and pNR2B levels induced by GQ1b. These results indicate that GQ1b not only increased BDNF but also upregulated NMDA receptor expression and activation through the NMDA receptor signaling pathways. Increased NR2A and NR2B expression and phosphorylation of the NR2B tyrosine 1472 residue, which is associated with NMDA receptor activation and enrichment of synaptic NMDA receptors (Grosshans et al., 2002; Lavezzari et al., 2003; Roche et al., 2001), may affect the positive response, as GQ1b regulated BDNF expression through the NMDA receptor. These findings support previous reports demonstrating that the enhancing effect of GQ1b on ATPinduced LTP is mediated by modulation of NMDA receptors (Fujii et al., 2002; Furuse et al., 1998). However, the specific mechanism of how GQ1b affects NMDA receptor functions is still unknown. One possible explanation is that GQ1b modulates NMDA receptor functions by interacting with ecto-protein kinases, as a previous study has show that the ecto-protein kinase inhibitor K252b blocks the effect of GQ1b on ATP-induced LTP in brain slices (Fujii et al., 2002). Also, it has been suggested that GQ1b might act as a Ca2þ donor with a sialic acid residue, allowing for an increase in Ca2þ influx through the NMDA receptor (Fujii et al., 2002; Furuse et al., 1998; Mizutani et al., 1996). Our data also suggest that different mechanisms might be involved in BDNF regulation between GQ1b and GM1, because monosialoganglioside GM1 activates the BDNF receptor, TrkB, which results in the induction of its ligand, BDNF (Pitto et al., 1998; Cheng et al., 2011). Intracerebroventricular GQ1b injection (one microgram) increased BDNF protein expression in the hippocampus and prefrontal cortex. Our previous reports have shown that hippocampal NR2A, NR2B expression and protein levels of pNR2B, pERK and pCREB were increased by the same concentrations of GQ1b ICV injection (Jung et al., 2010). In addition, GQ1b-injected rats exhibited improved spatial learning and memory performance in Ymaze and Morris water maze tests (Jung et al., 2008). Taken together, our previous reports and current study suggest that GQ1b might enhance cognitive function through activation of the NMDA receptor signaling pathway, which increases BDNF levels in rats.
Acknowledgments We thank Professor Gwag Byoung Joo (Ajou University, Korea) for help with the rat primary neuron cultures and Dr. Chung Tae Wook (Sungkyunkwan University) for helpful discussion relating to the subject. This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF2011-0011630) and Samsung Research Fund, Sungkyunkwan University, 2010. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.neuropharm.2013.10.022. References Barbour, S., Edidin, M., Felding-Habermann, B., Taylor-Norton, J., Radin, N.S., Fenderson, B.A., 1992. Glycolipid depletion using a ceramide analogue (PDMP) alters growth, adhesion, and membrane lipid organization in human A431 cells. J. Cell. Physiol. 150, 610e619. Bliss, T.V., Collingridge, G.L., 1993. A synaptic model of memory: long term potentiation in the hippocampus. Nature 361, 31e39. Cheng, P.L., Song, A.H., Wong, Y.H., Wang, S., Zhang, X., Poo, M.M., 2011. Selfamplifying autocrine actions of BDNF in axon development. Proc. Natl. Acad. Sci. U. S. A. 108, 18430e18435. Duchemin, A.M., Ren, Q., Mo, L., Neff, N.H., Hadjiconstantinou, M., 2002. GM1 ganglioside induces phosphorylation and activation of Trk and Erk in brain. J. Neurochem. 