Riluzole and novel naphthalenyl substituted aminothiazole derivatives prevent acute neural excitotoxic injury in a rat model of temporal lobe epilepsy

Abstract

Epileptogenic seizures, or status epilepticus (SE), leads to excitotoxic injury in hippocampal and limbic neurons in the kainic acid (KA) animal model of temporal lobe epilepsy (TLE). Here, we have further characterized neural activity regulated methylaminoisobutryic acid (MeAIB)/glutamine transport activity in mature rat hippocampal neurons in vitro that is inhibited by riluzole (IC₅₀ = 1 μM), an anti-convulsant benzothiazole agent. We screened a library of riluzole derivatives and identified SKA-41 followed by a second screen and synthesized several novel chlorinated aminothiazoles (SKA-377, SKA-378, SKA-379) that are also potent MeAIB transport inhibitors in vitro, and brain penetrant following systemic administration. When administered before KA, SKA-378 did not prevent seizures but still protected the hippocampus and several other limbic areas against SE-induced neurodegeneration at 3d. When SKA-377 - 379, (30 mg/kg) were administered after KA-induced SE, acute neural injury in the CA3, CA1 and CA4/hilus was also largely attenuated. Riluzole (10 mg/kg) blocks acute neural injury. Kinetic analysis of SKA-378 and riluzoles’ blockade of Ca²⁺-regulated MeAIB transport in neurons in vitro indicates that inhibition occurs via a non-competitive, indirect mechanism. Sodium channel Na_V1.6 antagonism blocks neural activity regulated MeAIB/Gln transport in vitro (IC₅₀ = 60 nM) and SKA-378 is the most potent inhibitor of Na_V1.6 (IC₅₀ = 28 μM) compared to Na_V1.2 (IC₅₀ = 118 μM) in heterologous cells. However, pharmacokinetic analysis suggests that sodium channel blockade may not be the predominant mechanism of neuroprotection here. Riluzole and our novel aminothiazoles are agents that attenuate acute neural hippocampal injury following KA-induced SE and may help to understand mechanisms involved in the progression of epileptic disease.

1. Introduction

Temporal lobe epilepsy (TLE) is a common and debilitating chronic neurologic disorder (Sirven, 2015; Devinsky et al., 2018). Currently, most antiseizure drugs (ASDs) used in the treatment of epilepsy are prescribed to prevent recurrent epileptic seizure activity and target mechanisms that increase inhibitory GABA transmission or reduce Glu excitatory transmission (Meldrum and Rogawski, 2007; Sills and Rogawski, 2020). Glu is the major excitatory neurotransmitter in the brain and is also a potent excitotoxin that after excessive synaptic release during seizure activity can lead to both acute and chronic neural injury (Holmes, 2002; Dudek and Sutula, 2007; Bittigau and Ikonomidou, 1997; Dodd, 2002; Rothstein et al., 1996). Here, we utilized the KA rat model of TLE to produce limbic seizures and hippocampal neural injury consistent with that seen in humans (Ben Ari, 1985; Sperk, 1994; Sharma et al., 2007; Levesque, 2013; Rusina et al., 2021). KA is a potent Glu excitotoxin that stimulates excessive presynaptic Glu release from dentate gyrus (DG) mossy fibers onto excitatory CA3 pyramidal neurons and mossy cells in the hilar region. Glu-induced excitotoxicity then proceeds to CA3-CA1 collateral subfields within the hippocampus and then via inter-connected excitatory hippocampal output projections to other limbic areas of the brain producing neural injury in multiple regions (Nadler, 1981; Ferkany and Coyle, 1983; Henze et al., 2000; Hopkins et al., 2000; Wang et al., 2005; Srivastava et al., 2020). The process of epileptogenesis is not clearly understood, but in the KA model of TLE it is initiated by epileptogenic seizures (Dudek and Sutula, 2007; Ben Ari, 1985; Sperk, 1994; Sharma et al., 2007; Levesque, 2013; Rusina et al., 2021; Wang et al., 2005), or status epilepticus (SE), as defined by stage 5 grade seizure activity using the Racine scale (Racine, 1972). SE results in acute hippocampal neural injury in the KA model of TLE and then during a ‘silent period’ (3–6 months) leads to hippocampal sclerosis and alterations in neuronal circuitry that eventually leads to spontaneous recurrent epileptic seizures (Bouilleret et al., 1999; Riban et al., 2002; Dunleavy et al., 2010), which most clinicians would define as epilepsy (Fisher et al., 1998; Watson, 2003; Janszky et al., 2005; Blumcke et al., 2012; Thom, 2014). Late-stage drug-resistant TLE results from the inability of ASD’s to prevent recurrent epileptic seizure activity (Tang et al., 2017; Loscher et al., 2020). In addition, many of the ASD’s used today have consequential negative side effects such as sedation, mental slowing, and motor reflex reduction (Walia et al., 2004; Perucca and Gilliam, 2012; Thigpen et al., 2013). A critical goal going forward is to identify new synaptic targets and to develop antiepileptic drugs (AEDs) specific to these targets to prevent acute hippocampal neural injury that can lead to epileptogenesis and epileptic disease, which would be most beneficial in clinical settings (White and Bialer, 2010; Loscher and Schmidt, 2011; Kaminski et al., 2014; Loscher, 2020; Benassi et al., 2021; Marques et al., 2022).

Riluzole is an anti-glutamatergic neuroprotective agent in several animal models of neural injury (Doble, 1996; Stutzmann et al., 1997) and is thought to act by reducing synaptic Glu release from axonal synapses (Martin et al., 1993; Lingamaneni and Hemmings, 1999; Wang et al., 2004), and thus excitotoxicity. Riluzole, also called Rilutek®, is an FDA approved drug for amyotrophic lateral sclerosis (ALS) (Salameh et al., 2015). Even though riluzole and other similar benzothiazole derivatives are also anti-convulsant agents (Mizoule et al., 1985; Romettino et al., 1991; Jimonet et al., 1994; Borowicz et al., 2004; Kim et al., 2007; Ugale et al., 2012; Ali and Siddiqui, 2013; Coleman et al., 2015; Jagbir et al., 2015), they are not utilized in epilepsy therapy in Europe or the United States. Riluzole’s ability to inhibit various activities varies with potency in vitro (μM - mM) among its multiple targets throughout the brain and spinal cord. Riluzole’s neuroprotective actions may include the inhibition of Na⁺ channel activity (Herbert et al., 1994; Song et al., 1997; Stefani et al., 1997; Prakriya and Mennerick, 2000; Spadoni et al., 2002); K⁺ and Ca²⁺ channel activity (Zona et al., 1998; Grunnet et al., 2001; Duprat et al., 2000; Huang et al., 1997), protein kinase activity (Noh et al., 2000), and Glu receptor activity (Debono et al., 1993; Zona et al., 2002). Riluzole has also been reported to stimulate small-conductance Ca²⁺-activated K⁺ channels (Sankaranarayanan et al., 2009) and the activity of synaptic Glu clearance transporters (Fumagalli et al., 2008) located on perisynaptic astrocytes that are critically important to resist direct Glu excitotoxicity (Sheldon and Robinson, 2007). The most potent target of riluzole is inhibition of spontaneous neuron specific, activity regulated, high affinity MeAIB/Gln transport activity in mature hippocampal neurons in vitro with an IC₅₀ ~1 μM (Erickson, 2017). Based on these multiple pharmacological activities, riluzole constitutes what has been termed a ‘privileged structure’, meaning a small, typically bicyclic ring system that is known to exert multiple biological activities and therefore can serve as a template for the design of drugs to inhibit or activate one of these targets selectively by appropriately decorating it with substituents (Horton et al., 2003; Priyanka and Jha, 2010; Sharma et al., 2012).

MeAIB is considered a selective substrate for a low affinity (K_m ~ 0.5 mM) neutral amino acid transport system originally defined as system A in mostly all cells, which prefers alanine described years ago (Christensen, 1990). System A transporters in the SLC38 gene family are called SNATs, which translates to Sodium-coupled Neutral Amino Acid Transporters or System N and A Transporters (Mackenzie and Erickson, 2004). SNAT1 and SNAT2 are the major Na⁺-coupled neutral amino acid transporters expressed in neurons (Varoqui et al., 2000; Yao et al., 2000; Reimer et al., 2000; Chaudhry et al., 2002a). Gln is a substrate of system A and the most abundant neutral amino acid in cerebral spinal fluid (CSF) (Gjessing et al., 1972; McGale et al., 1977). SNATs have been proposed to be a critical conduit from glia to provide Gln as precursor to neurotransmitter Glu synthesis and Ca²⁺-dependent Glu release from neuronal synapses (Mackenzie and Erickson, 2004; Chaudhry et al., 2002b), but see (Conti and Melone, 2006). Once released from axon terminals Glu is cleared from the synaptic cleft by Glu transporters on astrocytes (Danbolt, 2001), converted to Gln by Gln synthetase (Norenberg and Martinez-Hernandez, 1979), released from astrocytes (Deitmer et al., 2003; Leke and Schousboe, 2016), imported to axon terminals by unknown mechanisms, and then converted to Glu via phosphate-activated glutaminase (Kvamme et al., 2001; Masson et al., 2006) for exocytotic release. Replenishment and maintenance of cytoplasmic Glu levels in axon terminals by Glu/Gln cycling may be required for continued Glu synaptic release under prolonged Ca²⁺-regulated excitotoxic conditions (Marx et al., 2015; Pietrancosta et al., 2020), but see (Kam and Nicoll, 2007). Other metabolic pathways to neurotransmitter Glu synthesis and release have also been postulated (Schousboe et al., 1997; Hassel and Brathe, 2000; Takeda et al., 2012). Models of epilepsy in rodents support a role for MeAIB/Gln transport via the Gln-Glu cycle to sustain seizure activity because Glu-induced epileptiform activity is reduced in neurons by application of MeAIB in vitro and in vivo (Bacci et al., 2002; Tani et al., 2007, 2010, 2014; Kanamori and Ross, 2013).

