Anorectic and aversive effects of GLP-1 receptor agonism are mediated by brainstem cholecystokinin neurons, and modulated by GIP receptor activation

Abstract

Objective: Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are effective medications to reduce appetite and body weight. These actions are centrally mediated; however, the neuronal substrates involved are poorly understood.

Methods: We employed a combination of neuroanatomical, genetic, and behavioral approaches in the mouse to investigate the involvement of caudal brainstem cholecystokinin-expressing neurons in the effect of the GLP-1RA exendin-4. We further confirmed key neuroanatomical findings in the non-human primate brain.

Results: We found that cholecystokinin-expressing neurons in the caudal brainstem are required for the anorectic and body weight-lowering effects of GLP-1RAs and for the induction of GLP-1RA-induced conditioned taste avoidance. We further show that, while cholecystokinin-expressing neurons are not a direct target for glucose-dependent insulinotropic peptide (GIP), GIP receptor activation results in a reduced recruitment of these GLP-1RA-responsive neurons and a selective reduction of conditioned taste avoidance.

Conclusions: In addition to disclosing a neuronal population required for the full appetite- and body weight-lowering effect of GLP-1RAs, our data also provide a novel framework for understanding and ameliorating GLP-1RA-induced nausea - a major factor for withdrawal from treatment.

Keywords: Appetite; Area postrema; Brain; Cholecystokinin; Glucagon-like peptide-1; Glucose-dependent insulinotropic polypeptide; Nausea; Nucleus of the solitary tract.

Figure 1

CCK^AP/NTSneurons are responsive to GLP-1R agonist Exendin-4. (A)Exendin-4 (EX-4) suppresses food intake (n = 8–9; Treatment: F_{(3, 31)} = 21.81, p < 0.0001. Post hoc ∗p = 0.0006, ∗∗∗p < 0.0001), (B) Representative images, and (C) quantification of FOS expression in CCK^AP/NTS neurons, identified as eYFP-expressing cells in Cck^Cre::eYFP mice, following saline or EX-4 administration (n = 3–4; t (5) = 11.57, p < 0.0001). (D) Distribution of activated CCK^AP/NTS neurons between AP and NTS and in relation to total FOS-expressing cells. (E)Representative fluorescence in situ hybridization (FISH) labeling of endogenous Cck and Glp1r mRNAs in the AP and the NTS of the mouse (n = 3), (F) Representative trace, and (G) quantification of changes in membrane potential recorded upon bath application of EX-4 (1 μM; t (7) = 2.692, p = 0.0310). (H) Representative trace recorded upon application of EX-4 in the presence of tetrodotoxin (TTX; 1 μM). (I) Representative FISH labeling ofCck and Glp1r mRNAs in the NTS of the of the Cynomolgus monkey (n = 2). Data are presented as mean ± SEM. AP, area postrema; NTS, nucleus of the solitary tract; cc, central canal; DMX, dorsal motor nucleus of the vagus; and IHC, immunohistochemistry.

Figure 2

CCK^AP/NTSneurons are required for the anorectic and body-weight lowering effects of GLP-1R agonists. (A) Schematic of the strategy to inhibit CCK^AP/NTS neurons using Cre-dependent AAV expressing an eGFP-fused tetanus-toxin-light-chain (TeLC) and (B) TeLC-eGFP expression in CCK^AP/NTS neurons.(C) Food intake in control (CCK^AP/NTS-eGFP; n = 7) and CCK^AP/NTS-TeLC-eGFP (n = 8) mice injected with saline (F_(1,13) = 1.55, p = 0.2350) and (D) EX-4 (10 μg kg⁻¹, IP; F_(1,13) = 19.02, p = 0.0008).(E) EX-4 (20 μg kg⁻¹, IP, twice daily) suppresses body weight in control CCK^AP/NTS-eGFP (n = 7), but not in CCK^AP/NTS-TeLC-eGFP (n = 8) mice (Treatment: F_(1,13) = 56.23, p < 0.0001; Time: F_(7,91) = 22.21, p < 0.0001; Interaction: F_(7,91) = 7.48, p < 0.0001; two-way ANOVA. Delta BW: t (13) = 2.745,p = 0.0167). Data are presented as mean ± SEM. See also, Supplemental Fig. 1.

