High-throughput brain activity mapping and machine learning as a foundation for systems neuropharmacology.

Technologies for mapping the spatial and temporal patterns of neural activity have advanced our understanding of brain function in both health and disease. An important application of these technologies is the discovery of next-generation neurotherapeutics for neurological and psychiatric disorders. Here, we describe an in vivo drug screening strategy that combines high-throughput technology to generate large-scale brain activity maps (BAMs) with machine learning for predictive analysis. This platform enables evaluation of compounds' mechanisms of action and potential therapeutic uses based on information-rich BAMs derived from drug-treated zebrafish larvae. From a screen of clinically used drugs, we found intrinsically coherent drug clusters that are associated with known therapeutic categories. Using BAM-based clusters as a functional classifier, we identify anti-seizure-like drug leads from non-clinical compounds and validate their therapeutic effects in the pentylenetetrazole zebrafish seizure model. Collectively, this study provides a framework to advance the field of systems neuropharmacology.

In Vitro Screening for Seizure Liability Using Microelectrode Array Technology.

Drug-induced seizure liabilities produce significant compound attrition during drug discovery. Currently available in vitro cytotoxicity assays cannot predict all toxicity mechanisms due to the failure of these assays to predict sublethal target-specific electrophysiological liabilities. Identification of seizurogenic and other electrophysiological effects at early stages of the drug development process is important to ensure that safe candidate compounds can be developed while chemical design is taking place, long before these liabilities are discovered in costly preclinical in vivo studies. The development of a high throughput and reliable in vitro assay to screen compounds for seizure liabilities would de-risk compounds significantly earlier in the drug discovery process and with greater dependability. Here we describe a method for screening compounds that utilizes rat cortical neurons plated onto multiwell microelectrode array plates to identify compounds that cause neurophysiological disruptions. Changes in 12 electrophysiological parameters (spike train descriptors) were measured after application of known seizurogenic compounds and the response pattern was mapped relative to negative controls, vehicle control and neurotoxic controls. Twenty chemicals with a variety of therapeutic indications and targets, including GABAA antagonists, glycine receptor antagonists, ion channel blockers, muscarinic agonist, δ-opioid receptor agonist, dopaminergic D2/adrenergic receptor blocker and nonsteroidal anti-inflammatory drugs, were tested to assess this system. Sixteen of the seventeen seizurogenic/neurotoxic compounds tested positive for seizure liability or neurotoxicity, moreover, different endpoint response patterns for firing rate, burst characteristics and synchrony that distinguished the chemicals into groups relating to target and seizurogenic response emerged from the data. The negative and vehicle control compounds had no effect on neural activity. In conclusion, the multiwell microelectrode array platform using cryopreserved rat cortical neurons is a highly effective high throughput method for reliably screening seizure liabilities within an early de-risking drug development paradigm.

Nonimmobilizers and transitional compounds may produce convulsions by two mechanisms.

UNLABELLED: Some inhaled compounds cause convulsions. To better appreciate the physical basis for this property, we correlated the partial pressures that produced convulsions in rats with the lipophilicity (nonpolarity) and hydrophilicity (polarity) of 45 compounds: 3 n-alkanes, 18 n-haloalkanes, 3 halogenated aromatic compounds, 3 cycloalkanes and 3 halocycloalkanes, 13 halogenated ethers, and 2 noble gases (He and Ne). In most cases, convulsions were quantified by averaging the alveolar partial pressures just below the pressures that caused and slightly higher pressures that did cause clonic convulsions (ED50). The ED50 did not correlate with hydrophilicity (the saline/gas partition coefficient), nor was there an obvious correlation with molecular structure. For 80% of compounds (36 of 45), the ED50 correlated closely (r2 = 0.99) with lipophilicity (the olive oil/gas partition coefficient). Perhaps because they block the effect of GABA on GABA(A) receptors, five compounds were more potent than would be predicted from their lipophilicity. Conversely, four compounds may have been less potent than would be predicted because they (like conventional inhaled anesthetics) enhance the effect of GABA on GABA(A) receptors. IMPLICATIONS: Nonimmobilizers and transitional compounds may produce convulsions by two mechanisms. One correlates with lipophilicity (nonpolarity), and the other correlates with an action on GABA(A) receptors.