Here, we screened two riluzole based libraries (Sankaranarayanan et al., 2009; Shim et al., 2019) to identify compounds that possess the ability to inhibit spontaneous neural activity regulated, high affinity MeAIB/Gln transport in hippocampal neurons in vitro as potently as riluzole (Erickson, 2017), but are structurally unique. Based upon these screens, we have synthesized several novel chlorinated naphthalenyl substituted aminothiazoles that also potently inhibit activity regulated MeAIB/Gln transport activity at the plasma membrane in mature hippocampal neurons in vitro. The original compound identified in our first screen (SKA-41) and our most potent chlorinated derivatives in vitro (SKA-378, SKA-377 and SKA-379) exhibit excellent ‘drug-like’ properties with high oral availability and good brain penetration. We therefore tested riluzole and several of these novel aminothiazole derivatives in the KA-model of TLE.

The ability to prevent epileptic disease associated with TLE may depend on the ability of drugs to limit acute and subsequent chronic hippocampal neural injury following epileptogenic seizures. The neuroprotective effect of riluzole in the KA model of TLE has not previously been examined. We are the first to administer riluzole after SE in the literature; in this case, 1 h after the first KA-induced SE event. Our results demonstrate that riluzole, SKA-378 and SKA-379 are the most neuroprotective agents described when measured after 3 d in this acute model of KA neural injury. Riluzole and our novel aminothiazole derivatives of riluzole may be useful tools to further understand the mechanisms that prevent neural injury in the hippocampus and associated limbic regions following KA-induced SE that may lead to the progression of TLE. In addition, riluzole, SKA-378 and SKA-379 could have therapeutic value as AEDs for a potential new target.

2. Methods

2.1. Animals

We used Sprague-Dawley albino rats (Envigo, Indianapolis, IN or Charles Rivers, Wilmington, MA) for these studies. Primary hippocampal neuronal cultures were prepared from fetuses of both sexes on embryonic day 19 (E19) female rats. Adult (nine-to 11-week old, ~225 g) male rats were allowed to acclimatize to the new vivarium for 4–7 days and used for all pharmacokinetic, behavioral and in vivo experiments. All animal procedures were approved by the Institutional Animal Care and Use Committee at LSUHSC and UC Davis and complied with the NIH guidelines. All measures were taken to minimize pain, discomfort, or suffering of animals and were limited to that which is unavoidable in the context of scientifically sound research. Rats were housed 2 per cage in an environmentally controlled room (23 °C, 12-h light/12-h dark cycle) with food and water available ad libitum. Following KA administration rats were monitored daily by us and the Division of Animal Care staff.

2.2. Compounds

2.2.1. Library 1

The commercial sourcing or the synthesis of our first screening library consisting of compounds SKA-1 to SKA-56 was described previously (Sankaranarayanan et al., 2009). All compounds were stored as dry powder at RT and were checked for potential degradation by thin layer chromatography before shipping to New Orleans.

2.2.2. Library 2

The syntheses of the naphthalenyl substituted aminothiazoles SKA-190, SKA-193, SKA-198, SKA-219, SKA-220, SKA-230, SKA-232, SKA-247, SKA-251, SKA-255, SKA-257, SKA-260, SKA-265, and SKA-268 were described previously (Shim et al., 2019). SKA-77 (CAS no. 77815-14-6), SKA-88 (CAS no. 1826-16-0), SKA-89 (CAS no. 19968-59-3) and SKA-94 (CAS no. 28989-50.6) were purchased from Alfa Aesar. SKA-78 (CAS no. 2010-06-2) was purchased from Acros Organics. SKA-79 (CAS no. 2103-91-5), SKA-80 (CAS no. 2103-99-3), SKA-82 (CAS no. 30709-67-2) and SKA-232 (CAS no. 2834-79-9) were purchased from Sigma-Aldrich. SKA-255 (CAS no. 438233-93-3) was purchased from Oakwood Chemical. Scaled-up synthetic procedures for resynthesis or in-house synthesis of several compounds, which were used in larger quantities, are described in the Supplemental Data.

2.3. New chemical synthesis

The synthesis of several active novel chlorinated (SKA-377, SKA-378, SKA-379) or brominated (SKA-375) compounds, inactive/weakly-active (SKA-262, SKA-376) or fluorinated (SKA-382) compounds derived from SKA-75 and SKA-76 is described in detail in the Supplemental Data section.

2.4. Cell culture

Primary hippocampal neurons were prepared from embryonic day 19 Sprague Dawley rats of both sexes as described (Erickson, 2017; Brewer et al., 1993) with minor modifications. Pregnant dams were subjected to an overdose of compressed CO₂ administered in an anesthesia chamber and then decapitated using a guillotine. During dissection, the hippocampi were collected in Hanks Balanced Salt Solution (without Ca²⁺) containing 10 mM MgCl₂, 10 mM HEPES (pH 7.2), 0.2 mM Gln, penicillin (100 U/ml) and streptomycin (100 mg/ml) (pen/strep) at 4 °C. Hippocampi were then incubated in oxygenated Neurobasal medium (NB) without B27 (Invitrogen, Gaithersburg, MD) containing 0.25 mg/ml papain (Sigma, St. Louis, MO) and 0.25 mg/ml cysteine for 18 min at 37 °C with shaking. The tissue was transferred to a 15 ml conical tube containing DMEM plus 10% (v/v) fetal bovine serum (FBS) and allowed to settle. The medium was removed and the tissue was dissociated in same medium (~1 ml) by trituration with fire-polished Pasteur pipettes, the cells were counted, diluted in DMEM containing 10% (v/v) FBS, 0.5 mM glutamax, and 25 μM Glu, and then seeded in six-well dishes (~3 × 10⁵ cells/well) precoated with 25 μg/ml poly-d-lysine (>300,000 MW, Sigma). After 2 h at 37 °C the medium was removed and replaced with NB medium supplemented with B27, 1% (v/v) FBS, 0.5 mM glutamax, 25 μM Glu, and pen/strep. Neurons were fed by addition of 1 ml of NB medium supplemented with B27, 0.5 mM glutamax and pen/strep (without FBS) twice a week for 3 weeks. Cells were maintained at 37 °C in a humidified atmosphere containing 5% CO₂.

2.5. ¹⁴C-MeAIB transport studies in vitro

Uptake studies were performed using primary neuron-enriched hippocampal cultures at 19–21 days of maturation in vitro (DIV). Cells were first rinsed in buffer (2 ml) containing 125 mM NaCl, 4.8 mM KCl, 25 mM HEPES (pH 7.4), 1.2 mM KH₂PO₄, 2.5 mM MgSO₄, 5.6 mM glucose, and 1 mM EGTA (N buffer). The medium was then replaced with same buffer but without EGTA and containing 2.5 mM CaCl₂ (or with N buffer) and pre-incubated for 2 min ( ± inhibitors) at 37 °C. Spontaneous ¹⁴C-MeAIB transport measurements ( ± inhibitors) were then conducted over 15 min of uptake at 37 °C. For the initial drug screening assays, riluzole derivatives (20 μM, screen #1; 10 μM, screen #2) were directly mixed with ¹⁴C-MeAIB and added to the cells. For accurate determination of IC₅₀ values for these inhibitors, we preincubated the cells for 2 min in Ca²⁺-containing buffer with various concentrations of the SKA inhibitors and then replaced the medium with the same buffer (+inhibitor) containing ¹⁴C-MeAIB. Low concentrations of MeAIB (3.4–23.4 μM) were used in these experiments to enable a high-affinity transporter system to be more selectively studied. Total levels of ¹⁴C-MeAIB (56.4 mCi/mmol; PerkinElmer) were ~380,000 DPM/well. For substrate competition studies, various concentrations of unlabeled MeAIB (7.8 μM–250 μM) or Gln (31.25 μM–1000 μM) were mixed with ¹⁴C-MeAIB (3.4 μM). For kinetic analysis, we first measured the time-course of uptake to verify that 15 min uptake is within the linear range to record initial velocity and determine accurate K_m and V_max values. Following 15 min of spontaneous MeAIB uptake at 37 °C, the cultures were removed from the 37 °C incubator, placed on ice, and washed with ice-cold N buffer (2.5 ml). SDS (1%, 2 ml) was then added to solubilize the cells and the amount of radioactivity was determined by liquid scintillation counting (EcoScint, Nat. Diag.). Untreated cultures and cultures treated with TTX (1 μM), verapamil (25 μM), riluzole (10 μM) and uptake in the absence of Ca²⁺ ions, or at 4 °C, (i.e., background) were used to determine signal:noise to isolate activityregulated, high-affinity MeAIB/Gln transport activity. Background values were subtracted from total ¹⁴C-MeAIB uptake levels to determine levels of neural activity-regulated transport. The n values indicated represent data from independent neuronal cultures.

2.6. KA-induced SE and drug treatment

Rats were habituated for at least four days following their arrival and randomly assigned to groups (sham, KA-treated or drug-treated). We used two drug treatment protocols with rats receiving drug before KA administration or rats receiving drug after KA-induced SE. In the first drug treatment protocol, seizures were produced in rats that were injected intraperitoneally (i.p.) with KA (5 mg/kg/h x 2 h in saline; ~500 μL/injection). Drug-treated rats were injected with SKA-378 (30 mg/kg/h x 2 h in Caprylic/Capric triglyceride) beginning 30 min before KA administration. In the second drug treatment, rats were injected with KA (5 mg/kg/h) until SE was documented. Rats were then injected with 30 mg/kg of SKA-378, SKA-376, SKA-377 or SKA-379 or with 10 mg/kg of riluzole, SKA-41, SKA-378, SKA-377 or SKA-379 at 1 h and 4 h after rats reached SE and once per day for 2 d until perfusion on day 3. Seizure severity was assessed using the established five stages of seizure activity in the Racine Scale (Racine, 1972): (0) No behavioral change, (1) Mouth and facial movements (orofacial movements), (2) Head nodding (head myoclonus and/or severe orofacial movements), (3) Forelimb clonus, (4) Rearing and, (5) Rearing and falling with loss of postural control. Once SE is initiated (stage 5), SE events are observed recurrently for up to 4 h.

