Grape seed proanthocyanidin extract inhibits glutamate-induced cell death through inhibition of calcium signals and nitric oxide formation in cultured rat hippocampal neurons

Background Proanthocyanidin is a polyphenolic bioflavonoid with known antioxidant activity. Some flavonoids have a modulatory effect on [Ca2+]i. Although proanthocyanidin extract from blueberries reportedly affects Ca2+ buffering capacity, there are no reports on the effects of proanthocyanidin on glutamate-induced [Ca2+]i or cell death. In the present study, the effects of grape seed proanthocyanidin extract (GSPE) on glutamate-induced excitotoxicity was investigated through calcium signals and nitric oxide (NO) in cultured rat hippocampal neurons. Results Pretreatment with GSPE (0.3-10 μg/ml) for 5 min inhibited the [Ca2+]i increase normally induced by treatment with glutamate (100 μM) for 1 min, in a concentration-dependent manner. Pretreatment with GSPE (6 μg/ml) for 5 min significantly decreased the [Ca2+]i increase normally induced by two ionotropic glutamate receptor agonists, N-methyl-D-aspartate and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA). GSPE further decreased AMPA-induced response in the presence of 1 μM nimodipine. However, GSPE did not affect the 50 mM K+-induced increase in [Ca2+]i. GSPE significantly decreased the metabotropic glutamate receptor agonist (RS)-3,5-Dihydroxyphenylglycine-induced increase in [Ca2+]i, but it did not affect caffeine-induced response. GSPE (0.3-6 μg/ml) significantly inhibited synaptically induced [Ca2+]i spikes by 0.1 mM [Mg2+]o. In addition, pretreatment with GSPE (6 μg/ml) for 5 min inhibited 0.1 mM [Mg2+]o- and glutamate-induced formation of NO. Treatment with GSPE (6 μg/ml) significantly inhibited 0.1 mM [Mg2+]o- and oxygen glucose deprivation-induced neuronal cell death. Conclusions All these data suggest that GSPE inhibits 0.1 mM [Mg2+]o- and oxygen glucose deprivation-induced neurotoxicity through inhibition of calcium signals and NO formation in cultured rat hippocampal neurons.


Background
Proanthocyanidins are polymers of flavonoid molecules that are widely available in fruits, vegetables, nuts, seeds, flowers, and bark, and especially in grape seeds [1]. These compounds possess a broad spectrum of antioxidative properties that provide potent protection against free radical-induced diseases, such as ischemia and reperfusion injury [2][3][4], aging [5], and carcinogenesis [6]. Proanthocyanidins are also known to possess antibacterial, antiviral, anti-inflammatory, anti-allergic, and vasodilator properties [1,7].
Glutamate is a major neurotransmitter in the central nervous system. Glutamate increases intracellular free Ca 2+ concentration ([Ca 2+ ] i ) in neurons by activating ionotropic and metabotropic glutamate receptors. In pathological conditions, including epilepsy and ischemia, a massive glutamate release leads to glutamate neurotoxicity [8,9]. The neurotoxicity is mainly due to Nmethyl-D-aspartate (NMDA) receptors, which cause excessive elevation of intracellular Ca 2+ concentration ([Ca 2+ ] i ) and subsequent neuronal cell death [10]. Elevation of [Ca 2+ ] i following NMDA receptor activation stimulates nitric oxide synthase (NOS), an enzyme that induces formation of nitric oxide (NO) in neurons [11]. NO reportedly also mediates glutamate neurotoxicity [12,13].
The present study determined whether grape seed proanthocyanidin extract (GSPE) affected glutamateinduced Ca 2+ signalling and NO formation in cultured rat hippocampal neurons. It further examined whether GSPE protects neurons against neurotoxicity induced by low extracellular Mg 2+ concentration ([Mg 2+ ] o ) and oxygen glucose deprivation.
In addition to IP 3 receptors, ryanodine receptors can mobilize intracellular Ca 2+ stores [26]. Reproducible [Ca 2+ ] i increase was induced by treatment with caffeine (10 mM) for 2 min at 10-min intervals (peak 2/peak 1 = 98.8 ± 1.0% of control response, n = 14). Pretreatment with GSPE (6 μg/ml) for 5 min did not significantly affect caffeine-induced [Ca 2+ ] i response (92.5 ± 2.4% of control response, n = 9, P > 0.05) ( Figure 5).    . The frequency of [Ca 2+ ] i spikes was calculated from data collected during a 5 min window before GSPE application for control, and during a 5 min window 5-10 min after application of the drug for GSPE-treated samples. Data are expressed as the mean ± SEM. *P < 0.05 relative to control (unpaired Student's t-test).  Figure 9A). The effect of GSPE on oxygen glucose deprivation-induced cell death was examined further ( Figure 9B). The cells in glucose-free BSS, with and without GSPE (6 μg/ml), were gassed with 85% N 2 , 10% H 2 , and 5% CO 2 for 90 min, and then were regrown in DMEM supplemented with 10% horse serum and penicillin/streptomycin in a CO 2 incubator for 24 h. Oxygen glucose deprivation decreased neuronal cell survival to 57.8 ± 2.2% of the control. However, treatment with GSPE (6 μg/ml) increased cell survival to 81.3 ± 7.3% of the control.

