Fig 2: PVN OXT neurons as main orchestrator of social isolation induced social craving.(A) Representative confocal images of PVN stained with cFOS, OXT, and AVP antibody (scale bar: 20 µm). (B) cFos+ cells (as % of DAPI+ cells) (Unpaired samples t-test, t(22)=2.510, p=0.0199, n=12 slices from three mouse each group). (C) OXT+ cells (as % of DAPI+ cells) (Mann-Whitney U-test=33, p=0.0439, n=11–12 slices from three mouse each group). (D) OXT+/ cFos+ cells (as % of total OXT+ cells) (Unpaired samples t-test, t(22)=3.491, p=0.021, n=12 slices from three mouse each group). (E) Left: experimental paradigm, PVN neurons projecting to VTA (CTB-488 was injected in the VTA at P21) were subjected at multiple depolarizing current steps. Right: example traces from 250 pA depolarizing current injection. (F) Number of action potentials (APs) across increasing depolarizing current steps (Two-way RM ANOVA followed by uncorrected Fisher’s LSD post doc analysis, House condition main effect F(1, 30)=7.306, p=0.0112, Current step main effect F(1.857, 55.71)=30.47, p<0.001, House condition × Current step F(6, 180)=2.828, p=0.018). (G) Resting membrane potential of recorded cells (Mann-Whitney U-test=85.5, p=0.1483. Grouped n=13, Isolated n=19 from three mice each group). (H) Experimental design. R26-hM4Di/mCitrine mice were crossed with Oxytocin-Ires-Cre mice generating Oxt-hM4Di mice which express inhibitory DREAAD specifically in OXT neurons. (I) Experimental paradigm: Oxt-hM4Di mice were isolated between P28 and P35 and CNO was dissolved in the drinking water. After isolation mice were subjected whole-cell patch-clamp recordings. (J) Left: experimental paradigm, VTA pDA neurons were subjected at multiple depolarizing current steps. Right: example traces from 250 pA depolarizing current injection. (K) Number of APs across increasing depolarizing current steps (Three-way RM ANOVA followed by uncorrected Fisher’s LSD post doc analysis (see Figure 3—source data 1 for details), Current step main effect F(1.962, 98.08)=114.8, p<0.0001, House condition main effect F(1, 50)=3.456, p=0.0689, CNO treatment main effect F(1, 50)=5.826, p=0.0195, Current step × House condition F(6, 300)=1.866, p=0.0865, Current step × CNO treatment F(6, 300)=3.806, p=0.0011, House condition × CNO treatment F(1, 50)=2.293, p=0.1363, Current step × House condition × CNO treatment F(6, 300)=3.459, p=0.0026). (L) Resting membrane potential of recorded cells (Two-way ANOVA, CNO treatment main effect F(1, 50)=0.8745, p=0.3542, House condition main effect F(1, 50)=1.956, p=0.1681, CNO treatment × House condition F(1, 50)=3.531, p=0.0661, Grouped n=12, Grouped +CNO n=12, Isolated n=15, Isolated+CNO n=15 from 3 to 4 mice each group). (M) Experimental design. Oxt-hM4Di mice were isolated or kept grouped from P28 to P35. CNO was dissolved in drinking water and after isolation mice underwent to free direct social interaction task. (N) Social interaction time (Two-way ANOVA followed by Tukey’s multiple comparisons test, CNO treatment main effect F(1, 28)=7.088, p=0.0127, House condition main effect F(1, 28)=9.940, p=0.0038 CNO treatment × House condition F(1 ,28)=4.334, p=0.0466, n=8 each group). Data are represented as mean ± SEM. Figure 3—source data 1.Figure 3 - raw data and statistical output.

Fig 3: Adolescence acute social isolation induces synaptic scaling in adulthood mice.(A, B, G) Experimental paradigm. WT mice were isolated between P28 and P35. Then mice were regrouped, injected with 488-CTB in the mPFC (B) or NAc (G) between P45–P50 and at P60 were subjected at whole-cell patch-clamp recording. (C) Example traces of isolated AMPA and NMDA currents recorded at +40 mV. (D) AMPA-NMDA ratio of VTA-DA:mPFC projecting neurons (Unpaired samples t-test, t(8)=0.07544, p=0.9417, Grouped n=5, Isolated n=5 from two mice each group). (E) Example traces of Isolated AMPA current recorded at +40, 0, and –60 mV. (F) Rectification index (RI) of VTA-pDA:mPFC projecting neurons (Unpaired samples t-test, t(12)=3.545, p=0.004, Grouped n=8, Isolated n=8 from two mice each group). (H) Example traces of isolated AMPA and NMDA currents recorded at +40 mV. (I) AMPA-NMDA ratio of VTA-pDA:NAc projecting neurons (Mann-Whitney U-test=7, p=0.1602, Grouped n=5, Isolated n=6 from two mice each group). (J) Example traces of Isolated AMPA current recorded at +40, 0, and –60 mV. (K) RI of VTA-pDA:NAc projecting neurons (Unpaired samples t-test, t(10)=0.3720, p=0.7176, Grouped n=5, Isolated n=7 from two mice each group). (L). Experimental paradigm: Oxt-hM4Di mice were isolated between P28 and P35 or kept always grouped and CNO was dissolved in drinking water. Then mice were regrouped, injected with CTB at P45–P50, and kept group-housed until P60 when were subjected at whole-cell patch-clamp recording. (M) Example traces of isolated AMPA current recorded at +40, 0, and –60 mV. (N) RI of VTA-pDA:mPFC projecting neurons (Two-way ANOVA followed by Tukey’s multiple comparisons test, CNO treatment main effect F(1, 23)=11.19, p=0.0028, House condition main effect F(1, 23)=5.459, p=0.0285, CNO treatment × Hous condition F(1, 23)=11.22, p=0.028, Grouped n=8, Grouped CNO n=7, Regrouped n=7, Regrouped CNO n=5 from two mice each group). Data are represented as mean ± SEM. Figure 5—source data 1.Figure 5 - raw data and statistical output.