Introduction

Food craving, characterized by a strong desire to consume food, paired with the compulsion to eat not necessarily out of hunger, is a central feature of addictive disorders [1]. Numerous clinical studies demonstrate that an addictive tendency toward food is closely associated not only with psychological distress but also with metabolic disturbances that can lead to obesity and diabetes, thereby negatively impacting overall health [2,3,–4]. Although the mechanisms underlying food craving and compulsive eating are not yet understood, accumulating evidence from clinical and animal studies highlights the role of brain reward pathways [5,6,7,–8].

Alterations in the mesolimbic dopamine system can induce reward deficiency and are associated with impulsivity and compulsivity, which can lead to excessive intake of highly palatable foods that are high in sugar and fat despite negative consequences [9,10,–11]. This system, in conjunction with several brain sites, mediates reward-seeking and appetitive behavior by increasing the saliency of environmental cues and enhancing positive reinforcement [12,13,14,15,16,17,–18]. In particular, human brain imaging and genetic studies as well as animal models indicate the critical involvement of the dopamine D2 receptor (D2R) in obesity and compulsive eating [11, 19,20,21,22].

The central amygdala (CeA) modulates food intake [14, 15] by integrating motivation with salient environmental cues, which is important for conditioned stimulus-reward learning to obtain food and drug rewards within an emotional context [23, 24]. Functional magnetic resonance imaging of the human brain together with measurement of food addiction scores show that increased amygdala activation is associated with increased appetitive motivation and valuation of appetitive stimuli [25, 26]. Furthermore, genetic, anatomical, and optogenetic manipulations reveal that CeA D2Rs regulate impulsivity and compulsive-like eating [11].

While exploring how D2R-expressing neurons in the CeA regulate compulsive eating, we found significant colocalization of D2Rs and insulin receptors (InsRs) in the CeA. Here, we investigated the possible interaction between CeA D2Rs and InsRs in the control of compulsive-like eating. We found that proper regulation of crosstalk between these receptors is critical for fine-tuning brain insulin sensitivity, thereby contributing to the management of normal or maladaptive eating behaviors.

Materials and methods

Animals

Experiments were performed with male Drd2–/– mice (B6.129S2- Drd2tm1Low/J, Strain #:003190, Jackson Laboratory), Drd2-Cre bacterial artificial chromosome (BAC) transgenic mice (B6.FVB[Cg]-Tg[Drd2-cre] ER44Gsat/Mmucd, 032108-UCD, Mutant Mouse Resource & Research Centers), Drd2-EGFP BAC transgenic mice (STOCK Tg(Drd2-EGFP)S118Gsat/Mmnc, 000230-UNC, Mutant Mouse Resource & Research Centers), and Drd2flox/flox mice (detailed information was described in the Supplemental materials; Mouse Genome Informatics [MGI:8234914]). Mice were maintained in a specific pathogen-free barrier facility under constant temperature (23 ± 3 °C) and humidity (50 ± 10%) on a 12–12 h light-dark schedule. Animal care and all experimental procedures were performed in accordance with the guidelines of the Institutional Animal Care and Use Committees of Korea University (certificate number of approvals of animal protocol: KUIACUC-2024-0014).

Food self-acquisition task

Mice were habituated to individual housing for 1 week, and baseline body weight and food intake were measured for another week. Food intake was then restricted to ~85–90% of the baseline amount. The test apparatus (MED Chamber, Med Associates Inc., #ENV307W) consisted of a chamber (25 × 25 × 25 cm) housed within a ventilated wooden sound-attenuating box. The chamber contained two levers on the left and right sides 1.5 cm above a metal wire-grid floor. Mice received a sucrose pellet reward when they pressed the active lever with the correct reward value in the FR1, FR2, and FR3 training sessions. The FR3 session was used as a baseline, after which we performed foot shock (0.3 mA, 0.5 s) + reward testing on an FR3 schedule. Foot shock was delivered by scrambled electric shock through a metal grid using a shock generator (Med Associates Inc, #ENV-414).

Light/dark box test

We measured baseline mouse body weight and food intake for 1 week and performed a pre-test. On all test days, mice were habituated to the test room for 60 min. Next, mice were placed in the dark chamber. The test started with the removal of a screen door that divided the dark chamber from the light chamber. Time spent in the dark and light chambers and the number of crossovers were measured with video tracking software (Ethovision XT15).

Fiber photometry

We injected a Cre-dependent calcium indicator unilaterally into the CeA of Drd2-Cre male mice to specifically observe the activity of D2R-expressing neurons. During surgery, 1 μl AAV-EF1a-DIO-GCaMP6s-P2A-nls-dTomato (Addgene, #51082, final titer: 3–6 × 1011-12 VP/ml) was injected into the CeA unilaterally at 100 nl/min. For analysis of DA level in the CeA of D2R knockdown mice, we injected each 1 ul of AAV-hSyn-GRAB_DA2m (Addgene, #140553, final titer: 3–6 × 1011-12 VP/ml) and AAV-DIO-DSE-mCherry-PSE-shD2R into the CeA. AAV-EF1a-DIO-GCaMP6s-P2A-nls-dTomato was a gift from Jonathan Ting (http://n2t.net/addgene:51082 ; RRID:Addgene_51082) and pAAV-hsyn-GRAB_DA2m was a gift from Yulong Li (http://n2t.net/addgene:140553 ; RRID:Addgene_140553). A 400-μm optic fiber housed in a metallic ferrule (Doric Lenses, MFC_400/430-0.48_5mm_FLT) was implanted 0.2 mm above the injection site. We used a four-port Fluorescence Mini Cube (doric, ilFMC4-G2 IE(400–410)_E(460–490)_F(500–550)_S) for exciting GCaMP6s and delivering emission fluorescence. LED light was delivered at 460–490 nm to excite GCaMP6s and at 405 nm to isosbestic control.