Discussion
The present study used an in vitro rat hippocampal culture model to determine the inhibitory mechanisms of GSPE in low [ Glutamate depolarizes membranes by an influx of Na + (partly Ca 2+ ) through non-NMDA receptors, which secondarily activate voltage-gated Ca 2+ channels and induce Ca 2+ influx [23]. Glutamate also induces Ca 2+ influx directly through NMDA receptor channels and Ca 2 + -permeable non-NMDA AMPA receptor channels. In the present study, GSPE inhibited glutamate, AMPA, and NMDA-induced [Ca 2+ ] i increase, but it did not affect the depolarization-induced [Ca 2+ ] i increase from 50 mM K + HEPES-HBSS, suggesting that GSPE inhibits AMPA-induced [Ca 2+ ] i increase by inhibiting Ca 2+ influx directly through Ca 2+ -permeable AMPA receptors. In fact, Ca 2+ -permeable AMPA receptors are strongly expressed in hippocampal neurons, especially early in development [22]. All these data suggested that GSPE inhibited Ca 2+ influx through Ca 2+ -permeable AMPA channels and NMDA channels. This data are indirectly supported by other reports that flavonoids such as baicalin, baicalein, and EGCG, decreased glutamate or NMDA-induced [Ca 2+ ] i increase [15,31].
The group I metabotropic glutamate receptor agonist, DHPG, induces a release of Ca 2+ from IP 3 -sensitive stores by activating PLC [25,32]. In the present study, GSPE inhibited DHPG-induced [Ca 2+ ] i increase. Although the working mechanism of GSPE is not obvious, GSPE may inhibit DHPG-induced Ca 2+ release from IP 3 -sensitive stores or DHPG-induced activation of PLC. Therefore, further research is needed to determine whether proanthocyanidin inhibits release of Ca 2+ from    [27,29,33]. In the present study, GSPE inhibited low [Mg 2 + ] o -induced [Ca 2+ ] i spikes. All these data suggest a possibility that proanthocyanidin can inhibit glutaminergic synaptic transmission in hippocampal neurons. In the present study, GSPE did not affect the depolarizationinduced [Ca 2+ ] i increase induced by high K + , which is involved in neurotransmitter release in the synaptic terminal. Thus, it is not clear whether proanthocyanidin inhibited synaptic transmission by decreasing glutamate release in presynaptic sites.
In the present study, GSPE completely inhibited low [Mg 2+ ] o -induced NO formation, and it slightly inhibited glutamate-induced formation. GSPE reportedly has potent inhibitory action on NO production presumably through of the inhibition of Ca 2+ -dependent nitric oxide synthase [34]. In neuronal cells, NO was synthesized from Ca 2+ -dependent enzymes, neuronal nitric oxide synthase [35,36]. Therefore, the inhibition of excessive Ca 2+ influx or Ca 2+ release from intracellular stores and formation of NO by glutamate in the present study suggest that proanthocyanidin inhibits NO formation by inhibiting glutamate or low [Mg 2+ ] o -induced [Ca 2+ ] i increase.
Previous investigations have reported that proanthocyanidin protects multiple target organs from drug-and chemical-induced toxicity. GSPE protects cells against acetaminophen-induced hepato-and nephrotoxicity, amiodarone-induced lung toxicity, doxorubicin-induced cardiotoxicity, and dimethylnitrosamine-induced spleenotoxicity [37]. GSPE inhibited 12-O-tetradecanoylphorbol-13-acetate and O-ethyl-S,S-dipropyl phosphorodithioate-induced brain neurotoxicity [2,37]. Grape seed extract has also been reported to reduce brain ischemic injury in gerbils [4,38] and rats [39], suggesting that the neuroprotective effects of proanthocyanidin are mediated by its antioxidant effects and antiapoptotic effects, respectively. However, there have been no reports on the underlying roles of calcium signalling or NO formation in proanthocyanidin-induced neuroprotection. GSPE inhibited low [Mg 2+ ] o -and oxygen glucose deprivation-induced neuronal cell death as well as both [Ca 2+ ] i increase and Ca 2+ -dependent NO formation. Ischemic insults have reportedly induced [Ca 2+ ] i increase and formation of NO in neurons [10,12,40,41]. In addition, proanthocyanidin blueberry extract is reported to have reversed dopamine, Aβ 42 , and lipopolysaccharide-induced dysregulation of Ca 2+ buffering capacity, thereby inducing neuroprotection in hippocampal neurons [20]. These results suggest that proanthocyanidin might inhibit ischemia-induced neuronal cell death by inhibiting glutamate-induced [Ca 2+ ] i signalling and NO formation as well as antioxidant effects and antiapoptotic effects.
The daily intake of proanthocyanidins may vary from tens to several hundred mg/day depending on diet [42]. Proanthocyanidins, especially oligomeric proanthocyanidins, are more easily absorbed and are present in blood after oral intake [21,43]. Catechin and epicatechin are reportedly bioavailable to the brain after ingestion of oligomeric proanthocyanidin [43], which suggests that oligomeric proanthocyanidins can cross the blood-brain barrier and affect neuronal cells. In fact, the IH636 grape seed proanthocyanidin extract (GSPE) used in the present study was composed of more than 73% oligomeric polyphenolic compounds including monomeric, dimeric, trimeric, and tetrameric proanthocyanidin [44]. Although the biological efficacy of GSPE has been studied previously in humans [37,44], the bioavailablity of GSPE used in the present study remains unknown. However, it should be noted that this particular concentration of grape seed proanthocyanidin extract (GSPE) was less than or equal to the serum concentration in humans following intake of 200 mg/kg proanthocyanidins or oligomeric proanthocyanidins [21]. These data suggest a possibility that IH636 grape seed proanthocyanidin extract (GSPE) can induce neuroprotection after intake of oligomeric proanthocyanidin in humans as well as animals.

