Neurophysiologic Effects of Chemical Agent Hydrolysis Products on Cortical Neurons In Vitro

NeuroToxicology 22 (2001) 393±400

Neurophysiologic Effects of Chemical Agent Hydrolysis Products on Cortical Neurons In Vitro Joseph J. Pancrazio1,*, Edward W. Keefer2, Wu Ma1, David A. Stenger1, Guenter W. Gross2 1

Center for Bio/Molecular Science and Engineering, Code 6910, Naval Research Laboratory, Washington, DC 20375, USA 2 Department of Biological Sciences and Center for Network Neuroscience, University of North Texas, Denton, TX 76203, USA Received 21 November 2000; accepted 15 March 2001

Abstract The neurophysiologic effects of chemical agent hydrolysis products were examined on cultured cortical neurons using multielectrode array (MEA) recording and the whole-cell patch clamp technique. Measurement of neuronal network extracellular potentials showed that the primary hydrolysis product of soman, pinacolyl methylphosphonic acid (PMPA), inhibited network mean burst and spike rates with an EC50 of approximately 2 mM. In contrast, the degradation product of sarin, isopropyl methylphosphonic acid (IMPA), and the ®nal common hydrolysis product of both soman and sarin, methylphosphonic acid (MPA), failed to affect neuronal network behavior at concentrations reaching 5 mM. Closer examination of the effects of PMPA (2 mM) on discriminated extracellular units revealed that mean spike amplitude was slightly diminished to 95  1% (mean  S:E:M:, n ˆ 6, P < 0.01) of control. Whole-cell patch clamp records under current clamp mode also showed a PMPA-induced depression of the ®ring rate of spontaneous action potentials (APs) to 36  6% (n ˆ 5, P < 0.001) of control. In addition, a minor depression with exposure to PMPA was observed in spontaneous and evoked AP amplitude to 93  3% (n ˆ 5, P < 0.05) of control with no change in either the baseline membrane potential or input resistance. Preliminary voltage clamp recordings indicated a reduction in the occurrence of spontaneous inward currents with application of PMPA. These ®ndings suggest that PMPA, unlike MPA or IMPA, may more readily interfere with one or more aspects of excitatory synaptic transmission. Furthermore, the data demonstrate that the combination of extracellular microelectrode array and patch clamp recording techniques facilitates analysis of compounds with neuropharmacologic effects. # 2001 Published by Elsevier Science Inc.

Keywords: Cortical neurons; Extracellular recording; Hydrolysis; Patch clamp; Sarin; Soman; Multielectrode array

INTRODUCTION The extreme toxicity of chemical agents, such as methylphosphono¯uoridic acid, 1-methylethyl ester (sarin) and methylphosphono¯uoridic acid, 1,2,2-tri-

* Corresponding author. Tel.: ‡1-202-404-6026; fax: ‡1-202-767-9598. E-mail address: [email protected] (J.J. Pancrazio).

methylpropyl ester (soman), have been well-documented (for a review, see Munro et al. (1994) and Brown and Brix (1998)). In spite of their toxicity, sarin and soman can undergo rapid degradation yielding more long-lasting hydrolysis products (Munro et al., 1999). It has been recognized that elimination of production sites and cleanup operations present risk management concerns relative to chemical agent hydrolysis products. While several hydrolysis products have been examined for toxicity and mutagenicity (Munro et al.,

0161-813X/01/$ ± see front matter # 2001 Published by Elsevier Science Inc. PII: S 0 1 6 1 - 8 1 3 X ( 0 1 ) 0 0 0 2 8 - 6

