Oxytocin activation of paraventricular thalamic neurons promotes feeding motivation to attenuate stress-induced hypophagia

Abstract

The neuropeptide oxytocin (OT) regulates important brain functions including feeding through activating OT receptors in multiple brain areas. Both OT fibers and OT receptors have been reported in the paraventricular thalamus (PVT), an area that was revealed to be important for the control of emotion, motivation, and food intake. However, the function and modulation of PVT OT signaling remain unknown. Here, we used a progressive ratio (PR) schedule of reinforcement to examine the role of PVT OT signaling in regulating the motivation for food and patch-clamp electrophysiology to study the modulation of OT on PVT neurons in brain slices. We demonstrate that PVT OT administration increases active lever presses to earn food rewards in both male and female mice under PR trials and OT receptor antagonist atosiban inhibits OT-induced increase in motivated lever presses. However, intra-PVT OT infusion does not affect food intake in normal conditions but attenuates hypophagia induced by stress and anxiety. Using patch-clamp recordings, we find OT induces long-lasting excitatory effects on neurons in all PVT regions, especially the middle to posterior PVT. OT not only evokes tonic inward currents but also increases the frequency of spontaneous excitatory postsynaptic currents on PVT neurons. The excitatory effect of OT on PVT neurons is mimicked by the specific OT receptor agonist [Thr4, Gly7]-oxytocin (TGOT) and blocked by OT receptor antagonist atosiban. Together, our study reveals a critical role of PVT OT signaling in promoting feeding motivation to attenuate stress-induced hypophagia through exciting PVT neurons.

Fig. 1. Intra-PVT OT infusion increases motivation for HFHS food through activating PVT OT receptors.

a Schematics of the PVT location targeted by drug infusion. b Effects of intra-PVT OT (0.5 and 1.0 μg) infusion on breakpoints and reward earned in male mice (n = 13). One-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons. 0.5 and 1.0 μg OT vs. saline infusion: **p < 0.01, ***p < 0.001. c Effects of intra-PVT OT (1.0 μg) on breakpoints and reward earned in both male (n = 20) and female (n = 12) mice. Two-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons. OT vs. saline infusion: **p < 0.01, ***p < 0.001. d Effects of OT (1.0 μg) infusion on breakpoints and reward earned in the absence (n = 13 male mice) and presence (n = 13 male mice) of OT receptor antagonist Ato (1.0 μg). Two-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons. OT vs. saline infusion in the absence and presence of Ato: *p < 0.05, **p < 0.01, ***p < 0.001; OT plus Ato vs. OT infusion: ^#p < 0.05, ^##p < 0.01.

Fig. 2. Intra-PVT OT infusion attenuates acute stress-induced hypophagia and increases food intake of mice in light/dark conflict environment.

a Effects of intra-PVT OT (1.0 μg) infusion on both active and inactive lever presses during an FR1 task. n = 17 mice for each group. Two-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. OT vs. saline infusion on active lever presses: ***p < 0.001. b Effect of intra-PVT OT (1.0 μg) infusion on HFHS pellet consumed during an FR1 task. n = 17 mice. Paired t-test, p = 0.07. c Effect of intra-PVT OT (1 μg) infusion on regular chow intake. n = 9 mice for each group. Two-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. d Effect of intra-PVT OT (1.0 μg) infusion on 1 h palatable HFHS food intake in normal mice and mice after restraint stress of 30 min. n = 9 mice for each group. Two-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. OT vs. saline infusion in normal mice: ***p < 0.001; OT vs. saline infusion in stressed mice: n.s. no significant difference. e Effect of intra-PVT OT infusion on 2 h palatable food intake in both normal and stressed mice. n = 9 mice for each group. Two-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. OT vs. saline infusion in normal mice: *p < 0.05; OT vs. saline infusion in stressed mice: n.s., no significant difference. f Effect of intra-PVT OT infusion on 3 h palatable food intake in both normal and stressed mice. n = 9 mice for each group. Two-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. OT vs. saline infusion in both normal mice and stressed mice: n.s. no significant difference. g A diagram of the light/dark conflict box that was used for testing. h–l Effects of intra-PVT OT (1.0 μg) infusion on latency to the first exit from the dark compartment (h), total entries to the light compartment (i), percentage of time spent in the light compartment (j), total food approaches (k), and food intake (l) when food was placed in the food zone. n = 21 mice for each group. Unpaired t-test, OT vs. saline: *p < 0.05, **p < 0.01, ***p < 0.001. m Effect of intra-PVT (1.0 μg) infusion on the percentage of time spent in the light compartment in the absence of food in the food zone. n = 10 mice for each group. Unpaired t-test, n.s. no significant difference.

