|Home | About | Journals | Submit | Contact Us | Français|
Novel enzyme-luminescence method is used for the rapid and sensitive in vitro detection of natural neurotoxins (e.g., shellfish and mushroom toxins) using model brain cells. Paralytic shellfish poisons gonyautoxins (e.g., GTX2,3 and GTX1,4) were detected at 1 nM level by their inhibition of glutamate release from C6 glioma cells upon drug stimulation (IC50: GTX2,3 = 30 nM and GTX1,4 = 8 nM). Activation of glutamate release from C6 cells by ibotenic acid (a mushroom toxin) was also evaluated (EC50 = 10 nM). The method was tested for real-time detection of glutamate release from primary rat cortical neurons. Dose-dependent effects of KCl (0–200 mM) and NMDA on glutamate release from primary cortical neurons were studied. The effects of different culture conditions on K+-depolarization-induced glutamate release were also investigated. The method may be applicable to screening of drugs and toxins, and finding glutamatergic neurons in brain slices without in situ staining.
Marine poisons (e.g., paralytic shellfish poisons) and mushroom toxins (e.g., ibotenic acid) are extremely neurotoxic compounds due to their potent toxicity to the nervous systems of humans. A number of paralytic shellfish poisons (PSPs) are known to be responsible for the inhibition of glutamate release from cells. Ibotenic acid, isolated from Amanita muscaria and Amanita pantherina also stimulates glutamatergic neurons or glial cells through N-methyl-d-aspartic acid (NMDA) receptor-mediated elevation of intracellular Ca2+, which can lead to exocytosis of l-glutamate release from cells , , . Glutamate, a principal excitatory neurotransmitter , , , , is responsible for several neurological disorders, including strokes, epilepsy, schizophrenia, Alzheimer’s disease, and Parkinson’s disease , . Therefore, the natural toxins such as marine poisons and mushroom toxins can be detected indirectly by in vitro monitoring of extracellular l-glutamate release. The chemical structures of gonyautoxins (e.g., GTX1, GTX2, GTX3, GTX4) and ibotenic acid (a mushroom toxin) are shown in Scheme 1.
Voltage-dependent Na+ channel is critical for depolarization and conduction in most excitable cells. Paralytic shellfish toxins, a family of potent neurotoxins block the voltage-gated Na+ channel (VGSC). A schematic illustration of the induction mechanism of the activator (e.g., veratridine (VTRD) which activates voltage-dependent Na+ channels) and the inhibition mechanism of shellfish toxins (e.g., GTX2,3, GTX1,4 which block Nav1.4) of l-glutamate release from a glutamatergic nerve cell or a glial cell is depicted by numbers in squares in Scheme 2. The activation of the NMDA receptor is also important in the release of neurotransmitters, such as l-glutamate , , , , , . A schematic representation of the action mechanisms of the NMDA receptor agonists (e.g., ibotenic acid and NMDA) is depicted by numbers in circles in Scheme 2. NMDA and ibotenic acid are specific agonists for the NMDA receptors and they help to increase intracellular concentration of Ca2+ ions as well as the release of l-glutamate from cells. However, only a very few studies have been conducted on the effects of ibotenic acid on NMDA receptors on the release of glutamate , , . The effects of NMDA are normally determined using fluorescence microscopy with a calcium indicator . However, this technique is labor intensive as the indicator reagent should be loaded into the cell bodies, and any excess of the reagent should be removed by washing each time.
Several analytical techniques have been proposed to detect natural neurotoxins, such as HPLC with pre- and post-column derivatization , , thin-layer chromatography (TLC) , fluorometry , cell bioassay , flow cytometry , neurophysiological methods , capillary electrophoresis , , hemolysis , electrochemical methods , electrophoretic method , and ELISA , . An ELISA test kit for detecting saxitoxin in shellfish is also commercially available (RIDASCREEN® Saxitoxin and RIDASCREEN®FAST Saxitoxin; r-Biopharm AG). The pros and cons of these methods is outlined in Table 1. Although all of these systems have high selectivity and adequate sensitivity for natural neurotoxins, there still remains a strong interest in finding alternative methods for the detection of these toxins more rapidly and sensitively.
Shinohara and co-workers including the author have proposed a novel enzyme-luminescence technique to detect l-glutamate sensitively and rapidly based on model brain cells (C6 glioma cells) , , , . The principle of enzyme-luminescence technique, in which l-glutamate is oxidized with glutamate oxidase to produce H2O2 that reacts with luminol to generate chemiluminescence in the presence of horseradish peroxidase, is depicted in Scheme 3.