2.7. Fixation and tissue processing

Three days following KA administration the rats were deeply anesthetized using isoflurane and transcardially perfused with 0.9% saline for 5 min followed by 4% paraformaldehyde for 20 min. The brains were removed and post-fixed in paraformaldehyde overnight at 4 °C. The following day the brains were cut coronally anterior and posterior to the hippocampus and post fixed again in 4% paraformaldehyde overnight at 4 °C. The brains were then removed from the paraformaldehyde solution and transferred to 15% sucrose overnight at 4 °C and then followed by 30% sucrose overnight at 4 °C. Brains were then frozen in OTC mounting medium on dry ice and stored at −80 °C until use.

2.8. Fluoro-jade B (FJB) labeling

FJB labeling was performed in brain sections of rats saline perfused, paraformaldehyde-fixed, sucrose cryopreserved and frozen. Rat brains were sectioned at Bregma levels (−3.3 to −3.8) to obtain the entire medial hippocampus and other forebrain areas in this region including the cerebral cortex, pyriform cortex, entorhinal cortex, thalamus and amygdala. Brain sections were cut at 25 μm using a cryostat and then transferred to petri dishes filled with room temperature 1 x PBS until sectioning is completed and then mounted onto gelatin coated Fisher Superfrost Plus slides. Excessive liquid was removed, and the slides were airdried for 30 min at 37 °C and then incubated in 4% PAF vapor for 30 min at 37 °C. Slide mounted sections were then immersed in a solution of 1% sodium hydroxide in 80% ethanol for 5 min and then transferred to 70% ethanol for 2 min followed by deionized water for 1 min. Sections were then added to a solution of 0.06% potassium permanganate for 10 min on a rocker, rinsed in deionized water for 1 min then transferred to the FJB Working Solution (0.0004% solution dissolved in 0.1% acetic acid) containing 4’,6-diamidino-2-phenylindole (DAPI; 10 μg/ml) for 20 min. Following staining, slides were rinsed in deionized water (3 × 1 min), dried at 37 °C (~30 min), cleared in xylene (2 × 1 min) and cover-slipped with mounting medium DPX (Sigma-Aldrich).

2.9. NeuN immunohistochemistry (IHC)

Rat brains were sectioned at 20 μm using a cryostat and then transferred to petri dishes filled with room temperature 1 x PBS until sectioning is completed. Sections were then transferred to ice cold methanol for 10 min on ice and then to fresh 1 x PBS for rehydration. Sections were then transferred to an ice-cold sodium borohydride (0.1% in 1 x PBS) solution for 5 min followed by three rinses with 1 x PBS. After that, sections were transferred to 12-well plates and incubated for 1 h in blocking buffer containing 3% bovine serum albumin, 10% donkey serum and 0.25% Triton X in 1 x PBS at room temperature. The medium was then replaced with blocking buffer that contains the monoclonal primary antibody (1:1000) for NeuN (MAB377; Millipore) and incubated overnight at 4 °C. Sections were then washed (5 min intervals) with sequential washes (3 × 1 ml) of PBS-Tween (0.5%), 1 x TBS-Tween (0,5%), and 1 x PBS (3 × 1 ml) at room temperature. Sections were then incubated for 30 min in blocking buffer (diluted 3 x fold in PBS) at room temperature and the sections were then incubated in fresh buffer with a species-specific and highly cross-adsorbed secondary antibody (1:600) coupled to Alexa 594 (Molecular Probes) for 1 h at room temperature. The sections were washed again as described above, and then mounted onto Fisher Superfrost Plus slides and cover slipped using Prolong antifade mounting medium (Invitrogen).

2.10. Imaging/quantification

FJB and NeuN stained slides are imaged using a Leica (Nussloch, Germany) DMRXA automated upright inflorescent microscope, a Sensicam QE charge-coupled device digital camera (Cooke Corporation, Romulus, MI), and a filter set suitable for Alexa Fluorophores 488 and 594 (Molecular Probes). Images of sham, KA-treated and the drug-treated groups were obtained at the same exposure time for FJB (300 ms; green), NeuN (300 ms; red) or DAPI (3000 ms; blue). Quantification of neuronal cell damage by FJB staining and neural survival by NeuN fluorescence intensity was performed using Image J: FIJI. The images were uploaded to FIJI and then were split by color channels into separate images (green, blue, and red). The drawing tools selected fixed areas in the CA3, CA1, CA4 hippocampal areas and the DG itself for NeuN and then the fluorescent intensity of that area was quantitated and compared. The background fluorescent intensity was measured by selecting a similar fixed area free of labeled cells. The results were then transferred to Microsoft Excel and the Corrected Total Cell Fluorescence (CTCF) was calculated using the formula CTCF = Integrated Density (IntDen) - (Area of Selected Cell x Mean Fluorescence of background reading) for each hippocampal area. For the FJB + DAPI stained sections, DAPI is only used to visualize and locate hippocampal subregions. Quantification of FJB fluorescent intensity was performed on the pyramidal neuron cell body layers (CA1, CA3 and CA4) as indicated along with the entire hippocampus.

2.11. Statistical analysis

Statistical analysis was performed using GraphPad Prism software version 7.0 (GraphPad Software Inc). For initial drug screening of riluzole derivatives in vitro the differential potencies of the various compounds were analyzed by ANOVA. The most active compounds identified were used for detailed SAR relationship by analyzing their concentration-response and inhibitory potency (IC₅₀ values; ± SEM). All transport experiments were performed in at least 3 independent neuronal culture preparations and analyzed by repeated analysis of variance (ANOVA). Seizure severity was expressed as average Racine seizure score and assessed using ANOVA, with Tukey’s multiple comparison test to correct for multiple comparisons. Fluorescent intensity of NeuN and FJB staining in the CA1, CA3 and Hilar/CA4 regions are presented visually as representative sections. Mean fluorescent intensity of NeuN labeling and FJB staining in the CA pyramidal neuron subfields (CA1-4) and in the DG was quantitated. ANOVA was performed on these IHC data and Dunnett’s procedures to correct for multiple comparisons to analyze the results. A value of p < 0.05 was regarded as statistically significant.

2.12. Pharmacokinetic analysis

All compounds were dissolved in Miglyol 812 neutral oil (caprylic/capric triglyceride; Trade name Neobee M5, Spectrum Chemicals, Gardena, CA, USA) at 1 mg/ml. SKA-41 was gavaged at 10 mg (n = 3) and 30 mg/kg (n = 3). SKA-376, SKA-377 and SKA-378 were injected i.p. at 30 mg/kg (n = 3) or gavaged at 30 mg/kg (n = 3). At 30 min, 1 h, 2h, 4 h, 8h and 24 h after the gavage or i.p. administration, roughly 100–200 μL of blood was collected into EDTA blood sample collection tubes from a tail nick or the tail vein. Plasma was separated by centrifugation. After a washout period of 5 or 7 d the rats were reused, and compound administered again i.p. at 30 mg/kg (n = 6) and three rats sacrificed at 30 min and three rats sacrificed at 60 min to remove plasma and brain to obtain brain:plasma ratios. In another experiment SKA-41, SKA-376, SKA-377 and SKA-378 at 30 mg/kg were administered 1 h after KA-induced SE and plasma and forebrain tissue samples prepared at 4 h after drug administration (n = 3 per compound). For comparison brain and plasma samples were also obtained from rats not treated with KA at 4 h after compound administration. All brain and plasma samples were stored at −80 °C pending analysis. Sample preparation and LC/MS analysis are described in detail in the Supplemental Chemistry section. Plasma protein binding was determined with rat plasma using rapid equilibrium devices (RED, Fisher Scientific) according to the manufacture’s protocol.

2.13. Sodium channel electrophysiology

NIE neuroblastoma cells that endogenously express the Na_V1.2 subtype at high levels and HEK cells transiently transfected with Na_V1.6 were voltage clamped by whole-cell patch-clamp using an EPC-10 amplifier and Pulse software (HEKA, Lambrecht/Pfalz, Germany). All experiments were conducted at room temperature (22–24 °C). Cells were bathed in extracellular solution containing (in mM): 160 NaCl, 4.5 KCl, 1 MgCl₂, 2 CaCl₂ and 10 HEPES; pH was adjusted to 7.4 using NaOH (310 mOsm). Pipettes were pulled from 1.5 mm capillary tubing and filled with intracellular solution containing (in mM) either: 145 KF, 2 MgCl₂, 10 EGTA and 10 HEPES (pH adjusted to 7.2 with KOH; 300 mOsm) or 145 CsF, 3 KCl, 2 NaCl, 1 MgCl₂, 3 Na-ATP, 0.2 Na-GTP, 10 EGTA and 10 HEPES (pH adjusted to 7.4 with KOH; 303 mOsm). Pipette tip resistances were 2–4 MΩ. Series resistances of 3–10 MΩ were compensated 40–80%. Cells were voltage clamped to a holding potential of −80 mV. Na⁺ currents were elicited by 50-ms voltage steps from −90 mV to 0 mV applied every 10 s. The sampling frequency was 5 kHz. IC₅₀ values for inhibition of Na_V1.2 and Na_V1.6 sodium channel function by riluzole and our active in vitro compounds (SKA-41, SKA-75, SKA-376, SKA-377, and SKA-378) were fitted with Origin 9.0 (OriginLab, Northampton, MA). IGOR-Pro (Wavemetrics, Lake Oswego, OR) was used for current plotting.

