WO2021231945A1 - Assays for assessing anti-seizure activity of chemical compounds - Google Patents

Abstract

Methods for assessing anti-seizure activity of compounds in human neural cell co-cultures are disclosed. The co-cultures contain both functional excitatory and inhibitory human neuronal cells, as well as astrocytes. The co-cultures may make use of GABAergic inhibitory neurons, glutamatergic excitatory neurons, dopaminergic excitatory neurons, and/or serotonergic neurons. Assessment of the compounds anti-seizure activity is based on the compound's effect on the network activity of the co-culture.

Description

ASSAYS FOR ASSESSING ANTI-SEIZURE ACTIVITY OF CHEMICAL COMPOUNDS

CROSS-REFERENCE TO RELATED APPLICATION [0001] This application claims the benefit of U.S. Provisional Patent Application No. 63/025,819, filed May 15, 2020, which is incorporated herein by reference in its entirety.

FIELD

[0002] The herein disclosed embodiments relate to a novel in vitro assay for assessing anti-seizure activity of chemical compounds using iPSC-derived induced neurons.

BACKGROUND

[0003] In the past decade since the discovery of somatic reprogramming by Yamanaka and colleagues, many human induced pluripotent stem cell (iPSC)-derived neuronal model systems have been introduced in the academic field. However, broad applications in neuroscience drug discovery for the benefit of patients suffering from neurological disorders have repeatedly met challenges.

[0004] First, the process of generating human neurons following conventional differentiation methods is lengthy and afflicted by poor reproducibility due to heterogeneous target cell populations and their differential drift towards non-neural lineages. Moreover, specific cell types that are expanded from progenitor pools often do not exhibit a uniform maturation level or functional profile. Additionally, genetic, epigenetic, and transcriptional variances between iPSCs of individual donors can considerably impact differentiation timing and resulting cell ratios. Consequentially, in long-term conventional differentiation protocols, all these factors contribute to a large variability in phenotypic readouts of iPSC-derived in vitro models and limit their applicability for screening applications. Particularly, a consistent balance between excitation and inhibition important for neural network function, which requires stable ratios of excitatory and inhibitory neurons in a cell culture system, has rarely been achieved in scalable settings. Second, while many neural differentiation protocols allow for molecular signatures of neurons to look fairly similar to their in vivo counterparts in a brain, functionality remains limited. This is particularly important in light of neuronal network activity that is disrupted in many neurological disorders, including epilepsy.

[0005] In the academic field, the functionality issues are currently addressed by developing more complex protocols including generation of three-dimensional (organoid) cultures containing several relevant cell types for neural network function. However, the added complexity of these systems results in poorer reproducibility and throughput prerequisite for applications in the drug discovery field thus far have prevented broad implementation of organoid systems for therapeutic development.

[0006] Epilepsy is one of the most prevalent neurological disorders, affecting approximately 3 million people in the United States of America and about 50 million worldwide. Despite the availability of many anti-epileptic drugs (AEDs) in clinical practice, one third of the patients still suffer from recurrent and unprovoked seizures and are categorized as pharmaco-resistant. Significant amount of effort and resources have been put forward in the past three decades, seeking next generation drugs. Around 20 AEDs have been approved during this period, but the considerable unmet medical need remains to be addressed.

[0007] Current AED development relies heavily on preclinical animal models to establish efficacy and safety. While these models have been proven useful identifying potential AEDs, there are two major caveats that limit their ability to work at scale and to identify first-in-class next-generation drugs.

[0008] First, in vivo models are expensive and time-consuming and thus have difficulties fitting into large- scale phenotypic screenings. The resource burden is further exacerbated by the trend of moving from acute to chronic models with increased clinical relevance, which narrows the scope and number of candidate compounds to be tested. Secondly, because the mechanistic essence of epileptogenesis is largely unknown, instead of focusing on recapitulating disease etiology, current models (including all from the US National Institute of Neurological Disorders and Stroke’s Epilepsy Therapy Screening Program (ETSP) repertoire) exploit phenotypic interactions between candidate compounds and provoked hyperactivities in laboratory animals or their tissues. Depending on the induction methods, models may be biased to certain seizure types or produce false negative results. For instance, the rodent maximal electroshock seizure (MES) model is effective in identifying candidates that block generalized seizure but failed to predict positively for vigabatrin and levetiracetam that are efficacious against partial seizures. Thus, current framework tends to identify replacement for existing AEDs rather than first-in-class drugs.

[0009] To ameliorate these two issues, other models have been adapted using zebra fish and fruit flies. These models take advantage of benefits such as ease of genetic manipulation, fast reproduction cycle, lower maintenance cost, and higher throughput, but doing so at the expense of moving further away from human genetics and mammalian physiology. As a result, few have been used in mainstream AED development and none has been adapted by major programs, such as the ETSP.

[0010] There is a need for novel methods and assays for the screening and optimization of novel AED candidates. SUMMARY

[0011] Disclosed herein are compositions, methods and assays for the screening and optimization of novel AED candidates. These methods and assays make use of a defined population of cultured cells, unlike previously existing methods which rely on laboratory animal models, either in vivo or explanted brain slices in vitro. The assays can make use of a variety of chemicals or genetic alterations to induce hyperactivity of neural networks established in the cell cultures and can be validated for their purpose of modeling seizure like events with known AEDs. These compositions, methods, and assays disclosed herein enable cell culture-based screening and characterization of new anti-seizure medications for the first time. The assays support high throughput screening of candidate drugs and are capable of a substantial degree of automation.

[0012] In some embodiments the assay has been validated with a plurality of anti-seizure drugs. In some embodiments the assay has been validated with 10, 11, ... 17, 18, 19, 20 anti-seizure drugs. In some embodiments the anti-seizure drugs have known efficacious human total plasma concentrations. In some embodiments the known anti-seizure drugs used for validation comprise at least 1, 2, 3, 4, or more of: (-)- cannabidiol, (S)-carisbamate, carbamazepine, cenobamate, ezogabine, lacosamide, lamotrigine, levetiracetam, licarbazepine, mephenytoin, methsuximide, mexiletine, perampanel, phenytoin, rufinamide, safinamide, stiripentol, or valproate. In further embodiments the assays are counter-validated with failed anti-seizure candidate drugs. Whereas with the validation the assay can be used to identify drug candidates with potential anti-seizure activity, with the counter-validation the assay can be used to discriminate between those candidates likely to succeed and those likely to fail in clinical use.

[0013] Microelectrode array (MEA) assays involve culturing a mixture of excitatory induced neurons (iNs), inhibitory iNs, and astrocytes in wells of MEA plates. In some embodiments the ratio of excitatory iNs to inhibitory iNs is 70:30. In some embodiments the cell density of excitatory iNs, inhibitory iNs, and astrocytes is 140K, 60K, and 70K per well in a 48-well plate. In some embodiments the assay is conducted 23 days after seeding the plates. In some embodiments the assay is conducted 2 days after a full media exchange. Typically, an assay electrophysiological recording session begins with a period for acclimation of the cultures to the recording chamber. In some embodiments the acclimation period is 20 minutes. Typically, the assay next comprises a period for establishing a baseline for the electrical activity of the cultures. In some embodiments the period for establishing the baseline is 15 minutes. In acute, chemically- induced seizure models, the next step is the seizurogenic challenge. In some embodiments the challenge period is 30 minutes. In some embodiments involving acute, chemically-induced seizure models, the seizurogenic chemical is picrotoxin or bicuculline; in other such embodiments, the seizurogenic chemical is 4-aminopyridine. In some embodiments, the chemical agent is referred to as means for inducing acute seizure or means to induce increased network activity. With chronic seizure models, whether chemically or genetically induced, there is no acute challenge period. In some embodiments, the chronic seizure model is established by repeated treatment with a chemical agent, for example, kainic acid. In some embodiments, the chronic seizure model is established by genetic modification, for example, by introduction of loss-of- function mutations in one copy of the SCN1A gene encoding the sodium channel NaVl.l. Finally there is a treatment period, initiated by addition of the AED or AED candidate. In some embodiments, the treatment period is 30 minutes. In some embodiments the agent or modification establishing chronic seizure is referred to as means for inducing chronic seizure or means to induce increased network activity. In some of these embodiments the means for inducing chronic seizure is not a genetic modification. In other of these embodiments the means for inducing chronic seizure is not a chemical agent.

