|Home | About | Journals | Submit | Contact Us | Français|
Tetramethylenedisulfotetramine (TETS) is a potent convulsant that is considered a chemical threat agent. We characterized TETS as an activator of spontaneous Ca2+ oscillations and electrical burst discharges in mouse hippocampal neuronal cultures at 13–17 days in vitro using FLIPR Fluo-4 fluorescence measurements and extracellular microelectrode array recording. Acute exposure to TETS (≥ 2µM) reversibly altered the pattern of spontaneous neuronal discharges, producing clustered burst firing and an overall increase in discharge frequency. TETS also dramatically affected Ca2+ dynamics causing an immediate but transient elevation of neuronal intracellular Ca2+ followed by decreased frequency of Ca2+ oscillations but greater peak amplitude. The effect on Ca2+ dynamics was similar to that elicited by picrotoxin and bicuculline, supporting the view that TETS acts by inhibiting type A gamma-aminobutyric acid (GABAA) receptor function. The effect of TETS on Ca2+ dynamics requires activation of N-methyl-d-aspartic acid (NMDA) receptors, because the changes induced by TETS were prevented by MK-801 block of NMDA receptors, but not nifedipine block of L-type Ca2+ channels. Pretreatment with the GABAA receptor-positive modulators diazepam and allopregnanolone partially mitigated TETS-induced changes in Ca2+ dynamics. Moreover, low, minimally effective concentrations of diazepam (0.1µM) and allopregnanolone (0.1µM), when administered together, were highly effective in suppressing TETS-induced alterations in Ca2+ dynamics, suggesting that the combination of positive modulators of synaptic and extrasynaptic GABAA receptors may have therapeutic potential. These rapid throughput in vitro assays may assist in the identification of single agents or combinations that have utility in the treatment of TETS intoxication.
Tetramethylenedisulfotetramine (TETS), commonly called tetramine, is a highly toxic convulsant with a parenteral LD50 of 0.1–0.3mg/kg in mice or rats (Casida et al., 1976; Haskell and Voss, 1957; Voss et al., 1961). In adult humans, 7–10mg is estimated as a lethal dose (Guan et al., 1993). TETS was used as a rodenticide until banned worldwide in the early 1990s (Banks et al., 2012; Whitlow et al., 2005). It is, however, still available illegally, and is responsible for accidental and intentional poisonings, predominantly in China (Croddy, 2004; Wu and Sun, 2004; Zhang et al., 2011), but also in other countries, including the United States (Barrueto et al., 2003). Between 1991 and 2010 over 14,000 cases of TETS intoxication were reported in China with 932 deaths (Li et al., 2012). Extreme toxicity, history of intentional mass poisonings, and the absence of a specific antidote raise concern that TETS is a potential chemical threat agent that could cause mass casualties if released accidentally or intentionally (Jett and Yeung, 2010; Whitlow et al., 2005).
Mild-to-moderate poisoning with TETS leads to headache and dizziness, whereas severe intoxication produces status epilepticus and coma (Li et al., 2012; Whitlow et al., 2005). Animal studies demonstrate that TETS is active as a convulsant when administered orally, parenterally, and intraventricularly (Zolkowska et al., 2012). Sublethal seizures are not associated with evidence of cellular injury or neurodegeneration although there is delayed transient reactive astrocytosis and microglial activation (Zolkowska et al., 2012).
The primary convulsant mechanism of TETS has been thought to relate to blockade of type A gamma-aminobutyric acid (GABAA) receptors and the seizures induced in animals resemble those produced by other GABAA receptor antagonists including picrotoxin and pentylenetetrazol. Limited cellular physiological studies and results from [35S]t-butylbicyclophosphorothionate binding to brain membranes indicate that TETS inhibits GABAA receptors with an IC50 in the range of 1µM (Esser et al., 1991; Ratra et al., 2001; Squires et al., 1983) and it is therefore comparable in potency to picrotoxin as an inhibitor of GABAA receptors (Cole and Casida, 1986; Ratra et al., 2001; Squires et al., 1983).