3. Results

3.1. Development of novel aminothiazole riluzole derivatives that inhibit neural activity regulated, high affinity MeAIB transport in mature hippocampal neurons in vitro

A neural activity regulated MeAIB transport activity in mature neuron enriched hippocampal cultures that is potently inhibited by riluzole with ~1 μM affinity has recently been described (Erickson, 2017). Here, we report that the total level of spontaneous MeAIB transport after 15 min in Ca²⁺-containing (2.5 mM) Krebs-based medium into mature rat hippocampal neurons (DIV 19–21) at 23.4 μM external ¹⁴C-MeAIB concentration was 619±21 pmol/well and was inhibited in the presence of the Na⁺-channel blocker TTX (1 μM; 150 ± 15 pmol/well), the Ca²⁺ channel blocker verapamil (25 μM; 119 ± 9 pmol/well), riluzole (10 μM; 133 ± 8 pmol/well) or by the absence of extracellular Ca²⁺ (152 ± 8 pmol/well). Similar inhibition of transport by TTX, verapamil, riluzole, or the absence of external Ca²⁺ ions, was observed using 5.4 μM ¹⁴C-MeAIB concentration (Fig. 1A). Thus, ~80% of the total uptake of MeAIB observed here at either 5.4 or 23.4 μM external ¹⁴C-MeAIB concentration is eliminated by these inhibitors of presynaptic neurotransmission. Transport activity observed at 4 °C was minimal (DPM values ~600) compared to values for total ¹⁴C-MeAIB transport activity (DPM values ~14,000) at 5.4 or 23.4 μM and was subtracted from total uptake levels. Ca²⁺-dependent MeAIB transport is defined here as total uptake minus uptake in the absence of Ca²⁺ ion. Ca²⁺-independent transport has 4 °C DPM values subtracted as background. Transport experiments involving selective Na⁺ channel inhibition includes TTX (1 μM) as background and was subtracted. The % inhibition of activity regulated transport for TTX (1 μM), verapamil (25 μM) or in the absence of Ca²⁺ (−Ca²⁺) were not different from each other and thus, any of these blockers of activity regulated transport can serve as negative controls (i.e., background) to subtract from total uptake levels to isolate, characterize and define this neural activity regulated MeAIB/Gln transport system in mature hippocampal neurons in vitro.

Fig. 1.

Spontaneous high affinity MeAIB/Gln transport activity is regulated by neural activity in mature (DIV 19–21) neuron-enriched rat hippocampal networks in vitro. A, ¹⁴C-MeAIB transport (5.4 μM) is inhibited by tetrodotoxin (TTX; 1 μM), verapamil (Ver; 25 μM), omission of Ca²⁺ ion, or by riluzole (Ril; 10 μM). Transport experiments were conducted over a 15 min period and background levels determined at 4 °C were subtracted. B, Time-courses of Ca²⁺-dependent and Ca²⁺-independent ¹⁴C-MeAIB uptake. Ca²⁺-dependent values are total ¹⁴C-MeAIB (5.4 μM) uptake minus uptake in the absence of Ca²⁺. Ca²⁺ independent values have 4 °C values subtracted. C, Michaelis-Menten kinetic analysis of Ca²⁺-dependent and Ca²⁺-independent transport activity reveal high and low affinity systems. Background values (−Ca²⁺ or 4 °C) at each MeAIB concentration were subtracted. D, The relative high affinity of Gln for spontaneous Ca²⁺-dependent transport was determined by competition analysis (IC₅₀) with ¹⁴C-MeAIB (5.4 μM). Background is uptake in absence of Ca²⁺ and was subtracted from all values. All final determinations above were from 3 to 5 independent neuronal cultures and are presented as mean values±SEM with critical significance level of p < 0.05. E, Model of the Glu/Gln cycle between neurons and astrocytes that recycles Glu for neurotransmission indicates where the neuronal activity regulated MeAIB/Gln transport system that we describe here (blue transporter) may support the concept of a Ca²⁺-dependent pathway for cycling of this high affinity transporter by neural activity in synaptic terminals (modified from (Erickson, 2017)).

Here, we also report the kinetic analysis of spontaneous, activity regulated MeAIB transport in mature hippocampal neuron enriched cultures (DIV 19–21). We have previously reported a K_m value of ~40 μM of MeAIB for a brief (5 min) high KCl (60 mM) stimulated MeAIB transport activity in mature neurons in vitro using a Lineweaver-Burke analysis (Erickson, 2017). First, we performed a time-course of spontaneous Ca²⁺-dependent and Ca²⁺-independent MeAIB transport (5.4 μM) activity in mature hippocampal neural networks in vitro (DIV~21) to demonstrate that a 15-min uptake period is within the linear range of transport activity, which is required for accurate initial velocity measurements (Fig. 1B). We determine that the K_m for spontaneous Ca²⁺-dependent MeAIB transport is 61 ± 5 μM (n = 3) using Michaelis-Menten analysis (Fig. 1C). We also determined the kinetics of Ca²⁺-independent MeAIB transport under these conditions and find the K_m for MeAIB transport is ~8 times higher (474 ± 31 μM n = 3) likely representing uptake by low affinity MeAIB transporters such as SNAT1 and/or SNAT2 that are expressed in cell bodies and proximal dendrites of neurons in the brain (Mackenzie et al., 2003; Weiss et al., 2003; Melone et al., 2004, 2006). The maximal velocity (V_max) at 15 min of uptake for Ca²⁺-dependent and Ca²⁺-independent MeAIB transport is 1605 ± 139 and 1793 ± 131 pmol/well, respectively indicating that the maximal capacity of these distinct MeAIB transport systems are similar in these mature hippocampal neuronal cultures. To determine the affinity of Gln for spontaneous, neural activity regulated high-affinity MeAIB transport we examined the ability of unlabeled Gln to compete with MeAIB for this Ca²⁺-dependent activity (Fig. 1C). Spontaneous ¹⁴C-MeAIB (5.4 μM) transport is inhibited by unlabeled MeAIB with relatively high affinity (IC₅₀ 50 ± 6 μM; n = 6) in this assay, as expected. Gln also competes with ¹⁴C-MeAIB but at lower potency (IC₅₀ 162 ± 17 μM; n = 9), which is still within the range of reported values of Gln in brain extracellular space (Kanamori and Ross, 2004, 2011, 2013). These results further establish that a spontaneous, neural activity regulated, high affinity MeAIB/Gln transporter exists in mature hippocampal neuronal networks in vitro and that its transport activity is expressed on the plasma membrane in a neural activity regulated and Ca²⁺-dependent manner (Fig. 1D).

In this study, we used our assay of spontaneous, activity regulated high affinity MeAIB/Gln transport in mature hippocampal neurons as a screen to identify novel riluzole derivatives that inhibit this activity with similar affinity as riluzole. We first confirmed that riluzole inhibits spontaneous, neural activity regulated, high affinity MeAIB transport with micromolar affinity (IC₅₀ = 860 ± 20 nM), as previously reported (Erickson, 2017). We started by screening a library of riluzole derivatives (Library 1, Fig. 2A), which was originally developed to find activators of Ca²⁺-activated K⁺ channels (Sankaranarayanan et al., 2009). Compounds were added together with ¹⁴C-MeAIB (23.4 μM) in this assay and spontaneous transport activity was measured after 15 min in Ca²⁺-containing (2.5 mM) Krebs-based medium. This screen of Library 1 revealed 8 inhibitors that blocked high affinity MeAIB transport activity by more than 90% when tested at a concentration of 20 μM (Fig. 2B). Most of these compounds (SKA-3, SKA-7, SKA-11, SKA-19, SKA-32, SKA-45 and SKA-51) were discarded because of their close structural similarity to riluzole. SKA-41 was the only unique structure among these compounds being the only aminothiazole tested here as compared to the rest, which are all benzothiazoles. A search for reported biological activity for SKA-41 in the SciFinder Chemical Abstract System only revealed the previously published finding that SKA-41 is inactive on Ca²⁺-activated K⁺ channels (Sankaranarayanan et al., 2009). A preliminary study of SKA-41 pharmacokinetic properties revealed that the compound exhibits excellent brain penetration and oral availability (see below).

Fig. 2.

SKA-41 is an aminothiazole derivative of riluzole that inhibits spontaneous, neural activity regulated and high affinity MeAIB/Gln transport. A, We used a library of riluzole derivatives to screen in a first round for compounds that retain the ability to inhibit neural activity regulated high affinity MeAIB/Gln transport, yet may be structurally unique. All compounds (20 μM) were added together with ¹⁴C-MeAIB (23.4 μM). Most active structures presented are conventional benzothiazoles derived from riluzole. However, SKA-41 is structurally distinguished being an “open” 4-phenyl-aminothiazole. B, ¹⁴C-MeAIB transport is presented as a % of control neural activity regulated values. Red highlights reveal the active compounds in this first screen (n = 3 ± SEM). All values are results from independent neuronal cultures with critical significance level of p < 0.05.

We then tested a second compound library (Library 2, Fig. 3A) containing 27 derivatives of SKA-41 that were either purchased or synthesized previously (Shim et al., 2019) for their ability to block spontaneous, neural activity regulated MeAIB transport at 10 μM. Eight compounds, which all contained a larger, more lipophilic naphthalene substituent instead of the phenyl substituent of SKA-41, were identified as active (Fig. 3B). Indeed, all of these active compounds identified here are chemically derived from the two regioisomers SKA-75 or SKA-76 that exhibited intermediate potency of ~50% block at 10 μM and both have unsubstituted naphthalene rings (Fig. 3A). SKA-219, a brominated derivative of SKA-75, was initially a favorite because it inhibited activity regulated MeAIB transport in mature hippocampal neurons by >95% at 10 μM. The other active compounds we identified in this second screen were the SKA-75 derivatives SKA-190, SKA-193, and SKA-247 and the SKA-76 derivatives SKA-198 and SKA-220 (Fig. 3, Table 1).

