hPSC-derived microglia shape neuronal morphology and enhance network activity in vitro 173 5 Multiwell-Screen Software (Multi Channel Systems). Raw data was filtered using high-pass 2nd order Butterworth filter with 100 Hz cut-off and low-pass 4th order Butterworth with 3500 Hz cut-off. Dysfunctional electrodes were excluded after visual inspection. MultiwellAnalyzer Setup Software (Multi Channel Systems) was used to derive activity measures such as active channels, spike rate, burst count and network burst count. Spike detection threshold was set at -5.0 standard deviations from baseline. Active electrodes were defined as 10 spikes per minute with a minimum amplitude of 20 µV. Bursts started when at least 4 consecutive spikes were each less than 50 milliseconds (ms) apart and bursts ended when a spike was more than 50 ms after the previous. New burst detection started 100 ms after the last burst. Network bursts were detected if 25% or more of the electrodes in one well had simultaneous bursts and if 50% or more of the electrodes were active during this time window. Immunostaining Cultures were fixed, after 2, 4 and 6 weeks of microglia-like cell addition, using 2% PFA in culture medium for 20 minutes followed by 4% PFA in PBS for 20 minutes. Fixed cells were washed with PBS followed by blocking with blocking buffer (PBS + 5% NGS + 0.3% Triton X100 + 0.1% BSA) for 1 hour at room temperature. Primary antibodies, made up in blocking buffer, were incubated at 4°C overnight. After washing off excess of primary antibodies (See Supplementary Table II), secondary antibodies (Goat-anti-chicken/mouse/guinea pig/rabbit Alexa Fluor 568/647) were incubated for 2 hour at room temperature. Cells were washed again and nuclei were visualized using DAPI. High content screening was performed using the Cell Insight CX7 HCS platform (Thermo Fisher) in widefield mode with either a 10x or 20x objective with 25 images per well, an overview of number of analyzed images is displayed in Supplementary Table V. Cellomics Images were analyzed with Columbus analysis software (v2.5.2; PerkinElmer) by in-house developed scripts. From each well 25 images were acquired and quantified. Briefly, microglia numbers were determined by quantifying the GFP+ signal using the Find Image Region module. Identified regions that were excessively large (>5000 µm, probably caused by overlapping microglia), small (<400 µm), or on the border of the image were omitted. All remaining regions were considered microglia-like cells and their morphology properties were calculated using the Calculate Morphology Properties module, an average per well was obtained from the software. Nuclear debris were quantified by detecting all DAPI positive
RkJQdWJsaXNoZXIy MTk4NDMw