This method can help answer key questions in neuroscience regarding how intercellular communication between multiple human neural cell types contributes to synaptic circuit formation and function. This technique rapidly and systematically generates three-dimensional coculture spheres of functionally-mature astrocytes and neurons from human pluripotent stem cells. These can be utilized for rigorous experimental investigation.
The reproducibility and scalability of this protocol is a defined alternative to the emerging organoid technologies. These show promise for novel drug development and cellular-engraftment therapy to promote neuroregeneration after injury or disease. Begin by seating clusters of HPSCs in two milliliters of HPSC medium containing the rho-kinase inhibitor Y-27632 into each well of an ECM-coated six-well plate.
Maintain the stem cells until they are about 50%confluent, then change the medium to neural medium containing SB431542 and DMH1 to promote neural induction. When cells are about 95%confluent, split each well one-to-six into new ECM-coated wells. On day 14 dissociate the cells with detachment solution, and transfer the contents of each well to a non-coated T25 flask with medium containing Y-27632 to promote the formation of aggregates.
To generate astrocyte progenitors and astrocytes in spontaneously-formed 3D aggregates, switch to neural medium containing EGF and FGF2, and feed every three to four days until astrocyte identity is confirmed at four to six months. By default, this protocol produces dorsal cortical astrocytes. However, astrocyte subtypes can be regionally specified by the addition of patterning morphogens, if desired.
Once a week, when dark centers appear, collect the spheres by centrifugation, then gently dissociate the H-astro aggregates with detachment solution. Only replate those spheres that do not spontaneously attach to avoid passaging non-CNS and non-neural cells. After four to six months of expansion, confirm cell identity in either freeze and cryo-preservation medium as per manufacturer's instructions, or use immediately for experimentation.
For the generation of nontransgenic neuronal progenitors in neurons, or H-neurons, seed day-14 HPSC-derived neural progenitors into ECM-coated six-well plates with two milliliters of neural medium and Y-27632 per well. If working with doxycycline-inducible iNeurons, add neural medium containing doxycycline when the cells are about 35%confluent to induce differentiation and activate gene expression. After two days, feed transgenic and nontransgenic lines with two milliliters of neural media per well per day.
Continue until the cells are ready to lift at 70%confluency. For cocultures, first gently dissociate H-astros from 3D aggregates with detachment solution. Then proceed to dissociate H-neurons, or iNeurons, from a 2D monolayer.
First, remove the medium from the six-well plate, and add 500 microliters of detachment solution to each well containing monolayer cultures. Incubate at 37 degrees Celsius for five minutes. Following the incubation, gently add two to three milliliters of DMEM-F12 medium to remove the attached cells.
Then, collect the cells in medium in a 15-milliliter conical tube, and centrifuge at 300 times G for one minute. Aspirate the supernatant, add one milliliter of fresh medium, and pipette up and down gently with a 1, 000-microliter micro pipette to achieve a single-cell suspension. Count each cell type using a hemocytometer or automated cell-counter.
Add the desired ratio of dissociated H-astros and H-neurons, or iNeurons, in a total volume of two milliliters to each well of a microplate. Centrifuge the plate at 100 times G for three minutes, and return to the incubator for two days. After one day, densely-packed microspheres should have formed in the plate.
Use a 1, 000-microliter micro pipette to gently remove spheres from the micro wells. Lightly apply force to the bottom of micro wells with medium to remove any additional adhered spheres. After collecting all spheres in a 15-milliliter conical tube, allow to settle.
Then wash the spheres by first aspirating the old medium, and then adding at least two milliliters of fresh medium. Add the spheres to a spinner flask containing 50 to 60 milliliters of medium. Place the flask in the incubator on a magnetic stir plate set at 60 RPM.
A minimum of three weeks in culture is needed for synapse formation. Begin by coating the surface of each multielectrode array, or MEA, with one milliliter of polyornithine to render the surface hydrophilic, and incubate at 37 degrees Celsius for a minimum of four hours. Following the incubation, remove the polyornithine, and wash the surface of each MEA with deionized water.
Then add one milliliter of extracellular matrix, and return to the incubator for a minimum of four hours. When ready, remove the extracellular matrix, then apply the spheres in 1.5 milliliters of medium onto the MEA surface, ensuring that the spheres are positioned on top of the electrode array. Allow to adhere for two days.
To measure the electrical activity of neural spheres, place the MEA on a temperature-controlled head stage. Use a five-Hertz highpass filter, and a 200-Hertz lowpass filter, and a spike threshold of five times standard deviation, to record spontaneous electrical activity. Save raw data, including spike frequency and amplitude for statistical analysis.
Neural-cell-type restricted protein markers, such as MAP2 for neurons and GFAP for astrocytes, visually demonstrate the evenly-dispersed arrangement and maturity of astrocytes in neurons within 3D spheres. A maximum projection of astrocyte morphology and branching is shown in a representative 3D sphere with H-astros sparsely labeled with membrane-bound GFP. Mature H-astros express markers including GFAP.
Even though iNeurons express pre and postsynaptic proteins, as shown in the image on the left, synaptic density is significantly enhanced by the presence of H-astros and coculture. During recordings, healthy spheres of iNeurons placed on MEAs will spontaneously elicit voltage spikes greater than 40 microvolts with consistent firing frequencies. Coculture spheres display greater network connectivity with the presence of H-astros, resulting in an increase of synchronous network bursts of spikes.
Following this procedure, other methods such as calcium imaging can be performed in order to measure the physiological response of astrocytes to synaptic activity. In addition, transplantation into animal models can be conducted to investigate connectivity with preexisting synaptic circuits. By following this bioengineering approach, one can incorporate extracellular biomaterials and additional cell types such as oligodendrocytes, microglian, and ethelial cells in specific ratios in order to model the complex signaling that occurs in the brain.