[0014] Some embodiments are methods of assessing anti-seizure activity of a compound. Such methods include steps of providing a human neural co-culture comprising excitatory and inhibitory human neuronal cells, and astrocytes. In an aspect of these embodiments the neural co-culture exhibits network activity, such as synchronous network bursts. In some embodiments, the neural co-cultures have been previously exposed to a chemical agent that induces a chronic increase in network activity (as compared to unexposed cultures), or the neuronal cells have been genetically modified so that they exhibit an increased level of network activity (as compared to cultures of unmodified or non-mutant cells). In other embodiments, the neural co-cultures are exposed to a chemical agent that induces an acute increase in network activity (as compared to the unexposed culture). Neural co-cultures exhibiting increased network activity (however induced) are then exposed to a compound to be tested or characterized and the effect of the compound on network activity is measured. A decrease in network activity indicates anti-seizure activity and potential for the treatment of seizure associated disorders, such as epilepsy. In some embodiments, the effects of various concentrations of the compound being tested or characterized are measured and an EC50 is determined.

[0015] In various embodiments, the chemical agent used to induce increased network activity is a GABA receptor antagonist. In some embodiments the GABA receptor antagonist comprises picrotoxin. In some embodiments the chemical agent used to induce increased network activity comprises kainic acid or 4- aminopyridine. Increased network activity is considered induced seizure-like activity. In some embodiments, induced seizure-like activity is characterized by one or more of the following parameters: increased mean firing rate of spikes; increased total number of network bursts; increased network burst frequency; increased total number of bursts per recording period; increased burst frequency; increased burst duration; increased network burst percentage; decreased inter-network-burst interval; decreased inter-spike interval; and increased cross-correlation of detected spikes between all electrodes per well.

[0016] In some embodiments, the network activity assessed is one or more of total number of: spikes per recording period; mean firing rate of spikes; inter-spike interval; total number of bursts per recording period; burst frequency; number of spikes per burst; burst duration; inter-burst interval; the proportion of spikes occurring within a burst; total number of network bursts; network burst frequency; number of spikes per network burst; network burst duration; inter-network-burst interval; inter-spike interval within network bursts; the proportion of bursts occurring within a network burst; and cross-correlation of detected spikes between all electrodes per well. In some embodiments, the network activity assessed is mean firing rate of spikes. In some embodiments, network activity is measured using a microelectrode electrode array. In some embodiments, network activity is measured using ratio-metric measurements based on calcium imaging.

[0017] In some embodiments, the neural co-culture can comprise one or more of GABAergic inhibitory neurons, glutamatergic excitatory neurons, dopaminergic excitatory neurons, and serotonergic neurons. In some embodiments, the neural co-culture can comprise GABAergic inhibitory neurons and glutamatergic excitatory neurons. In an aspect of these embodiments, the ratio between said GABAergic inhibitory neurons and said glutamatergic excitatory neurons is between 20:80 and 70:30.

[0018] In some embodiments, functional human neuronal cells are obtained by in vitro differentiation by contacting a population of non-neuronal human cells with neuron reprogramming factors (NR), or agents to activate NR factors, wherein the NR factors are selected from the group consisting of: Neurogenin, Ascl ,NeuroD, Bm2, Bm3a, Emx, Cux2, Tbrl, Satb2, Dlxl/2/5, Nkx2.1, Nkx2.2, Lhx2/3/6/8, Sox2, Foxgl, Ctip2, Hb9, Isll/2, Klf7, Gata2, Foxa2, Fmxlb, Ptx, FEV, Fmxl, Foxa2, Nurrl, Pitx3, and En for a period of time sufficient to reprogram said non-neural cells. In various aspects, these non-neuronal human cells can be pluripotent cells, somatic stem cells, or induced pluripotent stem cells. In some embodiments, GABAergic inhibitory neurons are generated by contacting a population of non-neuronal human cells with neuron reprogramming factors Ascii and Dlx2. In some embodiments, GABAergic inhibitory neurons are generated by contacting a population of non-neuronal human cells with neuron reprogramming factor Ngn2. In some embodiments, the neuron reprogramming factors can be referred to as means for neuron reprogramming.

[0019] In some embodiments, human glial cells are derived by contacting a population of non-glial cells with one or more of whole serum, single serum components, insulin, BMP-inhibitor, TGF -beta-inhibitor, growth factors and morphogens such as EGF, CNTF, BMP2/4, and/or transcription factors such as Nfia, Nfib, Sox9, and Hes for a period of time sufficient to reprogram or step-wise differentiate non-glial cells to astroglial cells. In some embodiments, human astrocytes are derived by contacting a population of human pluripotent stem cells with an effective dose of a reprograming system comprising Nfia or Nfib for a period of time sufficient to reprogram said pluripotent cells . In one aspect, Sox9 can be overexpressed in the human pluripotent stem cells.

[0020] In some embodiments, the effect of a known or putative anti-seizure drug on the network activity of the neural co-cultures is validated by determining the effect of a plurality of known anti-seizure drugs, for example 10-20 known anti-seizure drugs. A correlation between the observed EC50 of the plurality of drugs and another known property of the drugs can be derived. In various aspects the known property can be effective dose or effective plasma concentration or effective brain concentration. The correlation can then be used to determine an expected property of the known or putative anti-seizure drug being tested or characterized with the assay. In various aspects the expected property is effective dose or effective plasma concentration or effective brain concentration.

BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The novel features of the herein disclosed embodiments are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present methods and assays will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which their principles of operation are demonstrated, and to the accompanying drawings of which:

[0022] Figure 1A-B depicts pure human-derived neural co-cultures for assessing anti-seizure effects. (Figure 1A) Schematic representation of workflow for direct reprogramming of iPSCs into excitatory and inhibitory neurons using lentiviral delivery of neurogenic transcription factors (upper panel). Primary astroglial cells are derived from brain tissue of aborted fetuses and expanded in vitro (lower panel). (Figure IB) Schematic representation of induced neuron/glia co-cultures grown in microelectrode array (MEA) wells (upper panel). Shown example of a 48-well MEA plate harboring 16 electrodes in each well (lower panel).

[0023] Figure 2 depicts the plate layout for compound testing.

[0024] Figure 3 depicts electrophysiological recording scheme for compound testing.

[0025] Figure 4 depicts parameter settings on the NeuralMetricTool.

[0026] Figure 5 depicts an example of MEA assay results using licarbazepine as the AED. [0027] Figure 6 depicts the correlation between AED human total plasma concentrations of various AEDs and their MEA EC_50S as determined using the PTX model. EC_50S were calculated based on the weighted mean firing rate data.

DETAILED DESCRIPTION

[0028] In general, the herein disclosed embodiments provide compositions and methods related to using iPSC-derived induced neurons which are useful for screening chemical compounds for the treatment of epilepsy.

[0029] Current anti-epileptic drug (AED) screening is largely limited to rodent models, which, despite considerable translational validity, are high in cost and low in throughput. No clinically relevant alternatives, for example from in vitro models, exist. Here we disclose a method to validate human-relevant in vitro seizure models for high-throughput, low-cost anticonvulsant screening. This method will provide a translationally relevant context of human neuronal network physiology for high-throughput, naive drug screening as well as for efficacy and neurotoxicity testing during AED candidate optimization.

[0030] The herein disclosed embodiments generally provide methods and assays for the screening and optimization of novel AED candidates. The methods robustly generate and efficiently utilize human excitatory and inhibitory neurons from human induced pluripotent stem cells that are grown in co-culture with human astrocytes. These cultures serve as a basis to develop a novel type of AED screening platform. This platform provides a translationally relevant context of human neuronal network physiology for high- throughput, naive drug screening as well as for efficacy and neurotoxicity testing during AED candidate optimization.

[0031] Embodiments disclosed herein provide novel and scalable in vitro seizure models that are based on human neurophysiology. These models provide high-throughput, cost-efficient entry points for AED candidate screening at a scale that will ultimately allow expanding discovery efforts towards novel, first- in-class AED candidates to address the considerable needs around refractory and rare types of epilepsy. These models will further allow gaining human-relevant insights into efficacy and neurotoxicity at an early step during the lengthy process of drug development.

