The synthetic neuroactive steroid SGE-516 reduces status epilepticus and neuronal cell death in a rat model of soman intoxication

Epilepsy & Behavior 68 (2017) 22–30

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

The synthetic neuroactive steroid SGE-516 reduces status epilepticus and neuronal cell death in a rat model of soman intoxication☆ Alison L. Althaus a,⁎, Hilary S. McCarren b, Aymen Alqazzaz b, Cecelia Jackson b, John H. McDonough b, Carl D. Smith b, Ethan Hoffman a, Rebecca S. Hammond a, Albert J. Robichaud a, James J. Doherty a a b

Drug Discovery, Sage Therapeutics, Inc., Cambridge, MA, USA United States Army Medical Research Institute of Chemical Defense, Aberdeen Proving Ground, MD, USA

a r t i c l e

i n f o

Article history: Received 21 September 2016 Revised 18 November 2016 Accepted 17 December 2016 Available online xxxx

a b s t r a c t Organophosphorus nerve agents (OPNAs) are irreversible inhibitors of acetylcholinesterase that pose a serious threat to public health because of their use as chemical weapons. Exposure to high doses of OPNAs can dramatically potentiate cholinergic synaptic activity and cause status epilepticus (SE). Current standard of care for OPNA exposure involves treatment with cholinergic antagonists, oxime cholinesterase reactivators, and benzodiazepines. However, data from pre-clinical models suggest that OPNA-induced SE rapidly becomes refractory to benzodiazepines. Neuroactive steroids (NAS), such as allopregnanolone, retain anticonvulsant activity in rodent models of benzodiazepine-resistant SE, perhaps because they modulate a broader variety of GABAA receptor subtypes. SGE-516 is a novel, next generation NAS and a potent and selective GABAA receptor positive allosteric modulator (PAM). The present study first established that SGE-516 reduced electrographic seizures in the rat lithium-pilocarpine model of pharmacoresistant SE. Then the anticonvulsant activity of SGE-516 was investigated in the soman-intoxication model of OPNA-induced SE. SGE-516 (5.6, 7.5, and 10 mg/kg, IP) significantly reduced electrographic seizure activity compared to control when administered 20 min after SE onset. When 10 mg/kg SGE-516 was administered 40 min after SE onset, seizure activity was still significantly reduced compared to control. In addition, all cohorts of rats treated with SGE-516 exhibited significantly reduced neuronal cell death as measured by FluoroJade B immunohistochemistry. These data suggest synthetic NASs that positively modulate both synaptic and extrasynaptic GABAA receptors may be candidates for further study in the treatment of OPNA-induced SE. © 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction Status epilepticus (SE) is a life-threatening condition characterized by unremitting seizures that last for at least 5 min. It is difficult to treat and can cause long-term brain damage or mortality [1]. Status epilepticus can arise spontaneously such as in persons with underlying epilepsy or brain lesions, but it can also be elicited in healthy individuals by exposure to convulsant stimuli. For example, organophosphorus

☆ The work described in this manuscript was funded by Sage Therapeutics. A.L.A., E.H., R.S.H., A.J.R, and J.J.D. are employees of Sage Therapeutics. Support for H.S.M., A.A., and C.J. was provided by appointments to the Research Participation Program for the U.S. Army Medical Research Institute of Chemical Defense administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and U.S. Army Medical Research and Materiel Command. The views expressed in this manuscript are those of the authors and do not reflect the official policy of the Department of Army, Department of Defense, or the U.S. Government. ⁎ Corresponding author at: Sage Therapeutics, 215 First Street, Cambridge, MA 02141, USA. E-mail address: [email protected] (A.L. Althaus).

nerve agents (OPNAs) are pro-convulsive chemicals which have been used in chemical warfare and terrorist attacks [2–6]. In addition, people can experience OPNA poisoning through contact with some pesticides [7,8]. These agents are typically inhibitors of acetylcholinesterase (AChE) and thus exert their proconvulsant effects via increased synaptic availability of acetylcholine [9]. Status epilepticus can arise very rapidly after exposure to OPNAs and quickly progress to involve multiple neurotransmitter systems. Three phases of OPNA intoxication have been described using the rat model of intoxication by the potent OPNA soman. The “cholinergic phase,” during which anticholinergic drugs are most effective, only lasts for a few minutes [9]. The “transition phase,” during which other neurotransmitter systems begin to contribute to aberrant spiking, lasts up to about 40 min after SE onset, and the “non-cholinergic phase,” after which seizures are being perpetuated by non-cholinergic mechanisms, begins by around 40 min after SE onset [9]. Current standard medical countermeasures (SMCs) for combat personnel exposed to OPNAs involve rapid intramuscular (IM) selfadministration of atropine sulfate (a cholinergic antagonist) and

http://dx.doi.org/10.1016/j.yebeh.2016.12.024 1525-5050/© 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

