(R,S)-Trihexyphenidyl, Acting Via a Muscarinic Receptor-Independent Mechanism, Inhibits Hippocampal Glutamatergic and GABAergic Synaptic Transmissions: Potential Relevance for Treatment of Organophosphorus Intoxication

Abstract

Preclinical studies have reported that, compared to the muscarinic receptor (mAChR) antagonist atropine, (R,S)-trihexyphenidyl (THP) more effectively counters the cholinergic crisis, seizures, and neuropathology triggered by organophosphorus (OP)-induced acetylcholinesterase (AChE) inhibition. The greater effectiveness of THP was attributed to its ability to block mAChRs and N-methyl-D-aspartate-type glutamatergic receptors (NMDARs) in the brain. However, THP also inhibits α7 nicotinic receptors (nAChRs). The present study examined whether THP-induced inhibition of mAChRs, α7 nAChRs, and NMDARs is required to suppress glutamatergic synaptic transmission, whose overstimulation sustains OP-induced seizures. In primary hippocampal cultures, THP (1–30 µM) suppressed the frequency of excitatory and inhibitory postsynaptic currents (EPSCs and IPSCs, respectively) recorded from neurons in nominally Mg²⁺-free solution. A single sigmoidal function adequately fit the overlapping concentration-response relationships for THP-induced suppression of IPSC and EPSC frequencies yielding an IC50 of 6.3 ± 1.3 µM. Atropine (1 µM), the NMDAR antagonist D,L-2-amino-5-phosphonopentanoic acid (D,L-AP5, 50 µM), and the α7 nAChR antagonist methyllycaconitine (MLA, 10 nM) did not prevent THP-induced inhibition of synaptic transmission. THP (10 µM) did not affect the probability of transmitter release because it had no effect on the frequency of miniature IPSCs and EPSCs recorded in the presence of tetrodotoxin. Additionally, THP had no effect on the amplitudes and decay-time constants of miniature IPSCs and EPSCs; therefore, it did not affect the activity of postsynaptic GABA_A and glutamate receptors. This study provides the first demonstration that THP can suppress action potential-dependent synaptic transmission via a mechanism independent of NMDAR, mAChR, and α7 nAChR inhibition.

1. Introduction

The morbidity and mortality resulting from deployment of organophosphorus (OP) nerve agents against civilians as recently as 2017–2018 in Syria, Malaysia, and England and during earlier terrorist attacks and wars are well documented (Coupland and Kobi-Renée, 2005; Hulse et al., 2019; Naughton and Terry, 2018; Romano and King, 2001). Equally well-documented are the poor health outcomes resulting from occupational, accidental, and intentional exposures to high levels of OP insecticides, especially in countries where use of these insecticides is not well regulated (Eddleston, 2019; Naughton and Terry, 2018).

Signs and symptoms of acute OP intoxication result primarily, though not exclusively, from the irreversible inhibition of acetylcholinesterase (AChE), the enzyme that catalyzes the hydrolysis of the neurotransmitter acetylcholine (ACh). The ACh build-up leads to a toxidrome characterized by miosis, profuse secretions, diarrhea, and bronchoconstriction due to overactivation of muscarinic receptors (mAChRs) in addition to skeletal muscle fasciculations and paralysis due to overstimulation followed by desensitization of nicotinic receptors (nAChRs). Central nervous system (CNS)-related clinical signs and symptoms of OP poisoning range from restlessness and confusion to tremors and seizures, and cardiorespiratory failure is the primary cause of death (see Pereira et al., 2014 and references therein). Thus, the standard OP antidotal therapy consists of high doses of atropine to block mAChR overactivation, an oxime (generally pralidoxime) to reactivate OP-inhibited, non-aged AChE, and benzodiazepines, as needed, to halt OP-induced seizures (Newmark, 2004).

Although the standard OP antidotal therapy saves lives, it has important limitations. Specifically, atropine and clinically used oximes do not cross the blood barrier well and do not effectively counter the effects of OP compounds in the CNS. In addition, uncontrolled OP-induced seizures quickly progress to status epilepticus and become refractory to pharmacotherapy (McDonough and Shih, 1997; McDonough and Shih, 1993). Therefore, efforts have long been underway for identification of more effective medical countermeasures against OP poisoning (Jett and Laney, 2021; Sakurada and Ohta, 2020).

Starting in the early 1990s, preclinical studies demonstrated that, used as a posttreatment, the mAChR antagonist (R,S)-trihexyphenidyl (THP) is more potent and effective than atropine in blocking the cholinergic crisis and halting the seizures induced by the nerve agent soman (McDonough and Shih, 1993; McDonough et al., 2000). While OP-induced seizures are initially triggered by neuronal hyperexcitability resulting from the cholinergic overdrive, they appear to be strengthened and sustained by the subsequent increased glutamatergic tone in different brain regions (McDonough and Shih, 1997). This premise gave rise to the hypothesis that the ability of THP to cross the blood brain barrier and block not only mAChRs but also N-methyl-D-aspartate receptors (NMDARs) accounts for its greater effectiveness compared to that of atropine as an OP antidote (McDonough et al., 2000). In fact, inhibitors of ionotropic glutamatergic receptors, including NMDARs and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs), help mitigate OP-induced seizures and neuropathology in preclinical models (Aroniadou-Anderjaska et al., 2020; Marrero-Rosado et al., 2020; Voorhees et al., 2017).

In the hippocampus, a brain region highly sensitive to the effects of OP nerve agents and insecticides (Apland et al., 2010; Kadar et al., 1995; Ojo et al., 2014), neuronal excitability and glutamatergic synaptic transmission are tightly regulated by mAChR and nAChR activities. Specifically, neuronal excitability and hippocampal synaptic transmission: (i) increase due to activation of neuronal nAChRs (particularly α7, α4β2, and α3β4 nAChRs) and M1, M3, and M4 mAChRs, and (ii) decrease following activation of M2 and M4 mAChRs (Albuquerque et al., 2009; Moran et al., 2019). Notably, THP has been reported to inhibit not only mAChRs, being relatively more selective towards M1/M3 mAChRs than M2 AChRs (Dörje et al., 1991; Richards, 1990), but also neuronal nAChRs. For instance, Gao et al. (1998) reported that, in mice, THP dose dependently and competitively inhibits nicotine-induced convulsions, which are thought to mediated by α7 nAChRs (Damaj et al., 1999). In addition, Strøm (2006) reported that THP inhibits nicotine (50 µM)-triggered increase in intracellular Ca²⁺ concentrations in neuroblastoma SH-SY5Y cells, which predominantly express α7 nAChRs (Peng et al., 1994). Thus, the goal of the present study was to test the hypothesis that THP suppresses hippocampal glutamatergic synaptic transmission via a multifactorial mechanism that involves inhibition of mAChRs, α7 nAChRs, and NMDARs.

