Abstracts

Rationale: The hippocampal dentate gyrus (DG) is an important brain structure in normal cognitive operations, and is known to be dysfunctional in temporal lobe epilepsy (TLE). DG is a highly inhibited structure, with granule cells (GCs; the principal cell of the DG) exhibiting sparse activation in response to stimulation of the perforant path (PP; the major input to the DG). Activation of the PP is known to inefficiently drive activity in downstream area CA3, a phenomenon known as dentate gating. Dentate gating and selective firing in the network are degraded via blockade of GABAergic inhibition, suggesting an important role for cortical GABAergic inhibitory interneurons (INs). We investigated the hypothesis that feedforward inhibition provided by defined subtypes of INs contributes to selective firing within DG networks.Methods: We studied the responsiveness to PP input and relative excitability of prominent defined cell types in DG, using multiple whole cell patch clamp recordings, extracellular stimulation, and large-scale dynamic calcium imaging in acute brain slices prepared from mice. INs, including parvalbumin (PV)-positive fast-spiking (FS) cells in the GC layer and somatostatin (Sst)-positive INs in the hilus, were targeted using specific mouse Cre recombinase knock-in driver lines. Calcium sensitive dye imaging was performed using bulk loading with Oregon Green BAPTA-1 (OGB-1) and AP-evoked calcium transients were visualized using two photon (2P) laser scanning microscopy. Results: GCs were markedly less responsive to PP-evoked stimulation than PV- or Sst-positive INs as measured by the stimulation intensity required to generate APs. GCs exhibited lower resting membrane potential (RMP) than PV- and Sst-positive INs; however, correction of the membrane potential with a DC offset to achieve a potential of -70 mV across cells had a minimal effect on this difference. Short-term synaptic dynamics of PP-evoked excitatory post-synaptic potentials (EPSPs) was not significantly different between the three cell types, exhibiting mild short-term synaptic depression across a range of physiologically relevant stimulation frequencies. Dynamic calcium imaging of GC activity confirmed on a larger scale that GCs exhibit sparse firing and require large stimulation intensities to evoke APs.Conclusions: GCs are markedly less responsive to PP simulation than two prominent subtypes of feedforward INs in the DG, PV-positive FS cells and Sst-positive hilar INs. The basis of this differential responsiveness remains unclear, but is not due to demonstrated differences in RMP between the cell types or to short-term dynamics of PP input to DG neurons. This differential responsiveness may be related to the phenomenon of dentate gating. Future studies are needed to further explore the mechanism(s) of sparse firing in DG and of dentate gating. Cell type specific manipulation of activity may be a useful strategy for such investigations.