Compositions and methods for the identification of therapeutics that improve satiety without aversion

Abstract

Compositions and methods for identifying compounds that improve satiety without aversion are provided herein. In certain embodiments, the method includes contacting a test compound and a postrema neuron that expresses glucagon-like peptide-1 receptor (APGLP1R neuron) and a nucleus tractus solitarius neuron that expresses glucagon-like peptide 1 receptor (NTSGLP1R neuron).

Description

[0001] COMPOSITIONS AND METHODS FOR THE IDENTIFICATION OF THERAPEUTICS THAT IMPROVE SATIETY WITHOUT AVERSION

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This application claims the benefit under 35 USC 119(e) of US Provisional Patent Application No. 63 / 554,097, filed February 15, 2024, which is incorporated herein by reference.

[0004] GRANT SUPPORT STATEMENT

[0005] This invention was made with government support under R00DK119574 and DP2AT011965 awarded by the National Institutes of Health. The government has certain rights in the invention.

[0006] BACKGROUND OF THE INVENTION

[0007] The established relationship between satiety and nausea is not fully understood. While nausea can occur independently of physiological satiety, the feeling of nausea is virtually always accompanied by loss of appetite. Despite its relevance to the lives of millions of people who experience nausea from drug treatment, disease, pregnancy, or other conditions, whether nausea and satiety arise from similar or different neural circuits remains unclear.

[0008] Addressing this question has become even more important as the effort to develop weight loss therapeutics intensifies. With 2.6 billion people worldwide classified as overweight or obese, the lack of effective weight loss strategies has become a pressing public health issue. One of the barriers to drug treatments for obesity is the incidence of aversive side effects, with nausea and vomiting at the top of this list. These undesirable effects of weight-loss therapeutics limit treatment adherence and increase attrition, compromising treatment efficacy.

[0009] To date, the most effective class of weight loss drugs is the long-acting glucagon- like peptide-1 receptor (GLP1R) agonists. Multiple generations of these therapies are clinically available, including exenatide (the synthetic exendin-4, Byetta®), liraglutide (Saxenda®), and the most effective and popular GLP1 -based drug, semaglutide (Ozempic® / Wegovy®), which results in an impressive -16% reduction in body weight in clinical trials. GLP1R agonists are also critical components of the newer dual- and triagonist strategies to treat obesity. However, nausea is by far the most common adverse event for each of these pharmacotherapies, with up to -60% of participants reporting this side effect.

[0010] The development of a drug that inhibits food intake without inducing nausea, or other aversive effects, would revolutionize our ability to treat obesity and associated diseases. However, this depends on the existence of dissociable GLPIR-mediated mechanisms for satiety and nausea / aversion, which remain unknown. Accordingly, there is a clearly an urgent need for screening assays to identify pharmacotherapies that inhibit food intake without inducing nausea.

[0011] SUMMARY OF THE INVENTION

[0012] The present invention comprises methods of identifying therapeutic compounds that do not cause an aversive response, the method comprises (a) contacting a test compound and a postrema neuron that expresses glucagon-like peptide- 1 receptor (ApGLpiRneuron); (b) contacting the test compound with a nucleus tractus solitarius neuron that expresses glucagon-like peptide 1 receptor (NTSGLP1Rneuron); (c) measuring a response of the APGLP1Rneuron to the test compound by comparing activity of the APGLpiRneuron in the presence and the absence of the test compound; and (d) measuring a response of the NTSGLP1Rneuron to the test compound by comparing activity of the NTSGLP1Rneuron in the presence and the absence of the test compound. In certain embodiments, the test compound does not cause an aversive response when the activity of the APGLP1Rneuron is not increased and the activity of the NTSGLP1Rneuron is not decreased in the presence of the test compound or is decreased by less than 50%. In certain embodiments, the methods further comprise (e) measuring the responses of the APGLP1Rneuron and the NTSGLP1Rneuron to a known aversive therapeutic compound; and (f) identifying test compounds that do not cause aversive responses based on the measured responses to the test compound and known aversive therapeutic compound. In certain embodiments, the known aversive therapeutic compound is a glucagon-like 1 protein receptor (GLP1R) agonist. In certain embodiments, the activity of the APGLP1Rneuron is decreased. In certain embodiments, the activity of the NTSGLP1Rneuron is the same or is increased. In certain embodiments, the aversive response is nausea or vomiting.