Fig. 3.

Screening of SKA-41 related compounds that inhibit spontaneous, activity regulated MeAIB/Gln transport reveals common structures. A, SKA-41 and related 4-phenyl- and naphthalenyl substituted aminothiazoles were screened in a 2nd round to evaluate their relative inhibitory potency at 10 μM concentration to inhibit ¹⁴C-MeAIB (5.4 μM) uptake and identify compounds that retain the ability to inhibit high affinity MeAIB/Gln transport like SKA-41, and are structurally related. Active compounds are mostly derived from SKA-75 (SKA-190, SKA-193, SKA-219, SKA-247) or from the regioisomer SKA-76 (SKA-198, SKA-220). SKA-219 is the most potent inhibitor evaluated in this 2nd screen. B, ¹⁴C-MeAIB transport is presented as a % of control neural activity regulated values. Red highlights reveal the active compounds in this second screen (n = 3 ± SEM). All values are results from independent neuronal cultures with critical significance level of p < 0.05.

Table 1.

IC₅₀ values of all active compounds tested for inhibition of spontaneous MeAIB transport activity, logP and % plasma protein binding.

Compounds	Ca2+-regulated MeAIB transport	logP	% plasma protein binding
Riluzole	0.86 ± 0.02 (5) (μM)	3.95	98 ± 0.5
SKA-41	5.4 ± 0.2 (4)	4.58	95.7 ± 0.4
SKA-219	1.6 ± 0.23 (4)	4.76	99.5
SKA-378	1.1 ± 0.17 (4)	4.2	99.2 ± 0.6
SKA-75	4.5 ± 0.45 (4)	4.05	ND
SKA-190	2.6 ± 0.2 (4)	4.53	ND
SKA-376	4.3 ± 0.4 (4)	4.53	98.5 ± 0.5
SKA-377	1.7 ± 0.26 (4)	4.91	99 ± 0.7
SKA-379	6.5 ± 0.25 (5)	4.42	ND
SKA-382	12.2 ± 1.7 (4)	4.14	ND
SKA-76	6.2 ± 0.9 (5)	4.05	ND
SKA-220	4.3 ± 0.4 (5)	4.42	ND
SKA-375	1.2 ± 0.1 (3)	5.25	ND
TTX	1.7 ± 0.2 nM (5)	–	–
4,9-anhydroTTX	61 ± 4 nM (3)	–	–

While SKA-219, the brominated SKA-75 derivative, was attractive because of its potency for inhibiting spontaneous, neural activity regulated MeAIB transport (Fig. 4A) we had concerns about its structure. Heavy halogens like bromine and iodine in drugs have been associated with phototoxicity, photoreactivity, reactive metabolite formation, etc. that can lead to safety issues (Stepan et al., 2011). Brominated compounds are also typically poorly soluble because of their high logP values, which makes formulation and drug administration challenging. We therefore synthesized the chlorine substituted SKA-376, SKA-377, SKA-378 and SKA-379 compounds that were obtained by reacting active compounds SKA-41, SKA-190, SKA-75 and SKA-76, respectively, with 2-chloro-1,3-bis(methoxycarbonyl)guanidine (Palau’Chlor^®), which is a highly reactive chlorinating agent that is ideal for introducing chlorine substituents in predictable locations on heterocycles (Rodriguez et al., 2014) (Supplemental data Scheme A). We found that SKA-378 inhibited spontaneous neural activity regulated MeAIB transport in vitro as potently as riluzole, SKA-219, and SKA-375, the brominated derivative of SKA-190 (Fig. 4).

Fig. 4.

Novel chlorinated aminothiazole entities derived from SKA-75 inhibit spontaneous, activity-regulated and high-affinity MeAIB/Gln transport. A, Screening strategy: Screen 1 of riluzole derivatives identified SKA-41 as an active aminothiazole compound. Screen 2 identified SKA-75 as an aminothiazole compound related to SKA-41 and was used for synthesis of most active agents described in vitro and in vivo here. B, Concentration response curves reveal that SKA-378 (red) and SKA-377 (green) similarly inhibit spontaneous, high affinity ¹⁴C-MeAIB transport activity like riluzole (blue). The curves are shifted to the right for SKA-376 (purple), which is indistinguishable from SKA-75 (orange); yet all are still in the low micromolar range, n = 3–5 ± SEM. All values are results from independent neuronal cultures with critical significance level of p < 0.05.

We report the concentration response relationship of all 12 of these active naphthalenyl substituted aminothiazoles derived from SKA-75 or SKA-76 that we have identified to block spontaneous activity-regulated ¹⁴C-MeAIB transport activity in mature hippocampal neurons in vitro, compared to riluzole (Table 1). We find that SKA-378 was as potent as riluzole to inhibit spontaneous high affinity MeAIB transport activity (IC₅₀ 1.1 ± 0.17 μM). SKA-377, the chlorinated derivative of SKA-190 exhibited potency similarly to SKA-378, (IC₅₀ 1.7 ± 0.26 μM). SKA-376, the chlorinated derivate of SKA-41 and SKA-379, the chlorinated derivative of SKA-76 were less potent (IC₅₀ ~4–6 μM) to inhibit spontaneous high affinity ¹⁴C-MeAIB transport activity similar to SKA-75 and SKA-76 (IC₅₀ ~5–6 μM) (Fig. 4, Table 1).

3.2. SKA-378 prevents neural injury in the hippocampus when given before KA-induced SE

The KA model of TLE in rats can be generated by i.p. injection of KA, the presynaptic glutamategic excitotoxin that produces stage 5 grade seizure activity (i.e., SE) based on the Racine scale (Racine, 1972). We confirm that KA-induced SE results in acute hippocampal neural injury by 3 d (Fig. 5). We used FJB staining in brain sections in situ to evaluate levels of KA-induced degeneration in the hippocampal pyramidal neuron layer subfields (Schmued and Hopkins, 2000). In sham rats, FJB labeling is not observed in the hippocampus. In KA treated rats, FJB fluorescence intensity is specifically increased in neurons of the CA3, CA1 and CA4 (within the DG hilus) subfields in the hippocampus after 3 d (Fig. 5A,B,C) indicating neural injury. High-power images of FJB labeling in these hippocampal subfields reveal specific cell body labeling (Fig. 5D). FJB fluorescence in degenerating neurons confirm that KA-induced injury includes neurons in the vulnerable CA subfields of the hippocampus in our hands.

Fig. 5.

KA-induced SE results in acute hippocampal neural injury at 3 d. A. Respective images of FJB labeling in neuronal cell bodies and processes in the rat hippocampus following KA-induced SE. Boxes correspond to the areas of quantification for all rats. Scale bar = 400 μM. B. Same image as above but background fluorescence and staining in processes was subtracted. C. Green color channel was converted to white for quantification. D. High power (40x) images for CA3, CA1 and CA4/hilar region. Scale bar = 40 μm.

SKA-378 was one of the most potent compounds that inhibit spontaneous, neural activity regulated MeAIB/Gln transport, as effectively as riluzole, and preliminary pharmacokinetic analysis revealed that SKA-378, like SKA-41, exhibited excellent bioavailability and brain penetrance (see below). Thus, SKA-378 was chosen to be examined first for neuroprotection studies in vivo. We examined if SKA-378 could prevent hippocampal neural injury produced by KA (5 mg/kg/h x 2 h). Here, SKA-378 (30 mg/kg) was given twice; 30 min before the first dose of KA (5 mg/kg) and 30 min after the second dose of KA (5 mg/kg). We find that such pretreatment with SKA-378 was only moderately effective to reduce seizure activity as >50% of the rats treated with SKA-378 still exhibited maximal stage 5 seizures on the Racine scale (i.e., SE), while the other SKA-378 pretreated rats reached at least stage 2–3 seizures that included wet dog shakes, staring, etc. Pretreatment of rats with SKA-378 did not show a significant reduction in seizure severity compared to non-treated KA-injected rats (Tukeys analysis; p = 0.5686) (Fig. 6D). Nevertheless, we examined neural injury in the hippocampus 3 d after KA administration in sham, KA treated, and SKA-378 pretreated rats. We report the results from sections containing the medial hippocampal region (Bregma −3.3 to −3.8). We find that SKA-378 pretreatment blocks neuronal hippocampal injury induced by KA as indicated by the absence of FJB staining in all of the CA pyramidal neuron cell layers (CA3: F (2,18) = 28.62; p < 0.0001; CA1: F(2,18) = 9.417; p = 0.0016; CA4: F (2,18) = 17.19; p < 0.0001) (Fig. 6A). The neuroprotection afforded by SKA-378 pretreatment in the hippocampus was also assessed by NeuN immunofluorescence. NeuN is a selective marker of neuronal cell bodies in the brain that delineate the hippocampal subfields including the CA and DG neuronal cell body architecture in situ and so its expression level reflects neuronal viability (Mullen et al., 1992). In sham rats, NeuN staining reveals intact neuronal architecture in all hippocampal subfields (Fig. 6C). NeuN expression in CA3, CA1, and CA4 subfields is markedly lower in KA-treated rats as measured by greatly decreased fluorescence intensity by IHC (Fig. 6C). DG neurons are not vulnerable to KA toxicity which is consistent with the absence of FJB labelling in these neurons (Fig. 6A) and NeuN fluorescent labeling is intact and not different from sham or SKA-378 pretreated rats (Dunnett’s analysis: F(2, 18) = 0.1387; p = 0.8715) following KA treatment (Fig. 6C). FJB fluorescence in degenerating neurons and NeuN labeling of surviving neurons confirm that KA-induced injury includes neurons in the vulnerable CA subfields of the hippocampus. NeuN labeling in SKA-378 treated rats showed clear pyramidal neuron delineation of CA4 (F(2,18) = 15.55; p < 0.0001) (DG hilar region), CA3 (F(2,18) = 27.18; p < 0.0001) and CA1 (F(2,18) = 20.59; p < 0.0001) subfields with fluorescent intensity in these areas nearly identical to sham levels (Fig. 6C). Quantification of the fluorescent intensity of FJB and NeuN labeling in the entire hippocampus and in the CA3, CA1 and CA4/hilar region are shown in Fig. 6B (FJB) and Fig. 6D (NeuN) and reveal that SKA-378 is a neuroprotective agent against acute KA-induced neural injury.