Conclusions
The results of the present study showed that IH636 grape seed proanthocyanidin extract protected neuronal cells against the low [Mg 2+ ] o -and oxygen glucose deprivation-induced neurotoxicity in cultured rat hippocampal neurons. The neuroprotective effects of proanthocyanidin might have been mediated by inhibition of glutamate-induced calcium signalling and NO formation. These results demonstrated that proanthocyanidin, and especially oligomeric polyphenolic compounds, may have future utility as neuroprotective agents or as supplements against glutamate excitotoxicity-related neurologic disorders such as epilepsy, traumatic brain injury, and ischemia.

Primary rat hippocampal cell culture
Rat hippocampal neurons were grown in primary culture as previously described [45] with minor modifications. Adult maternal Sprague-Dawley rats (250-300 g) were used in the present study. All experimental procedures performed on the animals were conducted with the approval of the Catholic Ethics Committee of the Catholic University of Korea and were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (revised 1996). Fetuses were removed on embryonic day 17 from maternal rats anesthetized with urethane (1.3 g/ kg b.w., i.p.). Hippocampi were dissected and placed in Ca 2+and Mg 2+ -free Hank's balanced salt solution, pH 7.4. Cells were dissociated by trituration through a 5-ml pipette and then a flame-narrowed Pasteur pipette. Cells were pelleted and resuspended in Dulbecco's modified Eagle's medium (DMEM) without glutamine and supplemented with 10% fetal bovine serum and penicillin/ streptomycin (100 U/ml and 100 μg/ml, respectively). Dissociated cells were then plated at a density of 50,000 cells/well onto 25-mm-round cover glasses that were coated with poly-L-lysine (0.1 mg/ml) and washed with H 2 O. The cells were grown in a humidified atmosphere of 10% CO 2 -90% air (pH 7.4) at 37°C. The medium was replaced 72-90 h after plating with DMEM supplemented with 10% horse serum and penicillin/streptomycin and fed every 7 days by exchange of 25% of the medium. The cells were cultured without mitotic inhibitors for a minimum of 12 days. The cells were used after 14-15 days in culture. During this period, neurons developed extensive neuritic networks, and formed functional synapses. . The cover glass was then mounted in a flow-through chamber that was superfused at a rate of 1.5 ml/min. Digital calcium imaging was performed as described by Rhie et al. [46]. The chamber containing the fura-2loaded cells was mounted on the stage of an inverted microscope (Nikon TE300, Tokyo, Japan), and alternately excited at 340 nm and 380 nm by rapidly switching optical filters (10 nm band pass) mounted on a computer-controlled wheel (Lambda 10-2, Sutter Instruments Inc., Novato, CA, USA) placed between a 100 W Xe arc lamp and the epifluorescence port of the microscope. Excitation light was reflected from a dichroic mirror (400 nm for fura-2) through a 20× objective (Nikon; N.A. 0.5). Digital fluorescence images (510 nm, 40 nm band-pass) were collected with a computer-controlled, cooled, charge-coupled device camera (1280 × 1035 binned to 256 × 207 pixels, Quantix, Photometrics, Tucson, AZ., USA). Image pairs were collected every 2-20 s using an Axon Imaging Work Bench 2.2 (Axon Instruments, Inc., Forster City, CA., USA); exposure to excitation light was 120 ms per image. [Ca 2+ ] i was calculated from the ratio of the background-subtracted digital images. Cells were delimited by producing a mask that contained pixel values above a certain threshold applied to the 380 nm image. Background images were collected at the beginning of each experiment after removing cells from another area to the coverslip. Autofluorescence from cells not loaded with the dye was less than 5% and thus not corrected. Ratio values were converted to free [Ca 2+ ] i by the equation [Ca 2+ ] i = K d β(R-R min )/(R max -R), in which R was the 340/380 nm fluorescence emission ratio and K d = 224 nM was the dissociation constant for fura-2. R min , R max , and β was determined in ionomycin-permeabilized cells in calcium-free and saturated solutions (R min = 0.325, R max = 9.23, β = 7.61).