In the first part of this study, the applicability of this system for in vitro model cell-based detection of natural neurotoxins (e.g., paralytic shellfish poisons and mushroom toxin) are demonstrated. In this procedure, racemic mixtures of the marine toxins GTX2,3 and GTX1,4 were administered to C6 glioma cells separately to evaluate the inhibition effects of these toxins on the release of glutamate from the cells upon stimulation by VTRD. Ibotenic acid, a NMDA receptor agonist was administered to C6 cells to induce the release of glutamate. The potency of shellfish poisons and the mushroom toxin was compared by calculating IC50 and EC50 values, respectively. In the second part of the study, the method is used for the real-time detection of glutamate release from primary rat cortical neurons instead of model cells. Dose-dependent excitatory effects of KCl and NMDA on the glutamate release are also presented. In addition, the effects of different culture conditions on the KCl-induced glutamate release are investigated.
Luminol (5-amino-2,3-dihydro-1,4-phthalazinedione), KCl, glucose, kanamycin sulfate and sodium glutamate monohydrate were purchased from Wako Pure Chemicals Industries, Ltd. (Osaka, Japan). l-Glutamate oxidase from Streptomyces sp. X-119-6 was obtained from Yamasa Corporation (Choshi, Japan). Peroxidase (POD) from horseradish was purchased from Toyobo Co., Ltd. (Osaka, Japan). Fetal calf serum was obtained from ICN Biomedical, Inc. Ham’s F10 medium powder, horse serum, DMEM, and penicillin–streptomycin were obtained from GIBCO. Racemic mixtures of the gonyautoxins (GTX), including GTX2,3 and GTX1,4 were received as a gift from Kitasato University, Japan. α-Amino-3-hydroxy-5-isoxazoleacetic acid (ibotenic acid) from mushrooms, modified Hank’s balanced salt (HBSS) powder, veratridine (VTRD), insulin-transferrin-sodium selenite (TIS), cytosine arabinoside (Ara-C), albumin from bovine serum, DNase I, trypsin inhibitor and N-methyl-d-aspartic acid (NMDA) were purchased from Sigma.
Rat model cells (C6 glioma cells- JCRB9096 and passage number 45) were obtained from Health Science Research Resources Bank, Osaka, Japan. Cells were grown in a T-25 flask with Ham’s F10 medium supplemented with 15% (v/v) horse serum, 2.5% (v/v) fetal calf serum and 1% penicillin–streptomycin at 37 °C with 5% CO2 in air. Passages of the cells were performed every fourth day using 0.25% trypsin—EDTA for detaching the cells for the duration of cell growth. 24 h prior to the cell assay, 4-day-old cells were inoculated at a density of 1 × 106 cells/mL into a 24-well plate.
Primary cultures of cortical neurons were prepared from Sprague–Dawley rats (Japan SLC, Hamamatsu, Japan) at embryonic day 17 or 18 (E17 or E18). In brief, small pieces of cerebral cortex were treated with 0.125% trypsin (DIFCO) and 1 mM EDTA for 15 min at 37 °C and were thoroughly dissociated in DMEM. After centrifugation, the cell pellet was treated with 0.004% DNase I and 0.03% trypsin inhibitor, and the suspended cells were seeded at 1 × 106 cells/mL in a poly-l-lysine-coated 24-well plate. The cells were grown for 2 days in DMEM (supplemented with 10% fetal calf serum and 100 mg/mL kanamycin sulfate) in a 10% CO2 incubator. The medium was then replaced with serum-free of DMEM supplemented with glucose (4.5 mg/mL), transferrin (5 μg/mL), insulin (5 μg/mL), sodium selenite (5 ng/mL), bovin serum albumin (1 mg/mL), and kanamycin sulfate (100 μg/mL) (TIS medium). Cytosine arabinoside (Ara-C) (final conc. 2 μM) was also added into TIS medium to prevent the proliferation of glial cells. After 1 d of incubation with the TIS medium, the medium was replaced again with fresh TIS medium but devoid of Ara-C. The morphology of cultured neurons is shown in Fig. 1. The l-glutamate release from primary cortical neurons was monitored by evoking KCl or NMDA using an enzyme luminescence method. A scheme depicting the use of different media during culturing is shown in Scheme 4. The rate of differentiation from primary neurons into glutamatergic neurons is thought to be high when three media are used sequentially.