Figure 3

Exendin-4 induces behavioral avoidance via CCK^AP/NTSneurons. (A) Schematic of the strategy to identify CCK^AP/NTS neurons projecting (→) to the parabrachial nucleus (PBN) or paraventricular nucleus of the hypothalamus (PVH) using Fluoro-Gold (FG) retrograde tracing. (B) Representative images showing CCK^AP/NTS→PBN neurons activated by EX-4. (C) Quantification of CCK^AP/NTS→PBN and CCK^AP/NTS→PVH neurons activated by EX-4 (n = 3). (D) Schematic of the optogenetic strategy to stimulate CCK^AP/NTS→PBN.(E)Stimulation of CCK^AP/NTS→PBN reduces food intake in fasted mice (n = 5–6; t (9) = 2.936, p = 0.0166) and (F) induces place avoidance in a real-time place preference assay (t (8) = 4.485, p = 0.0020), but (G) it does not induce conditioned taste avoidance (CTA).(H) Schematic of the strategy to silence CCK^AP/NTS using TeLC. (I) EX-4 (30 μg kg⁻¹, IP) induces CTA in wild-type mice (t (14) = 3.144, p = 0.0072). (J) EX-4 induces CTA in control CCK^AP/NTS-eGFP (n = 6), but not in CCK^AP/NTS-TeLC-eGFP (n = 7) mice (t (11) = 3.087, p = 0.0103). (K) EX-4 induces body weight loss in control CCK^AP/NTS-eGFP (n = 6), but not in CCK^AP/NTS-TeLC-eGFP (n = 7) mice (24 h: t (11) = 2.603, ∗p = 0.0246; 48 h:t (11) = 1.162, p = 0.270). Data are presented as mean ± SEM. See also, Supplemental Fig. 2.

Figure 4

GIPR agonism reduces the recruitment of Glp1r/CCK^AP/NTSneurons and the conditioned taste avoidance elicited by Exendin-4. (A) Quantification of CCK^AP/NTS neurons expressing FOS following an injection of EX-4, CCK, or [D-Ala2]-GIP (10, 10, and 100 μg kg⁻¹, IP). (B) FISH labeling of endogenous Gipr and Vgat mRNA in the AP of the mouse. (C) FISH labeling of endogenous Glp1r, Gipr, and Vgat in the AP of the Cynomolgus monkey. (D) Representative FISH labeling and quantification of endogenous Gipr, Glp1r, and Fos mRNAs in the mouse AP following EX-4 (30 kg⁻¹, IP) and [D-Ala2]-GIP (100 μg kg⁻¹, IP) alone or in combination (n = 5–6; F_{(3, 19)} = 222.9, p < 0.0001, one-way ANOVA. Post hoc ∗∗∗p < 0.0001). (E) Quantification of CCK^AP/NTS neurons expressing FOS following EX-4 and [D-Ala2]-GIP alone, or in combination (n = 3–5; AP: F_{(3, 10)} = 30.73, p < 0.0001, one-way ANOVA; NTS: F_{(3, 10)} = 29.17, p < 0.0001. Post hoc ∗p < 0.001).(F) Effect of EX-4 and [D-Ala2]-GIP alone or in combination on food intake (n = 8–9; Treatment: F_{(3, 31)} = 21.91, p < 0.0001; Time: F_{(1.853, 57.43)} = 1214, p < 0.0001; Interaction: F_{(9, 93)} = 7.445, p < 0.0001, two-way ANOVA. Post hoc ∗∗p < 0.005) and (G) CTA (n = 9–14; Conditioning: F_{(3, 43)} = 7.176, p = 0.0005, one-way ANOVA. Post hoc ∗∗p < 0.005. Test: F_{(3, 43)} = 6.026, p = 0.0016, one-way ANOVA. Post hoc ∗∗p = 0.0021, ^&p = 0.0321). Data are presented as mean ± SEM. See also, Supplemental Fig. 3.