394

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

1999), there is a paucity of information concerning the potential neurophysiologic effects of these compounds. Unlike chemicals such as carcinogens that may not produce deleterious effects for years, neurotoxicants have the capacity to impact function directly (Fiedler et al., 1996). Therefore, the possible effects of hydrolysis products on neuronal function warrants considerable attention. To address the question of whether or not chemical agent hydrolysis products alter neuronal function, we examined the effects of several representative agents on the function of cultured cortical neurons. In the present study, we have employed two electrophysiologic methods to assess changes in cortical neuronal function in vitro: extracellular recording via microelectrode arrays (MEA) and the whole-cell patch clamp method. Extracellular recordings accomplished from neuronal networks cultured on MEAs exhibit complex spatio-temporal spike and burst patterns that are highly sensitive to their chemical environment (Gross, 1994; Gross et al., 1992, 1995, 1997a,b). The patch clamp technique allows high-resolution measurements of membrane currents and potentials from identi®ed neurons with minimal membrane damage. We report that the sarin hydrolysis products, isopropyl methylphosphonic acid (IMPA) and methylphosphonic acid (MPA), exert no discernible electrophysiologic action; however, the soman hydrolysis product pinacolyl methylphosphonic acid (PMPA) depresses cortical neuron bursting, spike generation, and, at higher concentrations, spike amplitude. In addition, we discuss the advantage of utilizing both extracellular microelectrode array and patch clamp recording techniques to facilitate neuropharmacologic analysis of compounds. MATERIALS AND METHODS Chemicals MPA, IMPA, and PMPA were purchased from Sigma/Aldrich (St. Louis, MO). Immediately prior to use, the hydrolysis products were added to standard recording media to achieve the desired ®nal concentration. Cortical Culture for Extracellular Recording The techniques used to fabricate and prepare microelectrode arrays (MEAs) have been described elsewhere (Gross, 1979; Gross et al., 1985). The electrode

conductor pattern consisted of a central 0.8 mm2 recording matrix of 64 microelectrodes (Gross, 1994). MEA surfaces were activated by ¯aming and coated with poly-D-lysine and laminin (Lucas et al., 1986). Dissociated tissue cultures were prepared according to the basic method established by Ransom et al. (1977). Frontal cortical tissues were harvested from embryonic day 16 HSD:IRC mice (Charles River Laboratories, Wilmington, MA). The tissue was dissociated enzymatically and mechanically, seeded at a density of 0:2  106 cells/cm2 in DMEM with 5% horse serum onto MEA surfaces con®ned by a 4 cm2 silicone gasket (Gross, 1994). Cultures were incubated at 378C in a 10% CO2 atmosphere until ready for use, generally 3 weeks to 3 months after seeding. The medium was replenished twice a week with DMEM containing 5% horse serum (50% volume exchange). Spontaneous activity was apparent after approximately 7 days in the form of random spiking and stabilized in terms of coordinated spike and burst pattern by 15 days in vitro. Networks can remain spontaneously active and pharmacologically responsive for more than 6 months (Gross, 1994). Extracellular Recording Procedures and Data Analysis MEAs were placed into constant-bath recording chambers (Gross and Schwalm, 1994; Gross, 1994) and maintained at 378C on a microscope stage. The pH was maintained at 7.4 with a continuous stream of ®ltered, humidi®ed, 10% CO2 in air. Stainless steel chamber components were sterilized via autoclaving before each experiment. Activity was recorded by a multiampli®er system (Plexon Inc., Dallas, TX) consisting of 64 two-stage ampli®ers with a total system gain of approximately 10,000. Often, microelectrode sites showed activity from one or more units, which were statistically distinguished based on form and time course using commercially-available software (NEX, Version 2.2, Plexon Inc.). For identi®ed units, recordings demonstrated only a small degree of variability (Fig. 1A) in spike shape and amplitude indicating a stable coupling of the biological substrate to individual recording sites. Spike amplitudes of selected channels were monitored during the course of experiments to elucidate effects on action potential (AP) shape. Raster plots of unit activity show bursts (Fig. 1B), which have been shown to reveal modes of neuronal network behavior (Gross, 1994). Bursts were identi®ed using the NEX program and burst rate and duration were quanti®ed.

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

Fig. 1. Multielectrode array data: stability of single discriminated unit recording and firing patterns. (A) Two representative units during baseline recordings; 32 overlapping events are shown, which have been distinguished as the same unit. Consistency of discriminated units is evident from records at 5 min (upper row) and 100 min (lower row), indicating that the cell-electrode coupling was stable. Dotted lines indicate zero potential baseline. (B) 15 s raster plot of six discriminated units showing a normal frontal cortex pattern of short, coordinated bursts.

