A Multi-Electrode Array Platform for Modeling Epilepsy Using Human Pluripotent Stem Cell-Derived Brain Assembloids

All experimental procedures demonstrated below were conducted in accordance with the ethical guidelines of the University of Michigan Medical School Institutional Review Board and the Human Pluripotent Stem Cell Research Oversight Committee. The iPSC lines used in this protocol and the representative experiments were derived from human foreskin fibroblasts obtained from a commercial source. The details of the cell lines, reagents, and equipment used in this study are listed in the Table of Materials.

1. Derivation of fate-specific brain organoids from iPSCs

NOTE: This protocol contains information for preparing 3 wells in the microwell culture plate from 5 confluent (~85%) wells of a 6-well plate. iPSC maintenance protocol varies depending on cell lines.

1. Prepare the individual cell culture media using the composition detailed in Table 1. Store protected from light at 4 °C for up to 2 weeks.
2. Rinse all wells to be used in the microwell culture plate with 1 mL of DMEM/F-12 once. Add 1 mL/well mTeSR plus + 50 µM Y-27632 and spin the plate at 2000 x g for 5 min at room temperature to remove air bubbles from microwells. Store the plate in a 37 °C 5% CO₂ incubator.
3. Clean regions of differentiated iPSCs with a sterile P10 pipette tip under the microscope in a laminar flow clean bench. Wash off cell debris from the monolayer with 1 mL/well PBS without calcium or magnesium.
4. Dissociate the monolayer into single cells by adding 1 mL/well of pre-warmed cell dissociation reagent and placing the plate in a 37 °C 5% CO₂ incubator for 4-7 min (depending on characteristics and confluency of cell lines) until most cells lift off, with gentle agitation every 2-3 min.
5. Add 4 mL of DMEM/F-12 to each well to dilute the cell dissociation reagent, and moderately triturate with a 10 mL serological pipette. Collect cell suspension from all wells of the same line into a 50 mL conical tube.
6. Spin at 200 x g for 4 min at room temperature, aspirate the supernatant using a 10 mL serological pipette, and resuspend cells in 2 mL of mTeSR plus + 50 µM Y-27632. Take out 10 µL cell suspension, 1:1 (v/v) mix with trypan blue, and count cells with an automated cell counter.
7. Plate 3 x 10⁶ live cells per well, 3 wells per line. Add additional mTeSR plus + 50 µM Y-27632 as needed to bring the total volume to 2 mL/well. Spin the plate at 100 x g for 3 min at room temperature and put it back in the incubator.
   NOTE: The microwell culture plate used here contains 300 microwells per well, so plating 3 x 10⁶ cells results in 10,000 cells/microwell.
8. The next day, perform a half-volume media change by gently removing 1 mL of media and slowly adding 1 mL of mTeSR plus media (without Y-27632) with a P1000 pipette tip. Keep the tip near the media surface and aspirate as slowly as possible, as not to disturb the aggregating cells at the bottom.
9. Check under an inverted microscope one day after media change, and a clear outline of each aggregate can be seen. Remove the original media from each well with P1000 pipette tips.
10. Moderately spray 0.5 mL of neural induction media (NIM) with dorsomorphin (DM) and SB-4321542 (SB) into each well with sterile, cut P1000 tips. Immediately, gently pipet up aggregates suspended in media and transfer them to a sterile 10 cm suspension culture dish. Repeat until all aggregates are transferred. Do NOT triturate.
11. Add additional NIM + SB/DM as needed to bring the total volume to 10 mL/dish. Put the dish on an orbital shaker in a 37 °C 5% CO₂ incubator to prevent spontaneous fusion of aggregates. Mark the transfer date as Day 0 of organoid culture.
12. From Day 1 to the end, follow the media recipe in Table 1 and the media change timeline⁹ di Figure 1. Notably, organoids are divided into two suspension culture dishes designated as "dorsal" and "ventral" on Day 4.
    NOTE: Several key timepoints to keep in mind: organoid expansion begins on Day 6, when media is switched to Neural Differentiation Medium (NDM) with EGF/FGF. Neuronal differentiation is supported starting Day 25 by NDM + BDNF/NT3, and organoids are maintained in NDM without growth factors starting on Day 46.

2. Viral labeling and generation of assembloids

NOTE: Viral labeling^7,9is recommended for recognizing the identity of each region-specific organoid in an assembloid and assigning electrophysiological activity from specific electrodes to assembloid regions. It is optimal if only a single organoid is plated for each MEA well. This protocol can also apply to single organoid plating.