Cultured hippocampal neurons display spontaneous synchronous Ca2+ oscillations (Tanaka et al., 1996) that are driven by action potential-dependent synaptic transmission. Chemically diverse environmental toxicants have been reported to disrupt neuronal Ca2+ oscillations (Cao et al., 2010, 2011; Choi et al., 2010; Pereira et al., 2010; Soria-Mercado et al., 2009). Convulsant agents can also dramatically influence neuronal Ca2+. For example, the organophosphate diisopropylfluorophosphate produces long-lasting Ca2+ elevations in hippocampal neurons (Deshpande et al., 2010).
Hippocampal neurons also exhibit spontaneous electrical discharges as they form functional neuronal networks. These discharges, as detected in extracellular recordings, consist of infrequent synchronized field potentials, mixed with more frequent desynchronized random action potentials (Cao et al., 2012; Frega et al., 2012). Synchronous Ca2+ oscillations and neuronal electrical firing co-occur (Jimbo et al., 1993) and are important in mediating neuronal development and activity-dependent dendritic growth (Wayman et al., 2008). Genetic or environmental factors that interfere with neurotransmission influence the overall activity of neuronal networks (Frega et al., 2012; Kenet et al., 2007; Meyer et al., 2008; Shafer et al., 2008; Wayman et al., 2012). For example, picrotoxin, a GABAA receptor antagonist, produces striking changes in network electrical activity (Cao et al., 2012; Frega et al., 2012).
In this study, we used rapid throughput assays to characterize the influence of TETS on Ca2+ dynamics and electrical discharges in cultured hippocampal neurons. Inasmuch as TETS induces changes in Ca2+ dynamics that are similar to those produced by the GABAA receptor antagonists picrotoxin and bicuculline, our results support the view that TETS acts as a GABAA receptor antagonist. Using rapid throughput Ca2+ measurement, we demonstrate that the GABAA receptor-positive modulators diazepam and allopregnanolone reduce or prevent TETS effects on Ca2+ dynamics, suggesting these agents as potential treatment strategies for TETS-induced seizures.
Materials. Fetal bovine serum and soybean trypsin inhibitor were obtained from Atlanta Biologicals (Norcross, GA). DNase, poly-l-lysine, cytosine arabinoside, (+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate (MK-801), hydroxypropyl-β-cyclodextran, and 3,5-dimethyl 2,6-dimethyl-4-(2-nitrophenyl)-1,4-dihydropyridine-3,5-dicarboxylate (nifedipine) were from Sigma-Aldrich (St Louis, MO). The Ca2+ fluorescence dye Fluo-4, Pluronic F-127, and Neurobasal medium were purchased from Life Technology (Grand Island, NY). Tetramethylenedisulfotetramine (TETS) was synthesized as described previously (Zolkowska et al., 2012). Diazepam was from Western Medical Supply (Arcadia, CA). Allopregnanolone (3α-hydroxy-5α-pregnan-20-one) was custom synthesized and characterized as > 99% pure.
Primary cultures of hippocampal neurons. Animals were treated humanely and with regard for alleviation of suffering according to protocols approved by the Institutional Animal Care and Use Committee of the University of California, Davis. Hippocampal neuron cultures were dissociated from hippocampi dissected from C57Bl/6J mouse pups at postnatal day 0–1 and maintained in Neurobasal complete medium (Neurobasal medium supplemented with NS21, 0.5mM l-glutamine, and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid [HEPES]) with 5% fetal bovine serum. For Ca2+ imaging studies using FLIPR, dissociated hippocampal cells were plated onto poly-l-lysine-coated clear-bottom, black wall, 96-well imaging plate (BD, Franklin Lakes, NJ) at a density of 0.8 × 105/well. For microelectrode array (MEA) experiments, 120 μl of cell suspension at a density of 1.5 × 106 cells/ml were added to a 12-well Maestro plate (Axion BioSystems, Atlanta, GA). After 2h incubation, a volume of 1.0ml of serum-free Neurobasal complete medium was added to each well. The medium was changed twice a week by replacing half the volume of culture medium in the well with serum-free Neurobasal complete medium. The neurons were maintained at 37°C with 5% CO2 and 95% humidity.