Cortical Culture for Whole-Cell Patch Clamp Recording Cortical tissue was dissected from embryonic day 18 rats and cells were dissociated by papain digestion. Cells were plated in poly-D-lysine-coated 35 mm diameter culture dishes at a density of 1:14  104 cells/cm2. Cultures were maintained in serum-free neurobasal medium containing B27, 0.5 mM L-glutamine, and 25 mM glutamate in a cell culture incubator at 378C with 5% CO2/95% air. Neurons were cultured for 12 days prior to use in patch clamp experiments to ensure mature morphologic and electrophysiologic properties. Whole-Cell Patch Clamp Recordings Whole-cell patch clamp methods were employed as described by Hamill et al. (1981). While diffusible

395

messengers can be lost during whole-cell recording (Harata et al., 1997), this method is considered a standard approach for monitoring in mammalian neurons the electrophysiologic effects of toxicants (e.g. Scott et al., 1995; Lin et al., 1998) including organophosphates (Camara et al., 1997; Kao et al., 1999; Chebabo et al., 1999). Patch pipettes were pulled with a PP-83, two-stage puller and heat-polished with a CPM2 microforge (Adams List Associates, Westbury, NY). The patch electrode resistance was 5±10 MO when ®lled with an intracellular solution containing 50 mM KCl, 100 mM K-aspartate, 1 mM ethylene glycol-bis(b-aminoethyl ether)-N,N,N0 ,N0 -tetraacetic acid (EGTA)±KOH, 1 mM HEPES, 5 mM Na2ATP, pH 7.4 with 1N HCl. The extracellular bathing solution contained 140 mM NaCl, 5 mM KCl, 1.8 mM CaCl2, 0.8 mM MgCl2, 10 mM glucose, and 10 mM HEPES, pH 7.4. After achieving the cell-attached patch clamp con®guration, whole-cell recording was attempted only when the seal resistance exceeded 10 GO. Only neurons with baseline membrane potentials more negative than 50 mV were considered for subsequent drug application. For drug exposure, the bathing solution surrounding the cells was replaced with the hydrolysis product containing solution. Solution exchange was achieved within 30 s. An Axopatch 200B patch clamp ampli®er (Axon Instruments, Foster City, CA) coupled with the PCLAMP 6 data acquisition system was used for current clamp recording, where potentials were ®ltered at 2 kHz with a four pole, Bessel low-pass ®lter and digitized at 10 kHz. All patch clamp experiments were conducted at room temperature. Statistics Where appropriate, data are presented as mean percentage of control  S:E:M: and the number of cells/experiments performed (n). Statistical signi®cance was determined using Student's t-test with P < 0:05 considered as signi®cant. RESULTS Extracellular Recordings Spontaneous activity obtained from cultured frontal cortical networks combined phasic and tonic spiking with bursts of variable durations into complex temporal patterns. Under control conditions, spikes were typically biphasic with duration of 0.5±1.0 ms and amplitude ranging from 200 to 1200 mV (Fig. 1A). Bursting

396

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

Fig. 2. Comparative effects of chemical agent hydrolysis products on neuronal network activity. Graphs show mean burst (left ordinate) and spike (right ordinate) rates for 1 min periods averaged across all discriminated units (27 units for (A) and 16 units for (B)); data were smoothened using a three-point moving bin averaging window. (A) Pinacolyl methylphosphonic acid (PMPA) induced a concentration-dependent inhibition of frontal cortex burst and spike rates. (B) Methylphosphonic acid (MPA) exerted no apparent effect on network activity. Like MPA, isopropyl methylphosphonic acid failed to alter network behavior at concentrations up to 5 mM.

