@inproceedings{item_3176538,
  title = {{Effects of cellular excitatory{-}inhibitory composition on neuronal dynamics}},
  journal = {{BMC Neuroscience}},
  abstract = {Various brain regions have distinct and highly conserved ratios of excitatory and inhibitory neurons. For instance, cerebral cortex typically includes around 20{\textpercent} of inhibitory neurons. However, it is not clear whether unphysiological ratios would change collective neuronal dynamics or jeopardize the balance of excitation and inhibition on a synaptic level. To investigate this question, we developed a platform that allowed us to culture hippocampal networks with various fractions of inhibitory neurons. We also study how cellular composition affects neuronal dynamics in finite network models with balanced excitation/inhibition currents and neuronal adaptation. We used fluorescence-activated cell sorting to isolate inhibitory and excitatory neurons and seeded them while keeping prescribed inhibitory percentages. We recorded the calcium dynamics of these cultures. All of them developed spontaneous network activity manifested in full network bursts. The cultures with 10{\textendash}80{\textpercent} of inhibitory cells showed surprisingly similar mean inter-burst intervals, which were indistinguishable from unsorted control cultures that usually contain 20{\textendash}30{\textpercent} of inhibitory neurons. Fully excitatory and fully inhibitory cultures had significantly longer inter-burst intervals. The coefficient of variation of inter-burst intervals grew with the number of inhibitory neurons. To model the observed effects, we developed a set of networks with various fractions of excitatory and inhibitory neurons. The networks were comprised of adaptive leaky integrate-and-fire neurons driven by slow Poisson input. The relative strength of inhibitory and excitatory synapses was kept at balance. The model showed that the stable mean but increasing variance of inter-burst intervals can be achieved by the balance of excitation and inhibition that regulates effects of the adaptation. In a fully excitatory network, inter-burst intervals are determined by the adaptation alone. Adding inhibition to the network results in stopping bursts before the adaptation completely silences the activity. This, in turn, allows the next burst to start earlier, leading to shorter inter-burst intervals with higher variance. To further compare the behavior of the model and cultures, we disrupted the excitation/inhibition balance by decreasing the strength of inhibitory synapses. In the experimental setup, this corresponded to the application of bicuculline. In the cultures with 10{\textendash}80{\textpercent} of inhibitory neurons application of bicuculline led to prolonged interburstintervals and decreased variability. Under maximum concentration, the activity of these cultures was generally similar to the fully excitatory cultures. Similarly, in the model, blocking of inhibition resulted in stronger adaptation after a burst that led to longer and less variable inter-burst intervals. Overall, our results suggest that developed hippocampal cultures with artificial cellular excitatory and inhibitory composition tend to maintain the excitation/inhibition balance. This result in a constant mean activity but a growing variability of bursting in cultures with increasing numbers of inhibitory neurons.},
  volume = {20},
  pages = {94},
  publisher = {BioMed Central},
  address = {Barcelona, Spain},
  year = {2019},
  note = {28th Annual Computational Neuroscience Meeting (CNS{\textasteriskcentered}2019)},
  slug = {item_3176538},
  author = {Vinogradov, O and Sukenik, N and Moses, E and Levina, A}
}