Statistical analysis

Differences between two groups were analyzed using unpaired Student’s t-tests, and those among multiple groups were evaluated using one-way or two-way analysis of variance (ANOVA) followed by appropriate post hoc comparisons. A p-value of <0.05 was considered statistically significant.

Other methods

Additional experimental procedures are provided in the Supplementary Information.

Results

Absence of D2Rs increases compulsive-like eating

We evaluated compulsive-like eating in wild-type (WT) and dopamine D2R knock-out (Drd2−/−) mice by employing a palatable food (PF) self-acquisition task. When a mouse pressed the active lever, a sucrose pellet (i.e., PF) was delivered. After 7 days, the number of lever presses required to obtain PF was increased, shifting from a fixed ratio of 1 (FR1) to FR2 and FR3 training schedules (Fig. 1A). Both groups successfully acquired lever pressing as the sessions progressed (Fig. 1B, C), although Drd2−/− mice displayed slightly less active lever pressing than WT mice across sessions, probably due to decreased locomotor activity in these mice [27]. During FR1 to FR3 sessions, both WT and Drd2−/− mice exhibited significantly increased active lever pressing than inactive lever press, demonstrating that the increase in lever pressing is specific to PF-seeking (Supplementary Fig. 1B–E).

Fig. 1: CeA D2Rs regulate compulsive-like eating.
figure 1

A Experimental scheme for food self-acquisition task. FR: fixed ratio, FS: foot shock (scrambled shock, 0.3 mA, 0.5 s), PF: palatable food, A: active lever, I: inactive lever. B Number of active lever presses per hour by WT and Drd2−/− mice across days. Two-way repeated measures ANOVA, group×session interaction: F3,78 = 4.77, **p = 0.0042. C Number of active lever presses per hour from FR1 to FR3 sessions. One-way ANOVA followed by Bonferroni post hoc tests, ###p < 0.001 vs. FR1, ***p < 0.001 vs. WT. D Number of active lever presses per hour in the last 3 days of FR3 + FS sessions. Unpaired Student’s t-test, two-tailed, **p = 0.0060. E Number of active lever presses per hour in FS only sessions. WT n = 12, Drd2−/− n = 16. F Schematic diagram of AAV-GFP or AAV-Cre-GFP injection into the CeA of Drd2flox/fox mice. G Left: Representative images of AAV-GFP and AAV-Cre-GFP virus expression in the CeA at low magnification. Virus expression was observed in three sections per mouse used for experiments. GFP control n = 3, Cre-GFP n = 4. Right: Drd2 mRNA expression in the CeA of GFP and Cre-GFP virus-injected mice as detected by FISH. Red: Drd2, green: GFP or Cre-GFP, blue: DAPI. Scale bar: low magnification, 100 μm, high magnification, 20 μm. H Quantitative analysis of Drd2 mRNA expression in the CeA. Proportion of Drd2 mRNA in GFP and Cre- GFP virus-injected mice. Unpaired Student’s t-test, two-tailed, **p = 0.0054. GFP control n = 3, Cre-GFP n = 4. I Number of active lever presses per hour by GFP and Cre-GFP virus-injected mice across days. Two-way repeated measures ANOVA, group×session interaction: F5,75 = 5.63, ***p = 0.0002. J Number of active lever presses per hour from FR1 to FR3 sessions. One-way ANOVA followed by Bonferroni post hoc tests, ###p < 0.001 vs. FR1. K Number of active lever presses per hour in the last 3 days of FR3 + FS sessions. Unpaired Student’s t-test, two- tailed, **p = 0.0033. L Number of active lever presses per hour in FS only sessions. GFP n = 8, Cre-GFP n = 6. All values in the data represent mean ± SEM.

Full size image

After the FR3 sessions, additional FR3 sessions with delivery of electric foot shock (FS) were administered (FR3 + FS), and the perseverance of lever pressing to obtain PF despite the delivery of foot shocks was used as an indicator of compulsive-like food-seeking. Surprisingly, Drd2−/− mice exhibited significantly higher active lever pressing despite receiving foot shocks (Fig. 1B, D), showing perseverance in reward-seeking despite punishment. In contrast, WT mice exhibited significantly increased inactive lever press compared to active lever during FR3 + FS sessions (Supplementary Fig. 1F). When the rewards were omitted and only foot shocks were delivered (FS only), lever pressing was comparable between WT and Drd2−/− mice (Fig. 1B, E, Supplementary Fig. 1B, G), indicating similar sensitivity to foot shocks, but greater compulsivity in Drd2−/− mice, driving active lever pressing despite punishment. These data, together with previous observations [11], indicate that the absence of D2Rs promotes compulsive-like eating behavior.