Fig. 6.

SKA-378 prevents acute hippocampal injury in the KA model of TLE. FJB and NeuN labeling was measured in sham (n = 7), KA treated rats (n = 8), and KA treated rats (n = 6) that were given SKA-378 (30 mg/kg, i.p.) 30 min prior KA injections., A, Representative examples of FJB labeling are shown as black and white images. Treatment of rats with SKA-378 largely prevents FJB labeling in the hippocampus. B, Quantification of the FJB results reveal that fluorescence intensity in KA-treated rats pretreated with SKA-378 was minimal in most hippocampal subfields and in the entire hippocampus. C, NeuN immunostaining fluorescence was also measured in sham, KA treated rats, and KA treated rats pretreated with SKA-378. D. Quantification of the NeuN immunofluorescence intensity reveal that SKA-378 is neuroprotective in the CA1, CA3 and CA4 hippocampal subfields, n = 6–8 ± SEM. Repeated measures analysis of variance (ANOVA) followed by Dunnett’s procedures to correct for multiple comparisons were used to analyze the results. A value of p < 0.05 is regarded as statistically significantly different than values from KA-treated rats, with asterisks corresponding to different levels of significance (**p < 0.01, ***p < 0.001, ****p < 0.0001).

3.3. Riluzole and the novel chlorinated aminothiazoles SKA-377, SKA-378, and SKA-379 attenuate neural injury in the hippocampus when given after KA-induced SE

While SKA-378 could not prevent KA-induced SE in all animals, all SKA-378 pretreated animals were still protected from KA-induced hippocampal neural injury at 3 d. We therefore examined if SKA-378 could prevent neural injury if administered after KA-induced SE occurs (Fig. 7), which is more clinically relevant. Rats were treated with KA (5 mg/kg/h) and 1 h after SE was documented (Racine 5 stage) they were treated with SKA-378 (30 mg/kg), then again with SKA-378 4 h after SE, and then once a day for the next two days by i.p. administration. The duration of KA-induced SE events was between 4 and 6 h after the onset SE and was not affected by SKA-378. We examined neural injury by FJB staining and NeuN immunoreactivity by fluorescence microscopy 3 d post-KA treatment in the medial hippocampal sections as described above. Representative images of FJB staining in sham, untreated (KA) and post-SE drug treated (30 mgkg) rats (SKA-378, SKA-377, SKA-379 and SKA-376) are shown in Fig. 7A. In post-SE SKA-378 treated rats, FJB staining in hippocampal CA1-4 neurons was minimal (CA3: F(5,42) = 16.51; p < 0.0001; CA1: F(5,42) = 3.807; p = 0.0062; CA4: F(5,42) = 8.167; p < 0.0001) and similar to that observed in SKA-378 pre-treated rats (Fig. 7B). SKA-378 treatment after SE in this way also largely prevents the diminished NeuN fluorescence normally caused by KA in all pyramidal neuron subfields of the hippocampus (CA3: F(5,42) = 12.27; p < 0.0001; CA1: F(5,42) = 5.565; p = 0.0005; CA4: F(5,42) = 6.586; p = 0.0001) similarly to SKA-378 pretreated rats (Fig. 7C) further demonstrating the acute neuroprotection provided by SKA-378. SKA-376 did not significantly reduce neural injury in the hippocampal subfields as measured by both FJB staining and by NeuN fluorescence (Fig. 7B and C). On the other hand, SKA-377 and SKA-379 were neuroprotective against KA induced hippocampal neural injury when given post-SE, similar to SKA-378 (F(5,42) = 11.28; p < 0.0001) (Fig. 7B and C). We next examined the neuroprotection afforded by these novel compounds at 10 mg/kg, which is the average dose of riluzole that is generally used for neuroprotection studies (Fig. 8). Higher doses of riluzole (e.g., 30 mg/kg) render rats immobile and unresponsive, unlike SKA-378. Representative images of FJB staining in untreated (KA) and post-SE drug treated (10 mgkg) rats (SKA-378, SKA-377, SKA-379, SKA-41, and riluzole) are shown in Fig. 8A. We report that SKA-379, SKA-378, SKA-377 and SKA-41 administered at 10 mg/kg are neuro-protective, but not as protective as riluzole (F6,44) = 8.159a; p < 0.0001) (Fig. 8B) which prevents acute neural injury altogether. Riluzole also shortened the duration of KA-induced SE events to less than 2 h and the rats then appeared to be relatively immobile/sedated. A reduced level of neuroprotection observed at 10 mg/kg compared to 30 mg/kg of our most active aminothiazole compounds demonstrate a dose dependence, which is important to identify optimal dosage for acute administration.

Fig. 7.

SKA-378 prevents neural injury when administered at 30 mg/kg after KA-induced status epilepticus (SE). Rats were treated with KA (5 mg/kg/h) until SE was observed and then 1 h after animals exhibited the first SE event they were injected with SKA-378 (n = 9), SKA-377 (n = 7), SKA-379 (n = 8), or SKA-376 (n = 6) at 30 mg/kg and then again 4 h after SE initiation followed by i.p. injections once/day for the next 2 d. Image analysis was conducted at 3 d. A, Representative examples of FJB labeling are shown as black and white images. Post-treatment of rats with SKA-378 largely prevents FJB labeling in the hippocampus. B. Quantification of FJB staining in the entire hippocampus, and in the CA3, CA1 and CA4/Hilus subfields are presented. C. Representative examples of NeuN labeling. D. Quantification of NeuN immunofluorescence in the DG, and in the CA3, CA1 and CA4/Hilus subfields are shown. A value of p < 0.05 is regarded as statistically significantly different than values from KA-treated rats, with asterisks corresponding to different levels of significance (**p < 0.01, ***p < 0.001, ****p < 0.0001).

Fig. 8.

Riluzole prevents and SKA-378 attenuates neural injury when administered at 10 mg/kg after KA-induced status epilepticus (SE). Rats were treated with KA (5 mg/kg/h) until SE was observed and then 1 h after animals exhibited the first SE event they were injected with SKA-378 (n = 6), SKA-377 (n = 6), SKA-379 (n = 6), SKA-41 (n = 7) or riluzole (n = 8) at 10 mg/kg and then again 4 h after SE followed by i.p. injections once/day for the next 2 d. Image analysis was conducted at 3 d. A, Representative examples are shown as black and white images. B. Quantification of FJB staining in the entire hippocampus, and in the CA3, CA1 and CA4/Hilus subfields are presented. A value of p < 0.05 is regarded as statistically significantly different than values from KA-treated rats, with asterisks corresponding to different levels of significance (**p < 0.01, ***p < 0.001, ****p < 0.0001).

3.4. SKA-378 reduces neural injury in associated hippocampal limbic areas caused by KA-induced SE

We used FJB staining in situ to evaluate levels of KA-induced degeneration in hippocampal interconnected limbic regions. We observed extensive FJB labeling following KA-induced SE in neurons of the piriform and entorhinal cortex and in neurons of the midline thalamus and basolateral amygdala areas (Supplemental data), as reported previously (Hopkins et al., 2000; Srivastava et al., 2020). Pretreatment of rats with SKA-378 as described above greatly reduced FJB staining in these hippocampal limbic areas (Supplemental data). Riluzole (post-treated) largely prevented neural injury in these limbic areas (not shown).

3.5. Pharmacokinetics and plasma:brain ratios

The plasma time courses and plasma:brain ratios for SKA-41, SKA-378, SKA-377, and SKA-376 are shown in Fig. 9. The plasma concentration profiles of all tested compounds reveal excellent bioavailability for these compounds, with oral and i.p. application of 30 mg/kg achieving nearly identical exposure levels (= areas under the plasma curve) and plasma peak concentrations of 8–12 μM. All of these compounds were also highly brain penetrant as demonstrated by the total brain concentrations (>10 μM) that equaled or exceeded total plasma concentrations at 30 min, 1 h and 4 h after i.p. application. In keeping with their lipophilicity, SKA-376, SKA-377 and SKA-378 are all highly plasma protein bound with free concentrations ranging between 0.5 and 1% (see Table 1). In contrast to riluzole, which has a brain:plasma ratio of 1 (Coleman et al., 2015; Colovic et al., 2004) the original compound identified here, SKA-41, reached 7-fold higher concentrations in the brain than in the plasma (Fig. 9B). SKA-377 and SKA-378 were also found to have excellent oral availability when administered by gavage at 10 and 30 mg/kg (Fig. 9C,E).

Fig. 9.

Plasma time-course and brain/plasma levels of SKA-41, SKA-376, SKA-377, SKA-378 and SKA-379. Effect of KA-induced SE. A, Total plasma concentrations of SKA-41 following oral administration at 10 and 30 mg/kg. B, Matched brain and plasma concentrations of SKA-41 at 1 h after i.p. administration of 10 mg/kg and at 4 h after i.p. administration of 30 mg/kg after KA. C, E and G, Total plasma concentrations of SKA-378, SKA-377 and SKA-376 following oral and i.p. administration at 30 mg/kg. D, E and F, Matched brain and plasma concentrations at 30 min, 1 h and 4 h after i.p. administration of SKA-378, SKA-377 and SKA-376 at 30 mg/kg. The grey shaded points show brain and plasma levels 5 h after KA administration. All data are total concentrations ± SD, n = 3 rats per time point.

