New in vitro phenotypic assay for epilepsy: fluorescent measurement of synchronized neuronal calcium oscillations

Abstract

Research in the epilepsy field is moving from a primary focus on controlling seizures to addressing disease pathophysiology. This requires the adoption of resource- and time-consuming animal models of chronic epilepsy which are no longer able to sustain the testing of even moderate numbers of compounds. Therefore, new in vitro functional assays of epilepsy are needed that are able to provide a medium throughput while still preserving sufficient biological context to allow for the identification of compounds with new modes of action. Here we describe a robust and simple fluorescence-based calcium assay to measure epileptiform network activity using rat primary cortical cultures in a 96-well format. The assay measures synchronized intracellular calcium oscillations occurring in the population of primary neurons and is amenable to medium throughput screening. We have adapted this assay format to the low magnesium and the 4-aminopyridine epilepsy models and confirmed the contribution of voltage-gated ion channels and AMPA, NMDA and GABA receptors to epileptiform activity in both models. We have also evaluated its translatability using a panel of antiepileptic drugs with a variety of modes of action. Given its throughput and translatability, the calcium oscillations assay bridges the gap between simplified target-based screenings and compound testing in animal models of epilepsy. This phenotypic assay also has the potential to be used directly as a functional screen to help identify novel antiepileptic compounds with new modes of action, as well as pathways with previously unknown contribution to disease pathophysiology.

Figure 1. High-density cortical cultures develop spontaneous intracellular calcium oscillations.

Primary cultures from rat cortex were loaded with the intracellular calcium reporter dye Fluo-4 and imaged in a fluorescent plate reader. Representative images of 10-minute recordings are used for A-C. (A) Up to 9 days in vitro (DIV), the cellular population intracellular calcium level remained constant, with no peak observed during 10-minute recordings. (B) around days 9 and 10, fast elevations in intracellular calcium levels could be detected in some wells. (C-D) After 11-12 DIV the phenotype of calcium oscillations becomes established, with cultures displaying more frequent and rhythmic oscillations. (D) Graph showing the onset of calcium oscillations quantified as number of oscillations per 10-minute recording after following cultures from 3 separate batches of neurons (2 wells per batch and day, minimum 4 wells per time point) over a period of 8 days, from 6 to 14 DIV, and during their third week in vitro. Asterisk denotes sample statistically different from zero (one sample t-test, p<0.05).

Figure 2. Low magnesium-enhanced calcium oscillations are modulated by GABA and glutamate receptors, and ion channels.

(A) Representative recordings of cultures imaged in the presence of normal magnesium levels in the buffer (1 mM, center), high magnesium buffer (3 mM) or low magnesium buffer (0.1 mM). While high magnesium in the buffer arrested or severely reduced the number of oscillations, incubation in 0.1 mM magnesium led to more numerous and stable (similar amplitude) oscillations. (B-D) Response of the low magnesium-enhanced calcium oscillations to GABA receptor agonism (B), NMDA receptor inhibition (MK-501, C) and AMPA receptor inhibition (CNQX, D) injected 4 minutes into the recording (time exposed to the compound denoted by blue bar). (E-G) Response of the calcium oscillations to ion channel inhibitors.

Figure 3. 4-AP-enhanced calcium oscillations are modulated by GABA and glutamate receptors, and ion channels.

(A) Representative recordings of cultures exposed to vehicle or 4-AP 4 minutes into the recording (time exposed to the compound denoted by blue bar is 6 minutes). 4-AP induced a dose-response transient elevation of intracellular calcium baseline accompanied by an increase in frequency. (B-D) Response of the 4-AP-enhanced calcium oscillations to GABA receptor agonism (B), NMDA receptor inhibition (MK-501, C) and AMPA receptor inhibition (CNQX, D). (E-G) Response of the calcium oscillations to ion channel inhibitors.

Figure 4. Changes in Calcium oscillations in response to application of classic antiepileptic drugs.

(A-G) Representative recordings of cultures exposed to antiepileptic drugs that work by enhancing GABA receptor activity (A-B), inhibiting sodium channels (C-D) and inhibiting calcium channels (E-G). While other AEDs showed activity in the model, ethosuximide turned from inactive to excitatory at doses higher than 500µM. Time exposed to the compound denoted by blue bar is 6 minutes.

Figure 5. Changes in calcium oscillations in response to application of antiepileptic drugs working through a variety of modes of action.

(A-E) Representative recordings of cultures exposed to an agonist of cannabinoid CB1 receptors (A), a potassium channel opener (B), showing activity in this model Topiramate was inactive at the highest dose that could be tested (C). Levetiracetam also failed to show activity in this model at concentrations up to 5 mM (D-E) while valproate showed partial activity at concentrations up to 2 mM (G-H) and suppression of calcium oscillations at 5 mM (I). Time exposed to the compound denoted by blue bar is 6 minutes.