CeA D2Rs regulate compulsive-like eating

To further address the role of D2Rs in the control of compulsive eating, we targeted D2Rs in the CeA based on our previous finding that CeA D2Rs play a crucial role in reward-related impulsive and compulsive behaviors [11]. We generated Drd2 flox/flox mice (Supplementary Fig. 2) and injected AAV-Cre-eGFP virus into the CeA to selectively eliminate D2R expression in the CeA (CeA-Drd2KO mice) (Fig. 1F), resulting in a 73% reduction in D2R expression in CeA-Drd2KO mice (Cre-GFP) compared with control flox mice in which AAV-eGFP virus was injected (CeA-Drd2WT, GFP) (Fig. 1G, H). We then trained mice in the PF self-acquisition task.

Both control (GFP) and CeA-Drd2KO (Cre-GFP) mice displayed successful FR1-FR3 performance with no difference between groups (Fig. 1I, J, Supplementary Fig. 3B–E). However, CeA-Drd2KO mice (Cre-GFP) showed more active lever pressing than CeA-Drd2WT mice (GFP) during the FR3 + FS sessions (Fig. 1I, K) while CeA-Drd2WT mice exhibited significantly increased inactive lever press compared to active lever during FR3 + FS sessions (Supplementary Fig. 3F). In FS-only sessions, both CeA-Drd2WT and CeA-Drd2KO mice showed low levels of lever pressing (Fig. 1I, L, Supplementary Fig. 3G). Taken together, these data suggest that D2Rs in the CeA critically contribute to the compulsive-like perseverance of PF-seeking behavior.

CeA D2Rs regulate InsR signaling

Recent reports indicate that InsRs are highly expressed in the CeA and are associated with reduced food intake [28,29,–30]. Using Drd2-EGFP mice, we observed abundant expression of InsRs in the CeA and found that ~61% of CeA D2R- expressing neurons co-expressed InsRs (Fig. 2A, B). Interestingly, in Drd2−/− mice, InsR expression was decreased to 42% in the CeA compared with WT mice (Fig. 2C, D). In addition, in CeA-specific Drd2 KO (CeA-Drd2KO) mice, InsR expression was reduced by ~60% compared with control CeA-Drd2WT mice (Fig. 2E–G).

Fig. 2: D2R and InsR expression in the CeA and regulation of InsR signaling by D2Rs.
figure 2

A D2R and InsR expression in the CeA of Drd2-eGFP mice. Left: Low magnification. Right: High magnification of area in the yellow box shown in the left. Red: InsR, green: Drd2-eGFP, blue: DAPI. B Quantitative analysis of D2R and InsR expression in the CeA of Drd2-eGFP mice. Drd2-eGFP n = 6 (8 sections/mouse). C InsR expression in the CeA of WT and Drd2−/− mice. Green: InsR, blue: DAPI. D Proportion of InsR-expressing cells in the CeA of WT and Drd2−/− mice. Unpaired Student’s t-test, two-tailed,***p = 0.0005. WT n = 3, Drd2−/− n = 3 (6sections/mouse). E Schematic diagram of AAV-GFP or AAV-Cre-GFP injection into the CeA of Drd2flox/flox mice. F Left: Representative images of GFP or Cre-GFP virus expression in the CeA in low magnification. Right: InsR expression in the CeA of GFP and Cre-GFP virus- injected mice as detected by immunohistochemistry. Red; InsR, Green: GFP or Cre-GFP, Blue: DAPI. Scale bar: low magnification, 100 μm, high magnification, 20 μm. G Quantitative analysis of InsR expression in the CeA. The proportion of InsR expression in GFP and Cre-GFP-injected Drd2flox/flox mice. Unpaired Student’s t-test, two-tailed, *p = 0.0123. GFP n = 3, Cre-GFP n = 3. H Expression of pInsRTyr972 with GFP or Cre-GFP-injected Drd2flox/flox mice. Saline/Insulin (10 mU)/quinpirole (1 µg) was infused 30 min before sacrifice. Left: low-magnification of the CeA of saline(S), insulin(I), or quinpirole(Q)-infused mice. Middle/Right: high-magnification of pInsRTyr972 expression in the CeA detected by immunohistochemistry. Quantitative analysis: three sections of one hemisphere/mouse, 4mice/group. Red: pInsRTyr972, Green: GFP/Cre-GFP, Blue: DAPI, Arrowhead: pInsRTyr972-positive cells. I pInsRTyr972 expression in the CeA of GFP or Cre-GFP-injected Drd2flox/flox mice. Two-way ANOVA, Bonferroni posthoc tests, group×drug interaction: F2,18 = 9.09, **p = 0.0019. GFP S/I/Q n = 4, Cre-GFP S/I/Q n = 4. J pAktSer473 expression in the CeA of GFP or Cre-GFP-injected Drd2flox/flox mice. Two- way ANOVA, Bonferroni posthoc tests, group×drug interaction: F2,26 = 5.66, **p = 0.0091, GFP S/I/Q n = 6, Cre-GFP S/Q n = 5, Cre-GFP I n = 4. K Expression of pPTP1BSer50 in CeA of GFP/Cre-GFP-injected Drd2flox/flox mice. Quantitative analysis: three sections of one hemisphere/mouse, n=4mice/group. Red: pPTP1B, Green: GFP/Cre-GFP, Blue: DAPI, Arrowhead: pPTP1B-positive cells. L pPTP1BSer50 expression. Two-way ANOVA,Bonferroni posthoc tests, group×drug interaction: F2,18 = 5.13, *p = 0.0173, GFP S/I/Q n = 4, Cre-GFP S/I/Q n = 4. Scale bars in all images: low-magnification, 100μm, high-magnification, 20μm. All values in the data represent mean ± SEM.