pralidoxime (2-PAM; an oxime cholinesterase reactivator), as well as administration of a benzodiazepine (an anticonvulsant) [10]. However, data from rat and nonhuman primate models indicate that these measures are insufficient when treatment of SE is delayed [11–14]. Military personnel and civilians may encounter OPNAs in situations in which immediate treatment is unlikely to be available such as chemical warfare or terrorist attacks. Thus, treatments need to be effective even when administered after a significant delay. Studies have shown that adding other treatments to this SMC regimen can improve the response rate and outcome [15,16]. Allopregnanolone is an endogenous neuroactive steroid (NAS) and positive allosteric modulator of both synaptic and extrasynaptic GABAA receptors [17–19]. Consistent with its activity at GABAA receptors, allopregnanolone exhibits anticonvulsant activity in a variety of seizure models. Importantly, it retains anticonvulsant activity in the rat kainic acid model of refractory SE (RSE) [20]. In this model, many traditional antiepileptic drugs (AEDs) lose activity as SE progresses [21,22]. Allopregnanolone has a differentiated mechanism of action from many other AEDS affording the ability to potentiate a broad range of GABAA receptor subtypes which, as previously mentioned, may underlie its activity in models of RSE [19,20]. Synthetic NASs retain the ability to target multiple synaptic and extrasynaptic GABAA receptor isoforms, and can be designed with improved bioavailability over first generation NASs [19]. SGE-516 is a next generation NAS which has robust pharmacokinetic properties that support its use as an experimental therapeutic in a variety of animal disease models [19]. We hypothesized that SGE-516 would reduce OPNA-induced SE in the soman intoxication model even after a significant treatment delay, owing to its potent anticonvulsant properties and differentiated mechanism of action from benzodiazepines. We first show that SGE-516 reduced seizure activity when administered 60 min after SE onset in the lithium pilocarpine model. We next demonstrate that SGE-516 potently reduces seizure activity when administered 20 min after onset of soman-induced SE, and that 10 mg/kg SGE-516 reduces seizure activity even when administered 40 min after soman-induced SE onset. Finally, we report a significant neuroprotective action of SGE-516 in this model. 2. Methods 2.1. Lithium-pilocarpine experiments Experimental procedures were approved by the Institutional Animal Care and Use Committee and were in accordance with the National Research Council Guide for the Care and Use of Laboratory Animals. Status epilepticus was induced in EEG-tethered rats via the administration of lithium chloride and pilocarpine as previously described [21]. Experiments were performed at Melior Discovery (Exton, PA). 2.1.1. Animals Male Sprague–Dawley rats (Hsd:Sprague–Dawley, Harlan Laboratories, Indianapolis, IN) were surgically implanted with jugular vein catheters (Hilltop Labs, Scottdale, PA). Rats were maintained on a 12hour light/dark schedule with ad libitum access to food and water. Jugular vein catheters were flushed daily with 0.2 mL of heparin solution (50 units/mL) to maintain patency. 2.1.2. Electrode implantation Rats weighing 275–300 g were anesthetized with isoflurane (Aerane, Baxter Healthcare Corp., Deerfield, IL) using a nosecone adapter and Vetequip (Livermore, CA) inhalation anesthesia machine which delivered a 1.5% isoflurane concentration with a flow rate of 1.0 L/min of 30% O2/balance N2 gas. A topical anesthetic (2% lidocaine; Phoenix Pharmaceuticals, Phoenix, AZ) was used locally at the site of scalp incision. Two stainless steel screw electrodes (0–80 × 1/4″, PlasticsOne Roanoke, VA) were implanted flush with the inner skull surface,

23

one at + 3.0 mm AP and 2 mm left of the midline, and one at − 4.0 mm AP and 2.5 mm right of the midline. Wound edges were treated with triple antibiotic cream (Walgreens) and rats were allowed to recover from surgery for at least 1 week prior to the experimental recordings. Ibuprofen (30 mg/kg) was administered for post-surgery analgesia. 2.1.3. Status epilepticus Lithium chloride (127 mg/kg, IP; Sigma Aldrich, St. Louis, MO) was administered 18–24 h prior to EEG recording. On the test day, baseline EEG was recorded for 2 h prior to injection of scopolamine methyl bromide (1 mg/kg, IP; Sigma Aldrich, St. Louis, MO). Pilocarpine (50 mg/kg, IP; Sigma Aldrich, St. Louis, MO) was then administered 30 min after scopolamine. Status epilepticus onset was determined by EEG signal and behavioral scoring of seizure activity based on the Racine scale [23]. The first clonic seizure accompanied by continuous EEG abnormality was taken as the onset of SE. 2.1.4. SGE-516 treatment One hour after SE onset, rats were administered 1 mg/kg (n = 5), 3 mg/kg (n = 5), or 5 mg/kg (n = 4) SGE-516 (Sage Therapeutics) or vehicle (n = 5, 15% hydroxypropyl beta cyclodextrin – HPβCD Sigma Aldrich, St. Louis, MO) intravenously (IV). EEG activity was recorded for approximately 6 h after SE onset and rats were euthanized after EEG recording was terminated. 2.1.5. Data acquisition and analysis Cortical EEG signals were amplified (A-M Systems model 1700; 1000 gain), band pass filtered (0.3–1000 Hz), and digitized at 512 samples per second using ICELUS acquisition/sleep scoring software (M. Opp, U. Michigan) operating under National Instruments (Austin, TX) data acquisition software (Labview 5.1) and hardware (PCI-MIO16E-4). Data were analyzed off-line using Matlab (MathWorks, Natick, MA). Recordings were excluded from analysis if electrical noise obscured the signal or if the animal died prior to the end of recording. Two records from the 1 mg/kg SGE-516 cohort and one from the 5 mg/kg cohort were excluded from analysis. A spike event was defined as a short duration voltage deflection with amplitude N 10 times the root-mean-squared amplitude of the baseline EEG. Instantaneous spike probability was determined by calculating the running average of the total number of 3-second bins that contained a spike event in a train of 100 adjacent bins. Cumulative spike probability was determined for each animal by calculating the average of the instantaneous spike probability over the 150-minute recording period immediately after rescue injection was given. 2.1.6. Statistics Mean spike probability was compared with GraphPad (GraphPad Software Inc., La Jolla, CA) using a one-way ANOVA with Tukey's posthoc comparison. 2.2. Soman model experiments 2.2.1. Animals Male Sprague Dawley rats (Crl:CD(SD) Strain Code 001 Charles River Laboratories, Wilmington, MA) weighing 250–300 g prior to surgery were used for this study. They were housed in individual cages in a temperature and humidity controlled room with a 12-hour light–dark cycle (lights on at 06:00). Food and water were available ad libitum except during experimental periods. The experimental protocol was approved by the Institutional Animal Care and Use Committee (IACUC) at the United States Army Medical Research Institute of Chemical Defense and all procedures were conducted in accordance with the principles stated in the Guide for the Care and Use of Laboratory Animals, and the Animal Welfare Act of 1966 (P.L. 89-544), as amended.