To assess whether KA-induced seizure activity affects the brain penetration of these compounds we examined the brain:plasma ratios in rats 4 h after they exhibited KA-induced SE (Fig. 9). Rats were treated with SKA-41, SKA-376, SKA-377 and SKA-378 (30 mg/kg) once 1 h after SE and then harvested for plasma and brain after 4 h. This paradigm is similar to what we used for the in vivo neuroprotection studies post-SE KA treatment described above; although at 4 h post-SE in neural protection studies above we injected a second dose of compounds. Total brain levels of all four compounds at 4 h post SE (see grey shading in Fig. 9B,D,E,H) were somewhat higher than at 4 h after a single administration, but the presence of KA, which presumably opens up the blood brain barrier (Zucker et al., 1983; Pont et al., 1995), did not change the brain:plasma ratios of any of these lipophilic compounds.

3.6. On the mechanism of inhibition of spontaneous, activity-regulated high-affinity MeAIB transport by SKA-378 and riluzole

In this study, we also seek to identify potential mechanisms involved in the inhibition of Ca²⁺-regulated MeAIB/Gln transport activity in hippocampal neurons in vitro by riluzole and our novel aminothiazole derivatives of riluzole that may relate to their acute neuroprotective properties in vivo. For this, we first examined whether riluzole and SKA-378 are competitive or non-competitive inhibitors of spontaneous, neural activity regulated high affinity MeAIB transport activity in mature hippocampal neuronal circuits in vitro. We performed Michaelis-Menten kinetic analysis of SKA-378 and riluzole’s inhibitory action on Ca²⁺-regulated MeAIB transport activity in vitro and find that inhibition is noncompetitive because both riluzole and SKA-378 cannot compete at the substrate recognition site on the transporter (Fig. 10A). Eadie-Hofstee analysis indicates the K_m values for MeAIB in control (61 ± 5 μM), +SKA-378 treated (58 ± 6 μM) and +riluzole treated (63 ± 6 μM) are not different (n = 3). While SKA-378, and riluzole, at 1 μM concentration do not affect the K_m of spontaneous, activity-regulated ¹⁴C-MeAIB transport activity these compounds reduce the maximal velocity (V_max) of transport. The V_max values for SKA-378 (73 ± 6 pmol/min) and riluzole (54 ± 1 pmol/min) are roughly 50% less than observed for control Ca²⁺-regulated MeAIB transport values (107 ± 9 pmol/min) (Fig. 10B). These kinetic results indicate that riluzole and our novel aminothiazoles (e.g., SKA-378) act via indirect mechanisms to inhibit spontaneous, neural activity regulated high affinity MeAIB/Gln transport activity in mature hippocampal neuron-enriched cultures. Our results support the concept that neural activity regulated, Ca²⁺-dependent cycling of the transporter involved here, from intrasynaptic stores to the plasma membrane, exists in mature hippocampal neuronal cultures that could be targeted for neuroprotection (see Fig. 1D).

Fig. 10.

Non-competitive inhibition by riluzole and SKA-378 of spontaneous, neural activity regulated high affinity MeAIB/Gln transport activity in mature hippocampal neurons in vitro. A, Michaelis-Menten kinetic analysis of MeAIB uptake in the presence of riluzole or SKA-378 (1 μM). Background uptake was measured in the absence of Ca²⁺ ion at each MeAIB concentration and was subtracted. B, Eadie-Hofstee scatchard analysis shows that the affinity of MeAIB (K_m, -slope) is not affected by riluzole or SKA-378 while the initial velocity (V, V_max) of transport is reduced by ~50% by both riluzole and SKA-378. C, Inhibition of transport by the selective Na_V1.6 inhibitor 4,9-anhydroTTX. Background uptake was measured in the presence of 1 μM TTX and was subtracted. All values are results from independent neuronal cultures with critical significance level of p < 0.05.

A principal target of riluzole for neuroprotection is believed to be the inhibition of Na⁺ channel activity in neurons and we have previously reported that TTX (1 μM) eliminates spontaneous neural activity regulated MeAIB transport activity (Erickson, 2017), and that is also confirmed here (Fig. 1A). Therefore, we examined the inhibitory potencies of TTX itself and a derivative of TTX reported to be more selective for the Na_V1.6 isoform (Teramoto and Yotsu-Yamashita, 2015), which is enriched in mature forebrain excitatory neurons compared to the Na_V1.2 isoform that is expressed earlier but is replaced by Na_V1.6 during neuronal maturation (O’Malley and Isom, 2016; Meisler et al., 2021). We find that TTX exhibits an IC₅₀ inhibitory potency of 1.7 ± 0.2 nM (n = 5) to block spontaneous activity-regulated MeAIB transport activity in hippocampal neuron cultures. The Na_V1.6 inhibitor 4,9-anhydroTTX exhibits an IC₅₀ inhibitory potency of 61 ± 4 nM (n = 3) to block activity in neurons in vitro (Fig. 10C). The inhibitory potency of 4, 9-anhydroTTX on Na_V1.2 channels is IC₅₀ > 1 μM (Teramoto and Yotsu-Yamashita, 2015; Rosker et al., 2007). Our results indicate that Na_V1.6 Na⁺ channel activity may be critical to drive spontaneous neural activity regulated MeAIB/Gln transport activity in mature hippocampal neurons in vitro.

We further examined the inhibitory potencies of riluzole, and the original active aminothiazole identified SKA-41 in our first screen, SKA-75 and SKA-219 in our second screen, and three of our new chemical entities (SKA-376, 377, 378) to inhibit the activity of two major hippocampal Na⁺ channels Na_V1.2 and Na_V1.6 in heterologous cell line assays (Fig. 11) using whole-cell patch-clamp. In keeping with some literature reports (Herbert et al., 1994; Song et al., 1997; Stefani et al., 1997; Spadoni et al., 2002), we find that riluzole is a relatively weak Na_V1.2 blocker with an IC₅₀ of 67 ± 9 μM (Fig. 11A,D). The block is not frequency dependent (Fig. 11C). SKA-378 is even less potent on Na_V1.2 with an IC₅₀ of 118 ± 13 μM (Fig. 11B,D). This potency sequence was reversed on Na_V1.6, which was more potently blocked by SKA-378 (IC₅₀ 28 ± 4 μM) than by riluzole (IC₅₀ 115 ± 2 μM). The 4,9-anhydroTTX was less potent in our hands (IC₅₀ 138 ± 24 nM, Fig. 11E) than previously reported for Na_V1.6 expressed in oocytes (Rosker et al., 2007). TTX itself was significantly more potent (IC₅₀ 0.426 ± 0.003 nM, Fig. 11E) with values similar to the inhibition of spontaneous, neural activity regulated MeAIB transport (Fig. 10C). SKA-378 is the most potent agent to inhibit Na_V1.6 activity and least potent agent to inhibit Na_V1.2 activity compared to riluzole and the other active compounds in these heterologous cell assays in vitro. (Fig. 11D).

Fig. 11.

Inhibition of Na_V1.2 and Na_V1.6 by riluzole, SKA-41, SKA-75, SKA-376, SKA-377, SKA-378 and SKA-219 determined by whole-cell patch-clamp. A, Effect of riluzole on native Na_V1.2 currents in N1E-115 cells. Left, Time course of current inhibition by increasing concentrations. Middle, Fractional block of 100 μM. Right, Concentration response curve. B, Effect of SKA-378 on native Na_V1.2 channels in N1E-115 cells. Left, Time course of current inhibition by increasing concentrations. Middle, Fractional block of 100 μM. Right, Concentration response curve. C, Inhibition of Na_V1.2 current by riluzole does not depend on holding potential or pulse frequency. D, Table showing IC50 values ( ± SEM) in μM for Na_V1.2 and Na_V1.6 (holding potential −80 mV, inter-pulse interval 10 s (0.1 Hz). ^arecorded at 1 Hz to determine if there is any frequency dependence for riluzole on Na_V1.6. E, Concentration response curves for TTX (n = 2) and 4,9-anhydroTTX on Na_V1.6. All values are results from data of 3 or 4 independent cells with critical significance level of p < 0.05.

4. Discussion

Glutamate (Glu) is the major excitatory transmitter in synaptic vesicles that is released in the brain by Ca²⁺-regulated exocytosis (Maycox et al., 1990; Sudhof, 2012). Glu excitotoxicity and neural injury resulting from KA-induced SE in rodents is an accepted animal model of human TLE. Glu transmission in hippocampal neurons and Glu-induced excitotoxicity requires resupply of the synaptic Glu precursors for continued synthesis and release. Gln has long been thought to be an important precursor to synaptic Glu transmitter synthesis via the Glu/Gln cycle between neuronal synapses and astrocytes especially under intense depolarization conditions in the hippocampus; although the mechanisms involved still remain unclear. Excessive synaptic Glu transmission in the brain results in excitotoxicity, both acutely and chronically, that has long-term consequences for neural degeneration, in general, both in vitro and in vivo. Neuronal maturation of regulated presynaptic release of Glu and GABA and the excitatory/inhibitory balance required for stable hippocampal networks occurs after 3 weeks in vitro and after 3 postnatal weeks in hippocampal circuits in vivo. This is because the major synaptic vesicular transporters that control presynaptic excitatory and inhibitory synaptic neurotransmission in these neurons (i.e., VGLUT1 and VGAT) are also up-regulated during this period (Boulland et al., 2004; De Gois et al., 2005; Wilson et al., 2005) and they are required for Ca²⁺-regulated Glu and GABA synaptic transmission (Takamori et al., 2000; Moechars et al., 2006; Saito et al., 2010). The maturation of synaptic Glu neurotransmitter cycling after synaptic release to support continued excitatory transmission also develops during this period (Bolshakov and Siegelbaum, 1995; Wasling et al., 2004; Chowdhury et al., 2007; Hertz, 2013). SE induced neural excitotoxicity in the hippocampus and interconnected limbic regions by KA treatment also requires a mature hippocampal neural network that is similarly established and exists only after the third postnatal week (Nitecka et al., 1984; Wozniak et al., 1991; Haas et al., 2001). Activity regulated MeAIB/Gln transport activity develops during the second and third weeks in hippocampal neural cultures and is maximal after the third week (Erickson, 2017), as well. An important influence in neuro-protection here may arise from the selective expression of the specific transporter protein that mediates neural activity regulated high affinity MeAIB/Gln transport in mature hippocampal excitatory neurons. The vulnerability of the hippocampus to hyperactivation by KA treatment and the prevention of acute neural injury by riluzole and significant attenuation by SKA-378 and SKA-379 in this study suggests that these compounds target a common mechanism that could prevent the propagation of neural injury that leads to epileptic disease within hippocampal limbic circuits.