Full size image

We next examined InsR activity by assessing insulin-mediated phosphorylation of InsRs in the CeA of CeA-Drd2WT (Drd2flox/flox + GFP) and CeA- Drd2KO (Drd2flox/flox+Cre-GFP) mice as well as the possible regulation of InsR phosphorylation by a D2R agonist. We analyzed phosphorylation of Tyr-972 (pInsRTyr972), an important residue for the recruitment of IRS-1, phosphorylation of Tyr1162/1163 (pInsRTyr1162/1163) in the activation loop of the InsR catalytic domain [31, 32], and phosphorylation of Akt (pAktSer473) in the downstream InsR signaling cascade [32]. After 12 h of food deprivation, saline, insulin (10 mU), or D2R agonist quinpirole (1 μg) were infused into the CeA. A robust increase in the number of pInsRTyr972-positive cells was observed 30 min after insulin or quinpirole infusion in CeA-Drd2WT mice (Drd2flox/flox + GFP), whereas CeA-Drd2KO mice (Drd2flox/flox+Cre-GFP) showed higher basal levels of pInsRTyr972-positive cells, but no increase after insulin or quinpirole infusion (Fig. 2H, I, Supplementary Fig. 4A). Insulin or quinpirole infusion significantly increased the number of pInsRTyr1162/1163-positive cells in the CeA of CeA-Drd2WT mice (GFP). By contrast, CeA-Drd2KO mice (Cre-GFP) showed a higher basal level of pInsRTyr1162/1163-positive cells, but insulin or quinpirole infusion had little or no effect on their levels (Supplementary Fig. 4B, D, E). Insulin or quinpirole infusion significantly increased the number of pAKTSer473-positive cells in the CeA of CeA-Drd2WT mice but not in CeA-Drd2KO mice (Fig. 2J, Supplementary Fig. 4C, D).

To understand how the D2R activation induces the tyrosine phosphorylation of InsRs, we hypothesized that D2R activation with Gi protein coupling promotes the phosphorylation of InsRs. Several studies demonstrate that the cAMP-inhibiting Gi proteins, including Gαi2, positively regulate InsR activation [33,34,35,–36]. By contrast, the expression and activity of protein-tyrosine phosphatase 1B (PTP1B) is decreased in tissue expressing constitutively active Gαi2 [37]. We examined the involvement of PTP1B by analyzing the phosphorylation level of PTP1B at Ser50 (pPTP1BSer50), which negatively modulates its phosphatase activity, thus creating a positive feedback mechanism for insulin signaling [38]. We observed that D2R activation by quinpirole significantly increased the number of pPTP1BSer50-positive cells in CeA-Drd2WT (Drd2flox/flox + GFP) mice, whereas insulin or quinpirole infusion had little or no effect on number of pPTP1BSer50-positive cells in CeA-Drd2KO mice (Drd2flox/flox+Cre-GFP) (Fig. 2K, L and Supplementary Fig. 4F). These data suggest that D2R activation induces PTP1B inhibition and impairs its ability to dephosphorylate InsRs, leading to increased insulin signaling. Thus, the stimulation of CeA D2Rs may induce concomitant activation of InsR signaling in the CeA, and the absence of D2Rs severely impairs InsR signaling.

CeA InsRs impact compulsive-like eating

We next tested the effect of selective loss of InsR expression in CeA D2R- expressing neurons on compulsive-like eating behavior. To accomplish this, we used a gene knockdown strategy with Cre-dependent AAV vectors expressing a short-hairpin RNA (shRNA) targeting InsRs to induce cell type-specific loss of function while simultaneously visualizing recombination through mCherry labeling using AAV-DIO-DSE-mCherry-PSE-shInsR [39]. Selective expression of shInsR in D2R- expressing neurons in the CeA (CeA-D2RshInsR) resulted in an ~80% reduction in InsR expression compared with control mice in which AAV-DIO-eYFP was injected (CeA-D2ReYFP) (Fig. 3A–C). We found that D2R-specific loss of InsRs in the CeA exacerbated compulsive-like eating phenotype, as evidenced by robust perseverance of active lever pressing despite foot shock punishment (Fig. 3D–F, Supplementary Fig. 5B–F). In FS-only sessions, no difference in lever pressing was observed between control and CeA-D2RshInsR mice (Fig. 3D, G, Supplementary Fig. 5G). Together, these data suggest that InsRs and the interaction between InsRs and D2Rs in the CeA are critical for the control of compulsive-like eating behavior.

Fig. 3: Selective loss of InsRs in CeA D2R-expressing neurons results in stronger compulsive-like perseverance of PF-seeking behavior.
figure 3

A Schematic diagram of AAV-DIO-eYFP or AAV-DIO-DSE-mCherry-PSE-shInsR injection into the CeA of Drd2-Cre mice. B Representative InsR expression in the CeA of CeA-D2ReYFP control and CeA-D2RshInsR mice. Virus expression was observed in three sections per mouse used for experiments. CeA-D2ReYFP n = 3, CeA-D2RshInsR n = 3. Red: eYFP or shInsR, green: InsR, blue: DAPI. Scale bar: low magnification, 100 μm, high magnification, 20 μm. C Quantitative analysis of InsR expression in the CeA of CeA-D2ReYFP control and CeA-D2RshInsR mice. CeA-D2ReYFP n = 3, CeA-D2RshInsR n = 3. Unpaired student’s t-test, two-tailed, *p = 0.0344. D Number of active lever presses per hour by CeA-D2ReYFP and CeA-D2RshInsR mice across days. CeA-D2ReYFP n = 3, CeA-D2RshInsR n = 4. E Number of active lever presses per hour in FR1 to FR3 sessions. One-way ANOVA followed by Bonferroni post hoc tests, #p < 0.05, ###p < 0.001 vs. FR1. CeA-D2ReYFP n = 3, CeA-D2RshInsR n = 4. F Number of active lever presses per hour in the first 3 days of FR3 + FS sessions. Unpaired Student’s t-test, two-tailed, **p = 0.0061. CeA-D2ReYFP n = 3, CeA- D2RshInsR n = 4. G Number of active lever presses per hour in FS only sessions. CeA-D2ReYFP n = 3, CeA-D2RshInsR n = 4. All values in the data represent mean ± SEM.