was ubiquitous, and many networks displayed burst patterns that were coordinated among some or all of the channels selected for analysis (Fig. 1B). Typical

spontaneous baseline burst rates, averaged across the units recorded, ranged from 9 to 53 bursts/min (bpm) and burst durations ranged from 50 to 100 ms. To screen network activity for neuronal effects of the hydrolysis products, averaged activity across the units recorded from each culture experiment was evaluated for a range of concentrations. Fig. 2 summarizes changes in mean spike rate and burst rate for representative experiments with PMPA and MPA. Application of PMPA to neuronal networks elicited a depression in burst rate, and to a lesser extent, spike rate for all channels recorded (n ˆ 6; Fig. 2A). Overall, the EC50 for mean burst rate inhibition was 2:0  0:3 mM (n ˆ 5). In contrast to PMPA, the ®nal degradation product common to both soman and sarin, MPA, did not appear to alter neuronal function at concentrations up to 5 mM (n ˆ 4; Fig. 2B). The sarin hydrolysis product, IMPA, also failed to markedly affect bursting at concentrations reaching 5 mM (n ˆ 2; data not shown). Therefore, subsequent analyses and experiments were focused on PMPA. A subset of typical units was identi®ed for quantitation under control conditions and after exposure to PMPA. Under control conditions, the mean burst rate, mean burst duration, and peak-to-peak spike amplitude were 14:6  4:2 bpm (mean  S:E:M:; n ˆ 6), 0:26  0:14 s (n ˆ 6), and 160  21 mV (n ˆ 6), respectively. As shown in Fig. 3A, administration of PMPA resulted in a concentration-dependent depression in mean burst rate and, to a lesser extent, spike amplitude. For example, at 2 mM PMPA, burst rate was diminished to 61  5% (n ˆ 6; P < 0:001) of the control level, whereas the spike amplitude was reduced to only 95  1% (n ˆ 6; P < 0:01). No statistically-signi®cant effect on burst duration could be discerned. At higher concentrations (4 mM), spike amplitude was depressed, as shown in Fig. 3B. After the inhibition of bursting with 4 mM PMPA, the remaining spiking

Fig. 3. Inhibition of burst rate and spike amplitude by PMPA. (A) Comparative concentration-dependence of PMPA-induced reductions in burst rate and spike amplitude quantified from a subset of six discriminated units. Data are given as mean percent of control  S:E:M:;  P < 0:05; yP < 0:01; zP < 0:001. (B) Ensemble mean of 15 waveforms from a discriminated, single unit before and after exposure to 4 mM PMPA. Data were digitized at 40 kHz. Dotted line indicates zero potential baseline.

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

397

Fig. 4. Whole-cell patch clamp recordings of spontaneous activity from embryonic cortical neurons. (A) Under current clamp recording, spontaneous action potentials exhibiting burst-like activity could be resolved. Exposure to PMPA (2 mM) decreased cortical neuron firing, consistent with extracellular recordings. Dotted lines indicate zero potential baseline. (B) Under voltage clamp recording at a holding potential of 70 mV, spontaneous inward currents appeared reduced after exposure to 2 mM PMPA suggesting inhibition of excitation± secretion coupling.

units we observed were those units that ®red autonomously (13 of the 93 recorded units) and did not appear participate in the native bursting activity. Whole-Cell Patch Clamp Recordings Based on the results from extracellular recordings, we focused on the effects of PMPA on cultured cortical neurons measured under current clamp conditions. Under control conditions, from a baseline membrane

potential of 56  2 mV (n ˆ 8), neurons typically showed spontaneous 2±5 mV depolarizations with spontaneous APs observed at a rate of 1:6  0:5 s 1 (n ˆ 5). Administration of PMPA (2 mM) inhibited spontaneous AP ®ring to 36  6% (n ˆ 5; P < 0:001) of control levels, with no signi®cant change in the baseline membrane potential (Fig. 4A). The PMPAinduced depression in AP ®ring was reversed upon washing (data not shown). Under voltage clamp conditions neurons were maintained at a holding

Fig. 5. Whole-cell patch clamp recordings of evoked potentials from embryonic cortical neurons. Current injections as shown elicited hyperpolarizing and depolarizing potentials. Only slight effects of 2 mM PMPA were evident on membrane voltage (Vm). The first derivative of the membrane voltage, dVm/dt, calculated as the central difference from the digitized records, showed a reduction suggesting modest inhibition of inward currents involved in membrane excitability.