24

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

2.2.2. EEG implant surgery Surgery to implant cortical electroencephalographic (EEG) electrodes was performed 5–7 days prior to experimentation. Animal temperature, heart rate, respiratory rate, and oxygen saturation were monitored throughout the procedure. Subcutaneous meloxicam (1 mg/kg; Norbrook Laboratories Ltd., Co. Down, Northern Ireland) and intradermal bupivacaine (0.1 mL at 0.25%; Hospira, Inc., Lake Forest, IL) were used for analgesia. Rats were anesthetized with isoflurane (3–5% for induction, 1–3% for maintenance, RC2, Vetequip, Livermore, CA) and titrated to a plane at which they showed no response to painful stimuli. A midline incision of approximately 2.5 cm was made in order to expose the skull. Burr holes were created with a hand drill over the cerebellum and each hemisphere of the frontal cortex. Stainless steel screw electrodes (JI Morris, Southbridge, MA) were placed in the holes and attached to a connector plug (Vishay Intertechnology, Shelton, CT). The screws and plug were fixed in place with glass ionomer cement (GC Corporation, Tokyo, Japan), and the incision around the headpiece was closed with sutures. Following recovery from anesthesia, animals were returned to their home cage. An additional dose of 1 mg/kg meloxicam was provided 24 h after surgery.

2.2.3. Nerve agent exposure and SGE-516 treatment At approximately 07:00 on the experiment day, the animals were placed in individual recording chambers and connected to the EEG recording system via their headpiece plug. Following 60 min of normal baseline EEG recording, animals were pretreated with 125 mg/kg HI-6 (Kalexsyn Medicinal Chemistry, Kalamazoo, MI), delivered intraperitoneally (IP). Thirty minutes after pretreatment, 1.6 ∗ LD50 of the nerve agent soman (US Army Edgewood Chemical Biological Center, Aberdeen Proving Ground, MD) was delivered subcutaneously (SC). This dose elicited EEG seizure activity in 100% of animals studied. One minute after soman challenge, animals received 2 mg/kg atropine methyl nitrate (AMN, Wedgewood Pharmacy, Swedesboro, NJ) intramuscularly (IM). Along with HI-6, AMN protected animals from systemic toxicity of soman to allow survival until development of neurological symptoms. Onset of EEG seizure activity was defined by the appearance of repetitive spikes and sharp waves with an amplitude greater than twice that of the baseline and a duration of at least 10 s. At either 20 min or 40 min after seizure onset, animals were treated with 0.45 mg/kg atropine sulfate (Sparhawk Laboratories, Inc., Lenexa, KS) admixed with 25 mg/kg, 2-PAM (Baxter Healthcare Corporation, Deerfield, IL) (IM), 1.8 mg/kg, midazolam (Akorn, Inc., Lake Forest, IL) (IM) and either saline or 5.6, 7.5, or 10 mg/kg SGE-516 (IP). SGE-516 was synthesized by Sage Therapeutics and formulated in 30% HPβCD (Sigma Aldrich, St. Louis, MO) in water. Saline was used as a control instead of HPβCD vehicle to comply with IACUC regulations. In order to minimize pain and suffering, there is a limit placed on the number of animals that can be administered soman without any expectation of benefit from their experimental treatment. Thus, a single group of control animals was treated with saline (the most common vehicle for the test treatments screened at the facility), and this group served as controls for other drugs tested as well. The IP route of administration was used to reduce stress to the animal. The severity of convulsions that accompany soman-induced SE precludes acute IV administration and may cause pre-implanted IV catheters to dislodge. The possibility of catheter failure was considered an unnecessary risk by IACUC since the test treatment could be delivered IP. SGE-516 has robust bioavailability when administered IP, and thus target brain exposures can be achieved with this route. Doses of atropine-sulfate, and 2-PAM, were equivalent to the suggested prehospital limits for humans, as determined by guidelines for the ATNAA for military personnel (Meridian Medical Technologies, Inc., Columbia, MD) and the dose for midazolam was determined by personal communication and data from Silbergleit and colleagues [24]. Following administration of the test compound, EEG activity was

recorded for at least 4 h, after which the animals were euthanized for collection of brain tissue (see below). 2.2.4. EEG recording and analysis EEG signals were amplified with 1902 amplifiers, recorded with a Micro1401 data acquisition interface, and visualized with Spike2 software (all from Cambridge Electronic Design Limited, Cambridge, England). Data channels were sampled at 512 Hz and digitally filtered with a high-pass 0.3 Hz filter, a low-pass 100 Hz, and a 60 Hz notch filter. Spike probability was determined offline using Matlab as described above for the lithium-pilocarpine-treated rats. Group means were compared with a student's t-test or one-way ANOVA using GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, CA) and p ≤ 0.05 was considered significant. 2.2.5. Tissue preparation and histopathology Rats were perfused immediately following the EEG recording session. Rats were deeply anesthetized with sodium pentobarbital (≥75 mg/kg, Vortech Pharmaceuticals, Ltd., Dearborn, MI). They were then transcardially perfused with saline until fully exsanguinated, followed by 10% formalin for tissue fixation. Brains were removed and kept in formalin until being embedded in paraffin and sectioned into 5 μm coronal slices. The section corresponding to 3.24 mm posterior to bregma [25] was slide-mounted for FluoroJade B (FJB; EMD Millipore, Temecula, CA) staining according to published protocols [26]. Images of staining in regions of interest were captured using an Olympus BX50 microscope, DP71 camera, and cellSens Standard software (Olympus Corporation, Tokyo, Japan). Composite images of each region of interest were stitched together using FIJI and cropped to the following sizes for counting (width × height): amygdala (1000 μm × 1000 μm), piriform cortex (200 μm × 2000 μm), thalamus (1000 μm × 1300 μm), and parietal cortex (1500 μm × 2000 μm). The hippocampus counting region was defined by the anatomical boundaries of the structure. Cell counting was performed by a treatment-blinded technician. 2.2.6. Statistics Total spike probability was compared in GraphPad using a one-way ANOVA with Tukey's post-hoc comparison or student's t-test. Fluorojade B-positive cell counts for each region of interest were averaged across groups and compared by t-test or one-way ANOVA. For one-way ANOVAs, a Dunnett's post-hoc multiple comparison test was used to identify differences between experimental groups and the saline-treated group. In a separate analysis (Supplemental Fig. 1), Fluorojade B-positive cell numbers in each region were compared across treatment groups by two-way ANOVA with Tukey's or Sidak's multiple comparisons test (20 and 40-minute time points respectively). p-Values less than 0.05 were considered significant. 3. Results 3.1. SGE-516 reduced or eliminated LIPILO-induced SE The anticonvulsant efficacy of SGE-516 was first assessed in the LIPILO model of RSE. Pilocarpine is a muscarinic cholinergic agonist, and systemic exposure to it can cause SE through initial activation of cholinergic receptors and later recruitment of other neurotransmitter systems – similar to the progression of OPNA-induced SE [27,28]. When LIPILOinduced SE is allowed to progress for N15 min without intervention, seizures become refractory to many traditional anticonvulsants such as benzodiazepines, phenytoin, levetiracetam, carbamazepine, and valproic acid [21,22,29]. However, NASs such as allopregnanolone retain efficacy in a model of RSE even when administered 60 min after SE onset, perhaps due to the ability to modulate both synaptic and extrasynaptic GABAA receptors [20]. Thus, SGE-516, a novel synthetic NAS with a similar mechanism of action, was hypothesized to have anticonvulsant efficacy against LIPILO-induced RSE when administered after a 60-minute delay.

