TY - JOUR
AU - Haag, Daniel
AB - Abstract Assessment of neuroactive effects of chemicals in cell-based assays remains challenging as complex functional tissue is required for biologically relevant readouts. Recent in vitro models using rodent primary neural cultures grown on multielectrode arrays allow quantitative measurements of neural network activity suitable for neurotoxicity screening. However, robust systems for testing effects on network function in human neural models are still lacking. The increasing number of differentiation protocols for generating neurons from human-induced pluripotent stem cells (hiPSCs) holds great potential to overcome the unavailability of human primary tissue and expedite cell-based assays. Yet, the variability in neuronal activity, prolonged ontogeny and rather immature stage of most neuronal cells derived by standard differentiation techniques greatly limit their utility for screening neurotoxic effects on human neural networks. Here, we used excitatory and inhibitory neurons, separately generated by direct reprogramming from hiPSCs, together with primary human astrocytes to establish highly functional cultures with defined cell ratios. Such neuron/glia cocultures exhibited pronounced neuronal activity and robust formation of synchronized network activity on multielectrode arrays, albeit with noticeable delay compared with primary rat cortical cultures. We further investigated acute changes of network activity in human neuron/glia cocultures and rat primary cortical cultures in response to compounds with known adverse neuroactive effects, including gamma amino butyric acid receptor antagonists and multiple pesticides. Importantly, we observed largely corresponding concentration-dependent effects on multiple neural network activity metrics using both neural culture types. These results demonstrate the utility of directly converted neuronal cells from hiPSCs for functional neurotoxicity screening of environmental chemicals. multielectrode arrays, cell-based assays, induced pluripotent stem cells, induced neurons, neurotoxicity, neural network activity, neurotoxicity screening To date, thousands of chemicals with human exposure are still lacking toxicity information (Judson et al., 2009; Kavlock et al., 2009). A series of reports and legislation from the National Academy of Sciences (National Research Council [NRC], 2007), the implementation of the Registration, Evaluation, Authorization, and Restriction of Chemicals regulation (EU 2006), and the 2016 Lautenberg Chemical Safety for the 21st Century Act further highlight the immediate need for in vitro alternatives to standard animal testing to increase throughput of toxicity assessment (Collins et al., 2008). Consistent with these recommendations, the Administrator of the U.S. Environmental Protection Agency (U.S. EPA) recently issued a memo stating that “The EPA will reduce its requests for, and funding of, mammal studies by 30% by 2025 and eliminate all mammal study requests and funding by 2035” (Wheeler, 2019). This movement towards in vitro testing to assess hazard and use the data in the application of modern approaches to risk assessments (NRC, 2007) is driving the U.S. Tox21 collaboration (Thomas et al., 2018; Tice et al., 2013) as well as U.S. EPA programs (Thomas et al., 2019) that aim to develop cell-based quantitative screening assays for efficiently testing chemicals for potential health effects. Besides being low throughput in nature and requiring living animals current neurotoxicity evaluation according to regulatory guidelines is generally limited to behavioral observations (eg, Functional Observational Battery), neurophysiological measurements (eg, visual or auditory-evoked potentials), and tissue pathology and do not provide detailed mechanistic information (Buschmann, 2013). In contrast, cell-based models can be subjected to specified readouts and thus provide mechanistic insight into adverse effects (Sun et al., 2012). The reduced complexity and cellular accessibility of neural in vitro cultures permits direct measurements of various aspects of neuronal physiology (Xu et al., 2016) so that neuroactive effects that perturb neuronal function by interfering with a distinct signaling pathways can be readily distinguished from general or neuro-specific cytotoxicity (Wallace et al., 2015). Therefore, in vitro assays should be very useful to screen and prioritize chemicals for potential neurotoxicity and/or adverse neuroactive effects (Conolly et al., 2017). For developmental neurotoxicity and structural neurotoxicity testing, in vitro assays mostly examine relative numbers, morphological changes, or migration of individual cells (Barenys et al., 2017; Baumann et al., 2016; Clarke et al., 2017; Hofrichter et al., 2017), In contrast, assessment of potentially adverse neuroactive effects in fully developed brains requires measurements of neuronal function resulting from highly organized neuronal and glial interactions. Intracellular voltage/current and patch-clamp recording of brain slices and primary neural cultures from rodents represent the gold standard for electrophysiology analyses. Using imaging approaches, Ca2+ homeostasis or voltage-sensitive dyes can also provide information on chemical disruption of function. However, single-cell electrophysiology (current/voltage/patch clamp) is not suitable for medium- or high-throughput testing on mature neurons as it lacks the throughput and/or broad-based sensitivity to screen compounds with diverse modes of action (MOAs). In contrast, imaging approaches, while high throughput, lack the temporal resolution of electrophysiology approaches and therefore may be insensitive to subtle alterations in function such as changes in neuronal burst structure. Multielectrode arrays (MEAs) measure local field potentials of electrically active tissue grown on a grid of electrodes which provides multiple advantages for screening assays. First, in contrast to patch-clamping, the extracellular and nondisruptive recording of neuronal activity using MEAs does not constitute an endpoint measurement per se but allows the analyses of acute and chronic effects of single or repeated chemical exposure within the same sample. Moreover, the standardized parallel measurements across wells enable for (currently) up to 96-well format assays, thus approaching high-throughput levels. Finally, the cell population-based readout provides a comprehensive assessment of synchronized neuronal network activity, which can be regarded as a functional integration of underlying intra- and intercellular processes (including synaptic transmission, signal conductance, excitability for firing action potentials, electric membrane properties, and morphology). Standard recordings of neuronal activity support the parallel acquisition of over 30 parameters, to generate informative neuroactivity profiles upon exposure to test substances. Consequentially, primary rat cortical cultures grown on MEAs have successfully being used for neurotoxicological analyses of pesticides and pharmaceutical compounds (Bradley et al., 2018; Strickland et al., 2018; Vassallo et al., 2017; Wallace et al., 2015). Importantly, the effects on neural network activity are largely distinct between different substance classes and underlying mechanisms which are reflected in multiple metrics describing different characteristics of network architecture in a quantitative manner. Rodent models are important for physiological and molecular evaluation of neurotoxic effects. However, interspecies differences are increasingly being recognized as a major limitation for predicting adverse effects of human toxicant exposure, particularly for the central nervous system (CNS; Baumann et al., 2016; Gassmann et al., 2010; Legradi et al., 2018; Ongur and Price, 2000; Pak et al., 2015). Moreover, the recently passed Lautenberg Chemical Safety for the 21st Century Act amending the Toxic Substance Control Act calls for the use of human-based alternative models wherever possible (United States. Congress. Senate. Committee on Environment and Public Works, 2015). The use of human-induced pluripotent stem cells (iPSCs) and derivatives constitutes a powerful tool for toxicity evaluation and translational medical research as documented by a rapidly growing number of publications (Arber et al., 2017; Ardhanareeswaran et al., 2017; Brawner et al., 2017; Cobb et al., 2018; Devine and Patani, 2017; Grainger et al., 2018; Klatt et al., 2019; Prince et al., 2019; Silva and Haggarty, 2019; Tukker et al., 2018). The possibility to generate a large variety of specified cell types in a dish, combined with scalable readout techniques, allows in vitro screening of functional human tissue following toxicant exposure. The vast amount of available protocols to generate neural cell types from iPSCs through stepwise differentiation and efficient enrichment strategies as well as the derivation of self-organizing cerebral organoids provide a potential toolbox for these applications (Chambers et al., 2009; Douvaras and Fossati, 2015; Krencik and Zhang, 2011; Lancaster et al., 2013; Pamies et al., 2017; Pasca, 2018; Sloan et al., 2017; Wang et al., 2013). However, standard differentiation protocols can produce heterogeneous target cell populations due to epigenetic variability in local tissue domains during early steps of differentiation (eg, during embryoid body formation) and differential drift towards other lineages (Bauwens et al., 2008; Giobbe et al., 2012). Specific cell types that are expanded from progenitor pools often do not exhibit a uniform maturation level or functional profile (Hu et al., 2010). Moreover, genetic, epigenetic, and transcriptional variances between iPSCs of individual donors can considerably impact differentiation timing and resulting cell ratios (Cahan and Daley, 2013; Kim et al., 2010). Particularly 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 robust toxicological assessment. In contrast, direct reprogramming of iPSCs into defined neural cell types through exogenous expression of fate-determining transcription factors (TFs) utilizes dominant driving forces (Wapinski et al., 2013) which largely overcome or circumvent these issues and increased reproducibility of the target cell identity independent of the starting cell line. Pioneering this approach, the labs of Marius Wernig and Thomas Südhof successfully developed protocols to directly convert hiPSCs into induced neuronal (iN) cells that exhibit mature electrophysiology phenotypes and form functional synapses within only 2 weeks of culture period (Zhang et al., 2013). Surprisingly, a single neurogenic TF (Ngn2) was sufficient to rapidly generate excitatory glutamatergic neurons from human pluripotent stem cells resulting in a homogenous population of highly functional iN cells that resemble cortical layer II/III neurons (Zhang et al., 2013). Moreover, by applying different combinations of TFs, neuronal conversion could be efficiently driven into distinct subtypes such as gamma amino butyric acid (GABA)ergic interneurons (Yang et al., 2017). Here, the combined expression of Ascl1 and Dlx2 exclusively produced inhibitory iN cells that expressed calbindin, calretinin, or somatostatin (SOM) in a mostly nonoverlapping manner, while a smaller fraction of the generated cells was positive for parvalbumin (PV), cholecystokinin, or neuropeptide Y (NPY; Yang et al., 2017). Most importantly, this method produced consistent populations of distinct neuronal cell types at a relatively uniform maturation stage that quantitatively exhibit the same properties. This provides considerable advantage over conventional differentiation protocols regarding specification and reproducibility of neural tissue for in vitro assays. Here, we describe a new pure human neural coculture system consisting of iPSC-derived excitatory and inhibitory iN cells in a defined ratio together with primary fetal astrocytes. To evaluate whether these neural cocultures system is suitable for in vitro assessment of adverse neuroactive effects of chemical compounds we characterized the development of synchronized network activity in comparison to rat primary cortical cultures using multi-well MEAs. Finally, we tested a set of 9 chemicals with well-established neuroactive and neurotoxic effects, including 4 GABAA receptor antagonists, the fungicide/biocide tributyltin, and 4 insecticides representing types I and II pyrethroids, and 3 negative control compounds. We observed specific alterations of neural network activity for the different compound classes compared with control substances with an overall large correspondence between responses in human and rat cultures. MATERIALS AND METHODS Human neural cell cultures Glutamatergic excitatory and GABAergic inhibitory-iN cells were generated from hiPSCs as described previously (Yang et al., 2017; Zhang et al., 2013). Briefly, iPSCs from a female donor (hiPSC-C120 were maintained feeder-free in mTeSR1 medium (Stemcell Technologies)) on matrigel (Corning). At the day of infection cells were dissociated using accutase (Innovative Cell Technologies) and infected with lentiviruses for doxycycline-inducible (tet-on) expression of neurogenic TFs. For the generation of glutamatergic excitatory iN cells, dissociated iPSCs were infected with lentiviruses carrying Ngn2 (FUW-TetO-Ngn2-T2A-puromycin) and reverse tetracycline-dependent transactivator (rtTA) (FUW-rtTA) transgenes. For some imaging downstream applications, a fraction of excitatory iN cells were coinfected with lentivirus for tdTomato expression (FUW-TetO-tdTomato) in separate wells. For the generation of GABAergic inhibitory iN cells, dissociated iPSCs were infected with lentiviruses carrying Ascl1 (FUW-TetO-Ascl1-T2A-puromycin) andDlx2 (FUW-TetO-Dlx2-IRES-hygromycin), and rtTA (FUW-rtTA) transgenes. Infected iPSCs were plated in mTeSR1 supplemented with 10 µM Y27632 on matrigel and left to grow until 85–90% confluency. Subsequently, infected cells were dissociated into single cells using