Fig. 3. OT excites PVT neurons in both C57BL/6J and Swiss Webster mice.

a A representative current-clamp trace shows the membrane potential and action potentials of a PVT neuron in a C57BL/6J mouse before, during, and after OT (1 μM) application of 1 min. Expanded sections (10 s) are shown underneath the trace. b Percentage of recorded PVT neurons (n = 31) in slices of 15 C57BL/6J mice. If the depolarization was more than 1.5 mV and/or the firing rate was increased at least 20% by OT treatment, we considered those neurons as OT-excited or OT-responsive neurons. c Effect of OT on the resting membrane potential of responsive PVT neurons in C57BL/6J mice. n = 18 cells from nine mice. Paired t-test, ***p < 0.0001 compared with control before OT application. d Long-lasting effect on firing rate of responsive PVT neurons in C57BL/6J mice. n = 17 cells from nine mice. Firing rates of 1 min were measured at each time point for the recording of at least 30 min. One-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons: **p < 0.01, ***p < 0.001 compared with control at the time point before OT application. e Percentage of recorded PVT neurons (n = 25) in slices of 12 Swiss Webster mice. f A representative trace shows the membrane potential and action potentials of a PVT neuron in a Swiss Webster mouse before, during, and after OT (1 μM) application of 1 min. g A representative trace shows the membrane potential and action potentials of a PVT neuron before, during, and after OT (1 μM) application of 1 min in the presence of AP5 (50 μM), CNQX (10 μM), and Bic (30 μM). h A bar graph shows the effect of OT on the membrane potential of PVT neurons in Swiss Webster mice in the absence (n = 14 cells from nine mice) and presence of AP5, CNQX, and Bic (n = 8 cells from four mice). Two-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons: ***p < 0.001 compared with control before OT application. i OT effect on the firing rate of PVT neurons in Swiss Webster mice in the absence (n = 14 cells from nine mice) and presence of AP5, CNQX, and Bic (n = 8 cells from four mice). Two-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons: ***p < 0.001 compared with control before OT application.

Fig. 4. OT receptor is responsible for the OT excitation of PVT neurons.

a Representative voltage-clamp traces show the inward currents evoked by the application of OT (1 μM, top), selective OT receptor agonist TGOT (1 μM, second), OT (1 μM) in the presence of AP5 (50 μM), and CNQX (10 μM, third trace), and OT (1 μM) in the presence of OT receptor antagonist atosiban (1 μM, bottom). b The amplitudes of inward currents evoked by OT, TGOT, OT plus Ato, and OT in the presence of AP5 (50 μM), CNQX (10 μM), and Bic (30 μM) on all sampled PVT neurons. OT group: n = 6 cells from two mice, TGOT group: n = 12 cells from four mice, OT in the presence of AP5, CNQX, and Bic: n = 6 cells from two mice, OT + Ato group: n = 13 cells from three mice. One-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons: **p < 0.01 compared with OT application only. c Representative traces show sEPSCs in the absence (top) and the presence of OT (1 μM, bottom). d OT increases the sEPSC frequency of all sampled PVT neurons (n = 8 from two mice). Paired t-test, *p < 0.05 compared with control before OT application. e OT has no obvious effect on the sEPSC amplitude of all sampled PVT neurons (n = 8 from two mice). Paired t-test, p = 0.10. f Representative traces show sEPSC in the absence (top) and the presence of TGOT (1 μM, bottom). g TGOT increases the sEPSC frequency of all sampled PVT neurons (n = 16 from eight mice). Paired t-test, *p < 0.05 compared with control before OT application. h TGOT has no obvious effect on the sEPSC amplitude of all sampled PVT neurons (n = 16 from eight mice). Paired t-test, p = 0.09.

Fig. 5. Age and subregional differences in TGOT modulation of PVT neuron.

a Representative traces show TGOT (1 μM) excites PVT neurons of mice at postnatal 2–3 and 7–10 weeks. b Percentage of TGOT-responsive PVT neurons in both male and female mice at postnatal 2–3 and 7–10 weeks. Chi-square test, *p < 0.05 compared with postnatal 2–3 weeks. c TGOT evoked depolarization on TGOT-responsive neurons in both male and female mice at postnatal 2–3 weeks (n = 22 cell from ten male mice, n = 23 cells from ten female mice) and 7–10 weeks (n = 24 cells from 22 male mice, n = 28 cells from 20 female mice). Two-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. 2–3 vs. 7–10 weeks in both male and female mice: n.s., no significant difference. d Effect of TGOT on firing rate of TGOT-responsive PVT neurons in both male and female mice at postnatal 2–3 weeks (n = 22 cell from ten male mice, n = 23 cells from ten female mice) and 7–10 weeks (n = 24 cell from 22 male mice, n = 28 cells from 20 female mice). Three-way ANOVA followed by Bonferroni post hoc tests for multiple comparisons. TGOT vs. Ctrl: ***p < 0.001. e Representative traces show TGOT (1 μM) excites neurons in aPVT, mPVT, and pPVT of mice at postnatal 7–10 weeks. f Percentage of TGOT-excited neurons in aPVT, mPVT, and pPVT of mice at postnatal 7–10 weeks. Chi-square test, *p < 0.05 compared with aPVT area. g Effect of TGOT on the membrane potential of TGOT-responsive neurons in aPVT (n = 12 cells from eight mice), mPVT (n = 34 cells from 18 mice), and pPVT (n = 15 cells from 12 mice) of mice at postnatal 7–10 weeks. Two-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons. TGOT vs. Ctrl: ***p < 0.001. h Effect of TGOT on firing rate of TGOT-responsive neurons in aPVT (n = 12 cells), mPVT (n = 34 cells), and pPVT (n = 15 cells) of mice at postnatal 7–10 weeks. Two-way ANOVA repeated measures followed by Bonferroni post hoc tests for multiple comparisons. TGOT vs. Ctrl in various PVT subregions: *p < 0.05, ***p < 0.001. mPVT and pPVT neurons with TGOT treatment vs. aPVT neurons with TGOT treatment. ^#p < 0.05, ^##p < 0.01.