[0013] In another aspect of the invention, methods for screening therapeutic agents for treating GLP1R agonist related aversive response in a patient in need thereof, the methods comprising (a) expressing an area postrema neuron (AP) gene in a cell; (b) contacting the cell with a test compound; (c) determining the effect of the test compound on the expression level of the AP gene in comparison to that in the absence of the test compound; and (d) selecting the agents that can reduce or inhibit the expression levels of the genes as the therapeutic agent, are provided. In certain embodiments, the methods further comprise (e) expressing a dorsal vagal complex (DVC) gene in a cell; (f) contacting the cell with the test compound; (g) determining the effect of the test compound on the expression level of the DVC gene in comparison to that in the absence of the test compound; and (h) selecting the test compounds that do not significantly reduce the expression levels of the gene as the therapeutic agent. In certain embodiments, the DVC gene is a gene that expresses at the same level, or a greater level, in untreated NTS neurons when compared to AP neurons. In certain embodiments, the cell expressing the AP gene is an APGLP1Rneuron. In certain embodiments, the cell expressing the DVC gene is an NTSGLP1Rneuron.

[0014] In certain embodiments, the AP gene is Erbb4, Kcnabl, Unc5d, Csmd3, Grm7, Tenm3, Cdkl4, Robo2, Mgat4c, Kctd8, Slc24a3, Ntrk3, Luzp2, Ncam2, Rora, Slc2al3, Neill, Grin2b, Meisl, Spockl, Rmst, Lhfpl3, Epha5, Dpyd, Ccdc85a, Sgcz, Sema6d, Dabl, Gabra2, Pcdhl5, Cdhl2, Cntn4, Cntn6, Hs3st5, 01fm3, Nell2, Sorcs3, Nkain2, Grin2a, Grm8, Mpped2, Aff2, Sorcsl, Pcsk2, Asic2, Sytl, Brinp3, Kirrel3 Gabrbl, Rbfoxl, Slc8alTrpc5, Thsd7b, Ammercrl, Grik4, Zfpm2, Cdh4, Sema3e, Raplgap, Chnl, or Tmtc2. In certain embodiments, the NTS gene is Fstl4, Ephbl, Col25al, Synpr, Cadps2, Gria3, Ndst3, Klhll, Mania, Hs3st2, Pcdhl lx, Ptchd4, Ldb2, Glra3, Adcy8, Teadl, Rab27b, Fat3, Aik, Htr2c, Esrrg, Pdelc, Dgkg, Mei4, Cdh8, Hs3st4, Slcl7a6, Kcnhl, Nrpl, Esrl, Ak5, Fut9, Btbdl INckapS, Calnl, Rbfox3, Cntnap4, Limal, Mrap2, or Cdh7.

[0015] In another aspect of the invention, therapeutic compounds identified by the methods described herein are provided. Also provided herein are method of treating an obesity-related aversive response in a patient in need thereof, the method comprising administering to the patient the therapeutic compound. In certain embodiments, the patient does not have an aversive response to said therapeutic compound. In certain embodiments, the patient has fewer aversive responses when compared to a patient treated with the obesity therapeutic alone.

[0016] Still other aspects and advantages of these compositions and methods for making the compositions and using the compositions are described further in the following detailed description of the preferred embodiments thereof.