Measurement of synchronous intracellular Ca2+ oscillations. Hippocampal neurons between 13 and 17 days in vitro (DIV) were used to investigate how TETS alters synchronous Ca2+ oscillations that normally occur in healthy neurons at this developmental stage. This method permits simultaneous measurements of intracellular Ca2+ transients in all wells of a 96-well plate as described previously (Cao et al., 2010). Baseline recordings were acquired in Locke’s buffer (8.6mM HEPES, 5.6mM KCl, 154mM NaCl, 5.6mM glucose, 1.0mM MgCl2, 2.3mM CaCl2, and 0.0001mM glycine, pH 7.4) for 10min followed by addition of TETS and/or pharmacological agents using a programmable 96-channel pipetting robotic system, and the intracellular Ca2+ concentration ([Ca2+]i) was monitored for an additional 30min. Unless otherwise indicated, pharmacological interventions were introduced 10min prior to TETS. TETS triggered an immediate rise in [Ca2+]i that was quantified by determining the area under the curve (AUC) of the Fluo-4 arbitrary fluorescence units for a duration of 5min following TETS addition. TETS also altered the frequency and amplitude of neuronal synchronous Ca2+ oscillations, which were analyzed during the 10-min period after addition of TETS for 15min.
MEA recording. All MEA recordings were conducted on 13–17 DIV hippocampal neuronal networks at 37°C in culture medium without perfusion using a 12-well Maestro system (Axion). Each well contains 64 electrodes (30 µm diameter) in an 8 × 8 grid with interelectrode spacing of 200 μm. Before recording basal electrical activity, the cultures were equilibrated in freshly prepared, prewarmed Neurobasal complete medium for 1h. The 12-well Maestro plates were loaded onto a temperature-regulated headstage containing the recording amplifier, and raw extracellular electrical signals were acquired using Axis software (Axion BioSystems, Atlanta, GA). Signals from the amplifier were digitized at a rate of 25kHz and filtered using Butterworth Band-pass filter (cutoff frequency of 300 Hz). The Axis software was used to detect spontaneous events that exceeded a threshold of six times the noise. Raster plot and spike rate analysis were performed by exporting the raw data to the NeuroExplorer software (version 4.0, NEX Technologies, Littleton, MA).
Data analysis. Graphing and statistical analysis were performed using GraphPad Prism software (Version 5.0, GraphPad Software Inc., San Diego, CA). EC50 values were determined by non-linear regression using a three-parameter logistic equation. Statistical significance between different groups was calculated using Student’s t-test or by an ANOVA and, where appropriate, a Dunnett’s multiple comparison test; p values below 0.05 were considered statistically significant.
Cultured hippocampal neurons (13–17 DIV) exhibit spontaneous synchronous Ca2+ oscillations whose frequency and amplitude can be quantitatively assessed in real time using FLIPR (Fig. 1A). Addition of vehicle (0.01% dimethyl sulfoxide [DMSO]) had no significant effect on the properties of the synchronous Ca2+ oscillations during the 5-min phase I period or the 10-min phase II period (Fig. 1A, top trace). By contrast, exposure of the neurons to TETS caused an immediate increase in the amplitude of the Ca2+ oscillations, and at higher concentrations (3 and 10µM), a sustained plateau response that decayed slowly over the 5-min phase I period. The integrated Ca2+ signal (AUC) during the phase I period exhibited a concentration-dependent increase, with an EC50 value of 2.7µM (95% confidence interval [95% CI]: 1.4–5.2µM) (Fig. 1B). During phase II, TETS caused a concentration-dependent decrease in the frequency of the synchronous Ca2+ oscillations with an EC50 value of 1.7µM (95% CI: 0.69–4.12µM; Fig. 1C). Along with the reduction in the frequency, TETS increased the mean Ca2+ oscillation amplitude with an EC50 value of 1.8µM (95% CI: 1.12–2.80µM; Fig. 1D). TETS modestly prolonged the mean duration of individual Ca2+ transients compared with that measured from vehicle-exposed control neurons (data not shown). TETS-induced phase II Ca2+ responses (both frequency and amplitude) were reversible upon washout of TETS (Supplementary fig. 1).
For comparison, we studied the influence on Ca2+ dynamics in cultured hippocampal neurons of picrotoxin (PTX; 100µM), a noncompetitive blocker of GABAA receptors, and bicuculline (100µM), a competitive antagonist of GABAA receptors. Both antagonists elicited phase I and phase II responses that were similar to those induced by TETS (Fig. 2).