The results presented here provide the first demonstration that, concentration dependently and reversibly, THP inhibits action potential-dependent hippocampal glutamatergic and GABAergic transmissions. However, they also reveal that THP-induced inhibition of synaptic transmission: (i) occurs independently of mAChR, nAChR, and NMDAR inhibition, (ii) is not due to a reduction of the probability of transmitter release, and (iii) may be a result of inhibition of voltage-gated Na⁺ channels by THP. The ability of THP to inhibit action potential-dependent synaptic transmission via a cholinergic- and glutamatergic-independent mechanism can be an important determinant of its effectiveness to counter the clinical signs and symptoms of OP intoxication, including OP-induced seizures and neuropathology.

2. Material and Methods

2.1. Animals

Time-pregnant Sprague-Dawley rats [Crl:CD (SD)] were purchased from Charles River Laboratories (Raleigh, NC). Upon arrival at the central animal facility of the University of Maryland School of Medicine, pregnant rats were on gestation day 16–18 and were individually housed with food and water ad libitum under a 12:12 h light:dark cycle. On gestation day 17–19, pregnant rats were humanely euthanized by CO₂ asphyxiation followed by thoracotomy. Fetuses were removed by C-section and immediately decapitated. Six to eight fetuses from each of 32 pregnant rats were used to prepare primary hippocampal cultures. All procedures were approved by the University of Maryland School of Medicine Animal Care and Use Committee (Animal Use Protocol number 0919001) and were conducted in full accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.

2.2. Primary Hippocampal Cultures

Primary hippocampal cultures were the model of choice for the present study in part because the two-dimensional arrangement of cells in these cultures facilitates the accessibility of drugs to their cellular target(s), thereby allowing for an accurate resolution of the concentration and time dependence of potential effects of drugs on synaptic transmission. This is particularly relevant for lipophilic drugs such as THP, which are likely to slowly diffuse through the multiple cell layers in more complex systems (e.g., hippocampal slices and organotypic cultures). In addition, all experiments were carried out in high-density hippocampal cultures because, like the hippocampus in vivo (Cobb et al., 1997), these cultures have a higher prevalence of inter-neuronal compared to intra-neuronal (i.e., autaptic) synapses (Liu et al., 2013, 2009).

Primary hippocampal cultures were prepared according to the procedure described previously (Pereira et al., 1993). In short, hippocampi of 6–8 fetal rats were dissected out, pooled together (irrespective of sex of the fetuses), minced, and incubated in 0.25% trypsin at 37°C for 30 min. Subsequently, the tissue was mechanically dissociated in minimum essential medium (MEM high glucose, high bicarbonate, catalog number INV-A1451801; Thermo Fisher Scientific, Waltham, MA) supplemented with 10% heat-inactivated (56°C for 30 min) horse serum (catalog number INV-16050122, Thermo Fisher Scientific), 10% fetal bovine serum (catalog number SIG-F0926, Millipore Sigma, Rockville, MD), 2 mM glutamine (catalog number INV-25030081, Thermo Fisher Scientific), and deoxyribonuclease II (20 µg/ml, type V; catalog number SIG-D8764-30KU, Millipore Sigma).

Dissociated cells were plated at a density of approximately 65,000/cm² on 35-mm petri dishes (Nunclon^® Cell Culture Dishes) precoated with collagen (1:5 dilution in sterile water; type I collagen standard, catalog number 5015, Advanced Biomatrix, Carlsbad, CA). After 24 h of plating the cells and twice-a-week thereafter, the culture medium was replaced with MEM containing 10% heat-inactivated horse serum and 2 mM glutamine. On the 8^th day after plating, the antimetabolic mix of 5’-fluoro-2’-deoxyuridine (2 µg/ml; catalog number SIG-F0503, Millipore Sigma) and uridine (13 µg/ml; catalog number SIG-U3750, Millipore Sigma) was added to the medium for 24 h to halt glial cell proliferation. Neurons cultured for 15–30 days were used in the experiments described here.

2.3. Electrophysiological Recordings

Recordings of synaptic currents were obtained from voltage-clamped cultured hippocampal neurons by means of the conventional whole-cell mode of the patch-clamp technique. The physiological solution used to superfuse the cells was composed of (in mM): NaCl, 165; KCl, 5; CaCl₂, 2; 4-(2-Hydroxyethyl)-1-piperazine ethanesulfonic acid (HEPES), 5; and dextrose, 10 (pH = 7.3). The nominally Mg²⁺-free physiological solution was used to enhance the probability of transmitter release and to minimize Mg²⁺-induced inhibition of NMDAR activity in the neuron from which recordings were obtained and in neurons synapsing onto that neuron (Mangan and Kapur, 2004). In addition, as reported previously, the pattern of OP-induced increase in synaptic transmission in hippocampal slices resembles that observed in hippocampal preparations exposed to low Mg²⁺ concentrations (Kozhemyakin et al., 2010).

Patch pipettes were pulled from borosilicate glass capillaries (i.d. 1.5 mm; World Precision Instruments Inc., Sarasota, FL) with a horizontal puller (Sutter Instrument Company, Novato, CA). Pipettes were filed with a cesium methanosulfonate-based solution containing (in mM): Cs methanesulfonate, 130; CsCl, 10; MgCl₂, 2; 3,12-bis(carboxymethyl)-6,9-dioxa-3,12-diazatetradecanedioic acid (EGTA), 11; CsOH, 22; HEPES, 10; and lidocaine N-ethyl bromide (QX-314 bromide, catalog number SIG-552233, Millipore Sigma), 5 (pH adjusted to 7.3). When filled with the internal solution, pipettes had resistances ranging from 3.5 to 4.5 MΩ. All recordings were performed at room temperature (20–22°C).

Spontaneous excitatory postsynaptic currents (EPSCs) were recorded from neurons voltage clamped at −60 mV, near the reversal potential for GABAergic currents (−64 mV, under the present conditions). Spontaneous inhibitory postsynaptic currents (IPSCs) were recorded at the reversal potential for glutamatergic currents (0 mV). Miniature EPSCs and IPSCs (mEPSCs and mIPSCs, respectively) were recorded at −60 mV and 0 mV, respectively, from neurons continuously superfused with the nominally Mg²⁺-free physiological solution containing the Na⁺-channel blocker tetrodotoxin (TTX, 300 nM). Electrical signals were filtered at 3 kHz using an LM-EPC7 amplifier (List Electronic, Darmstadt, Germany), digitized at 10 kHz using a Digidata 1332A (Molecular Devices Corp., Union City CA), and stored in a PC using the pClamp 10 software (Molecular Devices Corp., Union City CA). Recordings were discarded if the seal resistance changed by more than 15%.

Synaptic currents were recorded under different experimental conditions before, during, and after 20-min exposure of the neurons to THP concentrations ranging from 1 to 30 µM. These concentrations were selected because they included THP concentrations reported to block mAChRs, nAChRs, and NMDARs (Giachetti et al., 1986; Olney et al., 1987; Richards, 1990; Sills and Loo, 1989; Strøm, 2006). After neurons in a culture dish were superfused once with THP-containing physiological solution, the dish was replaced to avoid carryover effects. No blinding and no formal randomization were used.