Riluzole is a benzothiazole anticonvulsant agent that suppresses presynaptic Glu transmission in the brain, by unclear mechanisms because it has multiple targets that could inhibit synaptic excitotoxicity under seizure activity to provide neuroprotection. The most potent action of riluzole and several of our novel naphthalenyl substituted aminothiazoles reported here is inhibition of spontaneous, high affinity MeAIB/Gln transport activity in mature hippocampal neuron enriched cultures in vitro. The blockade of Na⁺ channels in hippocampal neurons is one of the proposed mechanisms of action of riluzole, and Na⁺ channel blockade is considered a general target of many ASDs such as phenytoin and carbamazepine, and others that are used clinically (Errington et al., 2005; Pal et al., 2021). However, limbic seizures and neurotoxicity induced by KA in vivo are resistant to conventional ASDs such a phenytoin and valproic acid (Clifford et al., 1982; Fuller and Olney, 1981). In the KA model of temporal lobe epilepsy (TLE), a potential AED would be administered at least 1 h after the first SE event to prevent acute hippocampal neural injury. Here, we show for the first time that riluzole treatment after KA-induced SE blocks neural injury in the hippocampus. Post KA-induced SE treatment with SKA-378 and SKA-379 attenuate hippocampal neural injury in a similar way, but are less potent. Our data indicate that riluzole and the aminothiazole derivatives of riluzole described here may offer a new presynaptic approach to prevent neural injury in hippocampal circuits that occurs after KA-induced SE.

Naphthalenyl-substituted aminothiazole derivatives of riluzole are underdeveloped agents for neuroprotection. While the benzothiazole riluzole is relatively safe at low concentrations (Bellingham, 2011) it is not especially brain penetrant (Verma et al., 2016; Pelletier et al., 2018) and has side-effects such as sedation and impaired motor control in vivo in rats (Doble, 1996); many ASDs currently used clinically have similar associated side-effects (Perucca and Meador, 2005; Brodie, 2017; Loscher and Klein, 2021). Identifying the targets that protect vulnerable hippocampal neurons require pharmacological agents that can affect sites that are important for prevention of excitotoxic neural injury. Prolonged seizure events induced by KA treatment in vivo lead to excessive Glu release and NMDA Glu receptor activation but while postsynaptic Glu receptor therapy prevents neural injury in animal models of many excitotoxic disease states, including epilepsy (Loscher and Schmidt, 2012; Rogawski, 2013; Schidlitzki et al., 2017; Hanada, 2020), NMDA receptor antagonists are not well tolerated in humans because of unacceptable CNS side-effects (Muir, 2006; Lipton, 2004). In addition, while NMDA receptor blockade immediately after a 90 min KA-induced SE is an effective means to reduce the damage produced in several brain regions, including the hippocampus, this partial neuro-protection of the limbic system does not prevent the development of epilepsy (Ebert et al., 2002; Brandt et al., 2003). Thus, it is critical to further develop novel presynaptic neuroprotective agents that can reduce Glu-induced excitotoxic injury in these vulnerable hippocampal and limbic neurons. We report here that the administration of aminothiazoles SKA-377, SKA-378 and SKA-379 (30 mg/kg; i.p.) after KA-induced SE occurs significantly reduce acute neural injury in the hippocampus after 3d. Riluzole blocks neural injury at 10 mg/kg. Thus, riluzole and our novel aminothiazole derivatives of riluzole may be useful agents to further understand the mechanisms involved that can prevent acute Glu-induced neural injury in the hippocampus following SE events and, which otherwise, may lead to the initiation of chronic injury and progression of temporal lobe epilepsy.

The mechanism of acute neuroprotection of hippocampal neurons by riluzole and our novel napthalenyl substituted aminothiazoles following KA-induced SE is not known. Here, kinetic analysis reveals that riluzole and SKA-378 inhibit spontaneous, activity regulated, high affinity MeAIB transport in vitro (IC₅₀ ~1 μM) by indirect means because they cannot compete at the substrate recognition site for transport activity in neurons. That is, these compounds do not affect the K_m (affinity) but the V_max (transporter #) of transport activity at the plasma membrane; although an allosteric mechanism could be involved. TTX and the Na_V1.6 antagonist 4,9-anhydroTTX are potent (IC₅₀ ~ 1 and 60 nM, respectively) inhibitors of neural activity regulated MeAIB/Gln transport activity in hippocampal neuronal cultures. SKA-378 is the most potent agent to inhibit Na_V1.6 activity (IC₅₀ ~28 μM) and least potent agent to inhibit Na_V1.2 activity (IC₅₀ ~118 μM) in heterologous cell assays in vitro. Riluzole also inhibits Na_V1.6 and Na_V1.2 with similar μM affinities but in reverse order of potency in these cell assays. However, the high plasma protein binding of riluzole, SKA-378 and SKA-377 in serum (96–99%) could limit their free brain concentrations to the <1 μM range; although total brain concentrations of several of these active aminothiazoles compounds are in the >10 μM concentration. Thus, it is unlikely that Na_V1.6 inhibition is the major contributor to the neuro-protective effect by riluzole or our aminothiazole riluzole derivatives in vivo. However, the evaluation of % free concentrations of our active aminothiazoles, and riluzole, in brain CSF and extrasynaptic space, and compound distribution into phospholipid membrane compartments in synapses, may be necessary in future studies to better evaluate concentration response relationships of these lipophilic drugs in vitro and in vivo to their neuroprotective targets (Rizk et al., 2017; Treyer et al., 2018). Spontaneous, activity regulated MeAIB transport activity described here is also dependent on Ca²⁺ ion that is required for exocytosis of vesicles that contain Glu. The potency of riluzole and our novel active aminothiazoles in hippocampal neurons may result from their interference with the cycling of the neural activity regulated MeAIB/Gln transporter within Glu terminals to the plasma membrane. A similar scheme of activity regulated, Ca²⁺-dependent cycling of reuptake transporters between synaptic vesicles and the plasma membrane in nerve terminals has been described in other neurotransmitter systems in the brain (Ferguson et al., 2003; Deken et al., 2003).

The relationship between an initial limbic epileptogenic seizure event (i.e., SE) and the resulting neural injury in the hippocampus as a critical step in the progression to temporal lobe epilepsy is not understood. Presynaptic therapies that prevent acute limbic neural injury following an initial SE event (Pitkanen and Sutula, 2002; Guerriero et al., 2015; Saletti et al., 2019) are lacking. We report that riluzole prevents acute KA-induced hippocampal injury when administered after SE is observed for the first time. Novel naphthalenyl substituted aminothiazole derivatives of riluzole also provide acute neuroprotection in this way. However, SKA-378 does not prevent KA-induced SE, nor do the most active aminothiazole compounds (SKA-378 and SKA-379) described here limit the duration of SE following KA treatment, unlike riluzole. We would like to suggest that riluzole, SKA-378 and SKA-379 could serve as tool compounds that could clarify the role of SE-induced acute neural injury in epileptogenesis and epileptic disease using the KA-model of TLE. SE also induces acute and chronic inflammatory processes involving astrocytes and microglia in early and later stages (Vezzani et al., 2011, 2019; Mukhtar, 2020) and it will be important to establish whether riluzole and our novel aminothiazoles can also prevent acute and chronic inflammation following KA-induced SE in future studies. In conclusion, we have found that riluzole and several novel naphthalenyl substituted aminothiazole derivatives of riluzole developed here attenuate acute hippocampal injury following KA-induced SE. Riluzole and these novel aminothiazole compounds have significant scientific value to understand the mechanisms that underlie their prevention of acute hippocampal neural injury following KA-induced SE. In addition, riluzole and these novel aminothiazole derivatives of riluzole could have significant therapeutic value as a preventative measure to be used after an initial SE event occurs to reduce acute hippocampal neural injury that, otherwise, can lead to the progression of epileptic disease.

Abbreviation

ALS

Amyotrophic lateral sclerosis

ANOVA

Analysis of variance

ASDs

Antiseizure drugs

AEDs

Antiepileptic drugs

CTCF

Corrected total cell fluorescence

CSF

Cerebral spinal fluid

DAPI

4′,6-diamidino-2-phenylindole

DIV

Days in vitro

DG

Dentate gyrus

DMEM

Dulbecco’s modified eagle medium

EAAT2

Excitatory amino acid transporter 2

FJB

Fluoro-jade B

Glu

Glutamate

Glt1

Glutamate transporter 1

Gln

Glutamine

KA

Kainic acid

LC/MS

Liquid chromatography – mass spectrometry

V_max

Maximal velocity

MeAIB

Methylaminoisobutryic acid

K_m

Michaelis-Menton constant

NB

Neurobasal medium

NMDA

N-methyl-d-aspartate

Ril

Riluzole

SNATs

Sodium-coupled Neutral Amino Acid Transporters/System N and A Transporters

Na_V1.2

Sodium channel isoform 1.2

Na_V1.6

Sodium channel isoform 1.6

SLC38

Solute carrier family 38

SE

Status epilepticus

TLE

Temporal lobe epilepsy

VGAT

Vesicular GABA transporter

VGLUT1

Vesicular glutamate transporter-1

Data availability

Data will be made available on request.