Fig. 5: Reduced flexibility in updating reward-seeking behavior under conditions of silenced SOMIs in the DG.
From: Predictive goal coding by dentate gyrus somatostatin-expressing interneurons in male mice
a Left, confocal image stack of hM4Di expression in the DG hilus. Inset, antibody labeling against SOM (green) of an hM4Di-expressing cell (red). Middle, schematic representation of AAV-flex-hM4Di-mCherry or AAV-flex-hSyn-mCherry (control) bilateral injections in the dorsal DG of SOM-Cre mice (n = 4 and 5, respectively). Fraction of immunohistochemically labeled SOMIs coexpressing hM4Di. Right, schematic of the experimental set-up with the original trained reward zone (ORI) and the novel (NOV) reward area after reward translocation. b Running speed of control (black) and hM4Di-expressing (pink) SOMI mice after clozapine injection in the ORI area after translocation on day 1. Data are aligned to the original reward site (striped line). Linear regressions fitted to speed reduction in controls and hM4Di mice in pre-reward zone (n = 4 and 5 mice, respectively; two animal cohorts). Two-sided Mann–Whitney rank sum test, P = 3.66 × 10−4. c Lick rate on subsequent laps (upper) from a control and hM4Di-mouse upon clozapine injection. Dark and bright blue areas indicate pre- and post-reward ORI zones, respectively. Blue and black striped lines indicate the onset of the original and the novel reward, respectively. d Lick rate during subsequent laps plotted against lap number for control (gray, upper) and hM4Di-mice (pink, lower) on day 1/2. Data were fit to a least-squares linear regression to assess the correlation between lick rate and lap number as defined by Pearson r. Control P = 5.74 × 10−5, hM4Di P = 0.84. e Bar graphs summarize mean lick rates for baseline (bsl), pre- and the post-reward ORI areas, and pre-reward NOV areas for recording day 1 (left) during the first 4 runs, and pre- and the post-reward ORI for recording day 2 (right) for all runs. Lick rates decline in controls across days but stay elevated in hM4Di mice (controls from 32 ORI-4 and 37 ORI measures in n = 4 mice on day 1 and 2, respectively; two-way repeated measure ANOVA tukey posthoc; bsl vs pre paired t-test, P = 2.2963 × 10−5, bsl vs post P = 0.1606, pre vs post P = 0.0045, pre vs NOVpre P = 0.0003, post vs NOVpre P = 0.1442 on day 1; P = 0.5277, day 2; hM4Di mean lick rates from 40 ORI-4 and 106 ORI measures from 5 mice on day 1 and 2, respectively; paired t-test, bsl vs pre P = 0.0116, bsl vs post P = 8.7619 × 10−5, pre vs post P = 0.9909, pre vs NOVpre P = 0.0342, post vs NOVpre P = 3.723 × 10−6 on day 1; P = 3.17 × 10−4, day 2). Lines with shadows depict mean ± SEM; circles represent mean values for individual ORIs; boxes represent 25th–75th percentiles; red line, median; whiskers, the largest/smallest data point within the ±1.5 IQR range *P < 0.05; **P < 0.01; ***P < 0.001; ns not significant.