accutase and resuspended in N2/B27/insulin [10 µg/ml]/Dulbecco’s Modified Eagle’s – Medium (DMEM)-F12/NEAA/Y27632 [10 µM] medium containing doxycycline [2 mg/ml] to induce transgene expression (doxycycline was maintained in the iNs medium until the end of the experiments). Starting the day after induction, transgene-expressing cells were selected for 3 days in the presence of the corresponding selection markers (puromycin or puromycin/hygromycin)., After selection, remaining excitatory and inhibitory iN cells were detached from the plates and separately cryopreserved until coseeding with astrocytes. Primary human astroglial cells isolated from fetal cerebral cortices were expanded in astrocyte medium (DMEM, nonessential amino acids, glutaMAX, penicillin [100 U/ml], streptomycin [0.1 mg/ml], and 2% fetal bovine serum (FBS)) on poly-L-lysine-coated plates [2 µg/cm2] for 2 passages before being cryopreserved until coseeding with iN cells. For seeding of iN/glia co-cultures, excitatory and inhibitory iN cells as well as primary astroglial cells were thawed and seeded together in neural medium (neurobasal-A, B-27 supplement, glutaMAX, penicillin [100 U/ml], streptomycin [0.1 mg/ml], NT3 [10 ng/ml], and laminin [200 ng/ml]) supplemented with Y27632 [10 µM], cytosine arabinose (Ara-C, [2 µM], to inhibit astroglia proliferation), and 5% FBS. Half-medium changes were performed every other day until day 7, when media was switched to neural medium supplemented with 1% FBS. For imaging on cover slips, 1.2 × 105 excitatory iN cells, 0.6 × 105 inhibitory cells, and 0.6 × 105 astroglial cells were plated in 500 µl on 12 mm glass cover slips (Knittel Glass, Braunschweig, Germany) precoated with matrigel (Corning). This cell density allowed for good cell adhesion and spreading across the cover slips. For imaging on multi-well plates, around 0.6 × 105 excitatory iN cells, 0.3 × 105 inhibitory cells, and 0.3 × 105 astroglial cells were plated into each well of 96-well optical plate precoated with laminin (Sigma-Aldrich). For analyzing neuronal activity, 48-well MEA plates (Axion M768-KAP-48, Axion Biosystems) were first precoated with 0.1% polyethyleneimine (PEI) at 37°C, washed with water, and subsequently coated with 20 µg/ml laminin. To determine the optimal ratio between excitatory and inhibitory iNs in terms of producing robust neuronal network activity, different cell densities and ratios were seeded per well according to Supplementary Table 1. In each well, iN cells and astrocytes were seeded in a total volume of 50 µl, and 24 h after plating, the neural media was brought to a final volume of 500 µl. The cultures were maintained in a cell culture incubator at 37°C, 95% humidity, and 5% CO2. After the first week in culture, half-medium changes were performed every 3–4 days throughout the duration of the experiments. Based on the cell ratio optimization experiment, the number of iN cells and astrocytes seeded per MEA well for the remaining experiments performed in this study were 1.4 × 105 excitatory cells (70%), 0.6 × 105 inhibitory cells (30%), and 0.7 × 105 astroglial cells. As a complementary experiment to assess the potential contribution of electrical synapses to the spontaneous network activity propagation and the picrotoxin (PTX)-induced hyperactivity observed in the neuro/glia cocultures, the gap junction blocker carbenoxolone (CBX) was applied at a concentration of 25 µM. The excitatory and inhibitory iN cells used for this experiment were derived from a male donor iPSC line (Lonza) using the same protocol described earlier. iN/glia co-cultures were exposed to CBX and PTX at 25 days postplating (DPP) on MEA wells. Rat primary cortical cultures Primary neocortical cultures were prepared from postnatal day 0 male and female Long-Evans rat pups, as detailed in Strickland et al. (2018). All protocols involving animals were approved by the National Health and Environmental Effects Research Laboratory’s Institutional Animal Use and Care Committee. Briefly, neocortices were removed, dissociated with trypsin and resuspended in DMEM containing: GlutaMAX (2 mM) D-glucose (25 mM), sodium pyruvate (1 mM), with the addition of HEPES (10 mM) plus penicillin (100 U/ml), streptomycin (0.1 mg/ml) and 10% horse serum, pH = 7.4. Cortical cells were plated at a seeding density of 1.5 × 105 cells per well on 48-well MEA plates (Axion M768-KAP-48, Axion BioSystems) precoated with 0.05% PEI. Cells were placed in each well via a 25 µl media drop containing laminin administered directly onto the MEA. At 2 h after seeding, 475 μl neurobasal-A medium supplemented with B-27 was added to each well. The cultures were maintained in a cell culture incubator at 37°C, 95% humidity, and 5% CO2 and half-medium changes were performed every 3–4 days throughout the duration of the experiments. Immunofluorescence imaging Human iN/glial co-cultures were washed with PBS and fixed with 4% paraformaldehyde for 10 min at room temperature. Cocultures were then washed twice with PBS and permeabilized with 0.1% Triton X-100 in PBS for 15 min. Blocking of unspecific epitopes was performed using 10% goat serum, 1% bovine serum albumin (BSA), and 0.1% Triton X-100 in PBS for 45 min at room temperature. Primary antibodies were diluted in PBS-T (PBS, 0.1% Tween 20) supplemented with 1% BSA and 0.05% NaN3 for overnight incubation at 4°C. Following 2 washes with PBS-T, secondary antibodies diluted in PBS-T and 1% BSA were applied at room temperature for 1 h. After 2 additional washes, cell nuclei were counterstained with DAPI solution (Dako). For wide-field fluorescence imaging of neural cocultures, cover slips were mounted with vector shield medium (Dako) and images were obtained using a DM6000 B microscope equipped with a DFC365 FX digital camera (all from Leica). For general and subtype assessment of neuronal cells, cocultures were stained for microtubulin-associated protein 2 (MAP2, M4403, Sigma, 1:500), GABA (A2052, Sigma, 1:1000), and vesicular GABA transporter (vGAT, 131103, Synaptic Systems, 1:500, Synaptic Systems), respectively. Confocal fluorescence images were acquired directly from 96-well optical plates using an Operetta CLS High-Content Analysis System (PerkinElmer). Synapse formation was analyzed by staining for presynaptic solute carrier family 17 member 6 (vesicular-glutamate transporter 2 (vGLUT2), 135403, Synaptic Systems, 1:500) and postsynaptic discs large MAGUK scaffold protein 4 (postsynaptic density protein 95 (PSD-95), 6G6-1C9, Thermo Fisher Scientific, 1:500). Cell bodies of tdTomato-expressing excitatory iN cells were stained for tdTomato (tdTomato, ABIN6254170, antibodies online, 1:1000). Abundance of astroglial cells were visualized by glial fibrillary acidic protein (GFAP) staining (AB5804, Millipore, 1:1000). Chemical compounds A total of 9 compounds with well-established neurotoxic effects (Costa, 2015) were used to test and compare responsiveness in human and rodent neural cultures. This included the GABAA antagonists PTX, dieldrin, lindane, and bicuculline (BIC), as well as the biocide compound tributyltin. Moreover, we assessed the type I pyrethroid permethrin, the type II pyrethroids deltamethrin and cypermethrin, as well as esfenvalerate, a pyrethroid with “mixed” effects. Amoxicillin, salicylic acid, and glyphosate served as negative controls (Valdivia et al., 2014). Each compound was tested at 7 concentrations while keeping the indicated solvent volume at 0.1% of the culture medium (Table 1). Table 1. Chemical Compounds Used For Neurotoxicity Testing on MEAs Compound . CAS No. . DTXS ID . Class . Effect . Solvent . Purity(%) . Amoxicillin 26787-78-0 DTXSID303704 Penicillin-class antibiotic Negative control DMSO ≥90 Salicylic acid 69-72-7 DTXSID7026368 Nonsteroidal anti-inflammatory drug Negative control DMSO ≥99 Glyphosate 38641-94-0 DTXSID0034649 Organophosphorus herbicide Negative control water 96 BIC 485-49-4 DTXSID3042687 Isoquinoline alkaloid GABAA antagonists DMSO/ethanol ≥99 PTX 124-87-8 DTXSID7045605 Convulsant alkaloid GABAA antagonists DMSO 98 Lindane 58-89-9 DTXSID2020686 Organochloride insecticide GABAA antagonists Ethanol 99 Dieldrin 60-57-1 DTXSID9020453 Organochloride insecticide GABAA antagonists DMSO ≥95 Permethrin 52645-53-1 DTXSID8022292 Type I pyrethroid insecticide Modulation of VSSCs kinetics DMSO/ethanol ≥91 Deltamethrin 52918-63-5 DTXSID8020381 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Cypermethrin 52315-07-8 DTXSID1023998 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Esfenvalerate 66230-04-4 DTXSID4032667 Type I/II pyrethroid insecticide Intermediate modulation of VSSCs kinetics DMSO/ethanol 98.5 Tributyltin 56-36-0 DTXSID7043950 Fungicide/biocide Oxidative stress DMSO 96 Compound . CAS No. . DTXS ID . Class . Effect . Solvent . Purity(%) . Amoxicillin 26787-78-0 DTXSID303704 Penicillin-class antibiotic Negative control DMSO ≥90 Salicylic acid 69-72-7 DTXSID7026368 Nonsteroidal anti-inflammatory drug Negative control DMSO ≥99 Glyphosate 38641-94-0 DTXSID0034649 Organophosphorus herbicide Negative control water 96 BIC 485-49-4 DTXSID3042687 Isoquinoline alkaloid GABAA antagonists DMSO/ethanol ≥99 PTX 124-87-8 DTXSID7045605 Convulsant alkaloid GABAA antagonists DMSO 98 Lindane 58-89-9 DTXSID2020686 Organochloride insecticide GABAA antagonists Ethanol 99 Dieldrin 60-57-1 DTXSID9020453 Organochloride insecticide GABAA antagonists DMSO ≥95 Permethrin 52645-53-1 DTXSID8022292 Type I pyrethroid insecticide Modulation of VSSCs kinetics DMSO/ethanol ≥91 Deltamethrin 52918-63-5 DTXSID8020381 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Cypermethrin 52315-07-8 DTXSID1023998 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Esfenvalerate 66230-04-4 DTXSID4032667 Type I/II pyrethroid insecticide Intermediate modulation of VSSCs kinetics DMSO/ethanol 98.5 Tributyltin 56-36-0 DTXSID7043950 Fungicide/biocide Oxidative stress DMSO 96 Open in new tab Table 1. Chemical Compounds Used For Neurotoxicity Testing on MEAs Compound . CAS No. . DTXS ID . Class . Effect . Solvent . Purity(%) . Amoxicillin 26787-78-0 DTXSID303704 Penicillin-class antibiotic Negative control DMSO ≥90 Salicylic acid 69-72-7 DTXSID7026368 Nonsteroidal anti-inflammatory drug Negative control DMSO ≥99 Glyphosate 38641-94-0 DTXSID0034649 Organophosphorus herbicide Negative control water 96 BIC 485-49-4 DTXSID3042687 Isoquinoline alkaloid GABAA antagonists DMSO/ethanol ≥99 PTX 124-87-8 DTXSID7045605 Convulsant alkaloid GABAA antagonists DMSO 98 Lindane 58-89-9 DTXSID2020686 Organochloride insecticide GABAA antagonists Ethanol 99 Dieldrin 60-57-1 DTXSID9020453 Organochloride insecticide GABAA antagonists DMSO ≥95 Permethrin 52645-53-1 DTXSID8022292 Type I pyrethroid insecticide Modulation of VSSCs kinetics DMSO/ethanol ≥91 Deltamethrin 52918-63-5 DTXSID8020381 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Cypermethrin 52315-07-8 DTXSID1023998 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Esfenvalerate 66230-04-4 DTXSID4032667 Type I/II pyrethroid insecticide Intermediate modulation of VSSCs kinetics DMSO/ethanol 98.5 Tributyltin 56-36-0 DTXSID7043950 Fungicide/biocide Oxidative stress DMSO 96 Compound . CAS No. . DTXS ID . Class . Effect . Solvent . Purity(%) . Amoxicillin 26787-78-0 DTXSID303704 Penicillin-class antibiotic Negative control DMSO ≥90 Salicylic acid 69-72-7 DTXSID7026368 Nonsteroidal anti-inflammatory drug Negative control DMSO ≥99 Glyphosate 38641-94-0 DTXSID0034649 Organophosphorus herbicide Negative control water 96 BIC 485-49-4 DTXSID3042687 Isoquinoline alkaloid GABAA antagonists DMSO/ethanol ≥99 PTX 124-87-8 DTXSID7045605 Convulsant alkaloid GABAA antagonists DMSO 98 Lindane 58-89-9 DTXSID2020686 Organochloride insecticide GABAA antagonists Ethanol 99 Dieldrin 60-57-1 DTXSID9020453 Organochloride insecticide GABAA antagonists DMSO ≥95 Permethrin 52645-53-1 DTXSID8022292 Type I pyrethroid insecticide Modulation of VSSCs kinetics DMSO/ethanol ≥91 Deltamethrin 52918-63-5 DTXSID8020381 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Cypermethrin 52315-07-8 DTXSID1023998 Type II pyrethroid insecticide Prolonged modulation of VSSCs kinetics DMSO/ethanol ≥98 Esfenvalerate 66230-04-4 DTXSID4032667 Type I/II pyrethroid insecticide Intermediate modulation of VSSCs kinetics DMSO/ethanol 98.5 Tributyltin 56-36-0 DTXSID7043950 Fungicide/biocide Oxidative stress DMSO 96 Open in new tab MEA recordings Neuronal activity of human iN/glia cocultures and rat cortical cultures was measured on 48-well MEA plates using the Maestro 768 channel amplifier system (Axion BioSystems) equipped with an environmental control unit (37°C and 5% CO2) and operated by Axion Integrated Studios (AxIS) v2.3 software. All channels were recorded at a sampling frequency of 12.5 kHz and a Butterworth band-pass filter was applied to select signals between 0.2 and 5kHz. Subsequent spike detection was performed using AxIS adaptive threshold crossing method with a threshold of 8 × rms noise for each channel and a pre- and postspike duration of 0.84 and 2.16 ms, respectively. To track maturation and network formation of human iN/glia cocultures and rat cortical cultures, spontaneous neuronal activity was recorded weekly after plating for 10 min, following an equilibration period of 20 min. On day in vitro (DIV) 12 for the rat cultures, and DPP 37 for the human iN/glia cocultures, spontaneous neuronal activity in the absence of compound (baseline) was first recorded for 40 min following a 20 min equilibration period on the Maestro system. Compounds described in Table 1 were applied by preparing a 5× predilution of the desired final concentration in 100 µl media. For dosing, 100 µl of media were removed from each well, combined with the compounds, and then added back to the cultures for a final volume of 500 µl per well. Neuronal activity in the presence of compounds (dosed) was immediately recorded for 40 min, unless indicated otherwise. The 12 compounds were applied according to 2 main plate layouts (see Supplementary Figure 1) with 6 replicates per concentration measured