The culture medium was first removed from each well and C6 cells were washed twice with warm (37 °C) Hank’s solution. The cells were then incubated separately for 10 min in the presence and absence of paralytic shellfish poisons (e.g., GTX2,3 and GTX1,4) with 270 μL Hank’s solution (pH 7.4) supplemented with POD (64 U/mL), luminol (0.05 mM) and l-glutamate oxidase (17 U/mL) prior to stimulation. The cells were stimulated by auto-injection of 30 μL solution of the activator, veratridine, NMDA or ibotenic acid using a luminescence plate-reader (FLUOstar OPTIMA, BMG LABTECH, Germany). Hank’s solution (pH 7.4, 30 μL) was injected as a stimulant for the control experiments.
The culture medium was first removed from each well and the cortical neurons were washed twice with warm (37 °C) Hank’s solution. The cells were then incubated for 20 min with 270 μL measurement solution (205 μL Hank’s solution, 30 μL POD, 30 μL luminol and 5 μL l-glutamate oxidase) prior to stimulation. The cells were stimulated by injection of 30 μL solution of either K+ or NMDA using a luminescence plate-reader. Hank’s solution was injected as a stimulant for the control experiments.
After the assay of glutamate release, the solution was replaced with 300 μL Hank’s solution and the cells were disrupted by either sonication (power 7, 40 s) or freezing (−20 °C, overnight). The sample was separated following centrifugation (13.2 × 103 rpm, 6 min) by filtration using a membrane filter (pore size 0.22 μM). For recoding the l-glutamate residue, the measurement solution supplemented with Hank's buffer (pH 7.4, 205 μL), 64 U/mL POD (30 μL), 0.05 mM luminol (30 μL), and 17 U/mL l-glutamate oxidase (5 μL) was added to each well of a 24-well plate. Purified sample solution (30 μL) was injected into the wells of the 24-well plate using an automatic reagent injector to start the enzymatic reaction and luminescence was detected using a luminescence plate-reader equipped with a highly sensitive photomultiplier tube.
In order to investigate the effects of paralytic shellfish toxins, the cells were pretreated separately with racemic mixtures of the gonyautoxins, including GTX2,3 and GTX1,4, for 10 min prior to stimulation with veratridine (final conc., 0.5 μM), and the glutamate release from the cells was monitored in real-time. Fig. 2a shows time courses of l-glutamate response measured by the enzyme-luminescence method in the presence and absence of GTX2,3 at the indicated concentrations. Fig. 2b shows a semi-logarithmic plot of the luminescence peak intensity corresponding to veratridine-evoked l-glutamate release in the same experiments. Similarly, Fig. 3a and b shows the time courses of l-glutamate response and a semi-logarithmic plot of the luminescence peak intensity, respectively, in the presence and absence of GTX1,4 at the indicated concentrations. The results indicate that increasing concentrations of both GTX2,3 and GTX1,4 progressively inhibit the veratridine-induced release of glutamate. Several studies have suggested that gonyautoxins block Nav1.4 channel, which may decrease or delay neuronal depolarization as well as glutamate release , , ,  The data reported here support this phenomena.
The calculated IC50 values for GTX2,3 and GTX1,4 to inhibit veratridine-induced glutamate release are 30 nM and 8 nM, respectively, which are higher than those reported by Choudhary et al. . This discrepancy is probably due to the difference of both the method of analysis and the cells used. However, the results are consistent with the data reported previously  showing that GTX1,4 is the more potent paralytic shellfish toxin for blocking voltage-gated Na+ channels as compared to GTX2,3.
Ibotenic acid, isolated from mushrooms, is a powerful natural neurotoxin that can bind NMDA receptors and keep them open , , , . The activation of NMDA receptors promote the influx of cations (e.g., Na+, Ca2+), which can lead to membrane depolarization. This, in turn, can help trigger exocytosis of neurotransmitters such as l-glutamate , , , , . The C6 glioma cells were stimulated by ibotenic acid to evaluate the effects of the natural toxin on NMDA-receptor-mediated l-glutamate release. The potency of this mushroom toxin was compared with that of NMDA, a specific well-known agonist of the NMDA receptor. Fig. 4a shows the time courses of l-glutamate release evoked by ibotenic acid and NMDA at the indicated concentrations. The luminescence peak intensities are significantly higher for both cases than the control, which is consistent with that of the previous reports , , , , .
The higher luminescence peak intensity (peak height) is most probably due to either the increase of H2O2 produced by the l-glutamate efflux upon stimulation or the oxidation of stimulants (e.g., ibotenic acid, NMDA) themselves by l-glutamate oxidase. The stimulation experiment was conducted using either ibotenic acid or NMDA in the absence of cells to confirm the specificity. The results (not shown) are similar to those of the control experiments, indicating that ibotenic acid and NMDA are not oxidized by l-glutamate oxidase. Therefore, the observed high luminescence intensity is due to the release of l-glutamate from the cells.
