Aim: 

New neurons are generated from proliferating progenitors throug-hout life in mammalian dentate gyrus. Hippocampus-dependent learning rescues from death newborn cells in the first week after mitosis, a time at which cells do not receive synapses. Two mechanisms may be proposed: a paracrine message directly promoting survival or an anticipation of synapse formation onto newborn cells. The effect of spatial learning and of swimming on synapse formation onto immature neurons was studied.

Methods: 

Proliferating cells were labeled by a GFP-retroviral vector and, four-six days after, rats were trained for spatial learning in Morris water maze, or exercised in swimming. Spontaneous synaptic activity and synaptic responses to medial perforant pathway stimulation were revealed by patch clamp recordings carried out in GFP-positive cells in brain slices.

Results: 

In controls, GABAergic synapses appeared at 8 days after labeling and all visited cells showed a GABAergic response at 12 days, while glutamatergic synapses were not detectable before 10 days. In learner rats, about half of the visited cells received GABAergic synapses at 7 days; glutamatergic synapses appeared at 10 days as in controls, but the percentage of contacted cells was larger than controls at 10 and 12 days. In swimmer rats, GABAergic and glutamatergic synapse formation occurred at the same time as in learner rats, but cells contacted by glutamatergic synapses were higher in number with respect to both control and learner rats.

Conclusions: 

Learning and swimming anticipate the synapse formation onto immature neurons, thus possibly favoring a survival-promoting effect. Moreover, due to the involvement of neurogenesis in learning and in depression, these findings provide a possible mechanism for the effects of physical and cognitive activity on learning and depression.