Pattern separation is the computational process thought to underlie the ability to separate memories for similar events into discrete non-overlapping representations. Spontaneous Location Recognition (SLR) is a behavioural task designed to assess spatial pattern separation in rats. SLR has been previously shown to be dentate gyrus (DG)-dependent and sensitive to manipulations of plasticity mechanisms. Using SLR, we conducted a series of experiments to evaluate the role of immature neurons for spatial pattern separation.

First, we used a lentiviral approach to inhibit DG neurogenesis in adult male rats by inhibiting Wnt signaling, a principal regulator of adult hippocampal neurogenesis (Lie et al., 2005). Rats with inhibited DG neurogenesis demonstrated impaired performance on the SLR task, but only when objects were placed in similar spatial locations (i.e., when pattern separation was required).

Having previously identified BDNF as critical for encoding/consolidation of pattern-separated memories and performance on SLR (Bekinschtein et al., 2013), we then examined whether BDNF was acting on immature neurons specifically. Infusing BDNF into the DG improved performance on SLR in the control group, but not in the LV-dnWnt rats with inhibited neurogenesis. Together, these studies suggest that consolidation of "pattern-separated" memories requires the action of BDNF on immature neurons.

We then evaluated the effects of increasing neurogenesis, by elevating acyl-ghrelin levels in rats. Ghrelin is an orexigenic peptide known primarily for its growth hormone releasing properties, but also affects synaptic plasticity and hippocampal neurogenesis. The results revealed that treating rats with acyl-ghrelin (14 days, i.p. 10ug/kg) enhanced both adult hippocampal neurogenesis and performance on SLR when tested 8-10 days after the end of treatment. This suggests that sub-chronic systemic administration of ghrelin can produce long lasting improvements in spatial memory that persist following the end of treatment.