Full size image

PF and insulin modulate CeA D2R-expressing neuronal activity

To understand how CeA D2R-expressing neurons respond to PF, we used fiber photometry to perform in vivo calcium imaging. We virally expressed the Cre- dependent calcium indicator GCaMP6s (AAV-DIO-GCaMP6s-mCherry) in the CeA of Drd2-Cre mice (Fig. 4A) and recorded changes in GCaMP6s fluorescence signal through an optic fiber placed above CeA D2R-expressing neurons during exposure to PF in a light/dark box test. Histological analysis revealed that ~96% of GCaMP6-expressing cells overlapped with Cre-expressing CeA D2R-expressing cells (Fig. 4B, C). After measuring baseline body weight and food intake, mice underwent a pre-test without food for 15 min to measure their time spent in the light and dark chambers of the light/dark box [11]. Mice were then divided into two groups: one group was fed a NC diet, and the other group was fed a PF diet high in sugar and fat ad libitum for 14 days. After a 48-h PF withdrawal, mice were returned to the light/dark box for 15 min, with PF placed in the light chamber, to assess compulsive-like eating behavior [11], during which we recorded GCaMP signals from CeA D2R-expressing neurons (Fig. 4D). As shown by peri-event time histogram (PETH) analysis, there was a significant decrease in z-score (i.e., normalized ΔF/F) during PF consumption (Fig. 4E, G), whereas no significant change in GCaMP signal was observed during exposure to NC (Fig. 4E, F), indicating a decrease in CeA D2R-expressing neuronal activity in response to PF. This suppression was accompanied by increased PF intake in the light chamber, supporting a link between reduced D2R neuron activity and compulsive-like eating behavior (Fig. 4H).

Fig. 4: PF modulates CeA D2R neuronal activity.
figure 4

A Left: diagram of AAV-DIO- GCaMP6s injection into the CeA of Drd2-Cre mice. Right: Representative images of AAV-DIO-GCaMP6s virus expression in the CeA at low magnification. Virus expression was observed in three sections per mouse. Scale bar: 100 μm. Drd2- Cre n = 3. B Colocalization of Cre- and GCaMP6s-positive cells in the CeA. Red; Cre, green; GCaMP6s, blue; DAPI. Scale bar: 20 μm. C Quantification of Cre- and GCaMP6s-positive cells in the CeA of three Drd2-Cre mice. D Experimental scheme of the light/dark box test with fiber photometry. NC; normal chow, PF; palatable food. E Normalized z-score during NC and PF consumption. F Average z-score pre- ( − 2 to 0 s) and post- (0 to +10 s) NC consumption. Drd2-Cre n = 8. G Average z-score pre (-2s to 0 s) and post (0 s to +10 s) PF consumption. Drd2-Cre n = 12. Paired student’s t-test, two- tailed, *p = 0.0023. H Food consumption for 10 min. Drd2-Cre NC n = 7, Drd2-Cre PF n = 11. Unpaired student’s t-test, two-tailed, ***p = 0.0007. All values in the data represent mean ± SEM.

Full size image

We next recorded changes in the GCaMP6s signal from CeA D2R-expressing neurons during exposure to PF in the light/dark box test after infusion of saline, quinpirole, insulin, quinpirole+insulin, or the D2 receptor antagonist sulpiride into the CeA (Fig. 5A). Under saline infusion, CeA D2R neurons exhibited significantly reduced activity during PF consumption (Fig. 5B, C), consistent with the dynamics observed in Fig. 4E, and G. Quinpirole or insulin infusion, also induced significant reductions in z-scores during PF consumption (Fig. 5D–F) which was accompanied by a reduction in PF consumption (Fig. 5D–F, L). However, when quinpirole and insulin were co-infused, the suppression of neuronal activity was alleviated (Fig. 5D, G, L) corresponding to reduced PF consumption. In contrast, sulpiride infusion resulted in a significant suppression of neuronal activity and an increase in PF consumption (Fig. 5H, I, L), while its effect was slightly reversed by quinpirole (Fig. 5H, J, L) and fully alleviated by quinpirole+insulin, which also reduced PF consumption (Fig. 5H, K, L). These data indicate that D2R activation in the presence of insulin effectively enhances CeA D2R-expressing neuronal activity, thereby contributing to the suppression of PF consumption.