in separate MEA plates. Each MEA plate contained dedicated wells for single measurements (not cumulative) of 6 compounds at 7 different concentrations and their corresponding solvent controls. Solvents for each compound were selected based on previous experience and in no case did the presence of solvent alter the activity over the recording period. Replicate measurements for each compound concentrations were performed on different MEA plates with cells derived from the same cell production batch (human iN/glia cocultures) or cell preparation (rat cortical cultures). Only wells with 8 or more active electrodes (>5 spikes/min) during baseline recording were regarded for subsequent analysis. On average, 8% of the wells per plate had to be removed from the analysis because they had <8 active electrodes during the baseline recording conducted on the dosing day (which was the criteria for outlier removal). For the complementary experiments performed on the human iN/glia cocultures, including the excitatory and inhibitory iN ratio optimization experiment and the CBX application to assess the contribution of electrical synapses, cultures were equilibrated for 20 min before doing a baseline recording for 15 min. For the iN-ratio experiment, cultures were exposed to a single dose of PTX (3 µM) for 1 h at week 4 after plating. For the CBX experiment, cultures were exposed to CBX (25 µM) and PTX (10 µM) at week 4 after plating according to the experimental strategy described in Supplementary Figure 2. Briefly, after baseline recording, cultures were first pre-exposed to CBX or its solvent (H2O) for 30 min. Next, PTX was added to previously CBX or solvent-treated wells for 1 h. For dosing data analysis, a 15 min time window after increased activity in solvent control wells returned to baseline was selected. Data analysis For quantification of neuronal network activity, baseline and dosing spike files generated from recordings were further processed using Neural Metric Tool software (Axis BioSystems) to calculate a total of 69 activity metrics. Per channel, 5 or more consecutive spikes occurring with an interspike interval (ISI) of 100 ms or less were considered a burst. Network bursts were identified using the envelope algorithm and required a minimal of 50% participating electrodes and a burst inclusion of 80%. For analyzing neuronal responses elicited by the different chemical compounds, we decided to focus on the most informative and mostly independent activity metrics measured at a single-electrode and network (multiple electrodes) levels. Those included spiking (weighted mean firing rate [wMFR]) and bursting (Burst Duration and Frequency) metrics, as well as features of synchronized network events such as Network Burst Frequency and Synchrony Index. For each well, values of activity metrics for compound and solvent dosing were normalized to baseline values and presented as percentage change. Unless stated otherwise, all results are expressed as mean ± SEM. Significant differences for each metric between different compound concentrations and their corresponding solvent were identified using 1-way analysis of variance (ANOVA) with Dunnett’s method for multiple-to-one- comparisons, or Tukey’s test for multiple pairwise comparisons. The accepted level of significance was p < .05. Cell viability assays Cellular health and viability of dosed neural cultures were quantitatively assessed by measuring release of lactate dehydrogenase (LDH)-dependent conversion of tetrazolium salt to red formazan and cell-dependent conversion of resazurin to highly fluorescent resorufin, respectively (Brown et al., 2017). After 40 min of MEA recording in the presence of compound/solvent, 50 µl of culture media was removed from each well and transferred to a transparent 96-well plate to determine LDH concentrations using the CytoTox 96 Nonradioactive Cytotoxicity Assay kit (Promega). Fresh culture media (50 µl) and media from wells with lysed neural cultures served as blank and cellular LDH measurements, respectively. For LDH detection, 50 µl of reconstituted substrate mix was added to each 96-well and incubated in the dark, at room temperature for 30 min. After the reaction was stopped by adding 50 µl Stop Solution, raw LDH release was measured as single wavelength absorbance of converted formazan product at 490 nm using a Synergy 2 Microplate Reader (BioTek). Percentage of LDH release was calculated from the ratio of media-blanked release and cellular LDH. For fluorescence-based viability quantification using the CellTiter Blue Cell Viability Assay kit (Promega), treatment medium was removed from each well of the recorded MEA plate and replace with 200 µl fresh media containing a 1:6 dilution of CellTiter Blue (resazurin) reagent. After 60 min incubation at 37°C, 150 µl from each well were transferred to corresponding wells of an opaque 96-well plate. Fresh CellTiter Blue medium (150 µl) served as blank measurement. Fluorescence of converted resorufin was measured with Ex 544 nm at Em 590 nm using a Synergy 2 Microplate Reader (BioTek). Percentage cell survival was calculated from the ratio of blank-corrected compound and solvent control samples. RESULTS Pure hiPSC-Derived Neuronal/Glial Cocultures Excitatory and inhibitory neurons derived from hiPSCs through direct reprogramming using exogenous overexpression of Ngn2 and Ascl1/Dlx2, respectively, have been demonstrated to form functional synapses eliciting spontaneous inhibitory and excitatory postsynaptic currents (Yang et al., 2017). Here, we separately generated pure excitatory and inhibitory iN cells and coseeded both cell types together with human primary fetal astrocytes on MEAs to establish a screening platform for analyzing network activity of a defined mixed neuronal cell population (Figs. 1A and 1B). Immunofluorescence staining of co-cultures at 21 DPP showed MAP2-positive iN cells with predominantly mature neuronal morphology and abundant presence of synapses (Figure 1C). Cocultures further exhibited an even distribution of GABAergic inhibitory cells as visualized by staining for the neurotransmitter GABA and vGAT. Moreover, coseeded human primary glial cells showed strong signals for the astrocyte marker GFAP and absence of neuronal markers (MAP2, TuJ1), indicating the neuronal cell component of the mature cultures to be exclusively iPSC-derived (Figure 1D). Figure 1. Open in new tabDownload slide Pure human-derived neural cocultures for neurotoxicity assessment. A, Schematic workflow for direct conversion of induced pluripotent stem cells into excitatory or 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). B, Schematic representation of induced neuron/glia cocultures grown on 48-well multielectrode array plates. C, 40× confocal images of mixed excitatory/inhibitory iN coculture on astrocytes containing a small fraction of tdTomato-expressing induced neuronal (iN) cells. Proximity of the pre- and postsynaptic markers VGLUT2 and PSD-95 indicated formation of functional synapses. Boxes 1 and 2 show selected regions for higher magnification images below. (D) Characterization of human-derived neuron/glia cocultures by immunofluorescence imaging (10×). Upper panel: Pan-neuronal marker microtubule-associated protein 2 (MAP2), astroglial marker glial fibrillary acidic protein, and nuclear DAPI staining showed nonoverlapping and homogenous distribution of neurons and astroglial cells. Middle panel: Pan-neuronal marker MAP2, vesicular gamma amino butyric acid (GABA) transporter, and nuclear DAPI staining showed presence of inhibitory cells as a fraction of cocultured neurons. Lower panel: Pan-neuronal marker class III β-tubulin (TuJ1), inhibitory neurotransmitter GABA, and nuclear DAPI staining showed the presence of inhibitory cells as a fraction of cultured neurons. Cell numbers for imaged neural cocultures: 1.2 × 105 excitatory iNs, 0.6 × 105 inhibitory. iNs, and 0.6 × 105 astrocytes. Scale bar = 50 μm. Figure 1. Open in new tabDownload slide Pure human-derived neural cocultures for neurotoxicity assessment. A, Schematic workflow for direct conversion of induced pluripotent stem cells into excitatory or 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). B, Schematic representation of induced neuron/glia cocultures grown on 48-well multielectrode array plates. C, 40× confocal images of mixed excitatory/inhibitory iN coculture on astrocytes containing a small fraction of tdTomato-expressing induced neuronal (iN) cells. Proximity of the pre- and postsynaptic markers VGLUT2 and PSD-95 indicated formation of functional synapses. Boxes 1 and 2 show selected regions for higher magnification images below. (D) Characterization of human-derived neuron/glia cocultures by immunofluorescence imaging (10×). Upper panel: Pan-neuronal marker microtubule-associated protein 2 (MAP2), astroglial marker glial fibrillary acidic protein, and nuclear DAPI staining showed nonoverlapping and homogenous distribution of neurons and astroglial cells. Middle panel: Pan-neuronal marker MAP2, vesicular gamma amino butyric acid (GABA) transporter, and nuclear DAPI staining showed presence of inhibitory cells as a fraction of cocultured neurons. Lower panel: Pan-neuronal marker class III β-tubulin (TuJ1), inhibitory neurotransmitter GABA, and nuclear DAPI staining showed the presence of inhibitory cells as a fraction of cultured neurons. Cell numbers for imaged neural cocultures: 1.2 × 105 excitatory iNs, 0.6 × 105 inhibitory. iNs, and 0.6 × 105 astrocytes. Scale bar = 50 μm. Optimizing Ratio of Excitatory to Inhibitory in Cells for Screening Neuronal Network Activity The ratio of glutamatergic excitatory to GABAergic inhibitory neurons (exc:inh) largely varies between different brain regions. In the forebrain, medial ganglionic eminence and preoptic area derived GABAergic interneurons make up around 4–16% of the different thalamic structures (Çavdar et al., 2014), about 26% of the striatum (Bernacer et al., 2012; Graveland et al., 1985; Roberts et al., 1996), and approximately 6% of the neurons across the entire hippocampus (Botcher et al., 2014; Czeh et al., 2013). In the cortex, the fraction of GABAergic interneurons is more consistent and constitutes roughly 15–20% of the total neuronal population (Barinka et al., 2012; Gonchar et al., 2007; Salaj et al., 2015). However, the exc:inh synaptic ratio and thus the temporal dynamic of signal modulation largely differs between cortical regions and across circuits (Gittis et al., 2010; Gulyás et al., 1999; Liu, 2004; Medalla et al., 2017; Yang and Sun, 2018). Moreover, the distribution of different types of GABAergic interneurons, such as PV expressing basket and chandelier cells, SOM expressing Martinotti cells, or cells expressing either NPY, vasoactive intestinal peptide, or CK, considerably differs between cortical layers and regions (Pohlkamp et al., 2014; Qu et al., 2016; Rudy et al., 2011; Young and Sun, 2009). GABAergic inhibitory iN cells exhibit only a small fraction of PV-positive cells which largely overlaps with SOM expression (Yang et al., 2017), and glutamatergic excitatory iN cells mostly resemble cortical layer II/III neurons (Zhang et al., 2013). To that end, we sought to establish a neural coculture system consisting of iPSC-derived excitatory iN cells, inhibitory iN cells, and primary human astroglial cells that exhibit basic functional features of a cortical network such as spontaneous bursting events and coordinated firing. We therefore decided to first optimize exc:inh iN ratios of mixed neural cultures for their capacity to develop biologically relevant network activity. Moreover, to be suitable for in vitro screening applications, neural cocultures would ideally show network formation that is robust with homogenous distribution of firing events across each MEA well while still being highly sensitive to modulation of single network components. Therefore, we also tested different total cell densities regarding development of neuronal activity over time on 48-well MEA plates. Furthermore, we assessed the responsiveness to inhibition of GABA signaling by dosing with 3 µM PTX. Keeping astroglial cells constant at a total of 0.7 × 105 per well, 3 densities of excitatory iN cells (1.0 × 105, 1.4 × 105, and 1.8 × 105) were coseeded with varying numbers of inhibitory iN cells to attain cocultures with 20–30%, 30–40%, 40–50%, and 50–60% GABAergic cells, respectively (Supplementary Table 1). As expected, overall firing and the number of active electrodes (>5 spikes/min) generally increased over recordings at 7, 14, and 21 DPP in every condition (Figure 2A) except for cocultures comprising 1.0 × 105 excitatory and 50–60% inhibitory cells. The occurrence of bursts, indicating a more mature activity pattern, was detected after 14 days in the lowest cell density and GABAergic cell percentage but robustly developed until DPP 21 in coculture of medium excitatory cell densities as demonstrated by the percentage of spikes that fall within a burst (Figure 2B). Interestingly, cocultures with high excitatory cell densities showed comparably low burst percentages except when grown with high numbers of inhibitory cells. Coordination of firing and development of synchronized network bursts were observed first in the low density/low inhibitory condition and in cocultures consisting of medium number of excitatory and 40–50% inhibitory neurons (DPP14, Figure 2B). At DPP 21, network bursting was reached in all cocultures with medium excitatory cell numbers, low excitatory cocultures with up to 40% inhibitory cells and mostly high excitatory cocultures with over 50% inhibitory cells. Importantly, application of PTX increased spiking rates as well as number of active electrodes, and drastically augmented the number of synchronized network bursts in nearly all conditions. Based on the steady increase of active electrodes and spike frequency over time, as well as burst percentage and presence of network activity with most pronounced PTX responses, coculture conditions encompassing 1.4 × 105 excitatory