81, 696e707. Facci, L., Leon, A., Toffano, G., Sonnino, S., Ghidoni, R., Tettamanti, G., 1984. Promotion of neuritogenesis in mouse neuroblastoma cells by exogenous gangliosides. Relationship between the effect and the cell association of ganglioside GM1. J. Neurochem. 42, 299e305. Ferrari, G., Greene, L.A., 1998. Promotion of neuronal survival by GM1 ganglioside. Phenomenology and mechanism of action. Ann. N. Y. Acad. Sci. 19, 263e273. Fighera, M.R., Royes, L.F., Furian, A.F., Oliveira, M.S., Fiorenza, N.G., Frussa-Filho, R., Petry, J.C., Coelho, R.C., Mello, C.F., 2006. GM1 ganglioside prevents seizures, Naþ, Kþ-ATPase activity inhibition and oxidative stress induced by glutaric acid and pentylenetetrazole. Neurobiol. Dis. 22, 611e623. Figurov, A., Pozzo-Miller, L.D., Olafsson, P., Wang, T., Lu, B., 1996. Regulation of synaptic responses to high-frequency stimulation and LTP by neurotrophins in the hippocampus. Nature 381, 706e709. Fong, T.G., Neff, N.H., Hadjiconstantinou, M., 1997. GM1 ganglioside improves spatial learning and memory of aged rats. Behav. Brain Res. 85, 203e211. Fujii, S., Igarashi, K., Sasaki, H., Furuse, H., Ito, K., Kaneko, K., Kato, H., Inokuchi, J., Waki, H., Ando, S., 2002. Effects of the mono and tetrasialogangliosides GM1 and GQ1b on ATP-induced long-term potentiation in hippocampal CA1 neurons. Glycobiology 12, 339e344. Furuse, H., Waki, H., Kaneko, K., Fujii, S., Miura, M., Sasaki, H., Ito, K.I., Kato, H., Ando, S., 1998. Effect of the mono and tetra-sialogangliosides, GM1 and GQ1b, on long-term potentiation in the CA1 hippocampal neurons of the guinea pig. Exp. Brain Res. 123, 307e314. Grosshans, D.R., Clayton, D.A., Coultrap, S.J., Browning, M.D., 2002. LTP leads to rapid surface expression of NMDA but not AMPA receptors in adult rat CA1. Nat. Neurosci. 5, 27e33. Heldt, S.A., Stanek, L., Chhatwal, J.P., Ressler, K.J., 2007. Hippocampus-specific deletion of BDNF in adult mice impairs spatial memory and extinction of aversive memories. Mol. Psychiatry 12, 656e670. Hettmer, S., Ladisch, S., Kaucic, K., 2005. Low complex ganglioside expression characterizes human neuroblastoma cell lines. Cancer Lett. 225, 141e149. Huang, Y.H., Lin, Y., Brown, T.E., Han, M.-H., Saal, D.B., Neve, R.L., Zukin, R.S., Sorg, B.A., Nestler, E.J., Malenka, R.C., Dong, Y., 2008. CREB modulates the functional output of nucleus accumbens neurons: a critical role of N-methyl-Daspartate glutamate receptor (NMDAR) receptors. J. Biol. Chem. 283, 2751e 2760. Impey, S., Obrietan, K., Wong, S.T., Poser, S., Yano, S., Wayman, G., Deloulme, J.C., Chan, G., Storm, D.R., 1998. Cross talk between ERK and PKA is required for Ca2þ stimulation of CREB-dependent transcription and ERK nuclear translocation. Neuron 21, 869e883. Johannessen, M., Delghandi, M.P., Moens, U., 2004. What turns CREB on? Cell. Signal 16, 1211e1227. Jung, W.R., Kim, H.G., Kim, K.L., 2008. Ganglioside GQ1b improves spatial learning and memory of rats as measured by the Y-maze and the Morris water maze tests. Neurosci. Lett. 439, 220e225. Jung, W.R., Kim, H.G., Shin, M.K., Park, D.I., Kim, K.L., 2010. The effect of ganglioside GQ1b on the NMDA receptor signaling pathway in H19-7 cells and rat hippocampus. Neuroscience 165, 159e167.
M.K. Shin et al. / Neuropharmacology 77 (2014) 414e421 Lavezzari, G., McCallum, J., Lee, R., Roche, K.W., 2003. Differential binding of the AP2 adaptor complex and PSD-95 to the C-terminus of the NMDA receptor subunit NR2B regulates surface expression. Neuropharmacology 45, 729e737. Ledeen, R.W., 1984. Biology of gangliosides: neuritogenic and neuronotrophic properties. J. Neurosci. Res. 12, 147e159. Lewin, G.R., Barde, Y.A., 1996. Physiology of the neurotrophins. Annu. Rev. Neurosci. 19, 289e317. Lombardi, G., Moroni, F., 1992. GM1 ganglioside reduces ischemia-induced excitatory amino acid output: a microdialysis study in the gerbil hippocampus. Neurosci. Lett. 134, 171e174. Louis, J.C., Gorio, A., Massarelli, R., Harth, S., Dreyfus, H., 1983. Effect of gangliosides on the development of neurons in cell cultures. Birth Defects Orig. Artic. Ser. 19, 505e510. Lu, B., Pang, P.T., Woo, N.H., 2005. The yin and yang of neurotrophin action. Nat. Rev. Neurosci. 6, 603e614. Ma, Y.L., Wang, H.L., Wu, H.C., Wei, C.L., Lee, E.H., 1998. Brain derived neurotrophic factor antisense oligonucleotide impairs memory retention and inhibits longterm potentiation in rats. Neuroscience 82, 957e967. Mizutani, A., Kuroda, Y., Muramoto, K., Kobayashi, K., Yamagishi, K., Inokuchi, J., 1996. Effects of glucosylceramide synthase inhibitor and ganglioside GQ1b on synchronous oscillations of intracellular Ca2þ in cultured cortical neurons. Biochem. Biophys. Res. Commun. 222, 494e498. Monti, E., Bassi, M.T., Papini, N., Riboni, M., Manzoni, M., Venerando, B., Croci, G., Preti, A., Ballabio, A., Tettamanti, G., Borsani, G., 2000. Identification and expression of NEU3, a novel human sialidase associated to the plasma membrane. Biochem. J. 349, 343e351. Mutoh, T., Tokuda, A., Miyadai, T., Hamaguchi, M., Fujiki, N., 1995. Ganglioside GM1 binds to the Trk protein and regulates receptor function. Proc. Natl. Acad. Sci. U. S. A. 92, 5087e5091.
421
Okada, T., Wakabayashi, M., Ikeda, K., Matsuzaki, K., 2007. Formation of toxic fibrils of Alzheimer’s amyloid beta-protein (1e40) by monosialoganglioside GM1, a neuronal membrane component. J. Mol. Biol. 371, 481e489. Park, D.I., Kim, H.G., Jung, W.R., Shin, M.K., Kim, K.L., 2011. Mecamylamine attenuates dexamethasone-induced anxiety-like behavior in association with brain derived neurotrophic factor upregulation in rat brains. Neuropharmacology 61, 276e282. Pitto, M., Mutoh, T., Kuriyama, M., Ferraretto, A., Palestini, P., Masserini, M., 1998. Influence of endogenous GM1 ganglioside on TrkB activity, in cultured neurons. FEBS Lett. 439, 93e96. Roche, K.W., Standley, S., McCallum, J., Dune, Ly C., Ehlers, M.D., Wenthold, R.J., 2001. Molecular determinants of NMDA receptor internalization. Nat. Neurosci. 4, 794e802. Sadeghlar, F., Sandhoff, K., van Echten-Deckert, G., 2000. Cell type specific localization of sphingomyelin biosynthesis. FEBS Lett. 478, 9e12. Schengrund, C.L., 1990. The role(s) of gangliosides in neural differentiation and repair: a perspective. Brain Res. Bull. 24, 131e141. Shin, M.K., Kim, H.G., Kim, K.L., 2011. A novel trimeric peptide, Neuropep-1stimulating brain-derived neurotrophic factor expression in rat brain improves spatial learning and memory as measured by the Y-maze and Morris water maze. J. Neurochem. 116, 205e216. Silva, A.J., Kogan, J.H., Frankland, P.W., Kida, S., 1998. CREB and memory. Annu. Rev. Neurosci. 21, 127e148. Svennerholm, L., Fredman, P., 1980. A procedure for the quantitative isolation of brain gangliosides. Biochim. Biophys. Acta. 617, 97e109. Tsuji, S., Arita, M., Nagai, Y., 1983. GQ1b, a bioactive ganglioside that exhibits novel nerve growth factor (NGF)-like activities in the two neuroblastoma cell lines. J. Biochem. 94, 303e306. Zeller, C.B., Marchase, R.B., 1992. Gangliosides as modulators of cell function. Am. J. Physiol. 262, 21341e21355.