1. Calculate the volumes of virus stock and dilution media to be used: 200 µL media with virus is needed for each organoid to be labeled. Dorsal organoids are labeled with pAAV1-CAG-tdTomato, typically at a dilution ratio of 1:1000 (v/v), or approximately 2 x 10¹⁰ vg/mL. For ventral organoids, 1:300 (v/v), or 2 x 10¹¹ vg/mL pAAV1-mDLX-GFP virus is used.
   NOTE: Dilutions should be titrated to the desired effect and are affected by viral titer.
2. Warm up the appropriate amount of media to 37 °C. Thaw virus stock stored at -80 °C on ice. Label enough 48-well plates to accommodate all labeled organoids.
3. Put a beaker containing 10% bleach in the hood. Expel any materials in contact with viruses into the beaker. Dilute the fully thawed viruses in culture media in separate 15 mL conical tubes, labeled as "dorsal" and "ventral".
4. Move Day 56 organoids from culture dishes to 48-well plates. One organoid per well is recommended to prevent spontaneous fusion in static culture. Remove as much residual culture media as much as possible.
5. Add 200 µL of virus-media mixture to each well, respectively. Put the plates back in the 37 °C 5% CO₂ incubator. Tilt the plates on a sterile empty plate to help concentrate viral particles to the bottom of the well and optimize transduction efficiency.
   NOTE: Viral particles tend to sink to the bottom of the 15 mL conical tube, so frequently inverting the tube to mix viruses with media before adding them to each well is very important to ensure uniform transduction efficiency for all organoids.
6. The next day (Day 57), add an additional 800 µL of culture media to each well to bring the total volume to 1 mL/well.
7. Three days later (Day 60), perform a full media change on the labeled organoids. Bleach any materials in contact with viral particles.
8. Three days later (Day 63), check for the fluorescence expression under an epifluorescence inverted microscope. If the signal is strong, begin generating assembloids (see step 2.9).
9. Rinse all labeled organoids with 1 mL/well PBS twice to prevent cross-contamination by remaining viral particles during organoid fusion. Use a cut P1000 pipette tip to transfer one ventral organoid to each dorsal well (or vice versa) and relabel the assembloid plate accordingly. Tilt the plates as in step 2.5 for better fusion.
10. Carefully perform a half-volume media change 3 days after the fusion procedure (Day 66) without interrupting the assembloids. It usually takes about a week to form stable assembloids (Day 70).

3. Placing assembloids on pre-treated MEA plates

NOTE: It takes at least 4 days to complete the surface preparation procedures of MEA plates before plating assembloids. To save time, fusing dorsal and ventral organoids (i.e., step 2.9) and MEA pre-treatment can be performed in parallel.

1. Prepare reagents for surface pre-treatment and coating.
   1. To prepare 1% detergent/enzyme solution, dissolve 0.5 g of enzyme/detergent powder in 50 mL of deionized water. Vortex thoroughly until the powder is fully dissolved. Sterilize the solution with a sterile tube top vacuum filter in the biosafety cabinet before adding it to MEA wells. Make fresh before each use.
      NOTE: The detergent/enzyme sterilizes the MEA surface, removes any residual oils from the MEA manufacturing process, and ensures that the surface is hydrophilic enough to promote interactions with the hydrophilic compound polyethyleneimine (PEI)¹⁸.
   2. To prepare a 0.07% PEI solution¹⁹, begin with making a stock 7% PEI solution by mixing 1 mL of 50% PEI solution into a 15 mL conical tube with 6 mL of 1x borate buffer. Dilute 1 mL of stock 7% PEI solution in 99 mL of 1x borate buffer to achieve the final working concentration. For sterilization, filter through a 0.22 µm filter unit in the biosafety cabinet before use.
      NOTE: PEI, a positively charged polymer, is used in the primary coating process to facilitate the attachment of a negatively charged secondary coating with Basement Membrane Matrix (BMM) and the assembloids²⁰. Take out 1 mL of 50% PEI by pouring instead of pipetting into the conical tube, since the stock solution is very viscous. The amount needed can be more easily measured by weighing it into a conical tube, and the 50% PEI solution has a density of 1.08 g/mL. The 7% stock PEI solution can be stored in 1 mL aliquots at -20 °C for up to a month. Carry out all subsequent steps in the biosafety cabinet.
2. In consideration of the size of each assembloid, 6-well MEA plates with 64 electrodes per well are used to accommodate one assembloid per well. Calculate the number of MEA plates needed for the experiments to ensure enough biological replicates.
3. Add 1 mL of 1% detergent/enzyme solution to each well of a sterile MEA plate. Be careful not to damage electrodes with the pipette tip. If using a vacuum aspirator for washes, do not allow the tip to touch the electrode array. Incubate in a 37 °C incubator for 2 h.
4. Remove detergent/enzyme and wash each well five times with 1.5 mL of sterile deionized water. Since detergent/enzyme is not biocompatible, it is important to ensure that the remaining chemicals are completely washed off.
5. Fill each well with 1 mL of culture media for pre-conditioning. Put the plates back to 37 °C 5% CO₂ incubator for 2 days. Exposing the MEA surface to cell culture conditions promotes cell attachment.
6. Remove the culture media and cover the MEA electrodes in each well with 50 µL of 0.07% PEI solution for primary coating. Incubate at 37 °C for 1 h.
7. Aspirate PEI solution completely. Wash each well with 1 mL of sterile deionized water. Perform 3 quick washes in total. Dry the plates at room temperature for 15 min in the biosafety cabinet with the lid off.
8. Cover the electrode area in each well with 50 µL of 1:20 (v/v) BMM diluted in culture media for secondary coating. Put the MEA plates back in the 37 °C incubator overnight. Add sterile water to the compartments surrounding wells to ensure sufficient humidity throughout the course of coating.
9. The next day, aspirate diluted BMM with a vacuum, leaving a thin layer on the surface. If a thick layer has formed, forcefully spray the electrode contact area with culture media using a P1000 pipette tip. Wash as many times as needed to remove any thick chunks.
   NOTE: Residual thick BMM can interfere with electrode contact with the cells.
10. Collect an assembloid with a widely cut P1000 tip and place it on top of the electrodes in a well while transferring as little media as possible.
11. In a laminar flow hood under a dissection scope, carefully push the assembloid with a sterile dissection needle or P20 tip to the desired location, where both dorsal and ventral organoids can be mostly covered by electrodes. Use a P20 tip to remove liquid around the assembloid. Repeat this for all assembloids or a subset, which can be done within a few minutes.
    NOTE: Try NOT to touch or scratch the MEA surface with the pipette tip and/or needle. Try NOT to poke or tear the assembloid with pipette tips. Finish plating as quickly as possible to avoid the organoid completely drying out.
12. Incubate the assembloids at 37 °C for 10 min (without media). Transfer the plates from the incubator to a laminar flow hood and apply 2-3 small drops of 1:50 BMM diluted in culture media on top of the assembloids under a dissection scope. Incubate at 37 °C for another 30 min.
13. Carefully add 3 drops of culture media close to the assembloids under a dissection scope. This keeps the organoids hydrated. Incubate at 37 °C for 1 h.
    NOTE: If the assembloid moves or floats in any subsequent steps, plating should be restarted from step 3.11.
14. Every hour, carefully add 20-50 µL of cell culture media to each well under a dissection scope. This can be done over the course of a day. The goal is to reach at least 250 µL of total volume at the end of the day.
15. The next day, check whether all assembloids are settled and stabilized under the dissection scope. Add an extra 750 µL of culture media to reach a total volume of 1 mL/well. Leave the plates undisturbed for at least 2 days.
16. Carefully perform weekly media changes by taking out 500 µL of culture media and adding 700 µL of neurophysiological basal media. The additional volume of media added accounts for evaporation and can be adjusted depending on how much evaporation the MEA plate being used has. Add sterile water around the wells as needed to minimize evaporation from the MEA wells.