I. Human Neuronal Cultures

[0032] In one aspect, the herein disclosed embodiments utilize human neural cells, including neurons and astrocytes, that are co-cultured in vivo. Suitable co-cultures are described in WO2017223052A1, which is hereby incorporated by reference in its entirety for all that it teaches about generating the necessary neural cells and establishing and using such co-cultures. A. Induced excitatory neurons.

[0033] For the generation of induced excitatory glutamatergic neurons from human pluripotent stem cells (hPSCs), a direct differentiation protocol through exogenous expression of neurogenic transcription factors is used. The hPSCs are cultured in the presence of stem cell medium and induced to express an effective dose of Ngn2, e.g., by lentiviral infection (Figure 1A). The induced cells are cultured in neuronal medium until neuronal differentiation initiates, to generate committed immature induced neuronal cells, which can be replated in medium for neural co-cultures.

IT _ Induced inhibitory neurons.

[0034] For the generation of induced inhibitory GABAergic neurons from hPSCs, a direct differentiation protocol through exogenous expression of neurogenic transcription factors is used. The hPSCs are cultured in the presence of stem cell medium and induced to express an effective dose of Ascii and Dlx2, or Ascii, Dlx2 and Mytll, e.g., by lentiviral infection (Figure 1 A). The induced cells are cultured in neuronal medium until neuronal differentiation initiates, to generate committed immature induced neuronal cells, which can be replated in medium for neural co-cultures.

C. _ Astrocytes and Induced Astrocytes

[0035] Primary human glia cells are isolated from aborted fetuses and expanded ex vivo in glia medium until mature astroglial cells can be replated in medium for neural co-cultures (Figure 1A). In alternative embodiments, non-human animal primary (e.g., rat or mouse) astroglial cells are used.

[0036] For the generation of induced astrocytes from hPSCs, a direct differentiation protocol through exogenous expression of glial transcription factors is used. The hPSCs are cultured in the presence of stem cell medium and induced to express an effective dose of Sox9 and Nfib, e.g., by lentiviral infection. Induced astrocytes can further be generated by said transcription factor expression from neural stem cells derived from hPSCs by established differentiation methods, such as described in e.g. Palm et al., Sci Rep 5: 16321, 2015. The induced cells are cultured in glia medium until differentiation initiates, to generate committed mature astroglial cells, which can be re-plated in medium for neural co-cultures. Suitable protocols for generating induced astrocytes are described in PCT/EP2019/072065 and Canals et al., Nat Methods 15:693- 696, 2018, which are each hereby incorporated by reference in their entirety for all that they teach related to generating induced astrocytes. Alternatively, astroglial cells can be generated from hPSCs using stepwise differentiation e.g. by withdrawal of BMP and TGF to induce a neuroepithelial cell identity and subsequent culturing in neural medium supplemented with growth factors such as CNTF, FGF, and EGF following protocols such as described in Krencik et al., Nat Protoc 6: 1710-7, 2011. [0037] In some embodiments, the transcription factors and/or growth factors provided herein are human transcription factors. In some embodiments, the transcription factors provided herein are rodent transcription factors, such as mouse or rat transcription factors. In some embodiments, the transcription factors provided herein are human or rodent transcription factors, or functional homologues thereof.

[0038] In some embodiments, the astrocytes are generated using neural spheres, such as described in Shaltouki Al, et al., Stem Cells. 2013 May;31(5):941-52;, or in Tchieu J, et al., Nat Biotechnol. 2019 Mar;37(3):267-275.

[0039] The above alternatives for the astroglial component of the neural co-cultures may be referred to collectively as means for astroglial cell function or astroglial means.

D. Neuronal Astroglial Co-cultures.

[0040] The three cell types, that is excitatory neurons, inhibitory neurons, and astroglial cells, are combined in co-culture in devices for readout of neural activity, e.g. microelectrode array (MEA) plates (Figure IB). The ratio of excitatory/inhibitory neurons may be about 80:20, 70:30, 60:40, 50:50, 40:60, or 30:70. In some embodiments, the percentage of excitatory neurons in the combined populations of excitatory and inhibitory neurons is about 60%, 70%, or 80%. In some embodiments, the percentage of excitatory neurons in the combined excitatory/inhibitory neurons is from about 30% to about 70%, or about 80%, or from about 70% to about 80%. In some embodiments the ratio of excitatory to inhibitory neurons is from about 60:40 to about 80:20. In some embodiments the ratio of excitatory to inhibitory neurons is about 70 : 30 or about 66:33. In some embodiments the ratio of astroglial cells to neurons can be about 1: 10, 1:7.5, 1:5, 1:2.5, 1: 1, or 1:0.5. In some embodiments the number of neurons plated can be from about 10⁴, to 10⁵, 10⁶, or 10⁷ per well.

[0041] In some embodiments, the neural cells are seeded and maintained on MEA plates, which are specialized tissue culture plates comprising microelectrodes integrated into the well bottom for detection of extracellular currents and local field potentials (see, for example, the Maestro Platform from Axion BioSystems). The MEA plate may be pre-coated with a suitable substrate, including without limitation laminin, polyethylenimine (PEI), Matrigel®, or the like. Neuronal cells can be plated in neurobasal/B27 medium (Neurobasal/B27 medium: Neurobasal-A medium +B27 +0.5xGlutamax+NT3 [lOng/ml] +mouse laminin [200ng/ml]) supplemented with 2pg/ml doxycycline, 1% FBS, and IOmM Rock inhibitor (Y27632). The total number of seeded iN cells may be in the range between 300,000 and 600,000 cells per well for 12-well plate or between 100,000 and 250,000 cells per well for 48 well or 96 well plates, respectively. Glial cells are seeded in parallel, before, or after attachment of iN cells in the same medium at densities between 60,000 and 120,000 or between 20,000 and 50,000 cells per well for 12-well or 48 well plates, respectively. Two days after seeding, half of the medium is replaced and AraC is added to final concentration of 2 mM to 10 mM in order to prevent overgrowth of glial cells. During the first week after seeding, half-medium changes are performed every other day. During the second week, half-medium changes are performed every 3 days, and afterwards, half-medium changes are performed twice a week and at least two days before recording of neuronal activity. Neural co-cultures on MEA plates can be maintained at 37°C and 5% CO2 for over 6 weeks.

[0042] In one aspect, embodiments provide a human neural cell co-culture that provides synchronous network bursts, the co-culture comprising: in vitro differentiated functional human neuronal cells; and glial cells, such as mouse, rat, or human glia cells. In general, the neural cell co-culture provided herein is characterized by being capable of forming synapses, and preferably generate synchronous network bursts, which are observed about 2, 3, 4, or 5 weeks after the seeding of the co-culture. Bursts are considered synchronized network bursts if the first spikes of individual bursts are co-occurring within about 5, about 10, about 20, about 30, about 40 milliseconds; measured by at least about 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, or 95% of active electrodes in any single well on a MEA plate.

[0043] Further information on establishing and maintaining such human neuronal cultures can be found in WO2017223052A1, which is incorporated by reference in its entirety.

IL _ Quantitative Readout of Neural Activity

[0044] Of particular interest for the disclosed culture system are parameters related to the electrical properties and signal transmission characteristics of the cells and therefore directly informative about neuronal function and activity. Methods to measure neuronal activity may sense the occurrence of action potentials (spikes). The characteristics of the occurrence of a single spike or multiple spikes either in time- clustered groups (bursts) or distributed over longer time (spike train) of a single neuron or a group of neurons indicate neuronal activation patterns and thus reflect functional neuronal properties, which can be described by multiple parameters. Such parameters can be used to quantify and describe changes in neuronal activity in the herein disclosed systems.

[0045] Neuronal activity of human neural co-cultures can be assessed by electrical or imaging-based readout devices. In some embodiments the readout device measures extracellular currents and local field potentials, e.g. using MEA systems (Maestro MEA platform from Axion BioSystems). In other embodiments, the readout device measures calcium flux in neuronal cells by changes in fluorescence light emission using genetically encoded calcium indicators (GECI) or calcium-sensitive dyes. [0046] For detection of action potentials in neural cultures on MEAs, signals can be detected as spikes when exceeding a present voltage increase, e.g. 2X, 3X, 4X, 5X, 6X or more the standard deviation of voltages measured by each electrode. A set of sequential spikes may be defined as a burst if at least about 3, about 4, about 5 or more spikes are detected by one electrode within a defined period of time, e.g. from around about 10-500 milliseconds, around about 50 to about 250 milliseconds, or around about 100 milliseconds. Bursts detected across multiple electrodes per well can be defined as synchronized network bursts if the first spikes of individual bursts are co-occurring within about 5, about 10, about 20, about 30, about 40 milliseconds; measured by at least 25, 35, 45, 50, 65, 75% of active electrodes.

