|Title||Computer model of endocannabinoid effects in CA3|
|Publication Type||Conference Paper|
|Year of Publication||2015|
|Authors||Sherif, M. A., Skosnik P., Hajs M., & Lytton WW.|
|Conference Name||Society for Neuroscience 2015 (SFN '15)|
|Keywords||SFN, Society for Neuroscience|
Schizophrenia (SCZ) involves disturbances in neural function at different spatial and temporal scales. Theta (5-10 Hz) and gamma (30-100 Hz) oscillations have been shown to be abnormal in SCZ. In healthy volunteers, delta-9-tetrahydrocannabinol (THC), a cannabinoid-1 receptor (CB1R) partial agonist, induces symptoms similar to those reported in SCZ. CB1R agonism has been used to model some aspects of SCZ-related neural dysfunction in animals. Previous work has shown that CB1R agonism decreases theta (5-10 Hz), slow gamma (30-50 Hz), and fast gamma (60-90 Hz) power in CA3 region in rats. These converging data provide the impetus for developing a computer model of the CA3 region incorporating glutamatergic, GABAergic and endocannabinoid systems to investigate how CB1R agonism might explain the oscillatory changes that may in part underlie the various symptom domains of SCZ. The current model consisted of 2000 pyramidal cells, 100 parvalbumin-positive basket cells (PV cells), 100 cholecystokinin-positive basket cells (CCK cells) and 100 Oriens lacunosum Moleculare (OLM cells). Connectivity was based on the available literature and our previously published models. CB1R effects were implemented by varying GABA transmission at the synapses from CCK to pyramidal cells, as well as varying glutamatergic transmission at recurrent pyramidal connections. Experimentally, LFPs from CA3 region of 5 rats anaesthetized by intraperitoneal injection of chloral hydrate were obtained at baseline for model validation. Frequency bands within LFP signals were used to compare theta and low gamma oscillations with LFPs from the computer model. Model neurons were tuned to produce 8.3 ms membrane time constant for PV cells, 25 ms for CCK cells. In response to a depolarizing current of 100 pA, PV cells spiked at rate of 60 Hz, while CCK cells spiked at rate of 28 Hz. In response to a depolarizing current of 300 pA, PV cells spiked at rate of 132 Hz, while CCK cells spiked at rate of 52 Hz. In the LFPs from the model, emergent theta oscillations were seen at a frequency of 6 Hz, while slow gamma (high beta) oscillations were observed at 30 Hz, in agreement with rat controls from our experiments. These oscillation frequencies emerged from the much slower firing of pyramidal cell due to network interactions. Multiscale computer modeling is one suitable way to integrate findings from various modalities to reach a mechanistic understanding of the role the different elements play in the etiology and pathophysiology of brain disease. Such an understanding in SCZ can suggest potential treatments which would not be likely to be found through random ligand assessment.