2.4. Analysis of Electrophysiological Data

Frequency and amplitude of synaptic events were measured using the Mini Analysis 6.0.3 software (Synaptosoft Inc., Decatur, GA). The threshold amplitude for automated detection of spontaneous and miniature EPSCs and IPSCs was set at twice the baseline noise. Events that did not show a typical synaptic waveform were rejected manually.

The Clampfit module of the pCLAMP 10 software (Molecular Devices; San Jose, CA) was used to analyze the decay-time constants (τ_decay) of individual mEPSCs and mIPSCs that had a single peak, a sharp rising phase, and an exponential decay. The τ_decays of individual mIPSCs recorded from all cells were normally distributed. Thus, to estimate the mean τ_decay of mIPSCs, individual mIPSCs recorded from each cell before and during its perfusion with THP-containing physiological solution were aligned at half rise time and averaged. Then, Clampfit was used to estimate the τ_decays of the averaged mIPSCs recorded before and during exposure of each cell to THP.

The distribution of τ_decays of individual mEPSCs recorded from all cells was skewed to the right, with approximately 90% of the τ_decays being centered around 7.1 ms with a standard deviation (SD) of 3.1 ms and the remaining 10% being centered around 22.1 ms with an SD of 5.0 ms. Based on the mean and standard deviations of each distribution, 13 ms was taken as the threshold separating fast- from slow-decaying mEPSCs. To estimate the mean τ_decays of fast- and slow-decaying mEPSCs recorded from each cell before and during its perfusion with THP-containing physiological solution, individual mEPSCs that had τ_decays ≤13 ms and >13 ms, respectively, were aligned at half rise time and averaged. Subsequently, Clampfit was used to estimate the τ_decays of the averaged fast- and slow-decaying mEPSCs recorded before and during exposure of each cell to THP.

2.5. Statistical Data Analysis

Required sample sizes were estimated based upon our past experience performing similar experiments. Statistical analysis was conducted using SigmaPlot, version 12.0 (Systat Software Inc, San Jose CA) and GraphPad Prism, version 7.04 (GraphPad Software). The data were analyzed for normality (Shapiro Wilk) and heteroscedasticity (Levene’s test). Normally distributed data with equal variance were analyzed for statistical significance using one-way analysis of variance (ANOVA), repeated measures ANOVA (or mixed-effects ANOVA, if data points were missing), or two-tailed paired t-tests, as appropriate. If ANOVA was significant, it was followed by Tukey post-hoc test for multigroup comparisons. Data are presented as mean ± SD.

Cumulative distributions of amplitudes and inter-event intervals were plotted for mIPSCs and mEPSCs recorded during baseline (control) or in the presence of THP. Amplitudes and inter-event intervals recorded under each condition from each cell were randomized and 75 events/cell/condition were pooled to generate the cumulative histograms. The cumulative distributions of inter-event intervals and amplitudes were compared statistically using the Kolmogorov-Smirnov (K-S) test.

IC50 values were estimated using the GraphPad Prism built-in equation for analysis of inhibitor concentration vs. normalized response relationship. In all experiments, the null hypothesis was rejected if p values were < 0.05.

2.6. Chemicals

Atropine sulfate (catalog number SIG-A0257), D,L-2-amino-5-phosphonopentanoic acid (D,L-AP5, catalog number SIG-A5282), tetrodotoxin (TTX, catalog number SIG-T8024), and (R,S) trihexyphenidyl HCl (THP, catalog number SIG-T1516), (−)bicuculline methiodide (catalog number SIG-6889), and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) disodium salt hydrate (catalog number SIG-C239) were purchased from Millipore Sigma (Rockville, MD). Methyllycaconitine HCl (MLA) was a gift from Professor M. H. Benn (University of Calgary, Calgary, Alberta, Canada). Stock solutions of chemicals were made in distilled water, kept frozen, and diluted, on the day of the experiments, to the desired concentrations in the extracellular physiological solution used to superfuse the neurons.

3. Results

3.1. THP Inhibits Action Potential-Dependent GABAergic Synaptic Transmission

To assess the effects of THP on action potential-dependent GABAergic transmission, spontaneous IPSCs were recorded from neurons voltage clamped at 0 mV before, during, and after their 20-min exposure to THP in nominally Mg²⁺-free physiological solution. Recordings began after neurons had been superfused for at least 30 min with the drug-free nominally Mg²⁺-free physiological solution. As shown in Figure 1A, spontaneous synaptic currents recorded at 0 mV appeared as upward events. The finding that these synaptic events were not detected in the presence of the GABA_A receptor antagonist bicuculline (10 µM) confirmed their GABAergic nature (Figure 1B).

Figure 1. THP suppresses the frequency of IPSCs recorded from cultured hippocampal neurons.

A. Sample recordings illustrate synaptic events recorded at 0 mV from a neuron before, during, and after its exposure to THP (10 µM). Traces in the top, middle, and bottom of panel A are 4-min sweeps recorded immediately before, between 16 and 20 min during, and between 8 and 12 min after ending the THP exposure. Insets are short segments of the 4-min sweeps to illustrate individual IPSCs at an expanded time scale. Calibration bars of the compressed and expanded traces in the bottom panel apply to the traces in the top and middle panels. B. Synaptic events recorded at 0 mV are GABAergic because they are not detected in the presence of the GABA_A receptor antagonist bicuculline (10 µM). C. THP (10 µM) reduces the frequency of IPSCs, with the effect peaking around 12 min after onset of the THP exposure and being reversible following washout. For each neuron, the IPSC frequency recorded for 8 min before exposure to THP was taken as 100% (baseline) and used to normalize the IPSC frequency in 4-min bins throughout the entire recording time. Data points represent individual neurons, horizontal bars at midpoint represent means, and error bars represent SD. D, E. THP concentration dependently suppresses the frequency of IPSCs (D) and has no effect on their amplitude (E). In D and E, for each neuron the frequency and amplitude of IPSCs recorded between 16 and 20 min of the THP exposure were normalized to baseline. In absolute values, the frequency and amplitude of IPSCs recorded during baseline from all neurons were 9.79 ± 6.81 Hz and 61.30 ± 34.52 pA (mean ± SD, n = 14), respectively. Each data point represents normalized data from one neuron. Graph and error bars represent mean and SD, respectively, of the normalized values. Based on the post-hoc Tukey test: **, p < 0.01; *, p < 0.05.

Exposure of cultured hippocampal neurons to 10 µM THP reduced the frequency of spontaneous IPSCs (Figure 1C). Repeated measures ANOVA revealed that the effect of THP on the frequency of IPSCs was time dependent [F(3,9) = 9.851, p < 0.001]. Post-hoc analysis indicated that the THP-induced suppression of IPSC frequency reached maximum at approximately 12 min after the onset of the exposure of neurons to THP. As shown in Figure 1C, the IPSC frequency between 12 and 20 min after the beginning of the superfusion of the neurons with 10 µM THP (i.e., between 16- and 24-min recording times) was significantly lower than that recorded before (control) and during the first 8 min of the THP exposure. In addition, the post-hoc analysis indicated that effect of THP was fully reversible. As seen in Figure 1C, the IPSC frequency recorded 8–12 min after beginning of the washout phase was significantly higher than that recorded in the presence of THP and not significantly different from control.