and 30% inhibitory iN cells (0.6 × 105, Figs. 2A and 2B, middle panel) were considered optimal for general neurotoxicity testing applications. Furthermore, modulation of N-methyl-D-aspartate (NMDA)-, GABAA-, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)- and kainate receptor activity using specific agonist and antagonist treatment revealed dose-dependent responses in multiple network activity parameters (Supplementary Figure 3). Figure 2. Open in new tabDownload slide Comparison of different iN cell densities and glutamatergic excitatory to GABAergic inhibitory ratios based on the development of neural activity over time (7, 14, and 21 days postplating [DPP]) and after network activation on multielectrode arrays. A, Neuronal spiking (weighted mean firing rate) and number of active electrodes showed a steady increase over time and an additional rise after network activation (picrotoxin [PTX]) for both 100 and 140 K excitatory cell densities cocultured with 30–40% inhibitory neurons. B, At DPP 21, high percentages of spikes occurring within a burst (burst percentage) with the high numbers of total network bursts are observed for 140 K excitatory cells densities cocultured with 30–40%, 40–50%, and 50–60% inhibitory neurons. Out of these conditions, 140 K excitatory with 30–40% inhibitory neurons showed the largest increased in coordinated network activity upon PTX treatment. Each condition was measured in technical replicates (n = 6). Figure 2. Open in new tabDownload slide Comparison of different iN cell densities and glutamatergic excitatory to GABAergic inhibitory ratios based on the development of neural activity over time (7, 14, and 21 days postplating [DPP]) and after network activation on multielectrode arrays. A, Neuronal spiking (weighted mean firing rate) and number of active electrodes showed a steady increase over time and an additional rise after network activation (picrotoxin [PTX]) for both 100 and 140 K excitatory cell densities cocultured with 30–40% inhibitory neurons. B, At DPP 21, high percentages of spikes occurring within a burst (burst percentage) with the high numbers of total network bursts are observed for 140 K excitatory cells densities cocultured with 30–40%, 40–50%, and 50–60% inhibitory neurons. Out of these conditions, 140 K excitatory with 30–40% inhibitory neurons showed the largest increased in coordinated network activity upon PTX treatment. Each condition was measured in technical replicates (n = 6). Neuronal Network Development in Human and Rat Neural Cultures Direct neuronal conversion of iPSCs efficiently produced a homogenous cell population showing rapid cell cycle exit and neuronal commitment rapidly passing through a neural precursor stage (Zhang et al., 2013). When coseeded with astroglial supporter cells, such iN cells further mature to become electrically active and eventually form functional synapses (Zhang et al., 2013). Due to the unnatural process of reprogramming and low complexity of iN/glia cocultures, the formation of neuronal networks including the coordination of firing of multiple cells considerably differs from developing primary neural tissue. The ontogeny of neuronal network formation in vitro can be quantitatively assessed by parameters describing the frequency and structure of bursts as well as the synchrony of network activity. Particularly, the coefficient of variation (CV) of within-burst ISIs, the CV of interburst intervals (IBIs), the burst rate, and the network burst frequency have previously been demonstrated to be useful in describing the age or maturation level of neuronal networks (Cotterill et al., 2016). To be able to compare alterations of in vitro neuronal network activity induced by neurotoxic chemicals between human and rodent cultures, we therefore determined the earliest maturation time point of iN cell networks that most closely resemble fully developed primary rat cortical cultures using MEA recordings. As expected, human iN/glia cocultures exhibited a delay in the onset of basic neuronal firing compared with rat primary cultures, which was indicated by a slower increase of the number of active electrodes, ie, electrodes that detect spikes based on a defined threshold, over time (Figs. 3Ai and 3Bi). Human iN/glial cocultures reached a plateau of 12–14 out of 16 electrodes between DPP 28 and 37, whereas rat primary cortical cultures maintained an average of about 15 active electrodes from DPP 12 on. Along with neuronal maturation, the burst rate increased steadily in both culture types until reaching 3.7 ± 1.69 bursts/min at DPP 12 in rat primary and 2.18 ± 3.21 bursts/min at DPP 37 in human cultures (Figs. 3Aii and 3Bii). Moreover, the CVs of the within burst ISI and the IBI, describing the periodicity of single and grouped spiking events, respectively, plateaued at 0.29 ± 0.07 and 1.18 ± 0.22 after 12 days in rat primary cortical cultures and at 0.33 ± 0.09 and 1.22 ± 0.43 after 37 days in human iN/glia cocultures (Figs. 3Aiv, 3Av, 3Biv, and 3Bv) indicating a less regular discharge of iN cells particularly within groups of spikes. Finally, synchronous network bursts were observed in rat primary cultures starting at DPP 7 and steadily increased in frequency until DPP 12 (3.03 ± 1.06 network bursts/min). In human iN/glial cocultures, coordinated network bursts were first detected between DPP 21 and 28, but rapidly progressed from sporadic events to a stable network bust rate of 3.74 ± 0.57 network bursts/min by DPP 37 (Figure 3B). Overall, rat primary cortical cultures showed a fast development and rapid synchronization of network activity during the first 12 days after plating. In a separate set of experiments, activity in rat cultures remained stable for up to 50 DPP, when cultures were discontinued (not shown). In contrast, in human iN/glia cocultures synchronized neural network activity developed steadily over 5 weeks and plateaued around DPP 37. We therefore reasoned to use human iN/glia co-cultures at DPP 37 and rat primary cortical cultures at DPP 12 or older for subsequent assessment of neurotoxicant effects of network activity. Figure 3. Open in new tabDownload slide Characteristics of neural network development in rat primary cortical cultures and human induced neuronal (iN)/glia cocultures. A, Multielectrode array (MEA) measurements of network ontogeny in rat primary cultures showed a steady increase of active electrodes, burst rate, and network burst rate (i–iii) for the first 12 days in vitro (DIV), collected from multiple plates (n = 384). The coefficient of variation (CV) of inter-spike intervals (ISIs) occurring within bursts (iv) plateaued at around DIV 9. The CV of the interburst intervals (IBIs) slightly increased over time and reached a maximum at DIV 12 (v). B, MEA measurements of network ontogeny in human iN/glia cocultures displayed an increasing number of active electrodes (i) along maturation until reaching a maximum at 28 days post plating (DPP). Burst rates and network burst rates (i and ii) showed a steady increase until DPP 37, collected from multiple plates (n = 384). The CV of within burst ISI (iv) appeared very heterogenous across wells and stabilized over time until DPP 37 at value of 0.33, similar to rodent cultures (Aiv). The CV of IBI (v) slightly increased during culture duration and reach a maximum at DPP37. C, Coordination of neural network activity in human iN/glia cocultures (n = 8) is independent of direct electrical coupling through gap junctions, assessed at DPP 25. Application of the gap junction blocker carbenoxolone (CBX) did not alter single neuronal spiking (weighted mean firing rate) nor network burst frequency or synchrony at. Treatment with PTX significantly increased single spiking and most network activity parameters (network burst frequency, spikes per network burst and CV of ISI). Coapplication of CBX did not alter this evoked effect. Error bars represent SEM. Asterisks indicate significant mean differences with p-values (* ≤ .05, ** ≤ .01, and *** ≤ .005) determined by one-way ANOVA (n = 8) followed by Tukey’s test for multiple testing. Figure 3. Open in new tabDownload slide Characteristics of neural network development in rat primary cortical cultures and human induced neuronal (iN)/glia cocultures. A, Multielectrode array (MEA) measurements of network ontogeny in rat primary cultures showed a steady increase of active electrodes, burst rate, and network burst rate (i–iii) for the first 12 days in vitro (DIV), collected from multiple plates (n = 384). The coefficient of variation (CV) of inter-spike intervals (ISIs) occurring within bursts (iv) plateaued at around DIV 9. The CV of the interburst intervals (IBIs) slightly increased over time and reached a maximum at DIV 12 (v). B, MEA measurements of network ontogeny in human iN/glia cocultures displayed an increasing number of active electrodes (i) along maturation until reaching a maximum at 28 days post plating (DPP). Burst rates and network burst rates (i and ii) showed a steady increase until DPP 37, collected from multiple plates (n = 384). The CV of within burst ISI (iv) appeared very heterogenous across wells and stabilized over time until DPP 37 at value of 0.33, similar to rodent cultures (Aiv). The CV of IBI (v) slightly increased during culture duration and reach a maximum at DPP37. C, Coordination of neural network activity in human iN/glia cocultures (n = 8) is independent of direct electrical coupling through gap junctions, assessed at DPP 25. Application of the gap junction blocker carbenoxolone (CBX) did not alter single neuronal spiking (weighted mean firing rate) nor network burst frequency or synchrony at. Treatment with PTX significantly increased single spiking and most network activity parameters (network burst frequency, spikes per network burst and CV of ISI). Coapplication of CBX did not alter this evoked effect. Error bars represent SEM. Asterisks indicate significant mean differences with p-values (* ≤ .05, ** ≤ .01, and *** ≤ .005) determined by one-way ANOVA (n = 8) followed by Tukey’s test for multiple testing. Synchronized Neural Network Activity in iPSC-Derived Cultures Is Mediated by Synapse Signaling Synchronization of neuronal bursting is an important feature of network activity as it usually integrates mature cell-autonomous and synaptic signal processing and therefore provides valuable information on altered neuronal function. However, iPSC-derived neural cultures generated by stepwise differentiation techniques have recently been shown to primarily mediate coordination of firing through gap junctions (Makinen et al., 2018). Since direct electrical coupling of neurons to this extent does not reflect in vivo physiology this could lead to misinterpretation of altered neural activity upon chemical dosing. To assess the contribution of synaptic transmission to spontaneous and chemically challenged coordinated network activity of our iPSC-derived neural cocultures, we applied the gap junction blocker CBX in the presence and absence of PTX to our standard cocultures at DPP 25 and measured neural activity changes on MEAs. We observed no significant differences in single spiking or network metrics, and especially not in spontaneous synchronized network activity, when standard mixed neural cocultures were dosed with 25 μM CBX (Figure 3C). However, CBX showed a minor yet significant reduction in the number of spikes within a network burst compared with vehicle control. When the GABAA receptor blocker PTX (10 μM) was applied to untreated parallel cocultures, spiking rates, network burst frequency, spike density of network bursts, and ISI CV increased as expected, although the increase in network burst frequency did not reach significance. Importantly, addition of PTX to cocultures pre-exposed to CBX induced the same changes in single spiking and network activity as seen in unexposed cultures (Figure 3C). We therefore concluded that the coordination of neuronal firing in our iPSC-derived neural cocultures is predominantly mediated by synaptic signaling and not a result of direct electrical coupling through gap junctions. Alteration of Neuronal Network Activity in Responses to Neurotoxic Chemicals in Human and Rodent Neural Cultures To evaluate suitability of human iN/glia cocultures for identifying neuroactive effects of chemical compounds through alteration of network activity, we tested a set of 9 compounds with well-established effects together with 3 control compounds with demonstrated absence of neurotoxicity (Table 1). Dosing was performed at DPP 37 for human iN/glial cocultures and between DPP 12 and DPP 21 for rat primary cortical cultures according to the corresponding levels of network development determined in the previous ontogeny analysis. Recording of neuronal activity was conducted on MEA plates for 40 min before (baseline) and after (dosed) application of compounds at 7 concentrations plus corresponding solvent control. Metrics for neuronal function describing overall spiking, bursting behavior, neuronal network activity, and synchrony of coordinated firing in all dosed wells were calculated and normalized to baseline activity. In human iN/glia cocultures, 2 of the negative control compounds (salicylic acid and amoxicillin) showed no significant changes in any of the evaluated metrics at any of the applied concentrations (Figure 4A and Supplementary Figure 4A). Dosing with glyphosate also did not induce any apparent alteration in single neuronal firing or network activity except for a slight but significant increase in burst duration only at a medium concentration of 0.3 μM (Figure 4A). In rat primary cortical cultures, we did not observe any significant changes in neural network activity across all tested concentrations for amoxicillin and glyphosate (Figure 4A). However, we recognized a largely concentration-dependent decrease in neuronal spiking (wMFR), burst frequency and network burst frequency for salicylic acid (Figure 4A). As expected, the positive control tributyltin, which directly impairs neuronal