[0017] BRIEF DESCRIPTION OF THE FIGURES

[0018] FIGs. 1 A-1H: Hindbrain GLP1R neurons mediate the anorexic effects of GLP1- based obesity drugs and potently suppress food intake. (FIG. 1 A) The hindbrain dorsal vagal complex (DVC), hypothalamic arcuate nucleus (ARC) and nodose ganglion (NG) express GLP1R (in situ hybridization). Scale bar, 50 pm. (FIG. IB) Schematic for ablation of GLPIR-expressing neurons in each of these regions. (FIG. 1C) Food intake (4 h, food presented immediately after injection) following IP injection of saline or exendin-4 (Ex-4: 5, 10, or 20 pg / kg) in control (EGFP) or GLP1R neuron-ablated mice (n=8-14 / group, oneway ANOVAs, all ps<0.01 except for DVC where p=ns). (FIG. ID) Food intake (4 h, food presented 4 h post-injection) following SQ injection of vehicle (0.25% DMSO in saline) or semaglutide (10 pg / kg) in control (EGFP) or GLP1R neuron-ablated mice (n=6-7 / group, paired t-tests, all ps<0.05 except for DVC where p=ns). (FIG. IE) Left, percent body weight change over three weeks in diet-induced obese, high-fat diet-fed control (EGFP) or GLPIR-ablated neuron-ablated mice receiving every other day (week 1) followed by daily (weeks 2-3) SQ 40 pg / kg semaglutide treatment (n=7-l 1 / group, two-way repeated measures ANOVA, group x time interaction: p<0.001). Right, percent body weight change at 1, 2, and 3 weeks in all groups (n=7-l 1 / group, one-way ANOVAs, letters denote groups that are significantly different from each other). (FIG. IF) Food intake over 24 h (left, n=7-8 / group, two-way repeated measures ANOVA, control vs. hM3Dq: p<0.001) and in the dark period (right, n=7-8 / group, unpaired t-test, control vs. hM3Dq: p<0.001) in fasted mice with chemogenetic (hM3Dq) or control (EGFP) activation of DVCGLP1Rneurons (n=7-8 / group). (FIG. 1G) Body weight change across 40 days (left, n=9-l 1 / group, two- way repeated measures ANOVA, control vs. NaChBac: p<0.001) and cumulative body weight change (right, n=9-l 1 / group, unpaired t-test, control vs. NaChBac: p<0.001) in chow-fed mice with chronic (NaChBac) or control (EGFP) activation of DVCGLP1Rneurons (n=9-l 1 / group). (FIG. 1H) Body weight change across 40 days (left, n=9- 11 / group, two-way repeated measures ANOVA, control vs. NaChBac: p<0.001) and total lean and fat mass (right, n=8 / group, unpaired t-test, control vs. NaChBac: lean mass p=ns, fat mass p<0.001) in high-fat diet-fed mice with chronic (NaChBac) or control (EGFP) activation of DVCGLP1Rneurons (n=8 / group). Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001 for t-tests and post hoc comparisons, ***p<0.001 for two-way repeated measures ANOVA group x time interaction. Letters denote groups that are significantly different from each other.

[0019] FIG. 2A-2H: DVCGLP1Rneurons are required for the food intake suppression and weight loss effects of GLP -based obesity drugs. (FIG. 2A) Representative image and quantification of viral expression (EGFP) in GLPIR-expessing neurons of Glplr-ires-Cre mice (RNA in situ hybridization). (FIG. 2B) Representative images of GLP1R expression (RNA in situ hybridization) in the DVC, ARC, and NG of control mice or mice with GLP1R neuron ablation in each region. (FIG. 2C) Quantification of Caspase3- or DTA- mediated neural ablation: relative number (to controls) of GLP1R+ neurons in DVC (left), ARC (middle), and NG (right) of mice with viral injection into DVC, ARC, or NG, respectively (n=3-5 / group, unpaired t-tests, all ps<0.01). (FIG. 2D) Food intake from 0-8 h following IP injection of saline or exendin-4 (5, 10, or 20 pg / kg) in fasted control or GLPIR-ablated mice (n=8-14 / group, two-way repeated measures ANOVAs, all ps<0.01 except DVC where p=ns). (FIG. 2E) Percent body weight change during the onset of diet- induced obesity in high-fat diet-fed control (EGFP) or DVCGLP1Rneuron-ablated mice receiving biweekly SQ 12 pg / kg semaglutide (n=l 1-15 / group, two-way repeated measures ANOVA, control vs. DVC-Casp3 p<0.001). (FIG. 2F-2H) Fat mass (f, n=l 1-15 / group, two-way ANOVA, Control vs. DVC-Casp at 8 and 12 weeks: p<0.001), lean mass (g, n=l 1-15 / group, two-way ANOVA, all ps=ns) and food intake (h, n=l 1-13 / group, unpaired t-test, p<0.05) in high-fat diet-fed control or DVC GLPIR-ablated mice over 12 weeks of biweekly SQ semaglutide (12 pg / kg). Data are presented as mean ± S.E.M. **p<0.01, ***p<0.001 drug x time interaction; *p<0.05, **p<0.01, ***p<0.001 post hoc Bonferroni comparison.