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

Status epilepticus, defined by electrographic and clonic seizure activity, was elicited in 19 male Sprague Dawley rats. One hour after SE onset, rats were administered vehicle or 1, 3, or 5 mg/kg SGE-516 IV (Fig. 1A–C). Aberrant spike activity was measured for individual rats and mean spike rates calculated for each dose group. All three doses of SGE-516 reduced the rate of aberrant spiking rapidly after administration (Fig. 1B). A reduction in spike rate indicates that SGE-516 had an inhibitory effect on seizure activity, but this does not necessarily translate into a cessation of SE, if aberrant spiking is not eliminated. In this experiment, 3 out of 3 rats administered 1 mg/kg SGE-516 remained in SE for the duration of the recording period, despite exhibiting reduced spike rate (Fig. 1B–D). The instantaneous spike probability describes the likelihood that subjects exhibit any aberrant spike activity within a 5-minute time window, and is thus a very sensitive metric for ongoing seizures. Rats administered 1 mg/kg SGE-516 continued to exhibit nearly 100% aberrant spike probability for at least 3 h after SGE-516 administration. In contrast, aberrant spike probability was reduced for rats administered 3 mg/kg SGE-516 and nearly eliminated for rats administered

25

5 mg/kg SGE-516 (Fig. 1C). The cumulative spike probability over the 3-hour analysis window was determined for individual rats (Fig. 1D) and mean spike probability was compared across cohorts for statistical analysis. The 5 mg/kg SGE-516 cohort exhibited a significantly lower spike probability (1.41) than either the 1 or 3 mg/kg cohort (100 and 76.4 respectively; p ≤ 0.0007, one-way ANOVA with Dunnett's posthoc test). 3.2. SGE-516 reduced OPNA-induced seizure activity SGE-516 was hypothesized to reduce OPNA-induced SE due to its differentiated mechanism of action from midazolam and demonstrated anticonvulsant efficacy in the LIPILO model of RSE. The anticonvulsant efficacy of SGE-516 against OPNA-induced SE was evaluated using a rat model of soman exposure. This model included HI-6 (oxime cholinesterase reactivator) pretreatment and atropine methyl nitrate (cholinergic antagonist that does not cross the blood brain barrier) post-treatment to improve rat survival. Additionally, 20 min after SE onset, rats were administered SMCs (2-PAM, atropine sulfate, and

Fig. 1. SGE-516 reduced seizure activity in a rat model of LIPILO-induced SE. A. Sample traces from animals treated with each dose of SGE-516. The grey box indicates the time window used to calculate baseline. The maroon line indicates timing of the atropine injection. The first red line indicates the timing of the pilocarpine injection and the second red line indicates the time of SE onset as determined by clonic seizure activity and continuous EEG abnormality. The blue line indicates the timing of the SGE-516 rescue injection. Scale bar: length = 100 min, height = 5 μV. B. Composite traces show average instantaneous spiking events for each group (black line indicates mean spike rate, grey indicates 2 × standard deviation, blue line indicates rescue injection time). Scale bar: length = 50 min. C. Composite instantaneous spike probability (red line) for the first 3 h after SGE-516 rescue injection (blue vertical line). Scale bar: length = 50 min. D. Total spike probability for the 3-hour post-SGE-516 time period for each group. Black dots represent spike probabilities for individual rats. The spike probability for rats administered 5 mg/kg SGE-516 was significantly lower than that of the vehicle-treated cohort, but neither the 1 nor 3 mg/kg SGE-516-treated cohort differed from vehicle. ****p b 0.0001; one-way ANOVA with Dunnett's post-hoc comparison. We have elected to pay for color figures in print.

26

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

Fig. 2. SGE-516 reduced seizure activity when administered 20 min after the onset of soman-induced SE. A. Schematic representation of the experimental time course for each animal. HI-6 is an oxime cholinesterase reactivator, AMN (atropine methyl nitrate) is a peripheral cholinergic antagonist, SMCs are atropine sulfate (neuroactive cholinergic antagonist), 2PAM (oxime cholinesterase reactivator) and midazolam (a benzodiazepine). B. Sample traces from animals treated with SMCs + saline or each dose of SGE-516. C. Composite instantaneous spike probability for the first 3 h after rescue injection. Scale bar: length = 50 min. D. Total spike probability for the 3-hour post-SGE-516 time period for each group. The probability of spiking for rats administered all three doses of SGE-516 was significantly lower than the probability of spiking for the control cohort. ***p = 0.0003, ****p b 0.0001; one-way ANOVA with Dunnett's post-hoc comparison.