The finding that the IPSC frequency recorded 8–12 min after the beginning of the washout phase was not significantly different from that recorded pre-THP exposure further revealed that the significantly reduced IPSC frequency detected in the presence of THP could not be accounted for by spontaneous rundown of the events.

In absolute values, the frequency and amplitude of IPSCs recorded during baseline were 9.79 ± 6.81 Hz and 61.30 ± 34.52 pA (mean ± SD, n = 14), respectively. The IPSC frequency recorded from neurons between 16 and 20 min after onset of their superfusion with 1, 3, 10, and 30 µM THP was approximately 84%, 66%, 54%, and 10% of the baseline frequency, respectively (Figure 1D). One-way ANOVA revealed that the magnitude of the THP-induced suppression of the IPSC frequency was concentration dependent [F (3,10) = 22.417, p < 0.001]. Post-hoc analysis of the results further indicated that the effect increased as the concentration of THP increased from 1 µM to 30 µM (Figure 1D). By contrast, none of the test THP concentrations affected the amplitude of IPSCs [F (3,10) = 0.617, p = 0.620] (Figure 1E).

3.2. THP Inhibits Action Potential-Dependent Glutamatergic Synaptic Transmission

To determine whether THP inhibits action potential-dependent glutamatergic transmission, spontaneous EPSCs were recorded at −60 mV from neurons before, during, and after their 20-min exposure to THP. The holding potential was set close to the reversal potential for Cl⁻(−64 mV under the present experimental conditions) to allow the selective recording of EPSCs. Recordings began after neurons had been superfused for at least 30 min with the drug-free nominally Mg²⁺-free physiological solution. Synaptic currents recorded at −60 mV appeared as inward events (Figure 2A). The finding that these synaptic events were completely blocked by superfusion of the neurons with an admixture of the AMPAR antagonist CNQX (10 µM) plus the NMDAR antagonist D,L-AP5 (50 µM) confirmed their glutamatergic nature (Figure 2B).

Figure 2. THP Suppresses the frequency of EPSCs recorded from cultured hippocampal neurons.

A. Sample recordings illustrate synaptic events recorded from a neuron at −60 mV before, during, and after its exposure to THP (10 µM). Traces in the top, middle, and bottom of panel A are 4-min sweeps recorded immediately before, between 16 and 20 min during, and between 8 and 12 min after ending the exposure. Insets are short segments of the 4-min sweeps to illustrate individual EPSCs at an expanded time scale. Calibration bars of the compressed and expanded traces in the bottom panel apply to the traces in the top and middle panels. B. Synaptic events recorded at −60 mV are glutamatergic in nature because they are not detected in the presence of the glutamatergic receptor antagonists CNQX (10 µM)-plus-D,L-AP5 (50 µM). C. THP (10 µM) significantly and reversibly reduces the frequency of EPSCs. For each neuron, the EPSC frequency recorded for 8 min pre-THP exposure was taken as 100% (baseline) and used to normalize the EPSC frequency recorded 4 min before the THP exposure (control), 16–20 min during the THP exposure, and 8–12 min after onset of washout. D, E. THP suppresses the frequency of EPSCs in a concentration-dependent manner (D) and has no effect on their amplitude (E). For each neuron, the frequency and amplitude of EPSCs recorded at 16–20 min of the THP exposure were normalized to baseline. In absolute values, the frequency and amplitude of EPSCs recorded during baseline from all neurons were 18.45 ± 11.40 Hz and 62.53 ± 32.86 pA (mean ± SD, n = 13), respectively. Data points represent normalized data from individual neurons. Graph and error bars represent mean and SD, respectively, of the normalized values. Based on the post-hoc Tukey test: **, p < 0.01; *, p < 0.05.

Exposure of neurons to 10 µM THP reduced the frequency of EPSCs (Figure 2C). Repeated measures ANOVA revealed that the EPSC frequency recorded before, during, and after ending the superfusion of the neurons with THP differed significantly [F(3,2) = 20.690 p = 0.002]. Post-hoc analysis of the data indicated that THP significantly reduced the EPSC frequency (Figure 2C). It also showed that the effect of THP was fully reversible because the EPSC frequency recorded 8–12 min after beginning of washing the neurons with THP-free solution was: (i) significantly higher than that recorded in the presence of THP, and (ii) not significantly different from control (Figure 2C). The finding that the EPSC frequencies recorded during the washing phase and before exposure of the neurons to THP were comparable further suggested that spontaneous rundown of the EPSCs during the recording period could not account for the reduced EPSC frequency observed in the presence of THP.

In absolute values, the frequency and amplitude of EPSCs recorded during baseline from all neurons were 18.45 ± 11.40 Hz and 62.53 ± 32.86 pA, respectively (mean ± SD, n = 13). Between 16 and 20 min after onset of superfusion of the neurons with 1, 3, 10, and 30 µM THP, the EPSC frequency was approximately 74%, 60%, 41%, and 11% of the baseline frequency, respectively (Figure 1D). One-way ANOVA revealed that the magnitude of the THP-induced reduction of the frequency of EPSCs was concentration dependent [F (3,9) = 6.617, p = 0.012], and post-hoc analysis of the data indicated that the effect increased as the concentration of THP increased from 1 µM to 30 µM (Figure 2D). By contrast, THP had no significant effect on the amplitude of EPSCs [F(3,9) = 0.200; p = 0.894] (Figure 2E).

3.3. THP Equipotently Inhibits Action Potential-Dependent EPSCs and IPSCs

The concentration-response relationships for THP-induced reduction of the frequency of IPSCs and EPSCs overlapped. Statistical analysis did not reject the null hypothesis that a single sigmoid function with a slope (Hill coefficient) of −1 can fit the two data sets [F(1,25) = 1.874; p = 0.183]. The two concentration-response relationships were adequately fit by a sigmoidal function that yielded an IC50 of 6.3 ± 1.0 µM (r² = 0.70).

3.4. The Suppressive Effect of THP on Action Potential-Dependent GABAergic and Glutamatergic Transmissions Does Not Require NMDAR, mAChR, or α7 nAChR Activity

To determine whether NMDARs mediate the inhibitory effect of THP on action potential-dependent GABAergic transmission, IPSCs were recorded at 0 mV from neurons before, during, and after their 20-min exposure to THP (10 µM) in nominally Mg²⁺-free physiological solution containing the NMDAR antagonist D,L-AP5 (50 µM). As in the previous experiments, recordings began after neurons had been superfused for at least 30 min with the drug-free nominally Mg²⁺-free physiological solution. The mean frequency of IPSCs recorded during baseline in the presence of D,L-AP5 was 4.06 ± 1.92 Hz (n = 5).

A mixed-effects ANOVA revealed that the frequency of IPSCs recorded before, during, and after exposure to THP differed significantly [F(1.217,10.95) = 29.08, p < 0.001]. Data were statistically analyzed using mixed-effects ANOVA instead of a repeated measures ANOVA because washout data for 1 of the 5 neurons were missing. Post-hoc analysis of the data indicated that, despite the continuous NMDAR inhibition with D,L-AP5, THP (10 µM) significantly and reversibly suppressed the frequency of IPSCs (Figure 3A) as it did in the absence of D,L-AP5 (see Figure 1).