398

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

potential of 70 mV, spontaneous inward currents were observed which ranged from 10 to 500 pA. While detailed interpretation must await further experiments, it was clear that application of 2 mM PMPA decreased the frequency of spontaneous inward currents (n ˆ 2; Fig. 4B). We also examined evoked APs measured under current clamp mode. Considering only the ®rst AP of a burst, we observed small, yet statistically signi®cant reductions in the peak-to-peak AP amplitude, APP±P, and the maximum rate of AP rise, dVm/dtmax. Under control conditions, mean APP±P and mean dVm/ dtmax were 84  7 mV (n ˆ 5) and 124  30 V/s, respectively. With application of PMPA (2 mM), APP±P and dVm/dtmax decreased to 93  3% (n ˆ 5; P < 0:05) and 87  4% (n ˆ 5; P < 0:05) of control levels. As illustrated in Fig. 5, evoked potentials from quiescent cells revealed no change in input resistance, RIN, with application of PMPA from a control level of 303  22 MO (n ˆ 3), but reductions in APP±P and dVm/dtmax consistent with values from spontaneously ®ring cells. DISCUSSION We have used a combination of extracellular recording and the whole-cell patch clamp techniques to assess neuronal function at the network and cellular level, respectively. This report is the ®rst to evaluate the neurophysiologic effects of chemical agent hydrolysis products to soman and sarin. Our work demonstrates that PMPA, unlike IMPA and MPA, affects neuronal function, albeit at relatively high concentrations compared to the parent compounds, which have been shown to exert neurophysiologic effects at 0.01± 10 mM (Kumamoto and Shinnick-Gallagher, 1990; Heppner and Fiekers, 1991; Rocha et al., 1998). Neuronal network bursting was inhibited by 50% in the presence of 2 mM PMPA, whereas a reduction in spike amplitude was observed at higher concentrations. The decrease in spike amplitude with PMPA did not appear to be an artifact, since the other compounds did not affect spike amplitude and a similar depression in AP amplitude was observed during patch clamp recording. Such reductions in spike amplitude observed during extracellular recording could be attributed to a number of factors, including membrane depolarization. However, whole-cell patch clamp studies revealed no change in either the resting membrane potential or RIN with administration of PMPA, but a depression in AP ®ring consistent with the extracellular recording

results. Likewise, the small reduction in dVm/dtmax detected with patch clamp measurements was consistent with the reduction in spike amplitude detected with extracellular recording, and may suggest depression of inward currents (calcium and/or sodium) necessary for spike generation. Two observations suggest that PMPA may primarily interfere with one or more aspects of synaptic transmission. First, non-bursting, tonicallyactive spikes measured under extracellular recording were unaffected by PMPA at concentrations where channels showing coordinated bursts were inhibited. Regular synchronized bursts such as those observed in the present study have been suggested to be mediated by synapses (Maeda et al., 1998), in particular, those involving glutamatergic transmission (Canepari et al., 1997; Bacci et al., 1999). Second, during patch clamp recording, spontaneous membrane currents likely mediated by glutamatergic transmission appeared to be inhibited by PMPA. Further experiments will be necessary to clarify the nature of the spontaneous currents and quantify the PMPA-mediated effects. Previous work indicates that both MPA (Williams et al., 1987) and IMPA (Mecler and Dacre, 1982) possess very low oral toxicity in rats where lethal doses for each exceed 5000 mg/kg. In fact, rats that have been fed 350 mg/kg IMPA each day exhibited no toxicity after 90 days (Mecler, 1981). In contrast to IMPA and MPA, there are no previous data on the biologic activity of PMPA, with the exception of the present study. A simplistic estimate of functional neurotoxicity can be made with the following assumptions. Assuming that all consumed PMPA was available to the bloodstream (15 ml) of an average rat (250 g) and no clearance, we would estimate reaching a blood concentration of 1 mM PMPA after consumption of 10.8 mg/kg. This estimate must be considered with caution since it ignores potentially important pharmacokinetic considerations. For example, Reynolds et al. (1985) showed that PMPA concentrations in plasma, produced by soman hydrolysis after i.p. injection in mice, showed little decline over a period of 15 min to 8 h. However, the ratio of free PMPA to bound PMPA in plasma was 1:50 (Reynolds et al., 1985), suggesting that the majority of PMPA would be unavailable for neuronal substrates. Therefore, it is unlikely that PMPA would exert neurophysiologic effects under conditions of soman exposure, although we cannot exclude the possibility that during the handling of the degradation product PMPA, some neurophysiologic effects may result. Electrophysiologic methods offer the ability to evaluate putative neurotoxicants with metrics based on physiologic function (Pancrazio et al., 1999). Single

