Ipsilateral and contralateral hippocampal CA3-CA1 and CA2-CA1 projections were investigated in adult male Long-Evans rats by retrograde tracing. Injection of the retrograde tracer cholera toxin subunit B in the strata oriens and radiatum of dorsal CA1 resulted in labeling of predominantly pyramidal cells in ipsilateral and contralateral CA3 and CA2. The contralateral and ipsilateral anterior-posterior extents of CA3 innervation to CA1 were similar. Fifteen to twenty per cent of the hippocampus proper cells that give rise to CA1 stratum oriens innervation were CA2 pyramidal cells, whereas CA2 cells were a mere 3% for CA1 stratum radiatum innervation. The preferred projection of CA2 pyramidal cells to the CA1 stratum oriens was also manifested in transgenic mice that express GFP under the control of the CACNG5 promoter, in which CA2 cells express high amounts of GFP. The ratios of ipsilateral to contralateral projections were compared. For the CA3-CA1 connection, we found that dorsal CA1 stratum radiatum received more ipsilateral projections whereas CA1 stratum oriens received more contralateral innervation. Interestingly, ipsilateral connections dominated for both CA2-CA1 stratum oriens and CA2-CA1 stratum radiatum. These results demonstrate that the primary intrahippocampal target of CA2 pyramidal cells is the ipsilateral CA1 stratum oriens, in contrast to CA3 cells which project more diversely to bilateral CA1 regions. Such innervation patterns may suggest differential dendritic information processing in apical and basal dendrites of CA1 pyramidal cells.

Left-right asymmetry of human brain function has been known for a century, although much of molecular and cellular basis of brain laterality remains to be elusive. Recent studies suggest that hippocampal CA3-CA1 excitatory synapses are asymmetrically arranged, however, the functional implication of the asymmetrical circuitry has not been studied at the behavioral level. In order to address the left-right asymmetry of hippocampal function in behaving mice, we analyzed the performance of "split-brain" mice in the Barnes maze. The "split-brain" mice received ventral hippocampal commissure and corpus callosum transection in addition to deprivation of visual input from one eye. In such mice, the hippocampus in the side of visual deprivation receives sensory-driven input. Better spatial task performance was achieved by the mice which were forced to use the right hippocampus than those which were forced to use the left hippocampus. In two-choice spatial maze, forced usage of left hippocampus resulted in a comparable performance to the right counterpart, suggesting that both hippocampal hemispheres are capable of conducting spatial learning. Therefore, the results obtained from the Barnes maze suggest that the usage of the right hippocampus improves the accuracy of spatial memory. Performance of non-spatial yet hippocampus-dependent tasks (e.g. fear conditioning) was not influenced by the laterality of the hippocampus.

Gamma oscillations are implicated in higher-order brain functions such as cognition and memory, but how an animal’s experience organizes these gamma activities remains elusive. Here we show that the power of hippocampal theta-associated gamma oscillations recorded during urethane anesthesia tends to be greater in rats reared in an enriched environment than those reared in an isolated condition. This experience-dependent gamma enhancement is consistently larger in the right hippocampus across subjects, coinciding with a lateralized increase of synaptic density in the right hippocampus. Moreover, interhemispheric coherence in the enriched environment group is significantly elevated at the gamma frequency. These results suggest that enriched rearing sculpts the functional left–right asymmetry of hippocampal circuits by reorganization of synapses.

The gamma-filtered traces of the image in Fig. 3a were inadvertently duplicated from Fig. 2a. In addition, the legend for Fig. 3 incorrectly stated that slow gamma traces were displayed in Fig. 3a. In fact, these traces do not distinguish between fast and slow recordings. The correct version of the figure and its legend appear below. The slow and fast gamma power ratios of ENR þ ket were significantly lower than 1 (*Po0.05 and **Po0.01, respectively, t-test, N ENR ¼ 10,The ENR and ENR þ ket power ratios were significantly different (***Po0.001, t-test, N ENR ¼ 10, N ENR þ ket ¼ 8). Acute effect (within 90 min) of intramuscular ketamine administration (3.0 mg kg À 1 , ENR þ a-ket, N ENR þ a-ket ¼ 4) on slow and fast gamma power is also shown in black in the middle and the right panel. ENR þ a-ket did not show the inhibitory effects seen in ENR þ ket. Error bars represent s.e.m.

No abstract