Fig. 5: PF and Insulin modulate CeA D2R neuronal activity.
figure 5

A Experimental scheme of light/dark box test and fiber photometry with drug infusion. B Normalized z-score during PF consumption after saline infusion. C Average z-score during PF consumption after saline infusion. Saline: Drd2-Cre n = 7. Paired student’s t-test, two-tailed, **p = 0.0095. D Normalized z-score during PF consumption after quinpirole (1 µg), insulin (5 mU), or quinpirole+insulin infusion. E-G Average z-score during PF consumption after drug infusion. E Quinpirole: Drd2-Cre n = 10. Paired student’s t-test, two-tailed, *p = 0.0153. F Insulin: Drd2-Cre n = 10. Paired student’s t-test, two-tailed, **p = 0.0045. G Quin+Ins: Drd2-Cre n = 9. Paired student’s t-test, two-tailed, p = 0.3200. H Normalized z-score during PF consumption after sulpiride (1 µg), sulpiride+quinpirole, or sulpiride+quinpirole+insulin infusion. I-K Average z-score on pre (-2s to 0 s) and post (0 s to +10 s) during PF consumption after drug infusion. I Sulpiride: Drd2-Cre n = 12. Paired student’s t-test, two-tailed, **p = 0.0021. J Sulp+Quin: Drd2-Cre n = 13. Paired student’s t-test, two-tailed, *p = 0.0442. K Sulp+Quin+Ins: Drd2-Cre n = 11. Paired student’s t-test, two-tailed, p = 0.8766. L PF consumption for 10 min. in drug-infused mice. Saline, Quinpirole, Insulin, Quin+Ins Drd2-Cre n = 13. Repeated measure One-way ANOVA, Bonferroni posthoc tests **p = 0.0048, #p < 0.05 vs. saline. Sulpiride, Sulp+Quin, and Sulp+Quin+Ins Drd2- Cre n = 13. Repeated measure One-way ANOVA, Bonferroni posthoc tests, ***p = 0.0002, †††p < 0.001 vs. saline, **p < 0.01, ***p < 0.001 vs. group. Solid line: average z-score of food consumption. Shaded line: standard error of the mean. All values in the data represent mean ± SEM.

Full size image

We further explored the effect of insulin on CeA D2R neuronal activity by ex vivo electrophysiological recordings in brain slices from Drd2-EGFP mice. Bath application of quinpirole alone (10 µM) failed to significantly change the membrane excitability of D2R-expressing CeA neurons (Supplementary Fig. 6A–E). CeA D2R neurons did not show significant changes in rheobase or membrane potential. However, when co-treated with insulin (500 nM), quinpirole dramatically depolarized their resting membrane potential and increased their excitability, as shown by decreased rheobase and enhanced input resistance (Supplementary Fig. 6C–E). When insulin alone was perfused into the recording chamber, it significantly hyperpolarized the resting membrane potential of CeA D2R neurons and decreased their firing probabilities by increasing rheobase and reducing input resistance (Supplementary Fig. 6F–J). These ex vivo electrophysiological studies suggest that the activation of D2Rs and InsRs has a synergistic effect that substantially increases the excitability of D2R neurons in the CeA. Furthermore, this interaction may be critical for the regulation of compulsive-like eating behavior, supporting our GCaMP6s photometry analysis following quinpirole or insulin infusion.

Next, we examined the effect of selective optogenetic inhibition of CeA D2R-positive neurons expressing halorhodopsin in the CeA of Drd2-Cre mice by injecting AAV-DIO-eNpHR3.0-eYFP virus (Fig. 6A, B) and conducted the light/dark box test (Fig. 6C–H). Bilateral photoinhibition of CeA D2R neurons resulted in a significant and specific increase in palatable food consumption in the light compartment (Fig. 6G, H and Supplementary Video 1). We further tested whether InsR signaling contributes to the regulation of CeA D2R-mediated PF consumption by selectively activating CeA D2R neurons. To this end, we injected AAV-DIO-eYFP (eYFP) or AAV-DIO-ChR2-eYFP (ChR2) into the CeA of Drd2-Cre mice followed by CeA infusion of the InsR antagonist S961 (Supplementary Fig. 7, Supplementary Fig. 8A, B). Mice underwent the light/dark box test, and pre-test evaluations revealed no differences in basal body weight, food intake, or light/dark box preference between groups expressing eYFP (control) or ChR2 in CeA D2R neurons (Supplementary Fig. 8C–F). Following this, the mice were divided into two groups: one group was maintained on a NC diet, while the other was provided with PF for 14 days. Afterward, mice were reintroduced to the light/dark box for a 25-minute test, with PF placed in the light compartment. Fifteen minutes before testing, mice received CeA infusions of either saline or S961.

Fig. 6: Inhibition of CeA D2R neuronal activity enhances compulsive-like eating behavior.
figure 6

A Diagram of AAV-DIO-eNpHR3.0-eYFP injection into the CeA. B Experimental scheme of light/dark box test using optogenetic inhibition. Optic stimulation: 532 nm, 5 mW, continuous. C-D Basal body weight and food intake for 1 week. E-F Time spent in each box and the number of crossovers during 15-min pre-test. G Food consumption in light box during 10-min total laser-off and -on periods. Two- way ANOVA, Bonferroni post hoc tests, genotype × laser interaction: F1,20 = 18.86, p = 0.0003, **p < 0.01 vs. WT, ##p < 0.01 vs. laser-off. H Time course of food consumption in laser-off and -on 5-min periods. Green shaded-square: laser-on period. Unpaired student’s t-test, two-tailed, *p = 0.0261, **p = 0.0019. WT NC/PF n = 6, Drd2-Cre NC n = 5, PF n = 6. All values in the data represent mean ± SEM. I proposed interplay between D2Rs and InsRs in the CeA. A decrease in CeA D2R-expressing neuronal activity is associated with palatable food (PF) consumption. Activation of D2R induces InsR phosphorylation via Gi protein coupling and PTP1B inhibition as demonstrated in this study. The synergistic activation of D2Rs and InsRs enhances CeA D2R-expressing neuron activity, thereby controlling PF intake. Conversely, loss or blockade of D2Rs disrupts InsR signaling, leading to reduced CeA D2R-expressing neuron activity and increased PF consumption. Therefore, proper coordination between D2R and InsR signaling is critical for fine-tuning brain insulin sensitivity and managing normal or maladaptive eating behavior.