[0047] Neuronal activity parameters include, without limitation, total number of spikes (per recording period); mean firing rate (of spikes); inter-spike interval (distance (that is, time) between sequential spikes); total number of bursts (per recording period); burst frequency; number of spikes per burst; burst duration (for example, in milliseconds); inter-burst interval (distance (that is, time) between sequential bursts); burst percentage (the proportion of spikes occurring within a burst); total number of network bursts (spontaneous synchronized network activity); network burst frequency; number of spikes per network burst; network burst duration; inter-network-burst interval; inter-spike interval within network bursts; network burst percentage (the proportion of bursts occurring within a network burst); and cross-correlation of detected spikes between all electrodes per well (e.g., for MEA recordings, measure of synchrony).

[0048] Quantitative readouts of neuronal activity parameters may include baseline measurements in the absence of agents or a pre-defined genetic control condition and test measurements in the presence of a single or multiple agents or a genetic test condition in the presence or absence of a candidate agent. Quantitative readouts may include solvent control measurements. Furthermore, quantitative readouts of neuronal activity parameters may include long-term recordings and may therefore be used as a function of time (change of parameter value). Quantitative readout may further be acquired at multiple time points for a neural co-culture to measure latent effects, delayed effects, or long-term effects. Readouts may be acquired either spontaneously or in response to or presence of stimulation or perturbation of the complete neuronal network or selected components of the network. The quantitative readouts of neuronal activity parameters may further include a single determined value, the mean or median values of parallel, subsequent or replicate measurements, the variance of the measurements, various normalizations, the cross correlation between parallel measurements, etc. and every statistic used to a calculate a meaningful and informative factor.

[0049] For imaging-based high throughput measurement of single neuronal firing and neural network activity, iPSC-derived neurons can be infected with lentivirus for exogenous expression of calcium sensor proteins (e.g. GCaMP) and co-cultured with human primary astrocytes on 384-well optical plates. Upon cell membrane depolarization following action potential firing, increased calcium influx through voltage dependent calcium channels leads to increased intracellular calcium levels and therefor increased binding of the calmodulin domain of the calcium sensor proteins. This leads to a deprotonation of the chromophore domain resulting in bright fluorescence. Neuronal activity-mediated calcium dependent fluorescence signals are detected using a FLIPR Penta imaging system (Molecular Devices) and processed into parameters quantifying the electrophysiological activity, including synchronized neural network activity of the neuronal cell population. More generally, network activity can be measured using ratio-metric measurements based on calcium imaging. Calcium concentration, or changes therein, can be detected by means of chemical or genetically encoded calcium indicators (GECIs), such as Fluo-4 or GCaMP-family proteins, or by means of chemical or genetically encoded voltage sensors (GEVIs), such as ASAP2 and the like.

III. Chemical-Induced Seizure Model

[0050] The herein described methods and systems relate to the development of chemically induced in vitro models for seizure activity in human neuronal networks. Physiological characterization and definition of phenotypic endpoints for these models are based on intrinsic network activity of in vitro human neuronal networks as measured by, for example, micro-electrode array (MEA) recording. Furthermore, conventional patch-clamp electrophysiology analysis is employed to demonstrate mechanistic underpinnings and characteristics of seizure-like neural network activity in these cell cultures. These systems may be referred to as means for modeling a seizure or means for modeling a chemically-induced seizure.

[0051] Chemical seizure models described herein are based on two general mechanisms of neurotransmission: 1) inhibition of inhibitory, gamma-aminobutyric acid (GABA)-ergic neurotransmission and 2) agonism of excitatory, glutamatergic neurotransmission. Agents acting through these two general mechanisms induce a hyper-excitability of human in vitro neural networks, similar to their know action in the brain of various organisms including humans. This seizure-like hyper-excitability is counteracted in a dose-dependent manner by anti-epileptic activity of a variety of chemical compounds including FDA- approved AEDs used in clinical practice. The methods described herein utilize this system of chemical challenge and dose-dependent normalization response to create assay platforms that can be used for AED screening, drug optimization, and neurotoxicity assessment. Furthermore, the methods are used in conjunction with clinical AED efficacy data to predict dose ranges for novel AED candidates to be used in clinical trials [0052] Chemical-induced seizure is widely employed in in vivo and in vitro systems to model acute status epilepticus (single episode of epileptic seizure). Picrotoxin, along with metrazol, pilocarpine, 4- aminopyridine, and kainic acid, are some of the most common convulsants used.

[0053] The present embodiments include development of three distinct types of in vitro seizure models, two chemically induced, and one genetic model for the rare childhood epilepsy of Dravet syndrome. Physiological characterization and definition of phenotypic endpoints for all three models is based on intrinsic network activity as measured by multi -electrode array (MEA) recording. Furthermore, conventional patch-clamp electrophysiology analysis is used for detailed studies of cellular and synaptic mechanisms important to understand the properties of each in vitro seizure model.

[0054] In the human neuronal cultures composed of defined ratios of excitatory and inhibitory neuron types, picrotoxin induces acute hyperactivity of neural networks as measured by MEA recordings. This hyperactivity is counteracted in a dose-dependent manner by benchmark AEDs. Importantly, there is a good correlation of our in vitro efficacy data and human plasma levels of these AEDs at therapeutic doses.

A PTX Model

[0055] Gamma aminobutyric acid type A receptors (GABA_ARS) are ligand-gated chloride ion channels and represent the primary mediators of inhibitory neurotransmission, therefore playing an essential role in orchestrating central nervous system (CNS) function. Blocking of GABA_ARS leads to disinhibition of neuronal circuits and excessive neuronal firing. Picrotoxin (PTX) is a potent antagonist of GABA_ARS and can induced severe tonic -clonic seizures upon application in vivo. The hPSC-derived neural co-cultures described herein, relying on three cell types, excitatory glutamatergic iN cells, inhibitory GABAergic iN cells, and astrocytes derived from primary glial cells or induced from PCS, exhibit spontaneous electrical activity within 12 days after seeding and develop coordinated network activity within 21-28 days, as measured on microelectrode arrays (MEA). During this time period, application of PTX (e.g., about 10 mM) induces acute neuronal hyperactivity characterized by an increased rate of single action potentials as well as an increased frequency of network bursts and an elongation of network burst duration which resembles seizure-like neural activity during status epilepticus. This PTX-induced hyperactivity can therefore be regarded as a simplified proxy for seizure events in patients and can be used as an in vitro model for testing anti-seizure effects of therapeutic interventions. However, as this model relies on antagonism of GABA_AR by PTX, it is not well suited to the testing of drugs that modulate GABA signaling. B. Other GABA blockers

[0056] As an alternative to PTX, other GABA_AR antagonists such as bicuculline (BIC) or pentylenetetrazol (PTZ) have demonstrated similar responses and can also be used to induce acute seizure like activity in hPSC-derived neural co-cultures.

C _ Kainic acid model

[0057] Kainic acid (KA) is a chemical convulsant, acting through activation of glutamatergic transmission, broadly used for generation of rodent models for mesial temporal lobe epilepsy and in ex vivo brain slice preparations. Addition of kainic acid to the herein described neural cell co-cultures_resulted in an acute hyperactivity followed by a stably increased baseline activity measured on MEAs. This elevated baseline activity is likely due to synaptic changes in excitatory and inhibitory networks, a hypothesis supported by published findings from rodent in vivo and ex vivo (slice) electrophysiology. In vivo, functional changes induced by kainic acid exposure are accompanied by structural changes of neural networks, including synapses. In hippocampus, KA-dependent structural changes were found in form of excitatory mossy fiber sprouting to innervate excitatory granule cells and, to a greater extent, inhibitory intemeurons of the dentate gyrus, leading in sum to an interictal hyperinhibition of the excitatory granule cells. Furthermore, synaptic remodeling of entorhinal input was found to contribute to aberrant hippocampal network architecture in a KA-based mouse model for mesial temporal lobe epilepsy. The shift in synaptic distribution and formation of asymmetrical synapses on intemeurons disturbs the excitatory/inhibitory balance of hippocampal network activity and therefore likely contributes to the development of status epilepticus. The stably increased baseline activity in the described neural cell co-cultures can be used as a chronic in vitro seizure model fortesting AED candidates in the absence of an acute seizurogenic challenge. Thus the anti-seizure activity of AED candidates is observed as the normalization of elevated baseline activity towards the baseline activity of naive neural cell co-cultures. As this chronic seizure model does not depend on the presence of a GABA_AR antagonist, this model is more appropriate for testing drugs that operate by modulating GABA signaling.

