Hippocampus-dependent learning influences hippocampal neurogenesis

Abstract

The structure of the mammalian hippocampus continues to be modified throughout life by continuous addition of neurons in the dentate gyrus. Although the existence of adult neurogenesis is now widely accepted the function that adult generated granule cells play is a topic of intense debate. Many studies have argued that adult generated neurons, due to unique physiological characteristics, play a unique role in hippocampus-dependent learning and memory. However, it is not currently clear whether this is the case or what specific capability adult generated neurons may confer that developmentally generated neurons do not. These questions have been addressed in numerous ways, from examining the effects of increasing or decreasing neurogenesis to computational modeling. One particular area of research has examined the effects of hippocampus dependent learning on proliferation, survival, integration and activation of immature neurons in response to memory retrieval. Within this subfield there remains a range of data showing that hippocampus dependent learning may increase, decrease or alternatively may not alter these components of neurogenesis in the hippocampus. Determining how and when hippocampus-dependent learning alters adult neurogenesis will help to further clarify the role of adult generated neurons. There are many variables (such as age of immature neurons, species, strain, sex, stress, task difficulty, and type of learning) as well as numerous methodological differences (such as marker type, quantification techniques, apparatus size etc.) that could all be crucial for a clear understanding of the interaction between learning and neurogenesis. Here, we review these findings and discuss the different conditions under which hippocampus-dependent learning impacts adult neurogenesis in the dentate gyrus.

Keywords: cell survival; dentate gyrus; hippocampus; memory; neurogenesis; spatial learning.

Figure 1

Neurogenesis in the dentate gyrus. (A) Doublecortin labeling (green) shows the presence of immature neurons along the inner edge of the granule cell layer and in the subgranular zone. (B) Doublecortin (green) and zif268 (red) shows the immature neurons that have been activated in response to spatial memory retrieval.

Figure 2

Critical periods for spatial learning induced changes in immature cell survival in the dentate gyrus. Spatial learning does not impact the survival of immature neurons that are 1–5 days old at the time of learning. Survival of immature neurons that are 6–10 days old during training is selectively enhanced [although this can depend on task difficulty (Epp and Galea, 2009) and quality of learning (Epp et al., ; Sisti et al., 2007)]. Survival of immature neurons that are 15–20 days old at the time of learning is decreased. This effect cannot be detected if animals are perfused the day following training but can be observed if histological examination is delayed until day 20 following a probe trial 90 min before perfusion. Collected from findings from Epp et al. ( and 2011a). Described changes in neurogenesis are in comparison to rats trained on a cued version of the task.

Figure 3

Time course of activation of immature neurons in response to spatial memory. Spatial learning occurred either 1–5, 6–10, or 11–15 days following BrdU administration. The rats were then tested with a probe trial 5 days later and were then perfused 2 h later. No activation was seen in 10-day-old neurons. Rats trained on the spatial version of the task on days 6–10 had a small percentage of 15 day old neurons were activated but no difference existed between rats that received spatial versus non-spatial training (Epp et al., 2011a). Rats trained the spatial version of the task on days 6–10 (Chow et al., in press) or 11–15 showed enhanced activation of 20 day old neurons compared to rats that were trained on the non-spatial version of the task (Epp et al., 2011a). N/A, no activation; IEG, immediate early gene. Described changes in activation are in comparison to rats trained on a cued version of the task.