Figure 5. The LH^GABA to CA3^GABA projection regulates memory and learning behaviors.

. (a) Anterograde tracing scheme. (b) Microscopy analysis of WGA-GFP at the primary injection site (LH) or projected site (CA3) in WT mice. 3V, 3rd ventricle. Green: WGA-GFP, Blue: DAPI, Red: Vgat positive neurons. Arrows indicate cells positive for both WGA-GFP and tdTomato. The experiment was repeated independently once with similar results. n=5 mice. Male 4 months-old mice were used. (c) Quantification of GFP-positive (green), tdTomato-positive (red), and double-positive (yellow) cells in CA3. One-way ANOVA analysis was used followed by post hoc Tukey test. n=5 mice. Male 4 months-old mice were used. (d) Scheme of intracranial injection of AAV-ChR2 in the LH. (e) The amplitude of eIPSCs of CA3^GABA neurons in WT upon blue light stimulation of LH^GABA neurons. Brain slices were recorded in pure artificial CSF (aCSF), or in the presence of 4-aminopyridine (4-AP, a potassium channel blocker) plus tetrodotoxin (TTX, a sodium channel blocker), or GABA_A Receptor antagonist bicuculline (Bic). 2-way ANOVA analysis was used. Male 4 months-old mice were used. n=15 neurons/3 mice. (f) Representative traces of eIPSCs of CA3^GABA neurons. (g) Latency of eIPSCs of CA3^GABA neurons. 2-way ANOVA analysis was used. Male 4 months-old mice were used. n=13 neurons/3 mice. (h) Scheme of intracranial injection of AAV-FLEX-hM3Dq at CA3. (i) NOR test in hM3Dq-injected or mCherry-injected NS-V or WT (Vgat-Cre) mice with CNO treatment prior to the test. 2-way ANOVA analysis was used followed by post hoc LSD test. n_{WT_mCherry}=6, n_{WT_hM3Dq}=6, n_{NS-V_mCherry}=6, n_{NS-V_hM3Dq}=6 mice. Female 4 months-old mice were used. (j) MWM Probe test in hM3Dq-injected or mCherry-injected NS-V or WT (Vgat-Cre) mice with CNO treatment prior to the test. 2-way ANOVA analysis was used followed by post hoc LSD test. n_{WT_mCherry}=6, n_{WT_hM3Dq}=6, n_{NS-V_mCherry}=6, n_{NS-V_hM3Dq}=6 mice. Female 4 months-old mice were used. (k) AAV injection and laser probe implant scheme for optogenetic manipulation. (l) Novel Object Recognition (NOR) test with a blue laser stimulation in CA3 on mice infected with AAV-ChR2 at LH. Data were analyzed by two-tailed unpaired t test. n_{WT_light off}=7, n_{WT_light on}=7 mice. Male 4 months-old mice were used. (m-n) MWM test in the presence of CNO or saline treatment in NS-V or WT (Vgat-Cre) mice injected with AAV.FLEX.hM4Di at the LH. 2-way repeated ANOVA analysis was used followed by post hoc LSD test. n_{WT_Saline}=7, n_{WT_CNO}=7, n_{NS-V_Saline}=8, n_{NS-V_CNO}=7 mice. Male 4 months-old mice were used. (o) NOR test in hM4Di-injected NS-V or WT (Vgat-Cre) mice with or without CNO treatment. 2-way ANOVA analysis was used followed by post hoc LSD test. n_{WT_Saline}=11, n_{WT_CNO}=7, n_{NS-V_Saline}=11, n_{NS-V_CNO}=7 mice. Male 4 months-old mice were used. (p) Microscopy analysis of hM4Di-mCherry at the primary injection site. Male 4 months-old mice were used. The experiment was repeated independently once with similar results. (q) Scheme of AAV injection and laser probe implant for optogenetic manipulation. (r) NOR test of NS-V or WT (Vgat-Cre) mice with AAV.FLEX.EYFP or AAV.FLEX.eNpHR-EYFP injected at LH with yellow light on for all mice during the test. (Output power at the probe tip is around 7mW). 2-way ANOVA analysis was used followed by post hoc LSD test. n_{WT_EYFP}=7, n_{WT_eNpHR}=7, n_{NS-V_EYFP}=6, n_{NS-V_eNpHR}=7 mice. Male 4 months-old mice were used. (s-t) Spontaneous firing frequency and resting membrane potential of EYFP-infected or eNpHR-infected LH^GABA neurons in response to yellow light in brain slices. 2-way ANOVA analysis was used. Male 4 months-old mice were used. Firing Frequency: n_{WT_EYFP}=7 neurons/2 mice, n_{NS-V_EYFP}=7 neurons/2 mice, n_{WT_eNpHR}=8 neurons/2 mice, n_{NS-V_eNpHR}=7 neurons/2 mice; Membrane Potential: n_{WT_EYFP}=9 neurons/2 mice, n_{NS-V_EYFP}=8 neurons/2 mice, n_{WT_eNpHR}=9 neurons/2 mice, n_{NS-V_eNpHR}=8 neurons/2 mice. (u) Novel Object Recognition (NOR) test after LH-specific infusion of a selective GABA_A agonist SL651498. 2-way ANOVA analysis was used followed by post hoc LSD test. n_{WT_Vehicle}=9 mice, n_{NS-V_Vehicle}=9 mice, n_{WT_SL651498}=9 mice, n_{NS-V_SL651498}=10 mice. Male 4 months-old mice were used. Data is expressed as mean ± S.E.M. For detailed statistics results, see Supplementary Table 1. * P ≤ 0.05 is set as significance. Independently, the electrophysiological data was re-analyzed using the linear mixed-effects models (Supplementary Table S2).