Figure 3. NMDARs, mAChRs, and nAChRs are not required for THP to Inhibit GABAergic transmission.

IPSCs were recorded from neurons continuously superfused with physiological solution containing the NMDAR antagonist D,L-AP5 (50 µM) (A), the mAChR antagonist atropine (1 µM) (B), or the α7 nAChR antagonist MLA (10 nM) (C). For each neuron, the frequency of IPSCs recorded during 8 min before the exposure to THP (baseline) was taken as 100% and used to normalize the frequency of events recorded 4 min immediately before (control), 16–20 min during, and 8–12 min after ending the THP exposure. The mean IPSC frequencies recorded during baseline in the presence of D,L-AP5, atropine, and MLA expressed as mean ± SD were: 4.06 ± 1.92 Hz (n = 5), 11.21 ± 4.74 Hz (n = 8), and 10.78 ± 3.82 Hz (n = 9), respectively. In the presence of D,L-AP5 (A), atropine (B), or MLA (C), THP (10 µM) significantly and reversibly suppresses the frequency of IPSCs. Each data point represents one neuron, graph bars represent means, and error bars represent SD. Results of the mixed-effects ANOVA used to statistically analyze the data are described in the text. Asterisks represent the significance of differences according to multi-group comparisons by the post-hoc Tukey test: ***, p < 0.001; **, p < 0.01; *, p < 0.05.

The same experimental approach was used to determine whether mAChRs or α7 nAChRs underlie the inhibitory effect of THP on action potential-dependent GABAergic transmission. In short, IPSCs were recorded at 0 mV from neurons before, during, and after their exposure to THP (10 µM) in nominally Mg²⁺-free physiological solution containing the mAChR antagonist atropine (1 µM) or the α7 nAChR antagonist MLA (1 nM). The mean frequencies of IPSCs recorded during baseline in the presence of atropine and MLA expressed as mean ± SD were 11.21 ± 4.74 Hz (n = 8) and 10.78 ± 3.82 Hz (n = 9), respectively. Mixed-effects ANOVA revealed that, despite the presence of atropine or MLA, the frequency of IPSCs recorded before, during, and after exposure of the neurons to THP differed significantly [atropine: F(1.217,10.95) = 29.08, p < 0.001; MLA: F(1.401, 16.13) = 19.79, p < 0.001]. In addition, post-hoc analysis of the data indicated that, despite the continuous inhibition of mAChRs with atropine or α7 nAChRs with MLA, THP (10 µM) significantly and reversibly suppressed the frequency of IPSCs (Figure 3B, 3C) as it did in the absence of the receptor antagonists (see Figure 1).

To determine whether THP-induced inhibition of action potential-dependent glutamatergic transmission requires NMDARs, mAChRs, or α7 nAChRs, EPSCs were recorded at −60 mV from neurons before, during, and after their 20-min exposure to THP in nominally Mg²⁺-free physiological solution containing an admixture of D,L-AP5 (50 µM), atropine (1 µM), and MLA (1 nM). The frequency of EPSCs recorded in the presence of the three antagonists before exposure of the neurons to THP was 15.1 ± 6.41 Hz (mean ± SD, n = 5). Repeated measures ANOVA revealed that the frequency of IPSCs recorded before, during, and after exposure of the neurons to THP differed significantly despite the continuous inhibition of NMDARs, mAChRs, and α7 nAChRs [F(1.81,6.36) = 15.17, p = 0.0042]. In addition, post-hoc analysis of the data indicated that, in the presence of D,L-AP5, atropine, and MLA, THP (10 µM) significantly suppressed the frequency of EPSCs (Figure 4) as it did in the absence of the inhibitors (see Figure 2).

Figure 4. NMDARs, mAChRs, and nAChRs are not required for THP to inhibit glutamatergic transmission.

EPSCs were recorded from neurons continuously superfused with physiological solution containing an admixture of D,L-AP5 (50 µM), atropine (1 µM), and MLA (10 nM). For each neuron, the frequency of events recorded during baseline (8 min before exposure to THP) was taken 100% and used to normalize the frequency of events recorded 4 min immediately before (control), 16–20 min during, and 8–12 min after ending the THP exposure. The mean EPSC frequency recorded during baseline was 15.1 ± 6.41 Hz (mean ± SD, n = 5). In the presence of the NMDAR, the mAChR, and the α7 nAChR antagonists, THP significantly suppresses the frequency of EPSCs. Each data point represents one neuron, graph bars represent means, and error bars represent SD. Results of the repeated-measures ANOVA are described in the text. Asterisks represent the significance of differences according to multi-group comparisons by the post-hoc Tukey test: **, p < 0.01.

3.5. THP Does Not Affect Action Potential-Independent GABAergic and Glutamatergic Transmissions

THP-induced suppression of the frequency of IPSCs and EPSCs could have been a result of a presynaptic mechanism of action through which THP suppressed the probability of transmitter release. If THP were capable of reducing transmitter release from the synaptic vesicles, it would reduce the frequency of action potential-independent IPSCs and EPSCs without affecting their amplitude.

To assess the effects of THP on action potential-independent GABAergic transmission, mIPSCs were recorded at 0 mV from neurons before, during, and after their 20-min exposure to THP (10 µM) in nominally Mg²⁺-free physiological solution containing the Na⁺-channel blocker TTX (300 nM). Synaptic currents recorded at 0 mV appeared as upward events (Figure 5A), and their inhibition by bicuculline (10 µM) confirmed their GABAergic nature (data not shown). The mean τ_decay of mIPSCs was 60.9 ± 7.84 ms [range: 52.3–73.0 ms, n = 7] (Table 1).

Figure 5. THP does not affect the frequency or amplitude of mIPSCs recorded from neurons in the continuous presence of TTX.

A. Sample recordings illustrate mIPSCs recorded from a neuron at 0 mV before (control) and between 17 and 20 min during its exposure to THP (10 µM). Insets are short segments of the 3-min sweeps to illustrate individual mIPSCs at an expanded time scale. B, D. THP (10 µM) has no significant effect on the mean frequency (B) or amplitude (D) of mIPSCs. For each neuron, the IPSC frequency and amplitude recorded during baseline (i.e, during 6 min immediately before the THP exposure) were taken as 100% and used to normalize the frequency and amplitude of events recorded for 3 min immediately before (control) and between 17 and 20 min during the THP exposure. The mean frequency and amplitude of mIPSCs recorded during baseline from all neurons were 2.22 ± 1.18 Hz and 14.4 ± 3.75 pA (mean ± SD, n = 7), respectively. Each data point represents one neuron, graph bars represent means, and error bars represent SD. C, E. THP (10 µM) has no significant effect on the cumulative distributions of inter-event intervals (C) or amplitude (E) of mIPSCs. The cumulative histograms were generated with 75 events randomly selected from each neuron before and during its exposure to THP.

Table 1.