[0058] Deep characterization of network architecture in human neurons can be conducted by high-content imaging, for example on the Operetta system (PerkinElmer), to study the KA-treated neural cell co-cultures at cellular resolution. Changes in neurite morphology (Sholl analysis) and synapse density can be analyzed using protocols for sparse labelling of neural cell co-cultures in an optical 384-well plate format. For synaptic analysis, co-labeling with synaptic markers using immune-fluorescent staining can be performed, including against VGLUT2, synaptophysin, synapsin, gephyrin, and PSD95 proteins to distinguish pre- and post-synaptic compartments as well as excitatory and inhibitory synapses. D. 4-Aminopyridine Model

[0059] Another aspect provides a method to establish 4-aminopyridine (4-AP) model by using 4-AP to induce seizures in the in iPSC-derived neural co-cultures, and method of using such model to evaluate AED candidates.

[0060] Modeling acute seizures in iPSC-derived neural co-cultures by blocking GABA_A receptors constitutes a powerful assay for pre-clinical assessment of AED efficacies. However, a considerable number of AEDs exert their therapeutic effects by directly acting on GABA-mediated currents, such as vigabatrin, topiramate, stiripentol, as well as all benzodiazepines and barbiturates. The mechanism of action of these compounds is therefore likely to interfere with the artificial seizure -stimulus (e.g. PTX) thereby masking disease-relevant therapeutic effects on neuronal activity phenotypes. Consequentially, efficacies of AEDs acting through GABA receptor modulation can be over- or underestimated depending on the receptor binding site and affinities to compete with the stimulation chemical. In order to accurately assess potential antiepileptic efficacies of compounds acting on GABA signaling, we implemented an alternative, GABA receptor-independent acute chemical stimulation of seizure-like activity in the same iPSC-derived neural co-cultures. 4-aminopyridine is a potent potassium channel blocker that produces epileptiform activity in primary rodent neural cultures and causes convulsion in animals. Application of 4- AP is therefore widely used as an in vivo and in vitro seizure model. Exposure of iPSC-derived neural co cultures that develop spontaneous synchronized network activity on microelectrode arrays to 4-AP (e.g., about 10-300, such as about 30 mM or about 100 mM) induced acute hyperactivity. This hyperactivity is characterized by an increased rate of single action potentials as well as an increased frequency of single bursts and network bursts therefore recapitulating hallmarks of seizure-like events. This 4-AP-induced hyperactivity of iPSC-derived neural co-cultures can therefore be used as a GABA receptor modulation independent in vitro model for testing anti-seizure effects of therapeutic interventions.

D. Genetic Seizure and/or Epilepsy Models

[0061] In some embodiments, the neural cells comprise one or more gene variants that are related to epilepsy. Many rare epilepsies are genetically defined, including Dravet syndrome, which arises most commonly from loss-of-function mutations in one copy of the SCN1A gene encoding the sodium channel NaV 1.1. The monogenetic nature of this epilepsy type makes it particularly accessible for in vitro modeling using human stem cell derived neuronal culture systems. In some embodiments the neural cells carry a loss-of-function mutation in SCN1A, and serve as a general model for the screening and evaluation of candidate drugs for Dravet syndrome. In some embodiments the mutation is NaVl.l-p.S1328P, which has previously been shown to display specific functional defects in GABAergic inhibitory neurons. In other embodiments the neural cells carry patient-specific mutations so that they can be characterized as contributing the functional defects in neural cells, or not. These systems may be referred to as means for modeling a seizure or means for modeling a genetically-based seizure.

[0062] In some embodiments, the one or more gene variants that are related to epilepsy is selected from one of genes in Table 1. Wang et al., Seizure, 44: 11-20, January 2017, incorporated by reference.

Table 1. Epilepsy related genes.

AD, autosomal dominant; AR, autosomal recessive; UN, unknown; XL, X-linked; XLD, X-linked dominant; XLR, X-linked recessive.

[0063] In some embodiments, the neural cells are derived from patients with the gene variants.

[0064] In some embodiments, the neural cells are derived from cells genetically engineered to contain the gene variants.

[0065] In some embodiments, the genes variants are in the inhibitory neurons.

[0066] In some embodiments, the genes variants are in the excitatory neurons.

[0067] In some embodiments, the genes variants are in the astrocytes.

IV Compound Screening with Seizure Models [0068] The herein described induced seizure models using the herein describe neural cell co-cultures may be used to screen compounds and compound libraries for anti-seizure or anti-epileptic activity. In some embodiments, an initial screen is conducted using an imaging-based readout, for example, calcium imaging, Those compounds producing a positive signal can then be further characterized using an electrical-based readout. In some embodiments involving imaging-based readouts, the neural cells are seeded and maintained on plates with clear well bottoms, which are suitable for image-based analyses (e.g. high-content imaging, see, for example, Opera Phenix High-Content Screening System from Perkin Elmer). The clear- bottom plates may be pre-coated with a suitable substrate, including without limitation laminin, PEI, PO, PDL, Matrigel, etc. In other embodiments, screening is conducted using an electrical-based readout protocol alone or an imaging -based readout alone, such as calcium imaging.

V System and Kits

[0069] For convenience, the systems described herein, or components thereof, may be provided in kits. In various aspects, the kits can comprise for example cells, reagents, and/or apparatus useful for establishing the neural cell co-cultures provided herein. The kits could include the appropriate additives for providing the simulation, optionally include the cells to be used, which may be frozen, refrigerated or treated in some other manner to maintain viability, reagents for maintaining the neural co-culture system, reagents for measuring the parameters, and software for preparing the data analysis.

EXAMPLES

[0070] The following non-limiting examples are provided for illustrative purposes only in order to facilitate a more complete understanding of representative embodiments now contemplated. These examples should not be construed to limit any of the embodiments described in the present specification.

Example 1

Human Neurons and Cell Culture

[0071] Subtype-specific induced-neurons (iN) were derived from hiPSC, and co-cultured with established primary human astrocyte via proprietary protocols. Zhang et ah, Neuron 78:785-798, (2013); Yang et ak, Nature Methods 14:621-628 (2017); WO2011/091048A1; and WO2017/223052A1, all incorporated by reference in their entireties.

[0072] In brief, iPSCs were grown in mTeSRl medium (StemCell Technologies) on 6-well plates coated with Matrigel® (Coming) following a daily media change routine. For the generation of iN cells, iPSCs were harvested using TrypLE Select (Gibco) for enzymatic dissociation into single cells. Dissociated iPSCs were resuspended in mTeSRl medium containing 10 mM ROCK inhibitor Y-27632 (Stem Cell Technologies). For the generation of excitatory glutamatergic iN cells, iPSCs were infected with lentivirus for constitutive expression of reverse tetracycline transactivator (rtTA) and lentivirus for doxycycline inducible expression Ngn2 and a puromycin selection marker.

[0073] For the generation of inhibitory GABAergic iN cells, iPSCs were infected with lentivirus for constitutive expression of rtTA and lentivirus for doxycycline inducible expression of Ascii and a puromycin selection marker as well as Dlx2 and a hygromycin selection marker. Infected iPSCs for excitatory and inhibitory iN generation were separately seeded into Matrigel® -coated 6-well plates and further expanded in mTeSRl medium until reaching approximately 70% confluency. At the day of differentiation induction (day 0), media was changed to mTeSRl medium supplemented with 2 pg/ml doxycycline to activate transgene expression. The next day (day 1), iPSCs were dissociated again into single cells using TrypLE and reseeded in N3 medium (DMEM/F12, N2 supplement, B27 supplement, insulin [10 pg/ml], non-essential amino acids) containing doxycycline and 5 pg/ml puromycin (excitatory iN cells) or doxycycline, 5 pg/ml puromycin and 140 pg/ml hygromycin (inhibitory iN cells) on Matrigel® -coated plates. At days 2 and 3, complete media changes were performed. At day 4, iN cells were ready for co seeding with human primary astrocytes on MEA plates.

