Human iPSC-derived Neurogenin-2 (NGN2) Cortical Neurons Develop Functional Connectivity and Small-world Network Topology In Vitro
Mahsa Khayat-Khoei1, Nazanin Ershadinia1, David Oluigbo2, Timothy Sit3, Howard Weiner1, Susanna Mierau1
1Brigham and Women's Hospital, Harvard Medical School, 2Massachusetts Institute of Technology, 3Queen Square Institute of Neurology, University College London
Objective:
To investigate the development of functional connectivity in human induced pluripotent stem cell (iPSC)-derived neurogenin-2 (NGN2) cortical neurons and small-world network topology in vitro on microelectrode array (MEA) systems.
Background:
MEA recordings from human iPSC-derived neurons show disease-related phenotypes including differences in firing and burst rates.
Design/Methods:
To reveal microscale network activity, we investigated age-related differences in the functional connectivity and network topology. We cultured human NGN2 iPSC-derived neurons with human astrocytes on 64-electrode MEAs (n=15 cultures). We performed 10-minute recordings weekly from days-in-vitro (DIV) 14-35 and analysis with MEA-NAP (https://github.com/SAND-Lab/MEA-NAP/). We applied one-way ANOVA and the Tukey-Kramer method to adjust for multiple pairwise comparisons.
Results:
NGN2 cortical cultures showed a significant increase in the number of active electrodes (p=3e-7), mean firing rate (p=7e-6), and mean network burst rate (p=0.01) from DIV14-35. Significant functional connections were determined using the spike time tiling coefficient and probabilistic thresholding. Network features were compared from cultures at DIV14-35. The functional connectivity significantly increased from DIV14-35 including the network density (p=0.001) and top 10% of edge weights (p=3e-5). The number of modules significantly decreased (p=0.01) while the mean participation coefficient significantly increased (p=0.01) indicating a shift in activity from within subcommunities to network-wide. At DIV35, the mean small-worldness coefficient approached zero, indicating small-world topology, compared to more randomly connected networks at DIV14-28 (p=3e-4). 
Conclusions:
MEA recordings revealed that NGN2 neurons develop patterns of functional connectivity and network topology also seen at larger spatial scales in the brain and, thus, could inform mechanistic and therapeutic studies of neuronal development and disease.
10.1212/WNL.0000000000212109
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.