Early postnatal experience shapes both inhibitory and excitatory networks in the hippocampus. However, the underlying circuit plasticity is unclear. Using an enriched environment (EE) paradigm, we assessed the circuit plasticity of inhibitory cell-types in the hippocampus. We found that cholecystokinin (CCK)-expressing basket cells strongly increased somatic inhibition on the excitatory granular cells (GC) following EE while another pivotal inhibitory cell-type, parvalbumin (PV)-expressing cells did not show changes. By inhibiting activity of the entorhinal cortex (EC) using a chemogenetic approach, we demonstrate that the projections from the EC is responsible for the developmental plasticity of CCK+ basket cells. Our measurement of the input decorrelation by DG circuit suggests that EE has little effect on pattern separation despite of the altered CCK+ basket cell circuit. Altogether, our study places the activity-dependent remodeling of CCK+ basket cell innervation as a central process to adjust inhibition in the DG, while maintaining the computation in the circuit. Figure 5. PE housing has limited effect on pattern separation in the DG. (A) Top, cartoon depicting whole-cell recording of the GCs during simultaneous stimulation of the PP. Bottom, representative traces obtained in currentclamp mode in response to the stimulation with 5 input spike trains.To assess the level of pattern separation operated by single recorded GCs, the similarity of the inputs is compared to the similarity of the outputs using different comparative metrics. (B) Representative graphs of pattern separation with pairwise output similarity versus the pairwise input similarity measured by the three different metrics, including R, NDP and SF and using 10 ms bins. Data points below and above the dashed line correspond to pattern separation and pattern convergence, respectively. Solid lines represent the linear fits for the ANCOVA. Statistical comparisons were performed with an ANCOVA (the aoctool function in Matlab) and a two-sample t-test (*: p<0.05 and ***: p<0.001). (C) A summary for FR and p(Burst) (number of animals = 4 for both group, total 15 recordings for each group). Statistical comparisons were performed with a two-samples t-test for firing rate (t = 1.559, p = 0.13) and p(Burst) (t = 1.3627, p = 0.184). (D) Compactness, Occupancy or FR codes were measured for each recording sets. Statistical comparisons were performed with 2 sample t-test for binwise Compactness of output (t = 1.04, p = 0.307), the variations of Occupancy (t = 1.924, p = 0.073) and the variations in FR or spike trains (t = 1.129, p = 0.269). For each bars, the number of animals and the number of recorded cells used for the quantification are reported (a/n).