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

microelectrode techniques such as patch clamp yield detailed information concerning molecular mechanisms of action, but suffer with regard to throughput. The use of MEAs presents the capability of screening compounds for neurotoxicity in a comparatively rapid manner. In fact, initial drug evaluation with MEAs can allow immediate identi®cation of concentration ranges that warrant more detailed examination with molecular techniques including patch clamp. While certainly not high throughput in manner, this combined approach allows for a more rapid evaluation of neurophysiologic effects of potential toxicants. ACKNOWLEDGEMENTS The authors thank Ms. Kara M. Shaffer (Naval Research Laboratory) for excellent cell culture support. This work was funded by the Defense Advanced Research Projects Agency, the Defense Threat Reduction Agency, and the Of®ce of Naval Research. The opinions and assertions contained herein are strictly those of the authors and are not to be construed as of®cial or re¯ecting the views of the Department of the Navy or the military at large. REFERENCES Bacci A, Verderio C, Pravettoni E, Matteoli M. Synaptic and intrinsic mechanisms shape synchronous oscillations in hippocampal neurons in culture. Eur J Neurosci 1999;11:389±97. Brown MA, Brix KA. Review of health consequences from high-, intermediate- and low-level exposure to organophosphorus nerve agents. J Appl Toxicol 1998;18:393±408. Camara AL, Braga MF, Rocha ES, Santos MD, Cortes WS, Cintra WM, Aracava Y, Maelicke A, Albuquerque EX. Methamidophos: an anticholinesterase without significant effects on postsynaptic receptors or transmitter release. NeuroToxicology 1997;18:589±602. Canepari M, Bove M, Maeda E, Cappello M, Kawana A. Experimental analysis of neuronal dynamics in cultured cortical networks and transitions between different patterns of activity. Biol Cybern 1997;77:153±62. Chebabo SR, Santos MD, Albuquerque EX. The organophosphate sarin, at low concentrations, inhibits the evoked release of GABA in rat hippocampal slices. NeuroToxicology 1999;20:871±82. Fiedler N, Feldman RG, Jacobson J, Rahill A, Wetherell A. The assessment of neurobehavioral toxicity: SGOMSEC joint report. Environ Health Perspect 1996;104:179±91. Gross GW. Simultaneous single unit recording in vitro with a photoetched laser deinsulated gold multi-microelectrode surface. IEEE Trans Biomed Eng 1979;26:273±9. Gross GW. Internal dynamics of randomized mammalian neuronal networks in culture. In: Stenger DA, McKenna TM, editors.