Decay-time constant of mIPSCs and mEPSCs recorded before (control) and during superfusion of cultured hippocampal neurons with THP-containing physiological solution.

	τdecay (ms)
	mIPSC	mEPSCs - Fast	mEPSCs - Slow
Control	60.9 ± 7.84 [range: 52.3–73.0]	7.55 ± 1.05 [range: 5.86–8.74]	25.5 ± 8.35 [range: 13.8–35.8]
THP, 10 µM	58.3 ± 6.74 [range: 50.5–70.1]	7.56 ± 1.20 [range: 5.85–9.24]	26.1 ± 8.62 [range: 15.9–38.8]

As shown in Figure 5B, THP (10 µM) had no significant effect on the mean frequency of mIPSCs [two-tailed paired t = 0.083, df = 6, p = 0.94]. Statistical analysis using the KS test also revealed that THP had no effect on the cumulative distribution of inter-event intervals [KS = 0.072, p = 0.14] (Figure 5C). In addition, THP had no effect on the mean amplitude of mIPSCs [two-tailed paired t = 2.16, df = 6, p = 0.075] (Figure 5D), the cumulative distribution of the mIPSC amplitudes [KS = 0.079, p = 0.078] (Figure 5E), or the τ_decay of the mIPSCs [two-tailed paired t = 0.8702, p = 0.4176, df = 6] (Table 1).

To evaluate the effects of THP on action potential-independent glutamatergic transmission, mEPSCs were recorded at −60 mV from neurons before, during, and after their 20-min exposure to THP (10 µM) in nominally Mg²⁺-free physiological solution containing TTX (300 nM). Synaptic currents recorded at −60 mV appeared as inward events (Figure 6A), and their inhibition with an admixture of the glutamate receptor antagonists D,L-AP5 (50 µM)-plus-CNQX (10 µM) confirmed their glutamatergic nature (data not shown). Under the present experimental conditions, fast- and slow-decaying mEPSCs were recorded from hippocampal neurons in culture (Figure 6A, expanded traces). Their mean τ_decays were 7.55 ± 1.05 ms [range: 5.86–8.74 ms, n = 6] and 25.5 ± 8.35 ms [range: 13.8–35.8 ms, n = 6], respectively (Table 1). The fast- and slow-decaying mEPSCs are likely to be mediated by AMPARs and NMDARs, respectively.

Figure 6. THP has no effect on the frequency or amplitude of mEPSCs recorded from neurons in the continuous presence of TTX.

A. Sample recordings illustrate mEPSCs recorded from a neuron at 0 mV before (control) and between 17 and 20 min during its exposure to THP (10 µM). Insets are short segments of the 3-min sweeps to illustrate individual mEPSCs at an expanded time scale. B, D. THP (10 µM) had no significant effect on the mean frequency (B) or amplitude (D) of mEPSCs. For each neuron, the mEPSC frequency and amplitude recorded during baseline (i.e, during 6 min immediately before the THP exposure) were taken as 100% and used to normalize the frequency and amplitude of events recorded for 3 min immediately before (control) and between 17 and 20 min during the THP exposure. The mean frequency and amplitude of mEPSCs recorded during baseline from all neurons were 3.91 ± 1.19 Hz and 9.29 ± 2.77 pA (mean ± SD, n = 6), respectively. Each data point represents one neuron, graph bars represent means, and error bars represent SD. C, E. THP (10 µM) has no significant effect on the cumulative distributions of inter-event intervals (C) or amplitude (E) of mEPSCs. The cumulative histograms were generated with 75 events randomly selected from each neuron before and during its exposure to THP.

As shown in Figure 6B, 10 µM THP had no significant effect on the mean frequency of mEPSCs [two-tailed paired test: t = 0.084, df = 5, p = 0.94]. Statistical analysis using the KS test also revealed that THP had no significant effect on the cumulative distribution of inter-event intervals [KS = 0.082, p = 0.10] (6C). In addition, 10 µM THP did not affect the mean mEPS amplitude [two-tailed paired t = 0.083, df = 5, p = 0.94] or the cumulative distribution of the mEPSC amplitudes [KS = 0.038, p = 0.91] (Figure 6D, 6E). THP also had no effect on the τ_decays of the fast- and slow-decaying mEPSCs [fast τ_decay: two-tailed paired t = 0.0177, p = 0.987, df = 5; slow τ_decay: two-tailed paired t = 0.219, p = 0.835, df = 5] (Table 1).

4. Discussion

The present study is the first to demonstrate that THP suppresses action potential-dependent glutamatergic and GABAergic transmission in cultured hippocampal neurons. Data presented here also reveal that the effect of THP on synaptic transmission does not require inhibition of the activity of mAChRs, α7 nAChRs, or NMDARs and is not due to reduction of the probability of transmitter release. These findings and the potential beneficial effects of the multifaceted actions of THP during OP intoxication are discussed in this section.

4.1. THP Inhibits Action Potential-Dependent Synaptic Transmission Via a Cholinergic- and Glutamatergic-Independent Mechanism of Action

The finding that THP reduced the frequency of EPSCs and IPSCs recorded from cultured hippocampal neurons demonstrated, for the first time, that THP can inhibit action potential-dependent glutamatergic and GABAergic synaptic transmission between hippocampal neurons. It took approximately 12–16 min for the effect to reach its peak after onset of exposure of the neurons to THP and 8–12 min for the effect to be reversed following washing of the neurons with THP-free physiological solution. The relatively slow time-to-peak and slow reversibility of the effect may be accounted for by the time it takes for THP concentrations to reach an equilibrium at its site of action in part because, due to its lipophilic and alkaline nature, THP can be taken up by lysosomes and other acidic organelles (Ishizaki et al., 1998).

A single mechanism of action is likely to underlie the effect of THP on both neurotransmitter systems because the concentration-response relationships for THP to inhibit glutamatergic and GABAergic transmissions overlapped and could be adequately fitted by a single sigmoidal function. The IC50 for THP to reduce the frequency of EPSCs and IPSCs was found to be approximately 6.3 µM.