[0074] Primary human astrocytes (ScienCell) were grown in Astrocyte Basal Medium (ScienCell) containing 10% fetal bovine serum (Gibco) and Astrocyte Growth Supplement (ScienCell) on 15-cm dishes coated with poly-L-lysine. After expanding for two passages, primary astrocytes were used for neural co cultures on MEAs.

[0075] Microelectrode Array (MEA)

[0076] Microelectrode array (MEA) plates (Axion, classic-48 MEA plates) were coated with polyethyleneimine and laminin sequentially. Immature but committed excitatory and inhibitory induced neurons (NeuCyte, SynFire Co-Culture Kit (MEA)) were seeded on coated plates at a 70/30 ratio together with astroglial cells in neural medium on coated MEA plates. Neural medium was composed ofNeurobasal- A medium, B27 supplement, glutamine, NT3, doxycycline and mouse laminin supplemented with 5% FBS and ROCK inhibitor Y-27632. Cell density were 140K, 60K, and 70K per well for excitatory iNs, inhibitory iNs, and astrocytes respectively. Half-medium change with neural medium supplemented with 1% FBS and AraC was performed every other day for the first three weeks. After 1 week, half-media changes with neural medium supplemented with 1% FBS were perormed once every 3 days and full-media change was performed on day 21. All compound testing was scheduled on day 23 after seeding to avoid perturbation caused by media-change. [0077] Microelectrode Array Recording

[0078] Electrophysiological recordings were performed on Axion Maestro 768 channel amplifier systems with specialized 48-well MEA-plates (Axion, classic-48). Each well has an array of 16 electrodes (4x4) that record simultaneously. Acquisition rate was 12.5kHz, and analog filter was 0.2-5kHz band-pass. No digital filter was applied online during recording or offline during analysis. Concurrent to raw-data acquisition, local neuronal events were detected inline using adaptive threshold-search spike detection was performed using AxIS adaptive threshold crossing method with a threshold of 8 x standard deviation of system baseline noise for each channel and a pre- and post-spike duration of 0.84 ms and 2.16 ms, respectively. Spike lists were stored in separate files for later analysis (see below). The threshold for event detection was 8 standard-deviation of system baseline.

[0079] Before every recording session, MEA plates were acclimated for 20 minutes in the recording chamber of Maestro system. Environmental variables were controlled at 37°C and 5% CO2. To monitor the maturation of induced neuron culture, 15 -minute recordings were performed weekly on 7-, 14- and 21- days after seeding. These ontogeny data were used for quality control purpose.

[0080] For compound testing, an in-house established 15-30-90 recording scheme (Figure 3) was used. In brief, basal activity was recorded for 15 minutes, followed by the addition of PTX challenge to induce seizure-like activities. These elevated activities were monitored and let stabilized for 30 minutes. At which point, testing compounds were applied at various concentrations, and their effects were recorded for 90 minutes. Upon the completion of recording, cellular health of dosed neural cultures was quantitatively assessed by measuring release of lactate dehydrogenase (LDH)-dependent conversion of tetrazolium salt to red formazan. After MEA recording in the presence of compound/solvent, 50 pi of culture media was removed from each well and transferred to a 96-well plate to determine LDH concentrations using the CytoTox 96® Non-Radioactive Cytotoxicity Assay kit (Promega). Fresh culture media (50 mΐ) and media from wells with lysed neural cultures served as blank and cellular LDH measurements, respectively. For LDH detection, 50 pL of reconstituted substrate mix was added to each 96-well and incubated in the dark, at room temperature for 30 minutes. After the reaction was stopped by adding 50 mΐ of Stop Solution, raw LDH release was measured as single wavelength absorbance of converted formazan product at 490 nm using a Synergy TM 2 Microplate Reader (BioTek). The percentage of LDH release was calculated from the ratio of media-blanked release and cellular LDH. Conditioned medium was sampled, and standard LDH (lactate dehydrogenase) fluorescence assay was performed to evaluate cell damage (Released LDH levels). These fluorescence readings are positively correlated to compound cytotoxicity and were used to validate interpretations derived from electrophysiological results. Example 2

In vitro PTX-Seizure Model

[0081] Chemical-induced seizures are widely employed in in vivo and in vitro systems to model acute status epilepticus (single episode of epileptic seizure). Picrotoxin (PTX), along with metrazol, pilocarpine, and kainic acid, are some of the most common convulsants used.

[0082] Current anti-epileptic drug (AED) screening is largely limited to rodent models which are high in cost and low in throughput despite considerable translational validity. No clinically relevant alternatives are available from in vitro models. We have developed and validated human-relevant in vitro seizure models for high-throughput, low-cost anticonvulsant screening. This new method can provide a translationally relevant assessment of human neuronal network physiology for high-throughput, naive drug screening as well as for efficacy and neurotoxicity testing during AED candidate optimization.

[0083] Gamma aminobutyric acid type A receptors (GABA_ARS) are ligand-gated chloride ion channels and represent the primary mediators of inhibitory neurotransmission, therefore playing an essential role in orchestrating central nervous system (CNS) function. Blocking of GABA_ARS leads to disinhibition of neuronal circuits and excessive neuronal firing. Picrotoxin is a potent antagonist of GABA_ARS and can induce severe tonic-clonic seizures upon application in vivo. The hPSC-derived neural co-cultures described herein comprise three cell types, excitatory glutamatergic iN cells, inhibitory GABAergic iN cells, and astrocytes derived from primary glial cells. These cultures developed spontaneous electrical activity within 12 days after co-seeding on MEA plates and developed coordinated network activity within 21-28 days. As developing coordinated network activity represents a functional integration of glutamatergic and GABAergic synaptic transmission, inhibition of GABA_AR is expected to lead to a disturbance of the excitatory/inhibitory balance within the network with a disinhibition of excitatory neurons resulting in increased network activity.

[0084] Excitatory and inhibitory induced neurons were produced as described in Example 1 and co-seeded with primary human astrocytes on PEI-laminin-coated 48-well MEA plates in neural media (Neurobasal- A, B27⁺ supplement, glutaMAX, laminin, NT3, FBS, doxycycline) supplemented with ROCK inhibitor. For each MEA well, a total of 140K excitatory neurons, 60K inhibitory neurons and 70K astrocytes per seeded. Neural co-cultures were maintained grown in neural media supplemented with Ara-C for 28 days. At this time baseline neural network activity was recorded for 15 minutes, following a 20 minutes equilibration period in the environmental control unit of the MEA readout device (Maestro, Axion). After baseline recording and equilibration, 10 mM PTX was applied to sample wells which induced acute neuronal hyperactivity characterized by an increased rate of single action potentials as well as an increased frequency of network bursts and an elongation of network burst duration compared to baseline activity and vehicle- treated control wells. The induced acute increase in multiple activity parameters resembled seizure-like neural activity during status epilepticus and persisted for at least 90 minutes recording. This hyperactivity was counteracted in a dose-dependent manner by benchmark AEDs showing and high correlation between in vitro efficacy data and human plasma levels of these AEDs at therapeutic dose (Figure 6). PTX-induced hyperactivity in hPSC-derived neural co-cultures can therefore be regarded as a simplified proxy for seizure events in patients and be used as an in vitro model for testing anti-seizure effects of therapeutic interventions. As an alternative to PTX, other GABA_AR antagonists such as bicuculline (BIC) or pentylenetetrazol (PTZ) have demonstrated similar responses and can also be used to induce acute seizure like activity in hPSC-derived neural co-cultures.

Example 3 Model Validation

[0085] A total 18 antiepileptic drugs were tested in the in vitro PTX chemical -induced seizure model (Table 1), about 30 minutes after the PTX challenge. None of the tested compounds has primary targets in GABAergic pathways, except stiripentol. Any anti-convulsant property depending on activating/enhancing GABAergic route may not suitable to evaluate in the PTX chemical-induced seizure model.