viability (Ishihara et al., 2012; Nakatsu et al., 2006), decreased firing, bursting, and synchronized network activity with increasing concentrations in both human and rat cultures, representing an overall strong detrimental effect on neuronal function (Figure 4B). Figure 4. Open in new tabDownload slide Alteration of neuronal network activity in human induced neuronal (iN)/glia cocultures and rat primary cortical cultures in response to neurotoxic compound treatment on multielectrode arrays. Human iN/glia cocultures: 1.4 × 105 excitatory iNs, 0.6 × 105 inhibitory iNs, and 0.7 × 105 astrocytes. Depicted activity parameters were selected to represent overall spike rates (weighted mean firing rate [wMFR]) as well as bursting (burst duration, burst frequency) and network busting structures (network burst frequency, synchrony index). The left panels illustrate representative raster plots of dosed human iN/glia cocultures. A, Treatment with 3 nonneurotoxic negative control compounds (salicylic acid, amoxicillin, and glyphosate) did not change neural network activity in human iN/glia cocultures except for an increase in burst duration at a single medium concentration of glyphosate. In rat cortical cultures, treatment with amoxicillin and glyphosate did not considerably alter network activity at any concentration. In contrast, salicylic acid showed a concentration-dependent decrease in spiking (wMFR) and bursting (burst frequency, network burst frequency). B, Treatment with a neuro-cytotoxic control compound (tributyltin) impaired neuronal firing and network activity at medium to high concentrations in both human iN/glia cocultures and rat cortical cultures. C, Treatment with neurotoxicants that modulate gamma amino butyric acid receptor activity (picrotoxin, lindane, bicuculline, and dieldrin) showed an overall activating effect on neuronal spiking and network activity in both culture types with human iN/glia cocultures exhibiting a generally higher sensitivity. In rat cortical cultures, dieldrin only showed a tendency for increased burst rates at a 3 μM concentration. D, Treatment with chemicals that modulate voltage-sensitive sodium channel kinetics (permethrin, deltamethrin, cypermethrin, and esfenvalerate) showed a concentration-dependent disruptive effect on network synchrony in both human and rat neural cultures. In human iN/glia cocultures, an increase in bursting and network bursting frequencies was observed for all these chemicals at varying concentration ranges. In rat cortical cultures, only permethrin showed an enhanced bursting and network bursting rate at medium to high concentrations whereas deltamethrin, cypermethrin, and esfenvalerate exhibited reduced bursting activities at high concentrations. Asterisks indicate significant mean differences to vehicle control with a p-value ≤ .05 determined by 1-way ANOVA (n = 6) followed by Dunnett’s multiple comparison test. Figure 4. Open in new tabDownload slide Alteration of neuronal network activity in human induced neuronal (iN)/glia cocultures and rat primary cortical cultures in response to neurotoxic compound treatment on multielectrode arrays. Human iN/glia cocultures: 1.4 × 105 excitatory iNs, 0.6 × 105 inhibitory iNs, and 0.7 × 105 astrocytes. Depicted activity parameters were selected to represent overall spike rates (weighted mean firing rate [wMFR]) as well as bursting (burst duration, burst frequency) and network busting structures (network burst frequency, synchrony index). The left panels illustrate representative raster plots of dosed human iN/glia cocultures. A, Treatment with 3 nonneurotoxic negative control compounds (salicylic acid, amoxicillin, and glyphosate) did not change neural network activity in human iN/glia cocultures except for an increase in burst duration at a single medium concentration of glyphosate. In rat cortical cultures, treatment with amoxicillin and glyphosate did not considerably alter network activity at any concentration. In contrast, salicylic acid showed a concentration-dependent decrease in spiking (wMFR) and bursting (burst frequency, network burst frequency). B, Treatment with a neuro-cytotoxic control compound (tributyltin) impaired neuronal firing and network activity at medium to high concentrations in both human iN/glia cocultures and rat cortical cultures. C, Treatment with neurotoxicants that modulate gamma amino butyric acid receptor activity (picrotoxin, lindane, bicuculline, and dieldrin) showed an overall activating effect on neuronal spiking and network activity in both culture types with human iN/glia cocultures exhibiting a generally higher sensitivity. In rat cortical cultures, dieldrin only showed a tendency for increased burst rates at a 3 μM concentration. D, Treatment with chemicals that modulate voltage-sensitive sodium channel kinetics (permethrin, deltamethrin, cypermethrin, and esfenvalerate) showed a concentration-dependent disruptive effect on network synchrony in both human and rat neural cultures. In human iN/glia cocultures, an increase in bursting and network bursting frequencies was observed for all these chemicals at varying concentration ranges. In rat cortical cultures, only permethrin showed an enhanced bursting and network bursting rate at medium to high concentrations whereas deltamethrin, cypermethrin, and esfenvalerate exhibited reduced bursting activities at high concentrations. Asterisks indicate significant mean differences to vehicle control with a p-value ≤ .05 determined by 1-way ANOVA (n = 6) followed by Dunnett’s multiple comparison test. Application of test compounds with inhibitory effects on GABAA receptors, PTX, lindane, BIC, and dieldrin, generally increased activity in both types of cultures but showed specific differences in single parameters. PTX increased wMFR, burst duration, and burst frequency in a concentration-dependent manner in rat and human cultures with the percentage change being more pronounced in human cultures (Figure 4C and Supplementary Figure 4B). However, the network burst frequency was largely unchanged in human iN/glial cocultures, while clearly increasing with higher PTX concentrations in rat primary cultures (Figure 4C). For lindane, we observed an increase in wMFR as well as burst and network burst frequencies and a minor but nonsignificant decrease in burst duration at a higher concentration range in human cultures. In contrast, rat cultures exhibited minor increase in wMFR, burst duration, burst frequency, and network burst frequency (Figure 4C). Dosing of BIC enhanced wMFR, burst duration and burst frequency at medium to high concentrations in both human and rat, but showed elevated network burst frequency only in human iN/glia cocultures (Figure 4C) with no change in burst frequencies and showed a significant increase at higher compound concentrations in rat and human. Moreover, network burst frequencies were accelerated in response to BIC, lindane, and dieldrin, but showed no significant difference upon PTX dosing. Burst durations were increased in BIC and PTX following ascending doses but exhibited a decrease at high concentrations of lindane and dieldrin. Finally, the overall synchrony of neuronal firing (synchrony index) remained largely unaffected and only showed a significant reduction in response to high dieldrin doses. Thus, GABAA receptor blockers apparently increased firing of neuronal networks while maintaining or even enhancing bursting structures and coordinated network activity. In comparison, the tested types I and II pyrethroids permethrin, deltamethrin, cypermethrin, and esfenvalerate, showed a less consistent response pattern (Figure 4D and Supplementary Figure 4C). Here, the wMFR was significantly elevated in deltamethrin and cypermethrin solely at an intermediate concentration range but showed a concentration-dependent increase for permethrin. Whereas esfenvalerate reduced the firing rate at concentrations over 10 µM. Moreover, burst frequency and network burst frequency displayed particularly heterogenous changes with an increase at intermediate concentrations for deltamethrin and cypermethrin compared with a continuous increase along with the concentration range tested for permethrin. In contrast, esfenvalerate shows no considerable change in burst frequency but an enhanced network bust frequency in response starting at medium to high concentrations (Figure 4D). In all pyrethroids tested, burst duration and synchrony decreased with ascending concentrations, indicating a disruption of organized network activity despite an increase of individual neuronal firing. Most importantly, responses of human iN/glia cocultures showed a striking correspondence to rat primary cortical cultures for all compounds in the vast majority of analyzed metrics. Cytotoxic Effects of Neurotoxic Chemical in Human and Rodent Neural Cultures MEAs have proven to be a useful tool for studying neurotoxic compounds particularly with regard to distinguishing neuroactive effects from general or neuro-specific cytotoxicity (Wallace et al., 2015). Sequential analysis of electrophysiological readouts and chemical assays allows measurements of neuronal activity changes and cell survival in the same samples and under given dosing conditions. We therefore sought to assess how overall viability is impaired in rat primary cortical and human iN/glia cocultures after application of the test compounds at different concentrations. For this purpose, we quantified the release of LDH enzyme (a measure of membrane integrity) and the metabolic conversion of a fluorescence substrate (a measure of mitochondrial function/metabolism) for each MEA well immediately after recording. As expected, none of the negative control compounds, salicylic acid, amoxicillin, or glyphosate, impaired viability or altered neuronal firing at the tested concentrations (Figure 5A). In contrast, the tested GABAA blockers BIC, PTX, lindane, and dieldrin robustly increased firing along with ascending doses but exhibited no significant effect on cell survival within the examined time span of approximately 1 h (Figure 5C). Similarly, the pyrethroids permethrin, deltamethrin, cypermethrin, and esfenvalerate showed drastic and diverse effects on neuronal activity without affecting viability of neural cultures, even at higher concentrations (Figure 5D). These observations agree with the main MOA of GABAA blockers and pyrethroids to exert their neurotoxic effects through altered ion channel function and thus modulation of neuronal activity. On the contrary, the steady reduction of firing that occurred in response to increasing doses of tributyltin was mirrored by a decrease in cell viability (Figure 5B). This reflects the general, likely oxidative stress-mediated (Ishihara et al., 2012), cytotoxicity rather than a specific interference with neurophysiology. Notably, like the assessment of neuronal network alteration, viability testing showed overall correspondence of rat primary cortical cultures and human iN/glia cocultures in response to compound dosing. Figure 5. Open in new tabDownload slide Separate detection of functional and structural neurotoxic effects of tested chemicals in human induced neuronal (iN)/glia cocultures (1.4 × 105 excitatory iNs, 0.6 × 105 excitatory iNs, and 0.7 × 105 astrocytes; 140K excitatory iNs/60K inhibitory iNs/70K astrocytes) and rat primary cortical cultures. For ease of comparison neuronal spike rates (weighted mean firing rate, black lines) are plotted together with cell viability measurements represented by cell survival (red lines), determined by CellTiter Blue staining, and lactate dehydrogenase release (blue lines). A, Treatment with nonneurotoxic negative control compounds (salicylic acid, amoxicillin, and glyphosate) did not impair cell viability in human or rat neural cultures. In human iN/glia cocultures, none of these compounds considerably altered neuronal activity whereas in rat cortical cultures, salicylic acid decreased activity in the medium to high concentration range (see Figure 4A). B, Treatment with a neuro-cytotoxic control compound (tributyltin) decreased cell viability and neural activity in a concentration-dependent manner in both human and rat neural cultures. C, Treatment with neurotoxicants modulating gamma amino butyric acid receptor activity (picrotoxin, bicuculline, lindane, and dieldrin) showed no significant effect on cell viability in human or rat neural cultures. In contrast, neuronal spiking was increased for all these chemicals at medium to high concentrations, except for dieldrin in rat cortical cultures. D, Treatment with chemicals modulating voltage-sensitive sodium channel kinetics (permethrin, deltamethrin, cypermethrin, and esfenvalerate) did not affect cell viability in human or rat neural cultures. In human iN/glia cocultures, permethrin, deltamethrin, and cypermethrin increased neuronal spiking either high or at medium concentrations, whereas esfenvalerate decreased spiking at higher doses. In rat cortical cultures, deltamethrin, cypermethrin, and esfenvalerate decreased neuronal spiking at high concentrations, whereas high doses of permethrin slightly increased spike rates. Asterisks indicate significant mean differences to vehicle control with a p-value ≤ .05 determined by 1-way ANOVA (n = 6) followed by Dunnett’s multiple comparison test. Figure 5. Open in new tabDownload slide Separate detection of functional and structural neurotoxic effects of tested chemicals in human induced neuronal (iN)/glia cocultures (1.4 × 105 excitatory iNs, 0.6 × 105 excitatory iNs, and 0.7 × 105 astrocytes; 140K excitatory iNs/60K inhibitory iNs/70K astrocytes) and rat primary cortical cultures. For ease of comparison neuronal spike rates (weighted mean firing rate, black lines) are plotted together with cell viability measurements represented by cell survival (red lines), determined by CellTiter Blue staining, and lactate dehydrogenase release (blue lines). A, Treatment with nonneurotoxic negative control compounds (salicylic acid, amoxicillin, and glyphosate) did not