Fig. 4b shows a semi-logarithmic plot of the dose-dependence of ibotenic acid on the release of l-glutamate from C6 cells, in which the luminescence peak intensity (peak height) increases with increasing ibotenic acid concentration. The data also indicate that the minimum level of ibotenic acid that evokes a distinguishable luminescence response from cells over the control is 1 nM. Similarly, Fig. 4c shows the dependence of the luminescence peak intensity (peak height) on concentration upon stimulation with NMDA (0–200 μM). Saturated level is 50 μM NMDA and a much higher concentration of glutamate is released with this dose as compared with the control. Furthermore, the EC50 values for ibotenic acid- and NMDA-induced glutamate release were calculated to be 10 nM and 22 μM, respectively. The results demonstrate that ibotenic acid is a more potent stimulant for the activation of NMDA receptors than NMDA. Moreover, the binding affinity of ibotenic acid to NMDA receptors is higher than that of NMDA. The data are consistent with the results reported previously , .
The effects of two types of stimulants, such as KCl and NMDA on l-glutamate release from primary cortical neurons were also investigated. The luminescence peak appeared rapidly, within a few seconds, after the injection of either KCl or NMDA solution. Fig. 5a indicates that the luminescence peak intensity depends on the concentration of KCl. The data indicate that the luminescence intensity reaches a maximum at about 100 mM KCl, which is consistent with the observations in previous reports , , ,  Residue of glutamate was also measured after breaking the cells following centrifugation and filtration. The total amount of glutamate (released + unreleased) from cortical neurons determined in terms of luminescence peak intensities is shown in Fig. 5b. Fig. 5c shows the percentage of glutamate released as a function of the concentration of KCl, where the maximum glutamate release was observed at 100 mM KCl.
Fig. 6a shows the dependence of the luminescence peak intensity on concentration upon stimulation with NMDA, where the peak intensity reaches a maximum at an NMDA concentration of 100 μM. Fig. 6b shows a bar chart indicating the total glutamate response. Fig. 6c shows the percentage of glutamate released as a function of the concentration of NMDA, where the maximum amount of glutamate released was observed at 50–100 μM NMDA.
The primary cortical neurons were grown separately under different culture conditions, (i.e., DMEM + TIS with Ara-C + TIS, DMEM only, and DMEM + TIS), and the neurons were stimulated with a high concentration of KCl (100 mM) to monitor the glutamate released from the neurons. Fig. 7 shows the peak intensities due to the released and residual glutamate at different culture conditions. The percentage of glutamate released in the presence of all three media, two media, and one medium were calculated to be 38%, 32%, and 33%, respectively. The total amount of glutamate released in the presence of all three media is high compared to that of the other two culture conditions. This is probably due to the high rate of differentiation from primary neurons into glutamatergic neurons when three media are used sequentially. The increased glutamate released upon stimulation with an agonist may be a good marker of differentiation in primary cortical neurons, which is truly a dye free marker. A majority of conventional techniques used to monitor neuronal differentiation in vitro are based on immunostaining and are dependent on a fluorescent or radioactive dye , , . Thus, the enzyme-luminescence method may be a useful tool in the evaluation of neuronal differentiation ratio non-invasively, in real-time and independent of any fluorescent or radioactive probes.
The applicability of the enzyme-luminescence method reported earlier for in vitro screening of natural neurotoxins is demonstrated. Racemic mixtures of the gonyautoxins, containing GTX2,3 and GTX1,4, were used to evaluate their inhibition of the release of glutamate from the C6 glioma cells. The activation of the release of glutamate from C6 cells by ibotenic acid (a mushroom toxin) was also evaluated. In order to compare the potency of these toxins, EC50 or IC50 values were calculated. Furthermore, the enzyme-luminescence method is useful for the in vitro detection of the release of l-glutamate from primary cortical neurons. This novel detection technique may also be used to determine the neuronal differentiation ratio and to find glutamatergic neurons in real-time without immunostaining. Thus, the cell-response-based sensing technology may find widespread applications in screening of other drugs and toxins that activate or inhibit either ion-channels or receptors. This sensing tool may also have great potential for the investigation of the effects of various growth factors and chemicals on neuronal differentiation, neurotransmitter dynamics, neurodegeneration, and synaptogenesis.
I thank Professor Hiroaki Shinohara, University of Toyama, Japan for helpful suggestions and guidance. I also thank Professor M. Kodama and Professor S. Sato, Kitasato University, Japan for giving the racemic mixtures of gonyautoxins. This study was supported by research grants from the Shinminato Fisheries Cooperative, Tateyama Kagaku Group, and the Toyama Marine Biotechnology Consultation.