[0086] Stock solutions of the AEDs were prepared at 100 mM in DMSO, and aliquoted at 30 pL per vial and stored at -20 °C. Small aliquots prevented multiple freeze-thaw cycles and ensured the quality of test compounds. For all experiments, PTX (10 pM) was used to induce seizure-like events in induced neuron cultures, which is here referred to as the challenge or the challenge compound. PTX-treated wells reached a final DMSO concentration of 0.01%. Accordingly, designated vehicle control wells were treated at this point with DMSO diluted in culture media to reach a final concentration of 0.01%.

[0087] On the day of testing, AED stock solutions or vehicle (DMSO) were serially diluted with culture media. 10 pL of diluted solution was added to culture wells (after challenge had been initiated), so that final working concentration of the AED was between 0.1 and 1000 pM. Depending on the potency of test compounds reported in the literatures, dose-range was selected from one of the following three: low (0.1, 0.3, 1, 3, 10 pM), standard (1, 3, 10, 30, 100 pM), or high (10, 30, 100, 300, 1000 pM). At least 5 concentrations were tested for every AED, and every concentration had 6 technical (well) replicates (Figure 2), with each dose adding DMSO to the culture media to reach final concentration of 0.1% in the well. DMSO was also added to the vehicle control wells instead of AED to reach 0.1% concentration. [0088] Data analysis

[0089] To ensure the quality and consistency of the assay, wells showing low basal activity were excluded from analysis. “Low activity” is defined as “less than 8 active electrodes per well”, and the criterion for an active electrode is having more than 5 events per minutes. Definitions were based on empirical testing and recommendations from Maestro’s manufacturer.

[0090] Compound efficacy was determined by fitting data with dose-response curve where EC 50, E_max and Hill coefficient were obtained. EC50 calculated from weighted mean firing rate was used to compare potency across testing compounds; corresponding published efficacious human total plasma concentrations were also included for correlation plotting (Table 2).

Table 2 AEDs assessments

Compound 003 and its usefulness for treating epilepsy, etc. is disclosed in U.S. Provisional

Patent Application / _ (Atty. Docket number 1106950.00011), titled “Anti-epileptic

Pharmaceutical Compositions and Use Thereof’, and filed on date even with this application and which is incorporated by reference in its entirety).

**Plasma concentration achieved with a 20 mg/kg dosage in cynomolgus monkeys; efficacy not evaluated.

[0091] Correlation Plot of AED assessment

[0092] The data in Table 2 for the known AEDs was plotted in Figure 6. The Human Total Plasma concentrations for each drug represent the plasma concentration produced by the recommended dosage according to the Prescribing Information for each of these approved drugs. These values serve as surrogates of the in vivo effective concentration of each of the AEDs. PTX EC50 is the weighted mean firing rate EC50 as determined in the in vitro PTX-Seizure Model. That PTX EC50 is generally proportional to Human Total Plasma concentration validates that the model is predictive of efficacy (anti-seizure activity) and the line fit to the data allows one to predict the effective concentration of an AED candidate in humans from the AED candidate’s PTX EC50. As compared to the other AEDs used in this validation study, safinamide and (-)cannabidiol have unusually high protein binding and brain permeability, that is, the ratio of plasma concentration to brain concentration is unusually low. This explains the deviation below the line in Figure 6 of these two compounds.

Example 4

Phenotypic analysis in the kainic acid model

[0093] High content imaging is used to characterize synaptic structure and remodling in the kainic acid chronic seizure model. Synaptic phenotypes are identified as phenotypic endpoints for drug screening assays. The utility of previously established reporter constructs that are based on fusion proteins between specific synaptic markers and fluorescent proteins, including PSD95-EGFP, TEAL-gephyrin, or synaptophysin-EGFP are tested. Screening assays based on high-content imaging are validated with the panel of benchmark AEDs used above in the context of MEA assays. Major electrophysiological and morphological phenotypes observed in this model demonstrate concordance between human iN co-cultures and in vivo models.

Example 5

Model for Dravet syndrome

[0094] The patient-specific SCN1A mutation Navi . l-p.S1328P is introduced using commercially available CRISPR/Cas9 technology (Synthego). In the case of the Navl.l-p.S1328P mutation, a repair template containing the corresponding DNA variant (c.3982T>C) is provided for insertion during the DNA repair process triggered by site-specific nuclease activity. Individual iPSC clones are isolated and analyzed by genomic PCR and DNA sequencing for the presence of the heterozygous SCN1A patient mutation as previously described. Upon successful identification of the mutant clones, individual iPSC lines are expanded for genomic quality control by karyotyping, array comparative genomic hybridization, and whole exome sequencing. These studies allow the detection and exclusion of iPSC clones with genomic aberrations other than the targeted SCN 1 A mutation, which might be introduced during the cell engineering process. Finally, excitatory and inhibitory iN are generated according to the herein disclosed protocols. Navl.lp.S1328P-expressing mutant co-cultures and isogenic controls (from the parent iPSC line used for introduction of the mutation) are sparsely labelled with fluorescent TdTomato and EGFP in order to visually distinguish excitatory and inhibitory neurons. Co-cultures grown on glass coverslips are used in patch- clamp electrophysiology experiments to replicate the phenotypes previously reported specifically for GABAergic inhibitory neurons, as well as for identification of additional, potentially useful phenotypes for drug testing. Electrophysiological defects described for patient-derived human GABAergic neurons in culture included a reduction in voltage-dependent sodium currents. These previously reported neurons were selected for analysis based on expression of a lentiviral Dlxil/2b-GFP reporter. Given that the protocol for iN generation includes over-expression of the ventral forebrain transcription factor DLX2, orthologous regulatory sequences of which (from rodent) are part of the Dlxil/2b-GFP reporter, the iN should be highly comparable and therefore should display similar functional defects. In parallel, NavTl-p.S1328P mutant co-cultures on MEA plates are grown to assess the impact of the mutation on network activity.

[0095] Once phenotypes are identified, a selected panel of AEDs including those constituting current standard care for Dravet syndrome, are tested in the MEA assay or by patch-clamp electrophysiology. This panel includes clobazam and stiripentol, which have shown some efficacy in Dravet syndrome patients, as well as AEDs, which have shown no efficacy in the clinical setting or in Dravet syndrome animal models. Sodium channel blockers including lamotrigine, which have shown adverse effects, are also included in the assays, consistent with our current understanding of selective sensitivity to Navl.l insufficiency in inhibitory neurons as opposed to excitatory neurons. The assays using this model successfully distinguish those drugs known to provide therapuetic benefit from those that do not.

Example 6

Characterization of Compound 003 in the iN Dravet model

[0096] Compound 003 has successfully passed through the ETSP’ s animal model pipeline and importantly, demonstrated efficacy in the 6 Hz mouse model considered for treatment-resistant epilepsy. It is therefore a promising candidate not only for broad-spectrum anti -epileptic activity, but also for particularly refractory forms of rare epilepsies, such as Dravet syndrome. These results are recapitulated by testing Compound 003 in MEA experiments on the chemically induced seizure models described herein, establishing a dose response for calculation of an in vitro EC50 as described above. To elucidate the Compound 003 mechanism of action, an optimal (low) efficacious dose is selected and to test Compound 003 in competition experiments with modulators of specific cellular pathways. Target pathways for testing include sodium channels, monoamine oxidase, and androgen receptor signaling. The results establish a solid foundation for experimental design and data interpretation for Compound 003 efficacy in the Dravet syndrome model.

[0097] To directly evaluate Compound 003 efficacy on Dravet syndrome, Compound 003 is tested in the genetic model described herein using the NavTl-p.S1328P-expressing mutant iN co-cultures. MEA experiments are performed to establish a dose response curve and EC50. These measures are then directly compared to current standard of care AEDs, such as clobazam. A similar approach is taken for high-content imaging assays, to evaluate morphological phenotypes. Finally, Compound 003 efficacy is evaluated using patch-clamp electrophysiology. The data generated with this model is compared with data from a study of Compound 003 in a mouse model for Dravet syndrome to further validate it. Taken together, these experiments provide a comprehensive basis to support use of Compound 003 for treatment of Dravet syndrome.

[0098] In closing, it is to be understood that although aspects of the present specification are highlighted by referring to specific embodiments, one skilled in the art will readily appreciate that these disclosed embodiments are only illustrative of the principles of the subject matter disclosed herein. Therefore, it should be understood that the disclosed subject matter is in no way limited to a particular methodology, protocol, and/or reagent, etc., described herein. As such, various modifications or changes to or alternative configurations of the disclosed subject matter can be made in accordance with the teachings herein without departing from the spirit of the present specification. Lastly, the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the present invention, which is defined solely by the claims. Accordingly, the present invention is not limited to that precisely as shown and described.