impair cell viability in human or rat neural cultures. In human iN/glia cocultures, none of these compounds considerably altered neuronal activity whereas in rat cortical cultures, salicylic acid decreased activity in the medium to high concentration range (see Figure 4A). B, Treatment with a neuro-cytotoxic control compound (tributyltin) decreased cell viability and neural activity in a concentration-dependent manner in both human and rat neural cultures. C, Treatment with neurotoxicants modulating gamma amino butyric acid receptor activity (picrotoxin, bicuculline, lindane, and dieldrin) showed no significant effect on cell viability in human or rat neural cultures. In contrast, neuronal spiking was increased for all these chemicals at medium to high concentrations, except for dieldrin in rat cortical cultures. D, Treatment with chemicals modulating voltage-sensitive sodium channel kinetics (permethrin, deltamethrin, cypermethrin, and esfenvalerate) did not affect cell viability in human or rat neural cultures. In human iN/glia cocultures, permethrin, deltamethrin, and cypermethrin increased neuronal spiking either high or at medium concentrations, whereas esfenvalerate decreased spiking at higher doses. In rat cortical cultures, deltamethrin, cypermethrin, and esfenvalerate decreased neuronal spiking at high concentrations, whereas high doses of permethrin slightly increased spike rates. Asterisks indicate significant mean differences to vehicle control with a p-value ≤ .05 determined by 1-way ANOVA (n = 6) followed by Dunnett’s multiple comparison test. DISCUSSION The need for new approaches that enable rapid neurotoxicity screening of environmental compounds has lately spurred the development of neural in vitro assays. To this end, rodent primary neural cultures grown on MEA systems have successfully been used to investigate acute effects of compounds on neuronal function in conjunction with cytotoxicity testing to identify both functional and structural neurotoxicity (Strickland et al., 2018; Vassallo et al., 2017; Wallace et al., 2015). However, interspecies differences can limit extrapolation of animal-derived assays to the human situation as recently demonstrated by direct comparison of human versus rat sensitivity to chemicals in a neurosphere-based developmental neurotoxicity model which revealed drastic EC50 differences for some of the compounds (Baumann et al., 2016). Furthermore, the issue of insufficient translatability of results from testing CNS liabilities in animal models is increasingly recognized as a major constraint for preclinical safety assessment of pharmaceutical drug programs. Meta-studies identified 34% of drug failures in clinical trials between 2005 and 2010 to be related to CNS organ toxicities not identified in animals (Cook et al., 2014; Mead et al., 2016; Olson et al., 2000; Waring et al., 2015). One of the studies examining 141 compounds from different companies even showed that animal tests were not able to predict the 5 most common neurological adverse effects in clinical trials (Mead et al., 2016). Consequentially, more recent studies explored avenues to use hiPSC-derived neuronal cells on MEAs for the identification of neurotoxic and specific neuroactive effects (Bradley et al., 2018; Tukker et al., 2018). Multiple studies were able to separate seizure-inducing chemicals from negative controls by measuring alteration of neural firing as well as bursting and network bursting behavior effects (Baumann et al., 2016; Bradley et al., 2018; Tukker et al., 2016, 2018). Despite these encouraging results, further improvement of the robustness of overall neuronal activity and particularly coordinated network function in human-derived neural cultures will be necessary to develop reliable quantitative screening assays. In this study, we used directly iN from hiPSCs to establish a novel neural coculture system on MEAs for screening adverse neuroactive effects of chemical compounds on neural network activity. Direct neuronal conversion of iPSCs by forced overexpression of neurogenic TFs has been demonstrated to yield highly functional neurons of specified subtypes that exhibit considerably accelerated maturation compared with conventional differentiation approaches (Yang et al., 2017; Zhang et al., 2013). In contrast to available protocols applying extrinsic signaling cues to pluripotent cells for stepwise recapitulating neuronal development (Chambers et al., 2009; Lancaster et al., 2013; Pasca, 2018), this method exclusively produces homogenous populations of specified neuronal subtypes, depending of the respective TF combination, with a uniform maturation level sparing tedious clean-up or enrichment steps. Here, we combined glutamatergic excitatory iNs and GABAergic inhibitory iNs with primary astrocytes in a defined manner yielding pure human iN/glial cocultures. Admittedly, the necessary implementation of a human primary cell source might greatly limit the scalability of this neural coculture system for potential screening applications. Unfortunately, standard methods for deriving astrocytes from iPSCs follow stepwise differentiation and thus encounter similar inadequacies as standard neuronal protocols, ie, high variability, heterogeneous maturation levels, and undefined mixture of subpopulations. However, recently published approaches using direct reprogramming techniques to rapidly generate mature astrocytes from iPSCs (Canals et al., 2018) or neural stem cells (Tchieu et al., 2019) might provide a solution in the near future. Nevertheless, the human iN/glia cocultures used in the present study showed broad synaptic staining with an even distribution of GABAergic and glutamatergic neurons and consistent expression of pan-neuronal markers in all iPSC-derived cells. When testing different alterations of the coculture composition, we found that an excitatory to inhibitory iN ratio of 70:30 with a total number of 2 × 105 neurons per well of a 48-well MEA plate exhibited the steadiest increase of spiking events and coverage of electrodes over time. Moreover, cocultures of this kind showed the most robust development of coordinated network activity as well as the largest percentage of spiking to occur within bursts indicating high synaptic connectivity. Most importantly, this condition showed the most pronounced response to GABAA receptor inhibition with drastically enhanced overall firing and network burst frequency demonstrating functional integration of excitatory and inhibitory network components. Notably, the development of coordinated network activity in vitro was considerably delayed in iPSC-derived iN/glia cocultures as compared with primary cortical cultures derived from neonatal rats which reached a maturation plateau at around 12 days after plating. The rapid ontogeny of network activity in rodent neurons is well established (Cotterill et al., 2016; Illes et al., 2007; van Pelt et al., 2005; Wagenaar et al., 2006), and data in this study were concordant with previously published ontogeny data (Cotterill et al., 2016). Although the slower ontogeny of human networks can be easily explained by the different developmental starting points, ie, pluripotent stem cells versus postnatal neural tissue, this observation alludes to a general hurdle to overcome when working with iPSC-derived neurons: the immature nature of the resulting cell types, which affects cell-intrinsic properties and signal transmission. Particularly, the transition from early electrical coupling to chemical synapses that occurs during cortical development in vivo (Peinado, 2001; Rorig et al., 1996; Valiullina et al., 2016; Wang et al., 2012) needs to be assessed if neural network activity readouts are used to evaluate neuroactive effects in cell-based assays. Along these lines, a recent study showed that network synchronization in iPSC-derived neuronal cultures generated through conventional stepwise differentiation is largely mediated by electrical coupling via gap junctions (Makinen et al., 2018). Using the gap junction blocker CBX, we were able to demonstrate that baseline activity of neural networks in iN/glia cocultures did not depend on electrical coupling. Moreover, increasing network activity by inhibiting GABAergic synaptic transmission through PTX application was insensitive to gap junction blockage. We therefore concluded that coordinated neuronal firing in these cocultures truly reflect mature network activity. Ultimately, we investigated the responsiveness of human iN/glia cocultures to well-known neurotoxic chemicals and compared specific alterations of neural network activity side by side with rat primary cortical cultures. Upon exposure to types I and II pyrethroids, which interfere with voltage-sensitive sodium channel (VSSC) function, we observed an overall consistent response pattern of disrupted synchrony between human and rodent cultures. However, only human iN/glia cocultures cultures showed a concomitant increase in firing rate and appeared to be generally more sensitive to the neuroactive effects of these compounds. In contrast, the chemicals PTX, lindane, and BIC which mainly act through GABAA receptor inhibition increased spiking rates and bursting in both rat and human cultures while largely maintaining the coordinated activity of the network. Surprisingly, the organochloride insecticide dieldrin showed similar effects only in human iN/glia cocultures in the tested concentration range. It is tempting to speculate that the reduced complexity of represented neural cell types and circuitry in the iPSC-derived human cultures would lead to a more pronounced phenotype upon specific interference with cell autonomous and synaptic function. Future studies that directly compare neural cultures derived from both human and rodent iPSCs will be required to determine a potential species-specific effect underlying these observations. Importantly, neither of the 2 culture types showed major alterations of neural network activity when treated with the control compounds amoxicillin and glyphosate indicating specific responsiveness to neuroactive effects. Notably, however, rat primary cultures did show a clear response to dosing with salicylic acid, the active metabolite of aspirin, which reduced firing and bursting in a concentration-dependent manner. In contrast, neuronal activity in human cultures was not affected throughout the dosing range. Salicylic acid is well-known to have neuroactive effects at high concentrations in vivo and in brain slices from different cortical and subcortical regions where it suppresses current-evoked neuronal firing of GABAergic interneurons and reduces specific excitatory and inhibitory postsynaptic responses leading to decreased neural network activity in pyramidal neurons of the auditory cortex (Namikawa et al., 2017; Su et al., 2012; Wang et al., 2006). In vivo, this effect is commonly used as a rodent tinnitus model (Guitton et al., 2003; Yang et al., 2007). Since iPSC-derived iN/glia cocultures in our system do not comprise specified neuronal cell types of the auditory cortex circuitry salicylic acid is not expected to exert any neuroactive function, which corresponds to our observations from MEA recordings. Preparations of rat primary cortical cultures on the other hand may include such cell types to varying degrees which would explain the responsiveness on these cultures. This approach has the capability to generate concentration-response data on a large number of compounds for many different parameters that describe neuronal network function. In many cases, the concentration-response profiles follow classical Hill curves, whereas in other cases, biphasic responses are observed. For this reason, a robust approach to curve-fitting and analysis will be needed as larger sets of data are developed. One such approach is the use of the Toxcast pipeline (Filer et al., 2017), which allows curve fitting for 3 different types of responses (constant, Hill, Gain-Loss) and generates potency values that allow comparisons between compounds. Such values can then be used, eg, to screen and rank compounds by potency and/or identify different patterns of activity among groups of compounds (Kosnik et al., 2020). Other approaches rely less on estimates of potency, instead evaluating minimally effective concentrations of compounds along with identification of patterns of response for selected parameters after spike train analysis (Bradley et al., 2018; Gramowski et al., 2004, 2006). Ultimately, the approach to data analysis should be driven by the questions that the experiments are attempting to address. As briefly described here, there are multiple different approaches that can be taken. In summary, here we have successfully established pure human mixed cocultures of glutamatergic excitatory and GABAergic inhibitory neurons in a defined ratio and demonstrated suitability for identifying functional impairment of VSSCs and GABAA receptors for the purpose of neurotoxicity assessment. For future advancement of in vitro screening assays, it will be important to more carefully characterize represented neuronal identities, standardize the relative abundance of the different cell types, and determine the pharmacological responsiveness of the system to obtain conclusive results from neural network recordings. Human iN/glia cocultures generated through direct reprogramming into defined neuronal subtypes will therefore be useful to develop robust phenotypic network activity readouts to reliably identify specific domains of functional neurotoxicity. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. Funding Preparation of this document has been funded by the U.S. Environmental Protection Agency (U.S. EPA). This document has been subjected to review by the U.S. EPA Center for Computational Toxicology and Exposure and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. This work was supported in part by a Pathways Innovation Project award to T.J.S. and W.R.M. from the U.S. EPA Office of Research and Development. DECLARATION OF CONFLICTING INTERESTS None of the EPA authors (K.W, T.F.F., W.R.M., and T.J.S.) have any conflict of interest to declare. The authors L.S., J.D., and D.H. have been or are currently employed by NeuCyte Inc., a company that commercially distributes the hiPSC-derived neural cells described in this study and declared no potential conflicts of interest with respect to the research in this article. M.M and M.W. have no conflict of interest to declare. REFERENCES Arber C. , Lovejoy C. , Wray S. ( 2017 ). Stem cell models of Alzheimer's disease: Progress and challenges . Alzheimers Res. Ther. 9 , 42 . Google Scholar Crossref Search ADS PubMed WorldCat Ardhanareeswaran K. , Mariani J. , Coppola G. , Abyzov A. , Vaccarino F. M. ( 2017 ). Human induced pluripotent stem cells for modelling neurodevelopmental disorders . Nat. Rev. Neurol. 13 , 265 – 278 . Google Scholar Crossref Search ADS PubMed WorldCat Barenys M. , Gassmann K. , Baksmeier C. , Heinz S. , Reverte I. , Schmuck M. , Temme T. , Bendt F. , Zschauer T. C. , Rockel T. D. , et al. ( 2017 ). Epigallocatechin gallate (EGCG) inhibits adhesion and migration of neural progenitor cells in vitro . Arch. Toxicol. 91 , 827 – 837 . Google Scholar Crossref Search ADS PubMed WorldCat Barinka F. , Salaj M. , Rybář J. , Krajčovičová E. , Kubová H. , Druga R. ( 2012 ). Calretinin, parvalbumin and calbindin immunoreactive interneurons in perirhinal cortex and temporal area Te3V of the rat brain: Qualitative and quantitative analyses . Brain Res. 1436 , 68 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Baumann J. , Gassmann K. , Masjosthusmann S. , DeBoer D. , Bendt F. , Giersiefer S. , Fritsche E. ( 2016 ). Comparative human and rat neurospheres reveal species differences in chemical effects on neurodevelopmental key events . Arch. Toxicol. 90 , 1415 – 1427 . Google Scholar Crossref Search ADS PubMed WorldCat Bauwens C. L. , Peerani R. , Niebruegge S. , Woodhouse K. A. , Kumacheva E. , Husain M. , Zandstra P. W. ( 2008 ). Control of human embryonic stem cell colony and aggregate size heterogeneity influences differentiation trajectories . Stem Cells 26 , 2300 – 2310 . Google Scholar Crossref Search ADS PubMed WorldCat Bernacer J. , Prensa L. , Gimenez-Amaya J. M. ( 2012 ). Distribution of GABAergic interneurons and dopaminergic cells in the functional territories of the human striatum . PLoS One 7 , e30504 . Google Scholar Crossref Search ADS PubMed WorldCat Botcher N. A. , Falck J. E. , Thomson A. M. , Mercer A. ( 2014 ). Distribution of interneurons in the CA2 region of the rat hippocampus . Front. Neuroanat. 8 , 104 . Google Scholar Crossref Search ADS PubMed WorldCat Bradley J. A. , Luithardt H. H. , Metea M. R. , Strock C. J. ( 2018 ). In vitro screening for seizure liability using microelectrode array technology . Toxicol. Sci. 63 , 240 – 253 . Google Scholar OpenURL Placeholder Text WorldCat Brawner A. T. , Xu R. , Liu D. , Jiang P. ( 2017 ). Generating CNS organoids from human induced pluripotent stem cells for modeling neurological disorders . Int. J. Physiol. Pathophysiol. Pharmacol. 9 , 101 – 111 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Brown J. P. , Lynch B. S. , Curry-Chisolm I. M. , Shafer T. J. , Strickland J. D. ( 2017 ). Assaying spontaneous network activity and cellular viability using multi-well microelectrode arrays . Methods Mol. Biol. 1601 , 153 – 170 . Google Scholar Crossref Search ADS PubMed WorldCat Buschmann J. ( 2013 ). The OECD guidelines for the testing of chemicals and pesticides . Methods Mol. Biol. 947 , 37 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat Cahan P. , Daley G. Q. ( 2013 ). Origins and implications of pluripotent stem cell variability and heterogeneity . Nat. Rev. Mol. Cell Biol. 14 , 357 – 368 . Google Scholar Crossref Search ADS PubMed WorldCat Canals I. , Ginisty A. , Quist E. , Timmerman R. , Fritze J. , Miskinyte G. , Monni E. , Hansen M. G. , Hidalgo I. , Bryder D. , et al. ( 2018 ). Rapid and efficient induction of functional astrocytes from human pluripotent stem cells . Nat. Methods 15 , 693 – 696 . Google Scholar Crossref Search ADS PubMed WorldCat Çavdar S. , Bay H. H. , Yıldız S. D. , Akakın D. , Şirvancı S. , Onat F. ( 2014 ). Comparison of numbers of interneurons in three thalamic nuclei of normal and epileptic rats . Neurosci. Bull. 30 , 451 – 460 . Google Scholar Crossref Search ADS PubMed WorldCat Chambers S. M. , Fasano C. A. , Papapetrou E. P. , Tomishima M. , Sadelain M. , Studer L. ( 2009 ). Highly efficient neural conversion of human ES and iPS cells by dual inhibition of SMAD signaling . Nat. Biotechnol. 27 , 275 – 280 . Google Scholar Crossref Search ADS PubMed WorldCat Clarke K. E. , Tams D. M. , Henderson A. P. , Roger M. F. , Whiting A. , Przyborski S. A. ( 2017 ). A robust and reproducible human pluripotent stem cell derived model of neurite outgrowth in a three-dimensional culture system and its application to study neurite inhibition . Neurochem. Int. 106 , 74 – 84 . Google Scholar Crossref Search ADS PubMed WorldCat Cobb M. M. , Ravisankar A. , Skibinski G. , Finkbeiner S. ( 2018 ). iPS cells in the study of PD molecular pathogenesis . Cell Tissue Res. 373 , 61 – 77 . Google Scholar Crossref Search ADS PubMed WorldCat Collins F. S. , Gray G. M. , Bucher J. R. ( 2008 ). Toxicology. Transforming environmental health protection . Science 319 , 906 – 907 . Google Scholar Crossref Search ADS PubMed WorldCat Conolly R. B. , Ankley G. T. , Cheng W. , Mayo M. L. , Miller D. H. , Perkins E. J. , Villeneuve D. L. , Watanabe K. H. ( 2017 ). Quantitative adverse outcome pathways and their application to predictive toxicology . Environ. Sci. Technol. 51 , 4661 – 4672 . Google Scholar Crossref Search ADS PubMed WorldCat Cook D. , Brown D. , Alexander R. , March R. , Morgan P. , Satterthwaite G. , Pangalos M. N. ( 2014 ). Lessons learned from the fate of AstraZeneca’s drug pipeline: A five-dimensional framework . Nat. Rev. Drug Discov. 13 , 419 – 431 . Google Scholar Crossref Search ADS PubMed WorldCat Costa L. G. ( 2015 ). The neurotoxicity of organochlorine and pyrethroid pesticides . Handb. Clin. Neurol . 131 , 135 – 148 . Google Scholar Crossref Search ADS PubMed WorldCat Cotterill E. , Hall D. , Wallace K. , Mundy W. R. , Eglen S. J. , Shafer T. J. ( 2016 ). Characterization of early cortical neural network development in multiwell microelectrode array plates . J. Biomol. Screen. 21 , 510 – 519 . Google Scholar Crossref Search ADS PubMed WorldCat Czeh B. , Abraham H. , Tahtakran S. , Houser C. R. , Seress L. ( 2013 ). Number and regional distribution of GAD65 mRNA-expressing interneurons in the rat hippocampal formation . Acta Biol. Hung. 64 , 395 – 413 . Google Scholar Crossref Search ADS PubMed WorldCat Devine H. , Patani R. ( 2017 ). The translational potential of human induced pluripotent stem cells for clinical neurology: The translational potential of hiPSCs in neurology . Cell Biol. Toxicol. 33 , 129 – 144 . Google Scholar Crossref Search ADS PubMed WorldCat Douvaras P. , Fossati V. ( 2015 ). Generation and isolation of oligodendrocyte progenitor cells from human pluripotent stem cells . Nat. Protoc. 10 , 1143 – 1154 . Google Scholar Crossref Search ADS PubMed WorldCat Filer D. L. , Kothiya P. , Setzer R. W. , Judson R. S. , Martin M. T. ( 2017 ). tcpl: The ToxCast pipeline for high-throughput screening data . Bioinformatics 33 , 618 – 620 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Gassmann K. , Abel J. , Bothe H. , Haarmann-Stemmann T. , Merk H. F. , Quasthoff K. N. , Rockel T. D. , Schreiber T. , Fritsche E. ( 2010 ). Species-specific differential AhR expression protects human neural progenitor cells against developmental neurotoxicity of PAHs . Environ. Health Perspect. 118 , 1571 – 1577 . Google Scholar Crossref Search ADS PubMed WorldCat Giobbe G. G. , Zagallo M. , Riello M. , Serena E. , Masi G. , Barzon L. , Di Camillo B. , Elvassore N. ( 2012 ). Confined 3D microenvironment regulates early differentiation in human pluripotent stem cells . Biotechnol. Bioeng. 109 , 3119 – 3132 . Google Scholar Crossref Search ADS PubMed WorldCat Gittis A. H. , Nelson A. B. , Thwin M. T. , Palop J. J. , Kreitzer A. C. ( 2010 ). Distinct roles of GABAergic interneurons in the regulation of striatal output pathways . J. Neurosci. 30 , 2223 – 2234 . Google Scholar Crossref Search ADS PubMed WorldCat Gonchar Y. , Wang Q. , Burkhalter A. ( 2007 ). Multiple distinct subtypes of GABAergic neurons in mouse visual cortex identified by triple immunostaining . Front. Neuroanat. 1 , 3 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Grainger A. I. , King M. C. , Nagel D. A. , Parri H. R. , Coleman M. D. , Hill E. J. ( 2018 ). In vitro models for seizure-liability testing using induced pluripotent stem cells . Front. Neurosci. 12 , 590 . Google Scholar Crossref Search ADS PubMed WorldCat Gramowski A. , Jugelt K. , Stuwe S. , Schulze R. , McGregor G. P. , Wartenberg-Demand A. , Loock J. , Schroder O. , Weiss D. G. ( 2006 ). Functional screening of traditional antidepressants with primary cortical neuronal networks grown on multielectrode neurochips . Eur. J. Neurosci. 24 , 455 – 465 . Google Scholar Crossref Search ADS PubMed WorldCat Gramowski A. , Jugelt K. , Weiss D. G. , Gross G. W. ( 2004 ). Substance identification by quantitative characterization of oscillatory activity in murine spinal cord networks on microelectrode arrays . Eur. J. Neurosci. 19 , 2815 – 2825 . Google Scholar Crossref Search ADS PubMed WorldCat Graveland G. A. , Williams R. S. , DiFiglia M. ( 1985 ). A Golgi study of the human neostriatum: Neurons and afferent fibers . J. Comp. Neurol. 234 , 317 – 333 . Google Scholar Crossref Search ADS PubMed WorldCat Guitton M. J. , Caston J. , Ruel J. , Johnson R. M. , Pujol R. , Puel J. L. ( 2003 ). Salicylate induces tinnitus through activation of cochlear NMDA receptors . J. Neurosci. 23 , 3944 – 3952 . Google Scholar Crossref Search ADS PubMed WorldCat Gulyás A. I. , Megı’as M. , Emri Z. , Freund T. F. ( 1999 ). Total number and ratio of excitatory and inhibitory synapses converging onto single interneurons of different types in the CA1 area of the rat hippocampus . J. Neurosci. 19 , 10082 – 10097 . Google Scholar Crossref Search ADS PubMed WorldCat Hofrichter M. , Nimtz L. , Tigges J. , Kabiri Y. , Schröter F. , Royer-Pokora B. , Hildebrandt B. , Schmuck M. , Epanchintsev A. , Theiss S. , et al. ( 2017 ). Comparative performance analysis of human iPSC-derived and primary neural progenitor cells (NPC) grown as neurospheres in vitro . Stem Cell Res. 25 , 72 – 82 . Google Scholar Crossref Search ADS PubMed WorldCat Hu B. Y. , Weick J. P. , Yu J. , Ma L. X. , Zhang X. Q. , Thomson J. A. , Zhang S. C. ( 2010 ). Neural differentiation of human induced pluripotent stem cells follows developmental principles but with variable potency . Proc. Natl. Acad. Sci. U.S.A. 107 , 4335 – 4340 . Google Scholar Crossref Search ADS PubMed WorldCat Illes S. , Fleischer W. , Siebler M. , Hartung H. P. , Dihne M. ( 2007 ). Development and pharmacological modulation of embryonic stem cell-derived neuronal network activity . Exp. Neurol. 207 , 171 – 176 . Google Scholar Crossref Search ADS PubMed WorldCat Ishihara Y. , Kawami T. , Ishida A. , Yamazaki T. ( 2012 ). Tributyltin induces oxidative stress and neuronal injury by inhibiting glutathione S-transferase in rat organotypic hippocampal slice cultures . Neurochem. Int. 60 , 782 – 790 . Google Scholar Crossref Search ADS PubMed WorldCat Judson R. , Richard A. , Dix D. J. , Houck K. , Martin M. , Kavlock R. , Dellarco V. , Henry T. , Holderman T. , Sayre P. , et al. ( 2009 ). The toxicity data landscape for environmental chemicals . Environ. Health Perspect. 117 , 685 – 695 . Google Scholar Crossref Search ADS PubMed WorldCat Kavlock R. J. , Austin C. P. , Tice R. R. ( 2009 ). Toxicity testing in the 21st century: Implications for human health risk assessment . Risk Anal. 29 , 485 – 487 ; discussion 492 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Kim D. S. , Lee J. S. , Leem J. W. , Huh Y. J. , Kim J. Y. , Kim H. S. , Park I. H. , Daley G. Q. , Hwang D. Y. , Kim D. W. ( 2010 ). Robust enhancement of neural differentiation from human ES and iPS cells regardless of their innate difference in differentiation propensity . Stem Cell Rev. 6 , 270 – 281 . Google Scholar Crossref Search ADS WorldCat Klatt A. , Salzmann E. , Schneider L. J. , Reifschneider A. , Korneck M. , Hermle P. , Burkle A. , Stoll D. , Kadereit S. ( 2019 ). Toxicity of ionizing radiation (IR) in a human induced pluripotent stem cell (hiPSC)-derived 3D early neurodevelopmental model . Arch. Toxicol. 93 , 2879 – 2893 . Google Scholar Crossref Search ADS PubMed WorldCat Kosnik M. B. , Strickland J. D. , Marvel S. W. , Wallis D. J. , Wallace K. , Richard A. M. , Reif D. M. , Shafer T. J. ( 2020 ). Concentration-response evaluation of ToxCast compounds for multivariate activity patterns of neural network function . Arch. Toxicol. 94 , 469 – 484 . Google Scholar Crossref Search ADS PubMed WorldCat Krencik R. , Zhang S. C. ( 2011 ). Directed differentiation of functional astroglial subtypes from human pluripotent stem cells . Nat. Protoc. 6 , 1710 – 1717 . Google Scholar Crossref Search ADS PubMed WorldCat Lancaster M. A. , Renner M. , Martin C. A. , Wenzel D. , Bicknell L. S. , Hurles M. E. , Homfray T. , Penninger J. M. , Jackson A. P. , Knoblich J. A. ( 2013 ). Cerebral organoids model human brain development and microcephaly . Nature 501 , 373 – 379 . Google Scholar Crossref Search ADS PubMed WorldCat Legradi J. B. , Di Paolo C. , Kraak M. H. S. , van der Geest H. G. , Schymanski E. L. , Williams A. J. , Dingemans M. M. L. , Massei R. , Brack W. , Cousin X. , et al. ( 2018 ). An ecotoxicological view on neurotoxicity assessment . Environ. Sci. Eur. 30 , 46 . Google Scholar Crossref Search ADS PubMed WorldCat Liu G. ( 2004 ). Local structural balance and functional interaction of excitatory and inhibitory synapses in hippocampal dendrites . Nat. Neurosci. 