As alluded to earlier, both glutamatergic and GABAergic transmissions in different brain regions, including the hippocampus, are tightly regulated by the cholinergic tone established by the activity of the diverse types of mAChRs and nAChRs in glutamatergic and GABAergic neurons. In short, synaptic transmission is stimulated by activation of M1, M3, and M5 mAChRs and all nAChR subtypes, including α7 nAChRs, and suppressed by activation of M2 and M4 mAChRs (Albuquerque et al., 2009; Moran et al., 2019). In this context, there is evidence in the literature that THP can inhibit mAChRs and nAChRs, particularly α7 nAChRs. Specifically, binding studies report that THP binds with high, intermediate, and low affinities to M1/M4, M3, and M2/M5 mAChRs, respectively (Dörje et al., 1991). Likewise, functional in-vivo and in-vitro studies indicate that THP acts as a relatively more selective inhibitor of M1/M3 than M2 mAChRs (Giachetti et al., 1986; Joseph and Thomsen, 2017; Richards, 1990). For instance, with IC₅₀s of approximately 9.77 nM, 123.03 nM, and 3.55 nM, THP inhibits responses thought to be mediated by M1 mAChRs in hippocampal slices, M2 mAChRs in the left atrium, and M3 mAChRs in the ileum, respectively (Richards, 1990). An in-vivo study conducted in mice also showed that THP suppresses nicotine-induced convulsions (Gao et al., 1998), which are likely a result of nicotine-induced α7 nAChR activation because they can be blocked by the α7 nAChR antagonist methyllycaconitine (MLA) but not the α4β2 nAChR antagonist dihydro-β-erythroidine (Damaj et al., 1999). In addition, an in-vitro study conducted in SH-SY5Y cells, which predominantly express α7 nAChRs (Peng et al., 1994), revealed that, with an IC50 of 0.75 µM, THP inhibits nicotine (50 µM)-triggered increase in intracellular Ca²⁺ concentrations (Strøm, 2006). In the present study, however, THP-induced suppression of the frequencies of EPSCs and IPSCs was still detected in the presence of saturating concentrations of the nonselective mAChR antagonist atropine and the α7 nAChR antagonist MLA, and, as such, could not be accounted for by THP-induced inhibition of these cholinergic receptors. It is possible that, under the present experimental conditions, baseline activity of mAChRs and α7 nAChRs is too low and does not significantly contribute to the maintenance of the frequency of IPSCs or EPSCs recorded from hippocampal neurons in culture. This is in line with the finding that the baseline frequency of IPSCs recorded in the absence of receptor antagonists (9.79 ± 6.81 Hz) was comparable to that recorded in the presence of atropine (11.21 ± 4.75 Hz) or MLA (10.78 ± 3.82 Hz).

Although THP is known to act as an NMDAR antagonist (Olney et al., 1987; Sills and Loo, 1989) and NMDAR antagonists can decrease neuronal excitability and suppress synaptic transmission both in vivo and in vitro (Mangan and Kapur, 2004), NMDAR inhibition could not explain the THP-induced suppression of synaptic transmission. Specifically, the NMDAR antagonist D,L-AP5 did not prevent the effect of THP on synaptic transmission. This finding may be accounted for by the facts that: (i) in the present study, neurons were continuously superfused with nominally Mg²⁺-free physiological solution, which facilitates glutamate transmission (Mangan and Kapur, 2004), and (ii) glutamate reduces the interactions of THP with the NMDAR (Sills and Loo, 1989). For instance, in the absence of L-glutamate and Mg²⁺, the IC50 for THP to displace binding of the NMDAR ligand [³H]1-[1-(2-thienyl) cyclohexyl] piperidine ([³H]TCP) to rat forebrain membranes is 7.3 µM (Sills and Loo, 1989). However, in the presence of L-glutamate (0.5 µM) and Mg²⁺ (30 µM), the IC50 of THP to displace [³H]TCP binding increases to 12 µM (Sills and Loo, 1989). It is, thus, possible that extracellular concentrations of glutamate in the cultures superfused with nominally Mg²⁺-free physiological solution were too high and prevented inhibition of NMDARs by the test concentrations of THP.

4.2. A Presynaptic Mechanism of Action Does Not Account for THP-Induced Inhibition of Action Potential-Dependent Synaptic Transmission: Possible Involvement of Voltage-Gated Na⁺ Channels and Other Molecular Targets

At 10 µM, a concentration that was sufficient to reduce by more than 50% the frequency of EPSCs and IPSCs, THP had no significant effect on the frequencies of mEPSCs or mIPSCs recorded in the continuous presence of the Na⁺-channel blocker TTX. Therefore, under the present experimental conditions, THP did not affect the probability of transmitter release, and a presynaptic mechanism of action could not explain the reduction of the frequency of action potential-dependent EPSCs and IPSCs. The finding that, at 10 µM, THP had no significant effect on the amplitude and the τ_decays of mEPSCs or mIPSCs further indicated that THP did not affect the activity of postsynaptic AMPARs, NMDARs, and GABA_ARs.

Since TTX occluded the effects of THP on synaptic transmission, the effects could be a result of THP-induced suppression of action potentials. In this context, there are reports that THP has local anesthetic-like properties. Specifically, THP has been shown to decrease the frequency of action potentials induced by either glutamate or ACh in cortical neurons of cats and the amplitude of electrically evoked action potentials in the isolated frog sciatic nerve preparation (Clarke and Davies, 1973). In addition, with an IC50 of approximately 17.5 µM, THP inhibits the sodium conductance in squid giant axons (Wu and Narahashi, 1976).

Direct evidence that THP inhibits the activity of voltage-gated Na⁺ channels has been provided by a study conducted in a HEK293 cell line stably expressing Na_v1.6 channels harboring the gain-of-function R1872Q mutation (Atkin et al., 2018). Na_v1.6 channels are predominantly expressed in the nodes of Ranvier of axons in the central and peripheral nervous systems, and gain-of-function mutations of these channels, including the R1872Q mutation, have been associated with different epilepsy syndromes (Liu et al., 2019). The results of a high-throughput fluorescence-based Na⁺ flux assay applied to the HEK293 cell line revealed that, at 10 µM, THP inhibited by approximately 83% the Na_v1.6 channel activity (Atkin et al., 2018).

The finding that THP inhibits voltage-gated Na⁺ channels can be accounted for by the fact that its chemical structure has the basic features characteristic of Na⁺ channel blockers such as lidocaine and procainamide; these structural features include a lipophilic aromatic ring separated from an amine group by a linker (Ehring et al., 1988). In this context, diphenidol, which is structurally closely related to THP, has also been shown to block voltage-gated Na⁺ channels in neuroblastoma N2A cells with IC50s of 0.77 µM at −70 mV and 62.6 µM at −100 mV (Leung et al., 2010). The hypothesis that THP-induced inhibition of voltage-gated Na⁺ channels accounts for the suppression of action potential-dependent synaptic transmission in hippocampal neurons remains to be tested.

THP has also been shown to inhibit sigma 1 (σ1) receptor ligand binding to brain tissue (Hudkins and DeHaven-Hudkins, 1991). Since σ1 receptor interactions with voltage-gated Na⁺ and K⁺ channels can suppress neuronal excitability (reviewed in Ryskamp et al., 2019), the possibility cannot be ruled out that σ1 receptors contribute to THP-induced suppression of synaptic transmission in the hippocampus.

4.4. Potential Relevance of Suppression of Synaptic Transmission to the Effectiveness of THP as a Medical Countermeasure Against OP Intoxication

In the U.S., THP HCl was approved by the Food and Drug Administration in 2003 as an adjuvant oral treatment for patients presenting with all forms of parkinsonism. THP HCl is also indicated to control extrapyramidal disorders caused by centrally acting drugs such as phenothiazines and is used off-label for treatment of patients presenting with dystonia (Jilani et al., 2023). THP HCl has a half-life of 5–10 h, though some studies have reported half-lives as long as 33 h (Brocks, 1999). The most common adverse effects associated with the therapeutic uses of THP are primarily due to inhibition of mAChRs and include decreased sweating, dry mouth, abdominal discomfort, nausea, urinary retention, and constipation. In the context of OP intoxication, however, the effects that result from the mAChR inhibition are major determinants of the therapeutic effectiveness of THP. The safety and clinical data already available for THP, in addition to the fact that THP can cross the blood brain barrier and effectively suppress the toxic effects of OP compounds in the CNS, position THP as a potential drug to be repurposed as a medical countermeasure for posttreatment of patients presenting with OP intoxication.