[0099] Certain embodiments are described herein, including the best mode known to the inventors for carrying out the invention. Of course, variations on these described embodiments will become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventor expects skilled artisans to employ such variations as appropriate, and the inventors intend for the present invention to be practiced otherwise than specifically described herein. Accordingly, this invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described embodiments in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.

[00100] Groupings of alternative embodiments, elements, or steps of the present invention are not to be construed as limitations. Each group member may be referred to and claimed individually or in any combination with other group members disclosed herein. It is anticipated that one or more members of a group may be included in, or deleted from, a group for reasons of convenience and/or patentability. When any such inclusion or deletion occurs, the specification is deemed to contain the group as modified thus fulfilling the written description of all Markush groups used in the appended claims.

[00101] Unless otherwise indicated, all numbers expressing a characteristic, item, quantity, parameter, property, term, and so forth used in the present specification and claims are to be understood as being modified in all instances by the term “about.” As used herein, the term “about” means that the characteristic, item, quantity, parameter, property, or term so qualified encompasses a range of plus or minus ten percent above and below the value of the stated characteristic, item, quantity, parameter, property, or term. Accordingly, unless indicated to the contrary, the numerical parameters set forth in the specification and attached claims are approximations that may vary. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of the claims, each numerical indication should at least be construed in light of the number of reported significant digits and by applying ordinary rounding techniques. Notwithstanding that the numerical ranges and values setting forth the broad scope of the invention are approximations, the numerical ranges and values set forth in the specific examples are reported as precisely as possible. Any numerical range or value, however, inherently contains certain errors necessarily resulting from the standard deviation found in their respective testing measurements. Recitation of numerical ranges of values herein is merely intended to serve as a shorthand method of referring individually to each separate numerical value falling within the range. Unless otherwise indicated herein, each individual value of a numerical range is incorporated into the present specification as if it were individually recited herein.

[00102] The terms “a,” “an,” “the” and similar referents used in the context of describing the present invention (especially in the context of the following claims) are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g. , “such as”) provided herein is intended merely to better illuminate the present invention and does not pose a limitation on the scope of the invention otherwise claimed. No language in the present specification should be construed as indicating any non-claimed element essential to the practice of the invention.

[00103] Specific embodiments disclosed herein may be further limited in the claims using consisting of or consisting essentially of language. When used in the claims, whether as filed or added per amendment, the transition term “consisting of’ excludes any element, step, or ingredient not specified in the claims. The transition term “consisting essentially of’ limits the scope of a claim to the specified materials or steps and those that do not materially affect the basic and novel characteristic(s). Embodiments of the present invention so claimed are inherently or expressly described and enabled herein.

[00104] All patents, patent publications, and other publications referenced and identified in the present specification are individually and expressly incorporated herein by reference in their entirety for the purpose of describing and disclosing, for example, the compositions and methodologies described in such publications that might be used in connection with the present invention. These publications are provided solely for their disclosure prior to the filing date of the present application. Nothing in this regard should be construed as an admission that the inventors are not entitled to antedate such disclosure by virtue of prior invention or for any other reason. All statements as to the date or representation as to the contents of these documents is based on the information available to the applicants and does not constitute any admission as to the correctness of the dates or contents of these documents.

Claims

Claims

1. A method for assessing anti-seizure activity of a compound, comprising:

(1) providing a human neural co-culture comprising: in vitro differentiated functional excitatory and inhibitory human neuronal cells; and astrocytes, wherein said co-culture provides network activity, and wherein said network activity comprises synchronous network bursts;

(2) contacting said co-culture with an agent, wherein said agent modulates said network activity of said co-culture;

(3) contacting said co-culture with a compound;

(4) measuring effects of said compound on said network activity, thereby assessing anti-seizure activity of said compound.

2. The method of claim 1, wherein said neuronal cells comprise one or more of GABAergic inhibitory neurons, glutamatergic excitatory neurons, dopaminergic excitatory neurons, and serotonergic neurons.

3. The method of claim 1, wherein said co-culture comprises GABAergic inhibitory neurons and glutamatergic excitatory neurons.

4. The method of claim 1, wherein the ratio between said GABAergic inhibitory neurons and said glutamatergic excitatory neurons is between about 20:80 and about 70:30.

5. The method of claim 1, wherein said in vitro differentiated functional human neuronal cells are derived by the method comprising: contacting a population of non-neuronal human cells with neuron reprogramming factors (NR), or agents to activate NR factors, wherein the NR factors are selected from the group consisting of: Neurogenin, Ascl ,NeuroD, Bm2, Bm3a, Emx, Cux2, Tbrl, Satb2, Dlxl/2/5, Nkx2.1, Nkx2.2,

Lhx2/3/6/8, Sox2, Foxgl, Ctip2, Hb9, Isll/2, Klf7, Gata2, Foxa2, Fmxlb, Ptx, FEV, Fmxl, Foxa2,

Nurrl, Pitx3, and En for a period of time sufficient to reprogram said non-neural cells, wherein a population of functional human neuronal cells is produced.

6. The method of claim 5, wherein said non-neuronal cells are pluripotent cells.

7. The method of claim 5, wherein the non-neuronal cells are somatic stem cells.

8. The method of claim 5, wherein the pluripotent cells are induced pluripotent stem cells.

9. The method of claim 2, wherein said GABAergic inhibitory neurons are generated by contacting a population of non-neuronal human cells with neuron reprogramming factors Ascii and Dlx2.

10. The method of claim 2, wherein said GABAergic inhibitory neurons are generated by contacting a population of non-neuronal human cells with neuron reprogramming factor Ngn2.

11. The method of claim 1, wherein said human glial cells are derived by the method comprising: isolating glial cells from primary brain tissue.

12. The method of claim 1, wherein said human glial cells are derived by the method comprising: contacting a population of non-glial cells with one or more of whole serum, single serum components, insulin, BMP-inhibitor, TGF-beta-inhibitor, growth factors and morphogens such as EGF, CNTF, BMP2/4, and/or transcription factors such as Nfia, Nfib, Sox9, and Hes for a period of time sufficient to reprogram or step-wise differentiate non-glial cells to astroglial cells.

13. The method of claim 1, wherein said human astrocytes are derived by the method comprising: contacting a population of human pluripotent stem cells with an effective dose of a reprograming system comprising Nfia or Nfib for a period of time sufficient to reprogram said pluripotent cells, where a population of induced human astrocytes is produced.

14. The method of claim 13, wherein said human pluripotent stem cells further overexpress Sox9.

15. The method of claim 1, wherein said agent comprises a GABA receptor antagonist.

16. The method of claim 15, wherein said GABA receptor antagonist comprises picrotoxin.

17. The method of claim 1, wherein said agent comprises kainic acid.

18. The method of claim 1, wherein said agent comprises 4-AP.

19. The method of claim 1, wherein said network activity is assessed using one or more of: total number of spikes per recording period; mean firing rate of spikes; inter-spike interval; total number of bursts per recording period; burst frequency; number of spikes per burst; burst duration; inter-burst interval; the proportion of spikes occurring within a burst; total number of network bursts; network burst frequency; number of spikes per network burst; network burst duration; inter-network-burst interval; inter-spike interval within network bursts; the proportion of bursts occurring within a network burst; and cross-correlation of detected spikes between all electrodes per well.

20. The method of claim 1, wherein said network activity is measured using a microelectrode electrode array (MEA).

20. The method of claim 1, wherein said network activity is measured using ratio-metric measurements based on calcium imaging.

21. The method of claim 1, wherein induced seizure-like activity is characterized by one or more of the following parameters: increased mean firing rate of spikes; increased total number of network bursts; increased network burst frequency; increased total number of bursts per recording period; increased burst frequency; increased burst duration; increased network burst percentage; decreased inter-network-burst interval; decreased inter-spike interval; increased cross-correlation of detected spikes between all electrodes per well.

23. The method of claim 1, wherein modulation of network activity has been validated by measuring the effect of at least 10 known anti-seizure drugs.

24. The method of claim 22, comprising determining an EC50 of the compound according to one or more of said parameters is determined.