midazolam – a benzodiazepine) and saline, or 5.6, 7.5, or 10 mg/kg SGE516, IP (Fig. 2A). Intraperitoneal bioavailability of SGE-516 is 73%, with a brain to plasma ratio of 1.4 at 30 min after IP administration. Salinetreated rats were used as the reference control cohort instead of HPβCD-treated rats in order to comply with institutional animal welfare guidelines (see Methods for details). Status epilepticus, defined by unremitting electrographic seizure activity, was induced in 34 male Sprague Dawley rats via soman exposure. Instantaneous spike probability was reduced to ~90% after the rescue injection in the control cohort as a result of reduced spiking exhibited by 1 of 10 rats (Fig. 2C, D). In contrast, 5.6 mg/kg SGE-516 reduced instantaneous spike probability to ~50% by 30 min after the rescue injection and both 7.5 and 10 mg/kg SGE-516 doses reduced instantaneous spike probability to ≤ 25% by 30 min after the rescue injection (Fig. 2C). None of the three doses tested completely eliminated aberrant

spike probability across an entire cohort of rats; even when aberrant spiking was eliminated for some duration, it returned in most rats before the end of the 4-hour recording period. Cumulative spike probability was determined for each animal by the average of their instantaneous probability in the three-hour time window after rescue injection. Group means were compared across cohorts. All three doses of SGE-516 significantly reduced the mean spike probability compared to control ( p ≤ 0.0003, one-way ANOVA with Dunnett's multiple comparisons test; Fig. 2D). To investigate whether SGE-516 would retain anticonvulsant efficacy when the SE entered the “non-cholinergic phase,” two additional cohorts of rats were administered rescue injections (one control and one 10 mg/kg, IP SGE-516) at 40 min after SE onset (Fig. 3A). Control injection (SMCs + saline) at the 40-minute time point resulted in a slight reduction in instantaneous spike probability, while 10 mg/kg

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

27

Fig. 3. 10 mg/kg SGE-516 reduced seizure activity when administered 40 min after the onset of soman-induced SE. A. Schematic representation of the experimental time course for each animal. HI-6 is an oxime cholinesterase reactivator, AMN (atropine methyl nitrate) is a peripheral cholinergic antagonist, SMCs are atropine sulfate (neuroactive cholinergic antagonist), 2PAM (oxime cholinesterase reactivator) and midazolam (a benzodiazepine). B. Sample traces from animals treated with SMCs + saline 10 mg/kg SGE-516. C. Composite instantaneous spike probability for the first 3 h after rescue injection. D. Total spike probability for the 3-hour post-rescue injection for each group. The probability of spiking for rats administered 10 mg/kg SGE-516 was significantly lower than the probability of spiking for the control cohort. *p = 0.025; student's t-test.

SGE-516 resulted in a ~ 50% reduction in spike probability by 45 min after the injection. However, aberrant spike probability returned to ~ 90% by 3 h after the rescue injection (Fig. 3C). Even so, mean spike probability over the 3-hour time period was significantly lower for the cohort of rats treated with 10 mg/kg SGE-516 compared to control (p = 0.0245, student's t-test; Fig. 3D). 3.3. Administration of SGE-516 reduced neuronal cell death after somaninduced SE Widespread neuronal cell death is a common feature in preclinical models of prolonged SE, which can contribute to cognitive dysfunction, aberrant plasticity, and epileptogenesis [30,31]. Therapies that reduce the extent of SE-induced neuropathology may improve recovery in patients. To test the effect SGE-516 in preventing or ameliorating SEinduced neuropathology, dying neurons were identified in fixed brain tissue with FJB immunohistochemistry from a subset of rats in each of the soman-treated cohorts described above. Fluorojade B+ cells were counted in defined regions within the amygdala, piriform cortex, thalamus, parietal cortex, and hippocampus, brain regions that are particularly susceptible to seizure-induced neurotoxicity [30,32]. The average number of FJB+ cells from rats treated 20 min after SE onset with 5.6 (n = 5), 7.5 (n = 3), or 10 (n = 4) mg/kg SGE-516 was significantly lower than the average number of FJB + cells from control-injected rats (n = 9, p b 0.0001, one-way ANOVA with Dunnett's post-hoc test; Fig. 4A). Fluorojade B+ cell numbers were significantly reduced in all five brain regions measured (Supplemental Fig.

1A). In addition, the average number of FJB + cells from rats treated 40 min after SE onset with 10 mg/kg SGE-516 (n = 8) was significantly lower than the average number of FJB+ cells from control-injected rats (n = 8, p = 0.0002, student's t-test; Fig. 4B). Fluorojade B+ cell numbers were significantly reduced in the amygdala, piriform cortex, and thalamus, but not in the hippocampus or parietal cortex, suggesting that the latter two regions are more susceptible to cell death early after SE onset (Supplemental Fig. 1B). A separate subset of rats that was administered 7.5 mg/kg SGE-516 20 min after SE onset was not euthanized until 24 h later (n = 3 survived to this time point). The average number of FJB + cells from these rats was 325 ± 122 (data not shown), which suggests long-lasting amelioration of neuronal cell death, considering that the control rats that were euthanized 4 h after SE exhibited 1012 ± 80 FJB+ cells. 4. Discussion Organophosphorus nerve agents such as soman pose a real and serious threat to soldiers and civilians, and exposure to OPNAs is likely to occur under circumstances that preclude immediate medical attention [33]. Status epilepticus can arise within minutes of severe OPNA exposure and becomes treatment-resistant as the duration of SE progresses [9]. SGE-516 is an example of the class of next generation NASs which may be effective for the treatment of OPNA-induced SE on the basis of the results described in this paper. SGE-516 stopped seizures in the rat LIPILO model of RSE when administered 60 min after SE onset (Fig. 1). As more direct evidence that next generation NASs showed anticonvulsant

28

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

Fig. 4. All doses of SGE-516 tested were neuroprotective when degenerating neurons were counted 4 h after soman exposure. A. Sample images of a region of interest within the amygdala from rats treated at 20-minute post-SE onset. FJB+ cells in green. Scale bar = 200 μm. Ai. The average number of FJB+ cells (total # cells per rat across five brain regions) for rats treated with SGE-516 were significant lower than the number of FJB+ cells in rats treated with SMCs alone (one-way ANOVA with Dunnett's post-hoc test. p b 0.0001). B. Sample images of a region of interest within the amygdala from rats treated at 40-minute post-SE onset. FJB+ cells in green. Scale bar = 200 μm. Bi. Rats treated with 10 mg/kg SGE-516 had significantly fewer FJB+ cells than rats treated with SMCs alone (students t-test, p = 0.0002).

