CNS*2020 Online

Ashraya Samba Shiva, Bruce Graham
Virtual room: https://meet.google.com/ryc-nqoa-scy
Meeting code: ryc-nqoa-scy

Theta oscillations may act as carrier waves for synchronizing activities across neuronal regions. Several gamma cycles, each containing a specific pattern of activity correlated with an event e.g. animal location forming place fields, are encompassed in a single theta cycle. We have extended the septo-hippocampal population firing rate model proposed by Denham and Borisyuk (Hippocampus 10:698-716, 2000) to study the influence of inhibitory interneurons, specifically PV-containing basket cells(BCs) and bistratified cells (BSCs) on theta and theta-coupled gamma oscillations in both CA1 and CA3 hippocampal networks. Our CA1 microcircuit model is a combination of Denham and Borisyuk (2000) with the model of Cutsuridis et. al (Hippocampus 20:423-446, 2010). The CA3 model is adapted from the CA1 model on the basis of CA3-specific experimental data.

The theta-phase relationships of the neurons in these models are largely determined by the afferent and efferent connections of BCs in both CA1 and CA3. In CA1, BCs and BSCs are active on opposite phases of theta in a pure- theta tuning (no gamma), as per data from Klausberger et. al (Nature 421:844–848, 2003; Nat Neurosci 7:41–47, 2004) due to extra external drive to basket cells from CA3, enabling them to be active on the opposite cycle to CA1 pyramidal cells (PCs). As excitatory drive to BCs from PCs is increased there is a bifurcation in which BC activity switches to being in-phase with PCs, and BSC activity remains in-phase with PCs. A further increase in drive to BCs from PCs leads to gamma oscillations in both PC and BC activity, and forces BSC activity to shift to the opposite theta phase due to strong inhibition from BCs.

Varying strengths of external inputs also affects the strength of oscillations and phase relationships. Both in CA1 and CA3, PCs, BCs and BSCs reach steady state activity for very low or very high external inputs. BCs in CA1 rely on CA3 and EC input for their activity in theta-only tuning, so their activity reduces if these inputs are reduced. BSCs activity in CA1 actually reduces for increased CA3 and EC input, due to increased inhibition from BCs. In CA3, recurrent connections between PCs are far more likely (and hence stronger on a population scale) than in CA1. BSCs in CA3 are driven only by CA3 PCs and have no external sources of excitation. BCs in CA3 get dentate gyrus input in addition to PC input. Other interneurons (modeled as a single population) get excitation from CA1 and CA3 PCs, plus EC. In CA1, only CA3 and EC are external sources of input, apart from septum. The resultant main difference in activity in CA3, compared with CA1, is that BSCs are always in sync with BCs and PCs during theta. A minimum strength of septum input is required to generate theta, then increasing septum input raises the frequency of theta oscillations. Whereas, increasing the value of dentate gyrus input transforms the oscillatory activity into stable, non-oscillatory activity, and also decreases the septum activity. Strong EC and CA1 input to CA3 may silence CA3 PCs, BCs and BSCs through exciting other inhibitory interneurons.