Citation:

Mack, C., D. Hall, D. Herr, AND Tim Shafer. Low and High-Frequency Field Potentials of Cortical Networks Exhibit Distinct Responses to Chemicals. Experimental Biology 2017, Chicago, Illinois, April 22 - 26, 2017.

Impact/Purpose:

This is a report of an enhancement of an in vitro assay to detect neurotoxicant effects on neural function.

Description:

Neural networks grown on microelectrode arrays (MEAs) have become an important, high content in vitro assay for assessing neuronal function. MEA experiments typically examine high- frequency (HF) (>200 Hz) spikes, and bursts which can be used to discriminate between different pharmacological agents/chemicals. However, normal brain activity is additionally composed of integrated low-frequency (0.5-100 Hz) field potentials (LFPs) which are filtered out of MEA recordings. The objective of this study was to characterize the relationship between HF and LFP neural network signals, and to assess the relative sensitivity of LFPs to selected neurotoxicants. Rat primary cortical cultures were grown on glass, single-well MEA chips. Spontaneous activity was sampled at 25 kHz and recorded (5 min) (Multi-Channel Systems) from mature networks (14 days in vitro). HF (spike, mean firing rate, MFR) and LF (power spectrum, amplitude) components were extracted from each network and served as its baseline (BL). Next, each chip was treated with either 1) a positive control, bicuculline (BIC, 25μM) or domoic acid (DA, 0.3μM), 2) or a negative control, acetaminophen (ACE, 100μM) or glyphosate (GLY, 100μM), 3) a solvent control (H2O or DMSO:EtOH), or 4) a neurotoxicant, (carbaryl, CAR 5, 30μM ; lindane, LIN 1, 10μM; permethrin, PERM 25, 50μM; triadimefon, TRI 5, 65μM). Post treatment, 5 mins of spontaneous activity was recorded and analyzed. As expected positive controls BIC increased (224%) and DA decreased (30%) HF MFR compared to BL. The solvents and negative controls, ACE (110%), GLY (99%), H2O (124%) and DMSO:EtOH (94%), had no effect on MFR. The neurotoxicants, LIN (277%) and PERM (134%) increased MFR, while TRI (65%), and CAR (71%) decreased MFR. LF power spectra (PS) were obtained by averaging the 5 min recordings in 1 sec intervals across each network, with a Kaiser window applied (MatLab). Subsequently, spectral amplitudes were divided into 5 frequency bands, δ (1-4Hz), θ (4-8Hz), α (8-14Hz), β (14-30Hz), and γ (30-50Hz). LF PS total power is reported as the percent change from BL in amplitude (μV) per sec (Hz) per frequency band. From theta to gamma (4-50 Hz) H2O power increased 5-6% and DMSO:EtOH power was unchanged (<1%). Negative control GLY had no effect and ACE slightly increased power (2-7%) with increasing frequency. Positive control BIC significantly increased power (19-83%), however DA did not affect LF power. Other than a 3% decrease in delta, CAR did not affect LFPs. PERM increased spectral power 6-8% across all frequency bands. TRI exposure resulted in a concentration and frequency dependent decrease (3-10%) for all bands. LIN increased amplitude with increasing frequency from delta (10%, theta 18%) to alpha (28%), and then reversed from beta (24%) to gamma (15%). These results indicate LFP activity is sensitive to neurotoxic insult, and that LFP and HF signals may offer distinct information regarding neural network function. (This abstract does not represent Agency Policy)

Record Details: