Engineering and Characterization of an Optogenetic Model of the Human Neuromuscular Junction

This protocol describes how to engineer functional 3D in vitro human neuromuscular junctions that can be used to investigate neuromuscular pathologies and test potential therapeutics. This technique uses simultaneous optogenetic stimulation and video processing. It's quantified changes in neuromuscular junction function during development and due to external factors, such as neurotoxins and serum from myasthenia gravis patients.

The steps involving the muscle cell and motor neuron seeding into the bioreactor can be challenging at first. Keeping the collagen gel mixture on ice can help extend the time available to seed all of the cells before the gel solidifies. Begin by mixing a 10-to-1 base to curing agent mixture of polydimethylsiloxane, or PDMS, and place the mixture into a vacuum chamber.

Close all the valves and turn on the vacuum to degas the mixture for at least 30 minutes until no air bubbles remain. Pour the mixture into molds and degas the molds in the vacuum chamber for 1 hour, then close the molds with the top half of the mold in the correct orientation. Place a hexagonal steel rod over the center and clamp on both sides.

Next, refill the top with PDMS and cure the molds in a 65-degree Celsius oven for at least 4 hours. After the molds have cooled down to room temperature, remove the platforms from the molds. Meanwhile, so glass cover slips in 1%non-ionic surfactant polyol for 30 minutes, ensuring that the cover slips do not stack and are all properly coated, then rinse them well with distilled water and dry them overnight at 65-degrees Celsius.

Treat the glass cover slips and PDMS platform using a plasma cleaner on high with 6 liters per minute of oxygen for 1 to 2 minutes. Once treated, bond the two together by pressing the PDMS down onto the cover slip for at least 30 seconds. Prepare a 4-to-1 mix of 3 milligrams per milliliter collagen 1 and solubilized basement membrane matrix, then resuspend the myoblasts in this freshly prepared collagen mixture.

Next, add 15 microliters of this cell-collagen suspension to each muscle chamber of the bioreactor, making sure to spread the suspension across both pillars using the pipette tip. Allow the cell gel mixture to polymerize at 37 degrees Celsius for 30 minutes, then fill each bioreactor well with 450 microliters of myotonic growth media. On day 11 of motoneuron differentiation, transfer the cells and media to a 50-milliliter tube and allow the neurospheres to settle to the bottom for 5 minutes.

Aspirate the supernatant and resuspend the cells in 15 milliliters of motoneuron suspension culture medium supplemented with 20 nanograms per milliliter of brain-derived neurotrophic factor and 10 nanograms per milliliter of glial cell-derived neurotrophic factor. On day 14 of the differentiation protocols, seed the motoneurons into the platform. Prepare a 4-to-1 gel mixture of 2 milligrams per milliliter collagen 1 and solubilized basement membrane matrix.

Use a 400-nanometer cell strainer to select large neurospheres and resuspend them in the prepared gel mixture. Carefully aspirate media from the reservoir and the neurosphere well, then add 15 microliters of the gel mixture into the neurosphere channel. Load a 10-microliter pipette with 10 microliters of gel and pick one neurosphere.

Deposit the neurosphere into the neurosphere channel, ensuring that it is in the chamber, then slowly raise the pipette while releasing the remaining gel once the neurosphere is deposited. Allow the gel to polymerize for 30 minutes at 37 degrees Celsius, then add 450 microliters of NbActiv4 supplemented with 20 nanograms per milliliter of brain-derived neurotrophic factor and 10 nanograms per milliliter of glial cell-derived neurotrophic factor to the reservoirs. For imaging, use an inverted fluorescent microscope with a scientific complementary metal-oxide semiconductor camera software.

Set the Binning to 2-by-2, Exposure to 20 milliseconds, Rolling shutter on, Readout Rate to 540 megahertz, dynamic range to 12-bit and Gain 1 and Sensor Mode to Overlap. Use a 2x objective on the microscope to image the microtissues and attach a live-cell chamber to the microscope stage. Select the region of interest that contains the innervated skeletal tissues to minimize the file size and processing time, then place a 594 nanometer long-pass emission filter between the sample and the imaging objective to filter out blue light pulses from the camera.

Next, place a rectangular 4-well plate containing four bioreactors into the live-cell chamber, then click live view and center and focus the image with the desired ROI. Download the custom macro code from the GitHub folder to control the stage position, the Arduino board, and video acquisition, then set the output movie as day underscore tissue group underscore tissue name underscore experiment dot ND2. Run the macro code with the desired X and Y coordinates set on the stage and acquire a fast time lapse with 1700 frames at 50 frames per second.

After imaging, replace the media and return the samples to the incubator. Allow at least 24 hours between image acquisition sessions to avoid tissue fatigue. For movie processing, download the files from the GitHub folder and use the custom MATLAB code to process the videos in batch analysis.

Open recursive OS analysis script to have all the acquired videos in the same folder. Adjust the pre-analysis parameters and run the recursiveOSAnalysis script to analyze the movies through parallel processing. Adjust the post-analysis parameters like baseline time, peak threshold, min-min prominence, and min-min width as needed.

Finally, generate video and graph output by running recursiveOSMovie to generate a video file for each tissue with its respective contractility trace. This protocol was tested using two systems with different scales, a microfluidic device yielding 1-millimeter-long skeletal muscle tissues comprising approximately 20, 000 myoblasts and the open-well bioreactor system resulting in 4-millimeter-long muscle tissues comprising approximately 450 myoblasts. An optical setup was built to allow for controlled stimulation of motoneurons with blue light and the NMJ function was elevated using a custom microscope macro script paired with a custom Arduino code.

This code includes tunable adjustable parameters and determines if a contraction is triggered based on the time between a light pulse and the start of the contractility peak. The system captured improvement of NMJ function in tissues showing an increase in triggered responses one week after tissue fabrication, followed by a stable function for an additional week. Alpha-bungarotoxin stopped all triggered and spontaneous contractions resulting in a complete disruption of NMJ function, demonstrating that light stimulation of the tissues requires a functional NMJ.

Analysis of the tissues treated with 20%myasthenia gravis serum for 48 hours showed decreased function compared to control tissues treated with serum from healthy donors. 48 hours after removing and washing out the serum, the engineered NMJs showed recovery of function. Make sure all bioreactors are in the oven for equal time to avoid stiffness variability between batches.

Also, be sure to verify motor neuron seeding using the microscope. Following this procedure, metabalomics, proteomics, and CRISPR screening can be performed to gather additional information and to better understand the mechanistic changes caused by diseased serum.