Full size image

In the PF-fed group, optogenetic activation of CeA D2R neurons significantly suppressed PF consumption in ChR2-expressing mice. However, this suppression was abolished when S961 was infused into the CeA, demonstrating that InsR signaling is necessary for CeA D2R-mediated regulation of compulsive-like PF consumption (Supplementary Fig. 8G, H). These findings indicate that CeA D2R neurons, in conjunction with InsR signaling, are critical for modulating compulsive-like eating behavior specific to PF. In the NC group, however, optogenetic activation of CeA D2R neurons had no effect on NC consumption, even after six hours of food deprivation prior to testing. Furthermore, S961 infusion did not alter NC consumption in either eYFP control or ChR2-expressing mice (Supplementary Fig. 8G, H). Together, these data indicate that reduced CeA D2R-expressing neuronal activity is associated with increased PF consumption whereas activation of CeA D2R neurons suppresses compulsive-like eating in an InsR-dependent manner (Fig. 6I).

DA release during PF consumption and its modulation by CeA D2R knockdown

Next, we examined changes in DA release in the CeA in response to PF using the genetically encoded fluorescence DA sensor GRAB-DA2m [40]. We prepared two groups of Drd2-Cre mice: one receiving AAV- DIO-DSE-mCherry-PSE-shLacZ in the CeA as a control (shLacZ) and the other receiving AAV- DIO-DSE-mCherry-PSE-shD2R (shD2R) for selective D2R knockdown in CeA D2R neurons (Supplemental method). Selective expression of shD2R in D2R-expressing CeA neurons resulted in a ~ 77% reduction in D2R expression compared to the control group injected with shLacZ (Supplementary Fig. 9A–C). Both groups were injected with AAV-hSyn-GRAB_DA2m viruses bilaterally into the CeA and a fiber optic cannula was implanted unilaterally to enable real-time measurement of DA levels during PF-seeking behavior (Supplementary Fig. 9A, D). After one week of NC access, there was no significant difference in the GRAB signal between the shLacZ and shD2R mice (Supplementary fig. 9F, G). Subsequently, mice were randomly assigned to two groups: one with limited PF access (1 hr/day, PF limited) and the other with extended PF access (24 hr/day, PF extended) for two weeks (Supplementary Fig. 9D). In the PF-limited condition, shD2R mice consumed approximately 64% of their daily caloric intake during the 1-hour PF access period, significantly more than shLacZ mice, despite similar overall daily caloric intake between groups. This behavior indicates binge-like eating in shD2R mice with CeA D2R downregulation (Supplementary Fig. 9E). In the PF-extended condition, both groups displayed similar daily caloric intakes, derived almost exclusively from PF consumption (Supplementary Fig. 9E). In terms of body weight, PF-extended shD2R mice showed a slight, non-significant increase compared to shLacZ mice (Supplementary Fig. 9F).

After two weeks of PF exposure, DA release profiles during PF consumption were similar between shLacZ and shD2R mice in the PF-limited condition. However, in the PF-extended condition, shD2R mice exhibited a blunted DA signal in the CeA compared to shLacZ controls (Supplementary Fig. 9I–L), These findings indicate that CeA D2R deficiency leads to attenuated DA release under prolonged PF exposure, emphasizing the importance of CeA D2R in modulating reward-related DA signaling.

Discussion

Mounting evidence demonstrates that dopamine signaling closely interacts with brain insulin function not only in metabolic control but also in the regulation of reward-related behaviors, including eating and cognition [41,42,43,44,45,–46]. Here, we uncovered the molecular and cellular basis by which dopamine receptor signaling regulates brain insulin signaling ultimately shaping compulsive-like eating behavior.

Activation of D2Rs stimulates the phosphorylation of InsRs, resulting in D2R-mediated and insulin-independent InsR activation. D2Rs and InsRs likely interact in a push-pull manner to balance CeA D2R-expressing neuronal activity, which can tightly control eating. We found that CeA D2R-expressing neuron activity was decreased during PF consumption, which correlated with increased intake and persistent PF-seeking despite punishment. These results are in line with our previous finding that activation of D2R-expressing neurons in the CeA selectively suppresses impulsive behavior [11]. Our current data suggest that D2R activation stimulates InsR phosphorylation, leading to increased CeA D2R-expressing neuron activity, which in turn suppresses PF intake. Supporting this, quinpirole+insulin infusion into the CeA enhanced CeA D2R-expressing neuronal activity and decreased PF consumption. These results suggest that D2R-induced InsR activation is essential for regulating PF intake and that disruptions in this balance may contribute to maladaptive eating behaviors and impaired insulin sensitivity.