activity against OPNA-induced SE, administration of 5.6, 7.5, and 10 mg/kg SGE-516 doses 20 min after SE reduced aberrant spiking (Fig. 2), and 10 mg/kg reduced seizure activity even when administered 40 min after SE onset (Fig. 3). Finally, treatment with SGE-516 10 mg/kg enhanced neuronal survival, even though the reduction in aberrant spike activity was partial and transient (Fig. 4). SGE-516 is a next generation NAS with activity at synaptic and extrasynaptic receptors in vitro, and robust bioavailability across multiple routes of administration [19]. The ability of SGE-516 to positively modulate GABAA receptors is similar to that of allopregnanolone [19], which likely is important for its anticonvulsant activity. Allopregnanolone potently enhances inhibitory currents through GABAA receptors in a manner that is consistent with its demonstrated anticonvulsant activity in pre-clinical models [18,34,35]. Importantly, allopregnanolone also exhibits anticonvulsant efficacy in a rodent model of refractory SE induced by kainic acid [20]. While the underlying mechanisms of pharmacoresistance in RSE are not well-known, reduced synaptic availability of GABAA receptors may play a role [36]. Thus, the ability of NASs to modulate multiple synaptic and extrasynaptic GABAA receptor isoforms may be

important in treating refractory seizures. Clinical reports of the successful use of Althesin (a mixture of alphaxolone and alphadolone, both synthetic NASs) to treat patients with drugresistant SE suggest that the pre-clinical data showing anticonvulsant activity of NASs has potential translational value [37,38]. Indeed, there is much interest in exploring the therapeutic utility of NASs across multiple disease areas, especially those associated with GABAergic hypofunction [39]. Organophosphorus nerve agent-induced SE initially arises from massive potentiation of cholinergic synapses, but can progress very rapidly to include other neurotransmitter systems and become resistant to therapeutic intervention [9]. The efficacy of NASs in the LIPILO model of pharmaco-resistant seizure activity suggested that they might also have efficacy in animal models of OPNA-induced SE. Indeed, all doses tested significantly altered the rats' seizure activity, as measured by a reduction in aberrant spike probability in the 3 h after the rescue injection. Data from the control group showed that SMCs were insufficient to affect seizure activity in the majority of rats and underscore the need for improving therapeutic options.

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30

In the present study, the reduction of electrographic seizure activity in the soman-SE model by SGE-516 was temporary in most animals across treatment groups, while LIPILO-induced SE was completely controlled by the highest dose of SGE-516. Although LIPILO-induced SE involves a similar mechanism of seizure initiation as somaninduced SE, pilocarpine, utilized at the current dose, is less toxic than soman [28]. In addition, the different modes of administration utilized afford distinct pharmacokinetic exposure profiles. Intravenous administration of SGE-516 (which was used in the LIPILO experiments) allows for faster action and higher brain exposures per dose than IP administration (which was used in the soman experiments). Indeed, the time between SGE-516 administration and reduction in SE was longer after IP administration than IV. This can be seen in the instantaneous spike probability plots, and likely reflects the different time to maximum brain exposure of SGE-516 with the different administration routes. With increased dose or IV administration, it is possible that somaninduced SE could be even better controlled with NAS treatment. In addition to the superior performance of NASs in the LIPILO model of RSE, the significant reduction in neuronal cell death observed in all groups of animals treated with SGE-516 suggests a potential additional effect in OPNA-induced SE. Neuronal cell death is associated with many brain insults including stroke, traumatic injury, and SE [40,41]. The effects of neurotoxicity can be variable and hard to determine in humans but OPNA exposure has been correlated with long-term aberrant EEG activity and memory disturbances [42–44]. In laboratory models of OPNA-induced SE, cognitive decline and the development of epilepsy have been directly correlated with the extent of neuronal cell death across brain regions [45–48]. Reducing neuronal cell death may aid in recovery and rehabilitation for patients who survive OPNAinduced SE. Allopregnanolone has been shown to reduce neuronal cell death in other rodent models of SE, suggesting a potential class effect of GABAA receptor potentiating NASs [49,50]. However, additional studies are needed to determine the role of NASs and the long-term benefits of neuroprotection demonstrated in the soman model. 5. Conclusion Status epilepticus can be induced in rodent models with a variety of different noxious stimuli, but the underlying seizure activity can quickly progress to affect glutamatergic and GABAergic systems. Many experimental models of SE, including the LIPILO and soman models, require rapid intervention with AEDs in order to stop or reduce seizure activity [9,22], but rapid intervention is not always possible. The next generation NAS, SGE-516, reduced electrographic seizure activity in both LIPILO and soman models even at time points when benzodiazepines are not effective. Additionally, SGE-516 administration significantly reduced neuronal cell death compared to control treatment in the soman model. Together, the data presented in this study support the further development of next generation NASs to treat RSE and OPNA-induced SE. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2016.12.024. Conflicts of interest There are no conflicts of interest. References [1] Trinka E, Cock HR, Hesdorffer D, Rossetti AO, Scheffer IE, Shinnar S, et al. A definition and classification of status epilepticus - report of the ILAE Task Force on Classification of Status Epilepticus. Epilepsia 2015;56:1515–23. [2] United Nations mission to investigate allegations of the use of chemical weapons in the Syrian Arab Republic, final report; 2013. [3] Okumura T, Takasu N, Ishimatsu S, Miyanoki S, Mitsuhashi A, Kumada K, et al. Report on 640 victims of the Tokyo subway sarin attack. Ann Emerg Med 1996;28:129–35. http://dx.doi.org/10.1016/S0196-0644(96)70052-5.