4. MEA recording and data analysis

1. Start MEA recording about a week after switching to neurophysiological basal media (usually around Day 78).
2. Wait until the recording chamber reaches its set temperature (usually 37 °C). Transfer a plate to the device and allow them to equilibrate for at least 5 min prior to recording.
3. Hardware live settings
   1. Set the sampling frequency, i.e., the rate at which the raw voltage data is recorded. The default setting (12.5 kHz) is recommended.
   2. Specify analog mode (usually Neural Spikes, as in this case) to configure the hardware bandwidth and gain for particular applications. For visual monitoring of activity during recording, use Spike Detector and Burst Detector modules.
      NOTE: The analog settings affect how the raw voltage data are recorded and cannot be changed after data are recorded.
   3. (Optional) Apply a hardware digital filter to the recording data. Low pass filter options include a 2 kHz or 3 kHz Kaiser Window. Default settings can be modified after selecting the Analog settings.
   4. Set to referencing by median (default), which is highly recommended for neural recording, as in this case. Referencing improves the quality of low-amplitude signals by reducing the noise interference common to groups of channels.
      NOTE: The parameters above can be adjusted as needed depending on the activity seen in the assembloids, but should be kept consistent for all recordings in a given experiment.
4. Add plate maps and descriptions (if needed). Import the plate map from the prior recording for the same plate to ensure consistency and facilitate batch data processing. Ensure the files are named properly.
5. If signals look good (minimal noise and artifact), record the raw data. Generally 2-5 min of recording is enough for a representative sample.
6. Longitudinally record electrical activity once or twice a week, over 5-15 min, depending on the experimental design. Wait for at least 24 h after every media change before starting the next round of recording, as the electrical activity drops immediately after each media change.
7. (Optional) If electrical activity needs to be localized and distinguished between dorsal and ventral organoids, or if the activity of each electrode needs to be analyzed separately, obtain an image of each assembloid under bright field and epifluorescence (Figure 2B) after each recording.
8. When analyzing the data, apply an analysis configuration, including a spike/burst detector and statistics compiler. Use an adaptive threshold set to 5.5-6 standard deviations from baseline for spike detection. Identify bursts as an activity with ≥5 spikes in 100 ms¹³. Define network activity when over 25% of total active electrodes fire within 100 ms with a minimum of 50 spikes per network burst^13,14.
   NOTE: These parameters are modulated depending on the firing characteristics of a specific experiment. To increase sensitivity for low amplitude spikes, the adaptive threshold for spike detection can be lowered. For burst detection, various methods can be used, and the parameters can be adapted to detect bursts of spikes that differ significantly from the background activity. The network activity parameters may be modulated to a lower percentage of electrodes, especially when assembloids do not cover the entire field of electrodes. All parameters should remain consistent for the duration of an experiment to allow longitudinal comparisons.
9. Run the Statistics Compiler module for batch processing while replaying the recorded data. Export advanced metrics and/or spike output files for advanced analysis.
   NOTE: Advanced metrics output contains .csv file with electrode, well, and group averages for mean firing, bursting, and synchrony. Spike output (.spk file) includes waveform traces for all spikes, which can be imported to other platforms for advanced analysis, e.g., spike sorting.

5. End-point pharmacological assay and plate recycling

NOTE: Drug treatment assays can be performed after assembloids are mature enough and electrical activity peaks. Recording of baseline/pre-treatment activity and post-treatment activity need to be conducted on the same day.

1. Add 1-10 µL of drug solution right on top of the assembloid and gently swirl the plate to ensure a uniform mixture. Longer incubation time in drug-containing media is recommended in some cases, e.g., synapse-related drug treatment assays with glutamate/GABA receptor agonists/antagonists (Figure 2E) (see Results section for details).
   NOTE: Adding a small amount of drug solution to each well is recommended to achieve a final working concentration since a large volume change will temporarily decrease electrical activity in general. It is important to ensure that the addition of a vehicle in which the drugs of interest are dissolved does not provoke significant changes in electrophysiological parameters.
2. Perform enough PBS washes (typically 3) and wash out drugs after the post-treatment recording. Conduct an additional recording the next day after the drug wash-out to see whether electrical activity returns to the baseline level.
3. Optionally, forcefully pipet media to separate the assembloids from the MEA plate at the end of the recordings to perform further experiments, e.g., live imaging, fixation with 4% paraformaldehyde for whole-mount immunostaining, or lysis for RNA or protein extraction.
4. For analysis purposes, use at least three technical and biological replicates for each condition. Express electrophysiological data as a mean of n ≥ 3 replicates.
5. To recycle the plate at the end of the experiments, aspirate the remaining media after removing the attached assembloids. Wash once with sterile PBS. Fully cover the metal area with 200 µL/well of 0.25% trypsin-EDTA and put the plate back in a 37 °C incubator overnight.
6. The next day, aspirate trypsin, wash twice with 70% sterile ethanol and three times with sterile deionized water.
7. Air dry the plate in the biosafety cabinet for 15 min with the lid off or until it is completely dry. Seal the cleaned plate in a plastic bag and store it at room temperature until future use.
   NOTE: A well-cleaned MEA plate can be reused three times without significantly losing recording quality.