7 , 373 – 379 . Google Scholar Crossref Search ADS PubMed WorldCat Makinen M. E. , Yla-Outinen L. , Narkilahti S. ( 2018 ). GABA and gap junctions in the development of synchronized activity in human pluripotent stem cell-derived neural networks . Front. Cell Neurosci. 12 , 56 . Google Scholar Crossref Search ADS PubMed WorldCat Mead A. N. , Amouzadeh H. R. , Chapman K. , Ewart L. , Giarola A. , Jackson S. J. , Jarvis P. , Jordaan P. , Redfern W. , Traebert M. , et al. ( 2016 ). Assessing the predictive value of the rodent neurofunctional assessment for commonly reported adverse events in phase I clinical trials . Regul. Toxicol. Pharmacol. 80 , 348 – 357 . Google Scholar Crossref Search ADS PubMed WorldCat Medalla M. , Gilman J. P. , Wang J. Y. , Luebke J. I. ( 2017 ). Strength and diversity of inhibitory signaling differentiates primate anterior cingulate from lateral prefrontal cortex . J. Neurosci. 37 , 4717 – 4734 . Google Scholar Crossref Search ADS PubMed WorldCat Nakatsu Y. , Kotake Y. , Ohta S. ( 2006 ). Tributyltin-induced cell death is mediated by calpain in PC12 cells . Neurotoxicology 27 , 587 – 593 . Google Scholar Crossref Search ADS PubMed WorldCat Namikawa M. , Sano A. , Tateno T. ( 2017 ). Salicylate-induced suppression of electrically driven activity in brain slices from the auditory cortex of aging mice . Front. Aging Neurosci. 9 , 395 . Google Scholar Crossref Search ADS PubMed WorldCat National Research Council . (NRC). ( 2007 ). Toxicity Testing in the 21st Century: A Vision and a Strategy . The National Academies Press, Washington, DC . . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Olson H. , Betton G. , Robinson D. , Thomas K. , Monro A. , Kolaja G. , Lilly P. , Sanders J. , Sipes G. , Bracken W. , et al. ( 2000 ). Concordance of the toxicity of pharmaceuticals in humans and in animals . Regul. Toxicol. Pharmacol. 32 , 56 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat Ongur D. , Price J. L. ( 2000 ). The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans . Cereb. Cortex 10 , 206 – 219 . Google Scholar Crossref Search ADS PubMed WorldCat Pak C. , Danko T. , Zhang Y. , Aoto J. , Anderson G. , Maxeiner S. , Yi F. , Wernig M. , Sudhof T. C. ( 2015 ). Human neuropsychiatric disease modeling using conditional deletion reveals synaptic transmission defects caused by heterozygous mutations in NRXN1 . Cell Stem Cell 17 , 316 – 328 . Google Scholar Crossref Search ADS PubMed WorldCat Pamies D. , Barreras P. , Block K. , Makri G. , Kumar A. , Wiersma D. , Smirnova L. , Zang C. , Bressler J. , Christian K. M. , et al. ( 2017 ). A human brain microphysiological system derived from induced pluripotent stem cells to study neurological diseases and toxicity . ALTEX 34 , 362 – 376 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Pasca S. P. ( 2018 ). The rise of three-dimensional human brain cultures . Nature 553 , 437 – 445 . Google Scholar Crossref Search ADS PubMed WorldCat Peinado A. ( 2001 ). Immature neocortical neurons exist as extensive syncitial networks linked by dendrodendritic electrical connections . J. Neurophysiol. 85 , 620 – 629 . Google Scholar Crossref Search ADS PubMed WorldCat Pohlkamp T. , David C. , Cauli B. , Gallopin T. , Bouche E. , Karagiannis A. , May P. , Herz J. , Frotscher M. , Staiger J. F. , et al. ( 2014 ). Characterization and distribution of Reelin-positive interneuron subtypes in the rat barrel cortex . Cereb. Cortex 24 , 3046 – 3058 . Google Scholar Crossref Search ADS PubMed WorldCat Prince L. M. , Aschner M. , Bowman A. B. ( 2019 ). Human-induced pluripotent stems cells as a model to dissect the selective neurotoxicity of methylmercury . Biochim. Biophys. Acta Gen. Subj. 1863 , 129300 . Google Scholar Crossref Search ADS PubMed WorldCat Qu G. J. , Ma J. , Yu Y. C. , Fu Y. ( 2016 ). Postnatal development of GABAergic interneurons in the neocortical subplate of mice . Neuroscience 322 , 78 – 93 . Google Scholar Crossref Search ADS PubMed WorldCat Roberts R. C. , Gaither L. A. , Peretti F. J. , Lapidus B. , Chute D. J. ( 1996 ). Synaptic organization of the human striatum: A postmortem ultrastructural study . J. Comp. Neurol. 374 , 523 – 534 . Google Scholar Crossref Search ADS PubMed WorldCat Rorig B. , Klausa G. , Sutor B. ( 1996 ). Intracellular acidification reduced gap junction coupling between immature rat neocortical pyramidal neurones . J. Physiol. 490 (Pt 1) , 31 – 49 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Rudy B. , Fishell G. , Lee S. , Hjerling-Leffler J. ( 2011 ). Three groups of interneurons account for nearly 100% of neocortical GABAergic neurons . Dev. Neurobiol. 71 , 45 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat Salaj M. , Druga R. , Cerman J. , Kubova H. , Barinka F. ( 2015 ). Calretinin and parvalbumin immunoreactive interneurons in the retrosplenial cortex of the rat brain: Qualitative and quantitative analyses . Brain Res. 1627 , 201 – 215 . Google Scholar Crossref Search ADS PubMed WorldCat Silva M. C. , Haggarty S. J. ( 2019 ). Human pluripotent stem cell-derived models and drug screening in CNS precision medicine . Ann. N. Y. Acad. Sci. 1471 , 18 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat Sloan S. A. , Darmanis S. , Huber N. , Khan T. A. , Birey F. , Caneda C. , Reimer R. , Quake S. R. , Barres B. A. , Paşca S. P. ( 2017 ). Human astrocyte maturation captured in 3D cerebral cortical spheroids derived from pluripotent stem cells . Neuron 95 , 779 – 790.e6 . Google Scholar Crossref Search ADS PubMed WorldCat Strickland J. D. , Martin M. T. , Richard A. M. , Houck K. A. , Shafer T. J. ( 2018 ). Screening the ToxCast phase II libraries for alterations in network function using cortical neurons grown on multi-well microelectrode array (mwMEA) plates . Arch. Toxicol. 92 , 487 – 500 . Google Scholar Crossref Search ADS PubMed WorldCat Su Y. Y. , Luo B. , Jin Y. , Wu S. H. , Lobarinas E. , Salvi R. J. , Chen L. ( 2012 ). Altered neuronal intrinsic properties and reduced synaptic transmission of the rat’s medial geniculate body in salicylate-induced tinnitus . PLoS One 7 , e46969 . Google Scholar Crossref Search ADS PubMed WorldCat Sun H. , Xia M. , Austin C. P. , Huang R. ( 2012 ). Paradigm shift in toxicity testing and modeling . AAPS J. 14 , 473 – 480 . Google Scholar Crossref Search ADS PubMed WorldCat Tchieu J. , Calder E. L. , Guttikonda S. R. , Gutzwiller E. M. , Aromolaran K. A. , Steinbeck J. A. , Goldstein P. A. , Studer L. ( 2019 ). NFIA is a gliogenic switch enabling rapid derivation of functional human astrocytes from pluripotent stem cells . Nat. Biotechnol. 37 , 267 – 275 . Google Scholar Crossref Search ADS PubMed WorldCat Thomas R. S. , Paules R. S. , Simeonov A. , Fitzpatrick S. C. , Crofton K. M. , Casey W. M. , Mendrick D. L. ( 2018 ). The US Federal Tox21 Program: A strategic and operational plan for continued leadership . ALTEX 35 , 163 – 168 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Thomas R. S. , Bahadori T. , Buckley T. J. , Cowden J. , Deisenroth C. , Dionisio K. L. , Frithsen J. B. , Grulke C. M. , Gwinn M. R. , Harrill J. A. , Higuchi M. , Houck K. A. , et al. ( 2019 ). The Next generation blueprint of computational toxicology at the U.S. Environmental protection agency . Toxicol. Sci . 169 , 317 – 332 . Google Scholar Crossref Search ADS PubMed WorldCat Tice R. R. , Austin C. P. , Kavlock R. J. , Bucher J. R. ( 2013 ). Improving the human hazard characterization of chemicals: A Tox21 update . Environ. Health Perspect. 121 , 756 – 765 . Google Scholar Crossref Search ADS PubMed WorldCat Tukker A. M. , de Groot M. W. , Wijnolts F. M. , Kasteel E. E. , Hondebrink L. , Westerink R. H. ( 2016 ). Is the time right for in vitro neurotoxicity testing using human iPSC-derived neurons? ALTEX 33 , 261 – 271 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Tukker A. M. , Wijnolts F. M. J. , de Groot A. , Westerink R. H. S. ( 2018 ). Human iPSC-derived neuronal models for in vitro neurotoxicity assessment . Neurotoxicology 67 , 215 – 225 . Google Scholar Crossref Search ADS PubMed WorldCat United States. Congress. Senate. Committee on Environment and Public Works . ( 2015 ). Legislative hearing on the Frank R. Lautenberg Chemical Safety for the 21st Century Act (S. 697): Hearing before the Committee on Environment and Public Works, United States Senate, One Hundred Fourteenth Congress, first session, March 18, 2015. U.S. Government Publishing Office. For sale by the Superintendent of Documents, Washington, DC. S hrg. Valdivia P. , Martin M. , LeFew W. R. , Ross J. , Houck K. A. , Shafer T. J. ( 2014 ). Multi-well microelectrode array recordings detect neuroactivity of ToxCast compounds . Neurotoxicology 44 , 204 – 217 . Google Scholar Crossref Search ADS PubMed WorldCat Valiullina F. , Akhmetshina D. , Nasretdinov A. , Mukhtarov M. , Valeeva G. , Khazipov R. , Rozov A. ( 2016 ). Developmental changes in electrophysiological properties and a transition from electrical to chemical coupling between excitatory layer 4 neurons in the rat barrel cortex . Front. Neural Circuits 10 , 1 . Google Scholar Crossref Search ADS PubMed WorldCat van Pelt J. , Vajda I. , Wolters P. S. , Corner M. A. , Ramakers G. J. ( 2005 ). Dynamics and plasticity in developing neuronal networks in vitro . Prog. Brain Res. 147 , 173 – 188 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Vassallo A. , Chiappalone M. , De Camargos Lopes R. , Scelfo B. , Novellino A. , Defranchi E. , Palosaari T. , Weisschu T. , Ramirez T. , Martinoia S. , et al. ( 2017 ). A multi-laboratory evaluation of microelectrode array-based measurements of neural network activity for acute neurotoxicity testing . Neurotoxicology 60 , 280 – 292 . Google Scholar Crossref Search ADS PubMed WorldCat Wagenaar D. A. , Pine J. , Potter S. M. ( 2006 ). An extremely rich repertoire of bursting patterns during the development of cortical cultures . BMC Neurosci. 7 , 11 . Google Scholar Crossref Search ADS PubMed WorldCat Wallace K. , Strickland J. D. , Valdivia P. , Mundy W. R. , Shafer T. J. ( 2015 ). A multiplexed assay for determination of neurotoxicant effects on spontaneous network activity and viability from microelectrode arrays . Neurotoxicology 49 , 79 – 85 . Google Scholar Crossref Search ADS PubMed WorldCat Wang H. T. , Luo B. , Zhou K. Q. , Xu T. L. , Chen L. ( 2006 ). Sodium salicylate reduces inhibitory postsynaptic currents in neurons of rat auditory cortex . Hear Res. 215 , 77 – 83 . Google Scholar Crossref Search ADS PubMed WorldCat Wang S. , Bates J. , Li X. , Schanz S. , Chandler-Militello D. , Levine C. , Maherali N. , Studer L. , Hochedlinger K. , Windrem M. , et al. ( 2013 ). Human iPSC-derived oligodendrocyte progenitor cells can myelinate and rescue a mouse model of congenital hypomyelination . Cell Stem Cell 12 , 252 – 264 . Google Scholar Crossref Search ADS PubMed WorldCat Wang X. , Bukoreshtliev N. V. , Gerdes H. H. ( 2012 ). Developing neurons form transient nanotubes facilitating electrical coupling and calcium signaling with distant astrocytes . PLoS One 7 , e47429 . Google Scholar Crossref Search ADS PubMed WorldCat Wapinski O. L. , Vierbuchen T. , Qu K. , Lee Q. Y. , Chanda S. , Fuentes D. R. , Giresi P. G. , Ng Y. H. , Marro S. , Neff N. F. , et al. ( 2013 ). Hierarchical mechanisms for direct reprogramming of fibroblasts to neurons . Cell 155 , 621 – 635 . Google Scholar Crossref Search ADS PubMed WorldCat Waring M. J. , Arrowsmith J. , Leach A. R. , Leeson P. D. , Mandrell S. , Owen R. M. , Pairaudeau G. , Pennie W. D. , Pickett S. D. , Wang J. , et al. ( 2015 ). An analysis of the attrition of drug candidates from four major pharmaceutical companies . Nat. Rev. Drug Discov. 14 , 475 – 486 . Google Scholar Crossref Search ADS PubMed WorldCat Wheeler AR ( 2019 ). Memorandum: Directive to prioritize efforts to reduce animal testing. September 10, 2019. Available at: https://www.epa.gov/sites/production/files/2019-09/documents/image2019-09-09-231249.pdf. Accessed April 24, 2020). Xu J. C. , Fan J. , Wang X. , Eacker S. M. , Kam T. I. , Chen L. , Yin X. , Zhu J. , Chi Z. , Jiang H. , et al. ( 2016 ). Cultured networks of excitatory projection neurons and inhibitory interneurons for studying human cortical neurotoxicity . Sci. Transl. Med. 8 , 333ra48 – 333ra48 . Google Scholar Crossref Search ADS PubMed WorldCat Yang G. , Lobarinas E. , Zhang L. , Turner J. , Stolzberg D. , Salvi R. , Sun W. ( 2007 ). Salicylate induced tinnitus: Behavioral measures and neural activity in auditory cortex of awake rats . Hear Res. 226 , 244 – 253 . Google Scholar Crossref Search ADS PubMed WorldCat Yang N. , Chanda S. , Marro S. , Ng Y. H. , Janas J. A. , Haag D. , Ang C. E. , Tang Y. , Flores Q. , Mall M. , et al. ( 2017 ). Generation of pure GABAergic neurons by transcription factor programming . Nat. Methods 14 , 621 – 628 . Google Scholar Crossref Search ADS PubMed WorldCat Yang W. , Sun Q. Q. ( 2018 ). Circuit-specific and neuronal subcellular-wide E-I balance in cortical pyramidal cells . Sci. Rep. 8 , 3971 . Google Scholar Crossref Search ADS PubMed WorldCat Young A. , Sun Q. Q. ( 2009 ). GABAergic inhibitory interneurons in the posterior piriform cortex of the GAD67-GFP mouse . Cereb. Cortex 19 , 3011 – 3029 . Google Scholar Crossref Search ADS PubMed WorldCat Zhang Y. , Pak C. , Han Y. , Ahlenius H. , Zhang Z. , Chanda S. , Marro S. , Patzke C. , Acuna C. , Covy J. , et al. ( 2013 ). Rapid single-step induction of functional neurons from human pluripotent stem cells . Neuron 78 , 785 – 798 . Google Scholar Crossref Search ADS PubMed WorldCat Author notes Lorena Saavedra, Kathleen Wallace, and Theresa F. Freudenrich contributed equally to this study. © The Author(s) 2021. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Comparison of Acute Effects of Neurotoxic Compounds on Network Activity in Human and Rodent Neural Cultures
JF - Toxicological Sciences
DO - 10.1093/toxsci/kfab008
DA - 2021-02-04
UR - https://www.deepdyve.com/lp/oxford-university-press/comparison-of-acute-effects-of-neurotoxic-compounds-on-network-2NaSjApfe0
SP - 1
EP - 1
VL - Advance Article
IS - 
DP - DeepDyve
ER -