Extracellular recordings of electrical activity from multiple sites within the neuronal cultures at a high spatial resolution provide a robust measure of network activity and connectivity (Johnstone et al., 2010). After recording the basal electrical activity for 10min, increasing concentrations of TETS were serially introduced into the wells. A 10-min recording was collected at each TETS concentration. A control well was simultaneously recorded following introduction of vehicle (0.01–0.1% DMSO). Basal recordings for up to 60min showed that network firing activity was stable in the absence or presence of vehicle (Fig. 3A, left panel). Exposure to TETS concentrations of 2µM and greater produced a dramatic change in discharge pattern. Events became more highly clustered (Fig. 3A, right panel and Supplementary fig. 2) and the duration of clustered bursts induced by 6µM TETS lasted up to 10 s (Fig. 3A, right panel, fourth row). There was also an overall increase in the discharge rate (Fig. 3B). After washout of TETS, the neuronal network firing recovered to basal conditions.
We next examined the possible involvement of N-methyl-d-aspartic acid (NMDA) receptors and L-type Ca2+ channels in the effects of TETS on Ca2+ dynamics. Preincubation of neuronal cultures for 10min with MK-801 (1µM), an NMDA receptor blocker, attenuated both phase I and phase II effects of TETS (Figs. 4A–D). MK-801 slightly suppressed basal Ca2+ oscillations, which is consistent with an earlier report (Tanaka et al., 1996). By contrast, nifedipine (1µM), which inhibits L-type voltage-activated Ca2+ channels, was without effect on TETS-induced phase I or phase II Ca2+ responses (Figs. 4A–D). These results indicate that NMDA receptors but not L-type Ca2+ channels are required for the effects of TETS on Ca2+ dynamics.
We next determined whether the GABAA receptor-positive modulators diazepam and allopregnanolone could protect against TETS-induced Ca2+ dysregulation. Figure 5A (top trace) demonstrates that the oscillatory activity of neurons exposed to vehicle remained stable over the entire recording period. Introduction of diazepam (0.1, 0.3, or 1µM) attenuated the amplitude of basal spontaneous Ca2+ oscillations (Fig. 5A). Pre-exposure to diazepam caused a small concentration-dependent reduction of the phase I integrated rise in [Ca2+]i induced by TETS that reached statistical significance only at 1µM (Fig. 5B). Diazepam did not eliminate the phase I plateau response (Fig. 5A). Diazepam also caused a partial inhibition of the phase II frequency and amplitude effects of TETS, with the effect on amplitude reaching significance at 0.1µM (Figs. 5C and D).
As shown in Figure 6, allopregnanolone similarly attenuated the effects of TETS on Ca2+ dysregulation. Allopregnanolone (0.1–1µM) caused a concentration-dependent suppression of basal spontaneous Ca2+ fluctuations and it partially attenuated the response in phase I at 1µM without eliminating the plateau in Ca2+ levels (Figs. 6A and B). Allopregnanolone at 0.3 and 1µM also inhibited the phase II effect of TETS on the frequency and amplitude of Ca2+ oscillations, completely reversing phase II effects on transient amplitudes at 1µM (Figs. 6C and D).
We next evaluated the effect of a combination of diazepam and allopregnanolone, each at a low concentration (0.1µM) that by itself had minimal effects on phase I or phase II Ca2+ dysregulation. As shown in Figure 7, the combination strongly mitigated both phase I and phase II effects. In fact, the combination treatment was able to largely eliminate the plateau response obtained with acute TETS exposure (Fig. 7A), an effect not obtained with 10-fold higher concentrations of diazepam (Fig. 5) or allopregnanolone (Fig. 6) alone.