29

[4] Rosman Y, Eisenkraft A, Shiyovich A, Ophir N, Shrot S, Kreiss Y, et al. Lessons learned from the Syrian sarin attack: evaluation of a clinical syndrome through social media. Ann Intern Med 2014;160:644–8. [5] Proctor SP, Heaton KJ, Heeren T, White RF. Effects of sarin and cyclosarin exposure during the 1991 Gulf War on neurobehavioral functioning in US army veterans. Neurotoxicology 2006;27:931–9. http://dx.doi.org/10.1016/j.neuro. 2006.08.001. [6] Balali-Mood M, Saber H. Recent advances in the treatment of organophosphorous poisonings. Iran J Med Sci 2012;37:74–91. [7] Karalliedde L, Senanayake N. Organophosphorus insecticide poisoning. Br J Anaesth 1989;63:736–50. http://dx.doi.org/10.1093/bja/63.6.736. [8] Dassanayake T, Weerasinghe V, Dangahadeniya U, Kularatne K, Dawson A, Karalliedde L, et al. Long-term event-related potential changes following organophosphorus insecticide poisoning. Clin Neurophysiol 2008;119:144–50. http://dx. doi.org/10.1016/j.clinph.2007.09.134. [9] McDonough JH, Shih TM. Neuropharmacological mechanisms of nerve agentinduced seizure and neuropathology. Neurosci Biobehav Rev 1997;21:559–79. http://dx.doi.org/10.1016/S0149-7634(96)00050-4. [10] Cannard K. The acute treatment of nerve agent exposure. J Neurol Sci 2006;249: 86–94. http://dx.doi.org/10.1016/j.jns.2006.06.008. [11] Lallement G, Dorandeu F, Filliat P, Carpentier P, Baille V, Blanchet G. Medical management of organophosphate-induced seizures. J Physiol Paris 1998;92:369–73. http://dx.doi.org/10.1016/S0928-4257(99)80007-2. [12] Shih TM, McDonough JH, Koplovitz I. Anticonvulsants for soman-induced seizure activity. J Biomed Sci 1999;6:86–96. [13] Mazarati AM, Baldwin RA, Sankar R, Wasterlain CG. Time-dependent decrease in the effectiveness of antiepileptic drugs during the course of self-sustaining status epilepticus; 1998 13–5. [14] McDonough JH, McMonagle JD, Shih TM. Time-dependent reduction in the anticonvulsant effectiveness of diazepam against soman-induced seizures in Guinea pigs. Drug Chem Toxicol 2010;33:279–83. [15] Rossetti F, de Araujo FM, Pak T, Bailey K, Shields M, Chanda S, et al. Combined diazepam and HDAC inhibitor treatment protects against seizures and neuronal damage caused by soman exposure. Neurotoxicology 2012;33:500–11. http://dx.doi.org/10. 1016/j.neuro.2012.02.010. [16] Schultz MK, Wright LKM, Stone MF, Schwartz JE, Kelley NR, Moffett MC, et al. The anticholinergic and antiglutamatergic drug caramiphen reduces seizure duration in soman-exposed rats: synergism with the benzodiazepine diazepam. Toxicol Appl Pharmacol 2012;259:376–86. http://dx.doi.org/10.1016/j.taap.2012.01.017. [17] Belelli D, Lan NC, Gee KW. Anticonvulsant steroids and the GABA/benzodiazepine receptor-chloride ionophore complex. Neurosci Biobehav Rev 1990;14:315–22. http://dx.doi.org/10.1016/S0149-7634(05)80041-7. [18] Kokate TG, Svensson BE, Rogawski MA. Anticonvulsant activity of neurosteroids: correlation with gamma-aminobutyric acid-evoked chloride current potentiation. J Pharmacol Exp Ther 1994;270:1223–9. [19] Martinez Botella G, Salituro FG, Harrison BL, Beresis RT, Bai Z, Shen K, et al. Neuroactive steroids. 1. Positive allosteric modulators of the (gamma-aminobutyric acid)A receptor: Structure-activity relationships of heterocyclic substitution at C-21. J Med Chem 2015;58:3500–11. http://dx.doi.org/10.1021/acs.jmedchem.5b00032. [20] Rogawski MA, Loya CM, Reddy K, Zolkowska D, Lossin C. Neuroactive steroids for the treatment of status epilepticus. Epilepsia 2013;54:93–8. http://dx.doi.org/10.1111/ epi.12289. [21] Pouliot W, Bialer M, Hen N, Shekh-Ahmad T, Kaufmann D, Yagen B, et al. A comparative electrographic analysis of the effect of sec-butyl-propylacetamide on pharmacoresistant status epilepticus. Neuroscience 2013;231:145–56. http://dx. doi.org/10.1016/j.neuroscience.2012.11.005. [22] Morrisett RA, Jope RS, Snead OC. Effects of drugs on the initiation and maintenance of status epilepticus induced by administration of pilocarpine to lithium-pretreated rats. Exp Neurol 1987;97:193–200. [23] Racine RJ. Modification of seizure activity by electrical stimulation. II motor seizure. Electroencephalogr Clin Neurophysiol 1972;32:281–94. [24] Silbergleit R, Durkalski V, Lowenstein D, Conwit R, Pancioli A, Palesch Y, et al. Intramuscular versus intravenous therapy for prehospital status epilepticus. N Engl J Med 2012;366:591–600. http://dx.doi.org/10.1056/NEJMoa1109400. [25] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 6th ed. Burlington, MA.: Acadmic Press; 2006 [26] Schmued LC, Hopkins KJ. Fluoro-Jade B: a high affinity fluorescent marker for the localization of neuronal degeneration. Brain Res 2000;874:123–30. http://dx.doi.org/ 10.1016/S0006-8993(00)02513-0. [27] De Araujo FM, Rossetti F, Chanda S, Yourick D. Exposure to nerve agents: from status epilepticus to neuroinflammation, brain damage, neurogenesis and epilepsy. Neurotoxicology 2012;33:1476–90. http://dx.doi.org/10.1016/j.neuro.2012.09.001. [28] Tang FR, Loke WK, Ling EA. Comparison of status epilepticus models induced by pilocarpine and nerve agents - a systematic review of the underlying aetiology and adopted therapeutic approaches. Curr Med Chem 2011;18:886–99. http://dx.doi. org/10.2174/092986711794927720. [29] Zheng Y, Moussally J, Cash SS, Karnam HB, Cole AJ. Intravenous levetiracetam in the rat pilocarpine-induced status epilepticus model: behavioral, physiological and histological studies. Neuropharmacology 2010;58:793–8. http://dx.doi.org/10.1016/j. neuropharm.2009.12.007. [30] Dingledine R, Varvel N, Dudek F. When and how do seizures kill neurons, and is cell death relevant to epileptogenesis? Adv Exp Med Biol 2014;813:109–22. [31] Wasterlain C, Shirasaka Y. Seizures, brain damage and brain development. Brain Dev 1994;16:279–95. [32] Scholl EA, Dudek FE, Ekstrand JJ. Neuronal degeneration is observed in multiple regions outside the hippocampus after lithium pilocarpine-induced status epilepticus