We further demonstrated that the activation of D2Rs induces the ligand-independent phosphorylation of InsRs, which appears to be mediated by Gαi, which is known to be coupled to D2R [7, 47]. Previous studies show that Gαi2 enhances InsR signaling by suppressing PTP1B [36, 37]. Here, we observed that D2R activation via quinpirole significantly increased the level of pPTP1BSer50, thus representing enhanced insulin signaling, whereas in the absence of D2Rs, insulin or quinpirole infusion had little or no effect on pPTP1BSer50 level (Fig. 2K, L). These data suggest that D2R activation induces PTP1B inhibition and impairs its ability to dephosphorylate InsRs, leading to increased insulin signaling, thereby revealing a potential molecular mechanism underlying D2R-induced InsR phosphorylation and activation. Consistently, it was recently reported that PTP1B is elevated in the CeA of rats fed a high-fat diet and that knockdown of PTP1B reduces food intake and adiposity by improving insulin sensitivity [48]. Nonetheless, further studies on the close colocalization of and interaction between D2Rs and InsRs are necessary for untangling the interplay between these two receptors.

To gain more insight into the neuronal cell population expressing D2Rs and/or InsRs in the CeA, we reanalyzed single-nucleus RNA-sequencing (snRNA-seq) results from two recently published independent data sets [49, 50]. D2R- expressing neurons were predominantly observed in the lateral nucleus of the CeA (CeL) and the capsular nucleus of the CeA (CeC) [11, 15]. As previously reported [49], we predominantly identified clusters of inhibitory neurons (11 out of 16 clusters), throughout which InsR was ubiquitously expressed, but Drd2 was co-expressed in only two subtypes (inhibitory neuron clusters 1 and 7). In inhibitory neuron cluster 7, the Drd2-positive population co-expressed InsR in the CeC subregion based on expression of the calcitonin receptor-like receptor (Calcrl) marker gene (Supplementary Fig. 10A–C) [15, 49]. Notably, Drd2 and InsR co- expressing cells in inhibitory neuron cluster 1 appeared to derive from both CeC and CeL subregions, consistent with a previous report that CeL neurons express protein kinase Cδ (Prkcd) and function in appetitive/affective behavior [14, 15]. The same results were obtained in the second set of snRNA-seq data from the CeA (Supplementary Fig. 10D–F) [49]; Drd2 and InsR were co-expressed in neuronal cell populations in both the CeL (seq-c8 and seq-c6, equivalent to inhibitory neuron 1) and CeC (seq-c6 and seq-c5, equivalent to inhibitory neuron 7) [49, 50]. We also performed the same analysis to compare the co-expression of several feeding-related genes with Drd2 or InsR (Supplementary Fig. 10G). We found significant colocalization of Drd2 and InsR together with multiple genes related to energy homeostasis and feeding behaviors, such as cholecystokinin B receptor, gastric inhibitory polypeptide receptor, and neuropeptide Y receptor type 1 (Supplementary Fig. 10G). These findings provide strong neuroanatomical support for a functional interaction between D2Rs and InsRs in specific subtypes of CeA inhibitory neurons, further reinforcing their role in the regulation of appetitive behavior.

Könner et al. [44] report that selective deletion of InsRs in midbrain dopaminergic neurons increases body weight, fat mass, and hyperphagia in mice. Interestingly, in these mice, D2R expression in ventral tegmental area (VTA) dopaminergic neurons is lower than in control mice [44]. The precise reason for this decreased D2R expression is not clear, but we suspect that it is related to the colocalization of InsRs and D2Rs in the VTA based on our present findings. Indeed, when we analyzed the colocalization of D2Rs and InsRs in the VTA, we found that ~82% of VTA D2R- expressing neurons co-expressed InsRs (Supplementary Fig. 11). Therefore, the colocalization and interaction between D2Rs and InsRs do not appear to be a coincidental, but rather may reflect a broader homeostatic mechanism that integrates reward and metabolic regulation. In this context, we propose that D2R-InsR crosstalk plays a key role in modulating PF consumption (Fig. 6I, Supplementary Fig. 8), and that disruption of this interaction may contribute to excessive PF intake and associated metabolic dysregulation (Fig. 6I).

To assess the behavioral outcomes of D2R–InsR interaction, we employed two validated paradigms widely used to model compulsive-like eating: the light/dark conflict test and footshock-punished PF self-administration. These tasks assess reward-seeking persistence despite aversive stimuli, a core feature of compulsivity [22, 51,52,53]. Both global and CeA-specific D2R loss promoted aversion-resistant PF seeking, indicating impaired behavioral suppression of reward-driven behavior. While Song et al. [54] and other studies employing extinction–reinstatement paradigms have provided important insights into compulsive drug-seeking behavior [54, 55], our study focused on the persistence of reward-seeking despite aversive consequences, a complementary and equally well-validated dimension of compulsive-like behavior. This phenotype was strongly associated with disrupted D2R–InsR signaling in the CeA, highlighting its critical role in regulating PF consumption under adverse conditions.

Together, these findings establish the first demonstration of D2R-induced transactivation of InsRs in the brain, offering a mechanistic framework for how widely distributed brain InsRs may integrate with neuromodulatory systems. This D2R–InsR crosstalk not only fine-tunes CeA neuronal activity and feeding behavior but may also provide insight into why insulin resistance is frequently observed in dopaminergic disorders such as Parkinson’s disease and schizophrenia. Disruption of this interaction may contribute to both metabolic and neuropsychiatric dysfunction, with potential implications for therapeutic strategies targeting comorbid metabolic and behavioral symptoms [56,57,58,–59].