In this study, we characterized the effects of TETS on hippocampal neurons in culture using MEA field potential recording and Fluo-4 fluorescence measurements of Ca2+ dynamics in the neuronal network. Over time, hippocampal neurons in culture develop a rich network of processes and form numerous functional synaptic contacts (Arnold et al., 2005; Mennerick et al., 1995). Cultures that have developed for 13–17 DIV as used in this study are well organized and there is robust spontaneous electrical activity mediated by excitatory and inhibitory transmission between neurons. Neurons within such cultures exhibit spontaneous action potentials and cultures of sufficient cell density may show synchronized bursting of neurons throughout the entire culture (Arnold et al., 2005). Excitatory synaptic transmission is mediated by functional glutamate receptors of the NMDA and AMPA types (Abele et al., 1990). Importantly, the cultures contain GABAergic neurons, which comprise ~10% of the neuronal cell population. These GABAergic neurons form robust inhibitory synaptic connections mediated by GABAA receptors that exhibit physiological properties similar to those in intact preparations (Jensen et al., 1999, 2000). The GABAergic neurons impose tonic inhibition onto the network so that exposure of hippocampal cultures to GABAA receptor antagonists causes increased action potential firing, spontaneous rhythmic neuronal depolarizations, and bursting. The rhythmic depolarizations and bursting are dependent upon action potentials as they are eliminated by tetrodotoxin.
MEA recording allows the electrical activity of multiple neurons within the culture to be monitored, whereas FLIPR Fluo-4 fluorescence measurements provide a dynamic assessment of aggregate intracellular Ca2+ levels (Cao et al., 2010, 2012). Using these assays, we found that TETS dramatically increases intracellular Ca2+ levels and alters Ca2+ dynamics, initially causing a transient increase in the [Ca2+]i followed by a decrease in the frequency of synchronized Ca2+ oscillations but bigger transient amplitudes. MEA recordings of ongoing electric activity in the cultures showed an overall increase in discharge frequency and a change in the pattern of these discharges to a more clustered pattern interspersed by periods of electrical silence. The actions of TETS on neuronal Ca2+ dynamics and electrical discharge activity occur within the same concentration range, suggesting the two effects are mechanistically linked. The magnitude of the effects produced by TETS increase in a concentration-dependent manner with EC50 values of ~1–2µM; TETS inhibits GABAA receptor responses in other preparations at similar concentrations (Bowery et al., 1975; Dray, 1975; Roberts et al., 1981). Moreover, TETS-induced changes in Ca2+ dynamics and electrical discharges resemble those induced by the GABAA receptor antagonists bicuculline and picrotoxin (Arnold et al., 2005; Cao et al., 2012). Collectively, these observations support the view that the GABAA receptor-blocking activity of TETS is responsible for the effects on Ca2+ dynamics and electrical activity. Like picrotoxin, TETS is believed to be a reversible inhibitor of GABAA receptors, which is also consistent with the rapid reversibility of its effects on burst discharges and Ca2+ dynamics.
TETS-triggered alterations in electrical firing and synchronous Ca2+ oscillations appear to rely on spontaneous action potentials because they are prevented by tetrodotoxin block of Na+ channels (data not shown). The neuronal specificity of TETS in producing both phase I and phase II Ca2+ responses in hippocampal cultures is indicated by the observations that addition of TETS at concentrations of up to 3µM to the culture medium of skeletal myotubes alters neither basal Ca2+ levels nor electrically evoked Ca2+ transients (data not shown).
A key observation in this study is that the alterations in Ca2+ dynamics induced by TETS were largely inhibited by MK-801 demonstrating that NMDA receptors are required. The effect of TETS is unlikely to be due to direct activation of NMDA receptors inasmuch as bath application of NMDA, which directly activates NMDA receptors, fails to induce clustered burst discharges as observed with TETS and other GABAA receptor antagonists, although it does increase the overall discharge frequency (Cao et al., 2012). The NMDA receptor dependence for TETS-triggered Ca2+ responses is consistent with earlier in vivo reports that the NMDA receptor antagonist MK-801 inhibits picrotoxin or bicuculline-induced convulsion in mice (Czlonkowska et al., 2000; Obara, 1995). Moreover, ex vivo studies have demonstrated that the NMDA antagonist 2-APV suppresses picrotoxin-induced Ca2+ responses as well as the frequency and duration of the epileptiform discharges in the hippocampal slice preparation (Kohr and Heinemann, 1989). How NMDA receptor antagonists inhibit responses to GABAA receptor blockade remains to be determined. One possibility is that GABA antagonists cause enhanced synaptic glutamate release, leading to activation of synaptic NMDA receptors. The phase I Ca2+ response may therefore in part be generated by Ca2+ entry through these NMDA receptors. In support of this possibility, bicuculline-induced Ca2+ responses have been shown to involve synaptic but not extrasynaptic NMDA receptor activation (Hardingham et al., 2001, 2002).
