Detection of astrocytic slow oscillatory activity and response to seizurogenic compounds using planar microelectrode array

Abstract

Since the development of the planar microelectrode array (MEA), it has become popular to evaluate compounds based on the electrical activity of rodent and human induced pluripotent stem cell (iPSC)-derived neurons. However, there are no reports recording spontaneous human astrocyte activity from astrocyte-only culture sample by MEA. It is becoming clear that astrocytes play an important role in various neurological diseases, and astrocytes are expected to be excellent candidates for targeted therapeutics for the treatment of neurological diseases. Therefore, measuring astrocyte activity is very important for drug development for astrocytes. Recently, astrocyte activity has been found to be reflected in the low-frequency band < 1 Hz, which is much lower than the frequency band for recording neural activity. Here, we separated the signals obtained from human primary astrocytes cultured on MEA into seven frequency bands and successfully recorded the extracellular electrical activity of human astrocytes. The slow waveforms of spontaneous astrocyte activity were observed most clearly in direct current potentials < 1 Hz. We established nine parameters to assess astrocyte activity and evaluated five seizurogenic drug responses in human primary astrocytes and human iPSC-derived astrocytes. Astrocytes demonstrated the most significant dose-dependent changes in pilocarpine. Furthermore, in a principal component analysis using those parameter sets, the drug responses to each seizurogenic compound were separated. In this paper, we report the spontaneous electrical activity measurement of astrocytes alone using MEA for the first time and propose that the MEA measurement focusing on the low-frequency band could be useful as one of the methods to assess drug response in vitro.

Keywords: MEA - microelectrode array; astrocyte; culture; human; iPSC (induced pluripotent stem cell); seizure; slow-oscilatory activity; toxicology.

FIGURE 1

Spontaneous activity in human primary astrocytes detected by MEA. (A,B) Immunofluorescent images of astrocytes cultured on MEA at 8 days in vitro (8 DIV). Immunocytochemistry of GFAP (green), human nuclei (red), cell nuclei by Hoechst 33,258 (blue), merged images in A, and MAP2 (red) in (B) Scale bar = 200 μm. (C) Representative oscillation waveform at the spontaneous activity measurement for 10 min at 7 DIV. (D) The magnified waveform of the red underlined time in (C). (E) Plot of maximum amplitude in 10 min oscillation waveform of each well. Error bars indicate the SEM. (F) Waveforms (upper panel) and their power spectrogram (lower panel) at maximum spectral intensity. (a) astrocytes, (b) no cells, (c) fibroblasts, (d) after 10% DMSO treatment. The vertical axis of the spectrogram shows the linear frequency from 0.1 Hz to 50 Hz, and the color indicates power. A 20 s spectrogram was extracted from the 40 s waveform. (G) Oscillation power of the control experiments. (a) Human primary astrocytes, no cells, fibroblasts, and (b) human primary astrocytes before and after 10% DMSO treatment. unpaired two-tailed t-tests, ^***p < 0.001 versus astrocytes. Error bar, SEM.

FIGURE 2

Analytical parameters of spontaneous activity in human primary astrocytes. (A) Representative oscillation waveform of spontaneous activity in human primary astrocytes at 3 DIV (top), and waveform separated into seven frequency bands (DC, 0.1–1 Hz; delta, 1–4 Hz; theta, 4–8 Hz; alpha, 8–14 Hz; beta, 15–30 Hz; gamma, 35–50 Hz; high gamma, 80–150 Hz). (B) The waveforms in each frequency band were converted into RMS histograms. (C) Schematic diagram of analysis parameters. The red area are subject to analysis for each parameter.

FIGURE 3

Astrocyte activity depends on ion channel modulations. (A) high potassium experiments by cumulative administration of KCl (n = 5). (a) Representative oscillation waveform. (b) Average total RMS and (c) Total oscillations. (B) Inhibitory experiment by VU0134992 (n = 6). (a) Representative oscillation waveform. (b) Average total RMS and (c) Total oscillations. (C) Inhibitory experiment by cilnidipine (n = 4). (a) Representative oscillation waveform. (b) Average total RMS and (c) Total oscillations. KCl and inhibitors were added to astrocytes at 7 DIV. One-way ANOVA followed by Dunnett’s test, *p < 0.05, ^**p < 0.01, ^***p < 0.001 versus control/vehicle. Error bar, SEM.

FIGURE 4

Drug responses of primary astrocytes detected by MEA. (A) Representative oscillation waveform after the cumulative administration of seizurogenic compounds (a) 4-AP, (b) pilocarpine, (c) picrotoxin, (d) PTZ, and (e) chlorpromazine, and neutral compounds (f) acetaminophen and (g) DMSO. Compounds were added to astrocytes during 1–3 weeks of the culture. Vertical scale bar, 40 μV; horizontal scale bar, 20 ms. (B,C) Heatmaps of the analytical parameters of seizurogenic compounds and neutral compounds in DC potential (B) and in delta band (C). 4-AP (n = 6), pilocarpine (n = 6), picrotoxin (n = 6), PTZ (n = 6), chlorpromazine (n = 6), acetaminophen (n = 6), and DMSO (n = 6). (D) Dose-dependent changes of average total RMS in DC potential (upper) and in delta band (lower). One-way ANOVA followed by Dunnett’s test, *p < 0.05, **p < 0.01 versus vehicle. Error bar, SEM.

FIGURE 5

Scatterplots of principal component analysis (PCA) using the effective parameter set for detecting the drug responses of primary astrocytes. There was a clear separation between the seizurogenic and neutral compounds and between all seizurogenic compounds. In neutral compounds, DMSO and acetaminophen were not separated. DMSO (n = 6, blue), acetaminophen (n = 6, gray), 4-AP (n = 6, red), chlorpromazine (n = 6, orange), PTZ (n = 6, purple), picrotoxin (n = 6, green), and pilocarpine (n = 6, yellow). Higher concentrations are indicated by darker colored symbols.

FIGURE 6

Drug responses of human iPSC-derived astrocytes detected by MEA. (A) Representative oscillation waveform after the cumulative administration of seizurogenic compounds (a) 4-AP, (b) pilocarpine, (c) picrotoxin, (d) PTZ, and (e) chlorpromazine, and neutral compounds (f) acetaminophen and (g) DMSO. Compounds were added to astrocytes during 2–3 weeks of the culture. Vertical scale bar, 40 μV; horizontal scale bar, 20 ms. (B,C) Heatmaps of the analytical parameters of seizurogenic compounds (4-AP, n = 6; pilocarpine, n = 6; picrotoxin, n = 6; chlorpromazine, n = 6) and neutral compounds (acetaminophen, n = 6; DMSO, n = 6) in DC potential (A) and in delta band (B). One-way ANOVA followed by Dunnett’s test, *p < 0.05, **p < 0.01 versus vehicle.

FIGURE 7

Scatterplots of PCA using the effective parameter set for detecting the drug responses of iPSC-derived astrocytes. (A) DMSO (n = 6, blue), acetaminophen (n = 6, gray), 4-AP (n = 6, red), chlorpromazine (n = 6, orange), PTZ (n = 6, purple), picrotoxin (n = 6, green), and pilocarpine (n = 6, yellow). Higher concentrations are indicated by darker colored symbols. (B) The magnified scatterplots excluding the plots of pilocarpine and PTZ.