399

Enabling technologies for cultured neural networks. New York: Academic Press, 1994. p. 277±317. Gross GW, Schwalm FU. A closed chamber for long-term electrophysiological and microscopical monitoring of monolayer neuronal networks. J Neurosci Meth 1994;52:73±85. Gross GW, Wen W, Lin J. Transparent indium-tin oxide patterns for extracellular, multisite recording in neuronal cultures. J Neurosci Meth 1985;15:243±52. Gross GW, Rhoades BK, Jordan RS. Neuronal networks for biochemical sensing. Sens Actuators B 1992;6:1±8. Gross GW, Azzazy HME, Wu M-C, Rhoades BK. The use of neuronal networks on multielectrode arrays as biosensors. Biosens Bioelectron 1995;10:553±67. Gross GW, Harsch A, Rhoades BK, GoÈpel W. Odor, drug and toxin analysis with neuronal networks in vitro: extracellular array recording of network responses. Biosens Bioelectron 1997;12:373±93. Gross GW, Norton S, Gopal K, Schiffmann D, Gramowski A. Nerve cell network in vitro: applications to neurotoxicology, drug development, and biosensors. Cell Eng 1997;2:138±47. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch clamp techniques for high-resolution current recordings from cells and cell-free membrane patches. PfluÈgers Arch 1981;391:85±100. Harata N, Wu J, Ishibashi H, Ono K, Akaike N. Run-down of the GABAA response under experimental ischaemia in acutely dissociated CA1 pyramidal neurons of the rat. J Physiol (London) 1997;500:673±88. Heppner TJ, Fiekers JF. Soman reversibly decreases the duration of Ca2‡ and Ba2‡ action potentials in bullfrog sympathetic neurons. Brain Res 1991;563:303±5. Kao WY, Liu Q-Y, Ma W, Ritchie GD, Lin J, Nordholm AF, Rossi III, J, Barker JL, Stenger DA, Pancrazio JJ. Inhibition of spontaneous GABAergic transmission by trimethylolpropane phosphate. NeuroToxicology 1999;20:843±50. Kumamoto E, Shinnick-Gallagher P. Action of an irreversible acetylcholine esterase inhibitor, soman, on muscarinic hyperpolarization in cat bladder parasympathetic ganglia. Br J Pharmacol 1990;99:157±63. Lin X, Chen S, Tee DJ. Effects of quinine on the excitability and voltage-dependent currents of isolated spiral ganglion neurons in culture. Neurophysiology 1998;79:2503±12. Lucas JH, Czisny LE, Gross GW. Adhesion of cultured mammalian CNS neurons to flame-modified hydrophobic surfaces. In Vitro 1986;22:37±43. Maeda E, Kuroda Y, Kawana A. Modification of parallel activity elicited by propagating bursts in developing networks of rat cortical neurones. Eur J Neurosci 1998;10:488±96. Mecler FJ. Mammalian toxicological evaluation of DIMP and DCPD (Phase-3 IMPA), AD-A107574. Fort Detrick (MD): US Army Medical Research and Development Command, 1981. Mecler FJ, Dacre JC. Toxicological assessment of isopropyl methylphosphonic acid. Toxicologist 1982;2:35. Munro NB, Ambrose KR, Watson AP. Toxicity of the organophosphate chemical warfare agents GA, GB, and VX Ð implications for public protection. Environ Health Perspect 1994;102:18±38. Munro NB, Talmage SS, Griffin GD, Waters LC, Watson AP, King JF, Hauschild V. The sources, fate, and toxicity of chemical warfare agent degradation products. Environ Health Perspect 1999;107:874±933.

400

J.J. Pancrazio et al. / NeuroToxicology 22 (2001) 393±400

Pancrazio JJ, Whelan JP, Borkholder DA, Ma W, Stenger DA. Development and application of cell-based biosensors. Ann Biomed Eng 1999;27:1±15. Ransom BR, Neale E, Henkart M, Bullock PN, Nelson PG. Mouse spinal-cord in cell-culture. 1. Morphology and intrinsic neuronal electrophysiologic properties. J Neurophysiol 1977;40:1132±50. Reynolds ML, Little PJ, Thomas BF, Bagley RB, Martin BR. Relationship between the biodisposition of [3 H]soman and its pharmacological effects in mice. Toxicol Appl Pharmacol 1985;80:409±20. Rocha ES, Chebabo SR, Santos MD, Aracava Y, Albuquerque EX. An analysis of low level doses of cholinesterase inhibitors in

cultured neurons and hippocampal slices of rats. Drug Chem Toxicol 1998;21:191±200. Scott RH, Woods AJ, Lacey MJ, Fernando D, Crawford JH, Andrews PL. An electrophysiological investigation of the effects of cisplatin and the protective actions of dexamethasone on cultured dorsal root ganglion neurones from neonatal rats. Naunyn Schmiedebergs Arch Pharmacol 1995;352:247±55. Williams RT, Miller WR III, MacGillivray AR. Environmental fate and effects of tributyl phosphate and methyl phosphonic acid, CRDEC-CR-87103: NTIS AD-A184 959/5. Aberdeen Proving Ground (MD): US Army Armament Munitions Chemical Command, Chemical Research, Development and Engineering Center, 1987.