Digital [Ca 2+ ] i imaging
[Ca 2+ ] i measurement using fura-2-based-photometry [Ca 2+ ] i spikes were measured using fura-2-based-microfluorimetry [45]. The chamber containing the fura-2loaded cells was mounted on an inverted microscope (Nikon S-100F, Nikon, Tokyo, Japan). For the excitation of fura-2, light from a 75 W Xe arc lamp (LPS-220, Photon Technology International, NJ, USA) was passed through band-pass filters (340/20 and 380/20 nm, respectively). Excitation light was reflected sequentially from a dichroic mirror (400 nm) through a 40× phase contrast oil immersion objective (Nikon, Tokyo, Japan). Emitted light was reflected through a 510 nm filter to a photomultiplier tube (Model 710, Photon Technology International, NJ, USA) operating in photon-counting mode. Recordings were defined spatially with a rectangular diaphragm (D-104C, Photon Technology International, NJ, USA

Measurement of nitric oxide (NO)
To measure the formation of NO, the cells were incubated in an NO indicator DAF-2DA (20 μM) in HHSS without BSA for 60 min at 37°C. After DAF-2DA loading, the cells were rinsed with HHSS for 10 min and placed in a flow-through chamber. DAF-2T (the fluorescent triazolofluorescein produced by NO and DAF-2 reaction) images were obtained through excitation at 480 nm and emission at 535 nm/25 nm (DM 505 nm) [47] after treatment with or without GSPE.

Toxicity
For toxicity experiments, cells were plated on microetched coverslips (Belco Biotechnology, Vineland, NJ, USA) and at least 100 neurons were counted. In 0.1 mM Mg 2+ medium-induced excitotoxicity experiments, coverslips were exposed for 24 h to the 0.1 mM Mg 2+ medium with or without GSPE at 14 days in culture. After 20-24 h, the same fields of cells were recounted. In oxygen glucose deprivation-induced excitotoxicity experiments) [48], cultures were washed 3 times with a balanced salt solution (BSS: 116 mM NaCl, 5.4 mM KCl, 0.8 mM MgSO 4 , 1.0 mM NaH 2 PO 4 , 26.2 mM NaHCO 3 , 1.8 mM CaCl 2 , and 10 mg/L phenol red) lacking glucose and were aerated with an anaerobic gas mix (95% N 2 /5% CO 2 ) for 10 min to remove residual oxygen, then were transferred to an anaerobic chamber (1025/ 1029 Anaerobic System, ThermoForma, Ohio, USA) containing a gas mixture of 5% CO 2 , 10% H 2 , and 85% N 2 for 90 min. To terminate the oxygen glucose deprivation, cells were removed from the anaerobic chamber and then carefully washed with DMEM supplemented with 10% horse serum and penicillin/streptomycin. After 20-24 h, the same fields of cells were recounted.
Viable neurons were identified based on morphological criteria; they were phase-bright, had rounded somata, and extended long fine processes. Cell death was determined by comparing the number of viable neurons before and after treatment [30,49]. Viable neurons obtained were normalized and expressed as a percentage of sham-treated sister cultures (defined as 100%). Control experiments showed that the loss of viable neurons assessed in this manner was proportional to the number of neurons damaged. In control cells (medium exchange only), 28.4 ± 1.5% of the cells in the 0.1 mM Mg 2+ experiment ( Figure 8) and 27.3 ± 1.2% of the cells in the OGD experiment died.

Statistical analysis
Data are expressed as the mean ± SEM. Significance was determined using a Student's t-test or one-way analysis of variance (ANOVA) followed by a Bonferroni test.