Consistent with the role of GABAA receptors in restraining bursting and altered Ca2+ dynamics is our observation that the GABAA receptor-positive modulators diazepam and allopregnanolone are able to protect against the effects of TETS on Ca2+ dynamics. Allopregnanolone was more effective in mitigating the phase I response induced by TETS than by diazepam. It is well recognized that benzodiazepines such as diazepam only act on synaptic GABAA receptors, whereas neurosteroids such as allopregnanolone preferentially enhance extrasynaptic GABAA receptors although they act on synaptic receptors as well (Kokate et al., 1994; Lambert et al., 2003; Reddy and Rogawski, 2012). Therefore, the enhanced efficacy of allopregnanolone may be due to its additional action on extrasynaptic receptors. In addition, at higher concentrations, allopregnanolone is able to directly activate GABAA receptors in the absence of GABA, whereas benzodiazepines require the presence of GABA. Therefore, there are important pharmacological differences in the action of neurosteroids and benzodiazepines at GABAA receptors that could account for the enhanced activity of allopregnanolone. However, neither diazepam nor allopregnanolone alone was fully effective, even at the highest concentrations tested (1µM). Unexpectedly, we found that the combination of diazepam and allopregnanolone, each at a threshold concentration of 0.1µM, was highly effective at protecting against the effects of TETS on Ca2+ dynamics, causing a nearly complete inhibition of the phase I response, including the plateau in Ca2+, as well as the phase II changes. The combination of a benzodiazepine and a neurosteroid has not to our knowledge previously been studied in a simplified functional system. However, there is evidence from behavioral studies that neurosteroids can potentiate the actions of benzodiazepines (Gerak et al., 2004; Molina-Hernandez et al., 2003). The mechanism underlying the synergistic effect is not known. It is conceivable that the combined action on synaptic and extrasynaptic receptors accounts for the unique potency of the drug combination.
An alternative explanation of the synergism hypothesizes an interaction at the level of individual GABAA receptors. The recognition sites for neurosteroids on GABAA receptors are distinct from those that recognize benzodiazepines and barbiturates (Johnston, 1996). It is conceivable, however, that allopregnanolone and diazepam could produce a synergistic enhancement of GABAA receptors in a similar fashion as the synergism that occurs between barbiturates and benzodiazepines, where there is known to be allosteric coupling (DeLorey et al., 1993). Whether the in vitro synergism between diazepam and allopregnanolone observed in measurements of Ca2+ dynamics predicts efficacy of the combination in protecting against seizures induced by TETS or caused by other factors remains to be tested.
In summary, we have developed rapid throughput methods to detect TETS-induced Ca2+ dysregulation and altered electrical activity in cultured hippocampal neurons. We demonstrated that two GABAA receptors allosteric modulators, allopregnanolone and diazepam, when introduced either singly or in combination prior to TETS, mitigate TETS-induced Ca2+ dysregulation, suggesting that the in vitro methods described here have translational value to identify new therapies and optimize combinatorial strategies for the prevention of TETS poisoning. The basic approaches described here are likely to be of general utility for investigating chemically diverse threat agents that elicit changes in the electrical behavior or Ca2+ dynamics of neuronal networks in vitro. The rapid throughput approaches are also expected to be useful for identifying novel targeted interventions and for optimizing therapeutic strategies involving drug combinations.
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
National Institute of Neurological Disorders and Stroke grants 1U54 NS079202-01 (Z.C., B.D.H., M.A.R., P.J.L., and I.N.P.) and R21 NS072094 (M.A.R. and P.J.L.), National Institute on Aging grant 1R01 AG032119 (I.N.P), National Institute of Environmental Health Sciences grants 4P42 ES04699 (B.D.H., P.J.L., and I.N.P.), 1R01 ES 002710 (B.D.H.) and F32 ES007059 (M.M.). B.D.H. is a George and Judy Marcus Senior Fellow of the American Asthma Foundation.
We thank Susan Hulsizer for assistance with the neuronal cell cultures.