According to a pharmacokinetic study conducted in rats treated intravenously with THP HCl (3.2 mg/kg), concentrations of THP measured 1-h posttreatment were approximately 1,912 ng/g tissue in the brain and 140 ng/ml in plasma (Ishizaki et al., 1998). Considering 80% of the brain weight as water, the measured THP concentration in brain tissue would correspond to approximately 2,390 ng/ml, which is nearly 17-fold higher than the concentration measured in plasma. The high brain:plasma THP concentration ratio appears to be a result of the uptake of THP in cellular acidic organelles (Ishizaki et al., 1998).

In healthy subjects treated with the highest recommended oral dose of THP HCl for management of Parkinson’s disease (15 mg/day), plasma concentrations of THP are reported to be 50 ng/ml (Brocks, 1999). Assuming the brain:plasma ratio of THP concentrations in humans is comparable to that seen in rats, brain concentrations of THP in humans treated with a single oral dose of 15 mg/day may be as high as 850 ng/ml or approximately 2.8 µM. According to the data presented in the present study, 3 µM THP reduces by nearly 40% the frequency of EPSCs and IPSCs. Therefore, the effects of THP on synaptic transmission are likely to be observed at brain concentrations generated by clinically relevant doses of THP.

Suppression by THP of action potential-dependent glutamatergic transmission mediated by AMPARs and NMDARs can be an important mechanism to halt the progression of OP-induced seizures and neuropathology, which are sustained by increased glutamatergic signaling (Aroniadou-Anderjaska et al., 2020; Dorandeu et al., 2013; McDonough and Shih, 1993). By inhibiting glutamatergic transmission via a noncholinergic mechanism, THP can potentially suppress OP-induced seizures and neuropathology regardless of whether overstimulation of the glutamatergic system: (i) is secondary to the cholinergic overdrive resulting from OP-induced AChE (McDonough and Shih, 1997), or (ii) is induced by non-cholinergic actions of OP nerve agents and insecticides (Rocha et al., 1996; Torres-Altoro et al., 2011). In addition, THP-induced suppression of glutamatergic transmission via an NMDAR-independent mechanism can overcome the limiting effect that OP-induced overstimulation of glutamatergic signaling might have on the ability of THP to block NMDARs (Sills and Loo, 1989).

According to the data presented here, THP also suppresses GABAergic transmission, which is typically considered to serve as a break that helps prevent the propagation of seizure activity. In fact, the GABA_A receptor antagonists picrotoxin and pentylenetetrazol are proconvulsants, whereas positive allosteric modulators of GABA_A receptor activity, including the benzodiazepines used to treat OP-induced seizures, are anticonvulsants (Greenfield, 2013). It is, however, important to consider that GABAergic synaptic transmission plays an important role in synchronizing neuronal networks in different brain regions, including the OP-sensitive limbic structures such as the hippocampus and the amygdala (Avoli and de Curtis, 2011; Barker-Haliski and White, 2015). Thus, THP-induced suppression of GABAergic transmission may play a critical role in disrupting the GABArgic synchronization that has been proposed to contribute to the generation and maintenance of epileptiform activity (Avoli and de Curtis, 2011).

4.5. Study Limitations and Future Directions

While the present study is the first to demonstrate that, acting via a non-cholinergic and non-glutamatergic mechanism, THP can suppress hippocampal synaptic transmission, the study has some limitations. First, although enhanced glutamatergic transmission induced by OP compounds in the hippocampus in vitro resembles that seen in in-vitro hippocampal preparations in which recurrent bursting activity is induced by low Mg²⁺ (Kozhemyakin et al., 2010), it will be important to confirm that in the intact hippocampus, ex vivo or in vivo, THP can also suppress synaptic transmission enhanced by OP compounds. Second, since the cultures used in this study were prepared with hippocampi pooled from male and female fetal rats, it was not possible to determine whether the effect of THP on synaptic transmission is sex dependent. Considering that sex is an important biological variable in OP poisoning (Fawcett et al., 2009; Gage et al., 2020; Gage et al., 2023; Smith et al., 2015), it will be relevant to determine whether THP is equally effective in suppressing OP-induce neurotransmission overdrive and associated neuropathology in males and females. Finally, additional experiments are needed to identify the exact molecular mechanism(s) underlying the inhibitory effect of THP on synaptic transmission. As discussed under section 4.3, voltage-gated Na⁺ channels and σ1 receptors are known molecular targets for THP and may contribute to its ability to suppress action potential-dependent synaptic transmission.

4.6. Conclusions

The unique pharmacological profile of THP can account for its greater effectiveness compared to that of atropine as an OP antidote. In particular, the ability of THP to act via a mechanism independent of mAChR, nAChR, and NMDAR signaling to suppress action potential-dependent synaptic transmission, demonstrated for the first time in the present study, is likely to be an important determinant of its anticonvulsant and neuroprotective properties in OP intoxication. Through this mechanism, THP can effectively decrease excessive neuronal excitability and excitotoxicity during OP intoxication, and, thereby, overcome the limiting effects that excessive levels of extracellular ACh and glutamate may have on the degree to which it can block mAChRs, nAChRs, and NMDARs. This is a paradigm-shifting concept given that, to date, NMDAR block has been hypothesized to be the main mechanism accounting for the greater effectiveness of THP compared to atropine to mitigate OP-induced seizures and accompanying neuropathology.

Highlights.

• THP suppresses action potential-dependent GABAergic and glutamatergic transmission

• THP concentrations that inhibit synaptic transmission are clinically relevant

• THP does not affect the probability of transmitter release

• Inhibition of synaptic transmission by THP is not due to nAChR, mAChR, or NMDAR block

Abbreviations:

ACh

acetylcholine

AMPA

alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

ANOVA

analysis of variance

D,L-AP5

D,L-2-Amino-5-phosphonopentanoic acid

CNS

central nervous system

EGTA

3,12-bis(carboxymethyl)-6,9-dioxa-3,12-diazatetradecanedioic acid

EPSCs

excitatory postsynaptic transmission

HEPES

4-(2-Hydroxyethyl)-1-piperazine ethanesulfonic acid

IC₅₀

antagonist concentration that inhibits by 50% a given response

IPSCs

inhibitory postsynaptic currents

MEM

minimum essential medium

mEPSCs

miniature excitatory postsynaptic transmission

mIPSCs

miniature inhibitory postsynaptic currents

MLA

methyllycaconitine

nAChRs

nicotinic receptors

nH

Hill coefficient

NMDA

N-methyl-D-aspartate

QX-314 bromide

lidocaine N-ethyl bromide

τ_decay

decay-time constant

THP

trihexyphenidyl

TTX

tetrodotoxin

Data availability statement:

The data that support the findings of this study are available from the corresponding author upon request.