Skip to main content
. 2021 Aug 25;12:5116. doi: 10.1038/s41467-021-25356-2

Fig. 8. Burst-evoking Pv neuronal stimulation improves social interaction and cortical social representation.

Fig. 8

a Experimental scheme for social rescue by Pv neuronal stimulation. Pv-Cre;WT and Pv-Cre;Shank2–/– mice (8 weeks) were bilaterally injected in the mPFC with AAV-DIO-ChR2/ChETA/SSFO-EYFP or AAV-DIO-EYFP (control) for 3 weeks prior to three-chamber experiments. Experiments consisted of three rounds of three-chamber tests in which the second (Opto) round, but not the first (Pre) or third (Post) round, was performed in the presence of blue-light stimulation throughout the 10-min S–O session. Right, examples of mPFC neurons from a Pv-Cre;WT mouse injected with AAV-DIO-EYFP; DAPI (blue), nuclear marker. b–e Pv neuronal stimulation in the mPFC at 10 Hz (5 ms; ChR2), but not in a continuous mode (SSFO) or at 40 Hz (2 ms; ChETA), improves three-chamber social interaction in Pv-Cre;Shank2–/– mice infected in the mPFC with the indicated viruses (>12 weeks + 3 weeks), as shown by time spent sniffing (or nose-poking) social and object targets (n = 15 mice [WT-EYFP], 11 [KO-EYFP], 20 [WT-ChR2-10-Hz], 16 [KO-ChR2-10-Hz], 8 [WT-ChETA], 10 [KO-ChETA], 15 [WT-SSFO], 14 [KO-SSFO], *p < 0.05, **p < 0.01, two-way RM-ANOVA with Tukey’s test). f–i 10-Hz Pv neuronal stimulation modestly improves neuronal representation of social and non-social contexts in the mPFC of Shank2–/– mice, as shown by the proportions of neurons that show discriminative firings to target pairs (S–O, S–E, E–O, and L–R/left–right) and targets (S, E, O, Si, and total) and the total number of target pairs that a single neuron can discriminate (S–O, S–E, E–O, and L–R; ranging from 0 to 4). Two-way ANOVA could not be performed because the neuronal and target-number proportions were extracted from the calculation of target pair-discriminative or target-selective neurons using Venn diagrams; we thus attempted Chi-square analysis with Bonferroni correction and multiple comparisons with p-value adjustments (n = 194 neurons from 7 mice for WT-EYFP, 72/8 [WT-ChR2], 163/4 [KO-EYFP], and 364/9 [KO-ChR2], *p < 0.017 [comparison of WT-EYFP, KO-EYFP, KO-ChR2], Chi-square test with Bonferroni correction). See Source data for raw data values and Supplementary Table 1 for statistical details.