30

[33]

[34]

[35]

[36]

[37] [38] [39] [40]

[41]

A.L. Althaus et al. / Epilepsy & Behavior 68 (2017) 22–30 in the immature rat. Neuroscience 2013;45–59. http://dx.doi.org/10.1097/MPG. 0b013e3181a15ae8.Screening. Thiermann H, Worek F, Kehe K. Limitations and challenges in treatment of acute chemical warfare agent poisoning. Chem Biol Interact 2013;206:435–43. http://dx. doi.org/10.1016/j.cbi.2013.09.015. Kaminski RM, Livingood MR, Rogawski MA. Allopregnanolone analogs that positively modulate GABA a receptors protect against partial seizures induced by 6-Hz electrical stimulation in mice. Epilepsia 2004;45(7):864–7 bhttps://urldefense.proofpoint. com/v2/url?u=https-3A__www.ncbi.nlm.nih.gov_pubmed_-3Fterm-3DKaminski-262343-3BRM-26-2343-3BAND-26-2343-3BLivingood-26-2343-3BMR&d= CwMGaQ&c=WknmpdNpvrlj2B5K1aWVqL1SOiF30547pqSuOmtwXTQ&r= Ke2Ou4y6xYFa6e5FXO0Gpxk9gZqLbQACUg3BDgQOe7E&m=CZKl1flbUIP74HC3irk_ zNIfBSH5JltrwsV25bdAA8s&s=MvgMr5ze_ROsCtLZfcfQXC_pcM754r1H7gMyS-_ Mb9c&e=N. Lonsdale D, Burnham WM. The anticonvulsant effects of allopregnanolone against amygdala-kindled seizures in female rats. Neurosci Lett 2007;411:147–51. http:// dx.doi.org/10.1016/j.neulet.2006.10.023. Naylor DE, Liu H, Wasterlain CG. Trafficking of GABA(A) receptors, loss of inhibition, and a mechanism for pharmacoresistance in status epilepticus. J Neurosci 2005;25: 7724–33. http://dx.doi.org/10.1523/JNEUROSCI.4944-04.2005. Saady A, Wong D, Hicks RG. Case report: Althesin in status epilepticus. Anaesth Intensive Care 1979;Vll:267–70. Munari C, Casaroli D, Matteuzzi G, Pacific L. The use of Althesin in drug-resistant status epilepticus. Epilepsia 1979;20:475–84. Reddy DS, Estes WA. Clinical potential of neurosteroids for CNS disorders. Trends Pharmacol Sci 2016;37:543–61. http://dx.doi.org/10.1016/j.tips.2016.04.003. Bramlett HM, Dietrich WD. Pathophysiology of cerebral ischemia and brain trauma: similarities and differences. J Cereb Blood Flow Metab 2004;24:133–50. http://dx. doi.org/10.1097/01.WCB.0000111614.19196.04. Kumar G, Mittal S, Moudgil SS, Kupsky WJ, Shah AK. Histopathological evidence that hippocampal atrophy following status epilepticus is a result of neuronal necrosis. J Neurol Sci 2013;334:186–91.

[42] Sekijima Y, Morita H, Yanagisawa N. Follow-up of sarin poisoning in Matsumoto. Ann Intern Med 1997;127:1042. [43] Yanagisawa N, Morita H, Nakajima T. Sarin experiences in Japan: acute toxicity and long-term effects. J Neurol Sci 2006;249:76–85. http://dx.doi.org/10.1016/j.jns.2006. 06.007. [44] Miyaki K, Nishiwaki Y, Maekawa K, Ogawa Y, Asukai N, Yoshimura K, et al. Effects of sarin on the nervous system of subway workers seven years after the Tokyo subway sarin attack. J Occup Health 2005;47:299–304. http://dx.doi.org/10.1539/joh.47.299. [45] Myhrer T, Andersen JM, Nguyen NHT, Aas P. Soman-induced convulsions in rats terminated with pharmacological agents after 45 min: neuropathology and cognitive performance. Neurotoxicology 2005;26:39–48. http://dx.doi.org/10.1016/j.neuro. 2004.07.011. [46] Filliat P, Baubichon D, Burckhart M, Pernot-Marino I, Foquin A, Masquellez C, et al. Memory impairment after soman intoxication in rat: correlation with central neuropathy. Improvement with anticholinergic and antiglutamatergic therapeutics. Neurotoxicology 1999;20:535–49. [47] Collombet JM, Beracochea D, Liscia P, Pierard C, Lallement G, Filliat P. Long-term effects of cytokine treatment on cognitive behavioral recovery and neuronal regeneration in soman-poisoned mice. Behav Brain Res 2011;221:261–70. http://dx.doi.org/ 10.1016/j.bbr.2011.03.006. [48] Filliat P, Coubard S, Pierard C, Liscia P, Beracochea D, Four E, et al. Long-term behavioral consequences of soman poisoning in mice. Neurotoxicology 2007;28:508–19. http://dx.doi.org/10.1016/j.neuro.2006.11.004. [49] Singh S, Hota D, Prakash A, Khanduja KL, Arora SK, Chakrabarti A. Pharmacology, biochemistry and behavior allopregnanolone, the active metabolite of progesterone protects against neuronal damage in picrotoxin-induced seizure model in mice. Pharmacol Biochem Behav 2010;94:416–22. http://dx.doi.org/10.1016/j.pbb.2009. 10.003. [50] Leskiewicz M, Budziszewska B, Jaworska-Feil L, Lason W. Effects of neurosteroids on kainate-induced seizures, neurotoxicity and lethality in mice. Pol J Pharmacol 1997; 49:411–7.