[0020] FIG. 3 A-3H: DVCGLP1Rneuron activation suppresses food intake and energy expenditure. (FIG. 3 A) Schematic for continuous food intake and energy expenditure measurements using mouse metabolic chambers. (FIG. 3B) Cumulative food intake over 48 h in high-fat diet-fed control mice or mice with chronic NaChBac-mediated DVCGLP1R neuron activation (n=7-8 / group, two-way repeated measures ANOVA, control vs. NaChBac: p<0.001). Shaded areas represent dark periods. (FIG. 3C) Average inter-meal interval of high-fat diet-fed control mice or mice with chronic NaChBac-mediated DVCGLP1Rneuron activation (n=7-8 / group, unpaired t-test, control vs. NaChBac: p<0.01). (FIG. 3D) Average meal size in high-fat diet-fed control mice or mice with chronic NaChBac-mediated DVCGLP1Rneuron activation (n=7-8 / group, unpaired t-test, control vs. NaChBac: p=ns). (FIG. 3E) Energy expenditure in high-fat diet-fed control mice or mice with chronic NaChBac-mediated DVCGLP1Rneuron activation (n=7-8 / group). Shaded area represents dark period. (FIG. 3F) Dark period energy expenditure normalized to lean mass by ANCOVA in high-fat diet-fed control mice or mice with chronic NaChBac-mediated DVCGLP1Rneuron activation (n=7-8 / group, ANCOVA, p=0.06). (FIG. 3G) Energy expenditure in fasted chow-fed control mice or mice with chemogenetic DVCGLP1Rneuron activation (n=8 / group). Shaded area represents dark period. (FIG. 3H) Dark period energy expenditure normalized to body mass by ANCOVA in chow-fed control mice or mice with chemogenetic DVCGLP1Rneuron activation (n=8 / group, ANCOVA, p<0.001). Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0021] FIG. 4A-4P: In vivo activity dynamics of APGLP1Rand NTSGLP1Rneurons in response to semaglutide and nutritive and aversive stimuli. (FIG. 4 A) Schematic of AP (darker circles) and NTS (lighter circles) subregions of the DVC at bregma -7.48. (FIG.4B) 3D reconstruction of DVCGLP1Rpopulations in the AP, NTS, and cuneate nucleus (Cu) after Glplr-ires-Cre mice were injected with AAV5-EFla-DIO-EGFLP-10a, which expresses a soma-restricted fluorophore. Arrows indicate directionality: R, rostral; D, dorsal; C, caudal; V, ventral. (FIG. 4C) Schematic of setup for simultaneous in vivo two-photon imaging of APGLP1Rand NTSGLP1Rneurons in mice that are implanted with intravenous (IV) and intraduodenal (ID) catheters as well as a duodenal exit port. For experiments, mice received a single administration of each stimulus while calcium activity was recorded. (FIG. 4D) Representative two-photon maximum projection image of GCaMP6s expression in DVCGLP1Rneurons. (FIG. 4E) Heat maps depicting z-scores of responses of APGLP1Rneurons to vehicle (0.25% DMSO in saline, IV) or semaglutide (60 pg / kg, IV) (n=787 neurons, 5 mice). Dashed white lines indicate start of stimulus. (FIG. 4F) Heat maps depicting z-scores of responses of NTSGLP1Rneurons to vehicle or semaglutide (n=614 neurons, 5 mice). Dashed white lines indicate start of stimulus. (FIG. 4G) Average activity responses (z-scores) in APGLP1Rneurons (left bar) and NTSGLP1Rneurons (right bar) to vehicle or semaglutide administration (n=5, two-way ANOVA, AP: vehicle vs semaglutide: p<0.001, NTS: vehicle vs. semaglutide p<0.01). (FIG. 4H) Individual data points comparing APGLP1Rand NTSGLP1Rneuron activity responses (average z-scores) to vehicle and semaglutide. Dotted lines represent threshold for statistically significant neural activation (z=1.64, see Methods for additional details). (FIG. 41) Proportions of APGLP1Rneurons (left) and NTSGLP1Rneurons (right) responsive to Ensure (2 / 3 kcal / ml, 300 pl, ID), cinacalcet (0.5 mg / kg, IV), or both (n=7, two-way ANOVA, all ps<0.01). (FIG. 4J) Average activity responses (z-scores) in APGLP1Rneurons (left bar) and NTSGLP1Rneurons (right bar) to Ensure or cinacalcet administration (n=7, two-way ANOVA, all ps<0.05). (FIG. 4K) Individual data points comparing APGLP1Rand NTSGLP1Rneuron activity responses (average z-scores) to Ensure and cinacalcet. Dotted lines represent threshold for statistically significant neural activation (z=1.64, see Methods for additional details). (FIG. 4L) Representative two-photon images of GCaMP6s with neurons color-coded based on average responsivity to nutrients (Ensure) or the aversive stimulus cinacalcet, with the most nutrient-responsive neurons (index -10) and the most aversion-responsive neurons (index 10). Boundary between AP and NTS indicated by dashed line. (FIG. 4M) Proportions of APGLP1Rneurons (left) and NTSGLP1Rneurons (right) responsive to Ensure (2 / 3 kcal / ml, 300 pl, ID), LiCl (150 mg / kg, IV), or both (n=6, two-way ANOVA, all ps<0.001 except AP: Ensure vs. LiCl: p=ns and NTS LiCl vs both: p=ns). (FIG. 4N) Average activity responses (z-scores) in APGLP1Rneurons (left bar) and NTSGLP1Rneurons (right bar) to Ensure or LiCl administration (n=6, two-way ANOVA, all ps<0.01 except Ensure: AP vs. NTS p=ns). (FIG. 40) Individual data points comparing APGLpiRand NTSGLP1Rneuron activity responses (average z-scores) to Ensure and LiCl. Dotted lines represent threshold for statistically significant neural activation (z=1.64, see Methods for additional details). (FIG. 4P) Representative two-photon images of GCaMP6s with neurons color-coded based on average responsivity to nutrients (Ensure) or the aversive stimulus LiCl, with the most nutrient-responsive neurons (index -10) and the most aversion-responsive neurons (index 10). Boundary between AP and NTS indicated by dashed line. Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001. FIG. 5A-5B: (FIG. 5A) Images of soma-restricted AAV5-EFla-DIO-EGFPL10a in Glplr-ires-Cre mice across the rostral-caudal axis. (FIG. 5B) Quantification of neurons in AP vs. NTS across the rostral-caudal axis (n=3).

[0022] FIG. 6A-6C: Simultaneous in vivo imaging of APGLP1Rand NTSGLP1Rneurons. (FIG. 6A) Coronal images of soma-restricted Glplr-ires-Cre;Ai9(tdTomato) mice injected with AAV5-EFla-DIO-EGFPL10a in the DVC. Images depict the dense GLP1R projection into the NTS but not into the AP. (FIG. 6B) Maximum projection two-photon images of the DVC from Glp Ir-ires-Cre ;Ai9(tdTomato) mice injected with Cre-dependent GCaMP6s depicting the halo of GLP1R projections into the NTS, part of the strategy to draw boundaries between AP and NTS (see Methods for additional details). (FIG. 6C) Two-photon images of the DVC from Glp Ir-ires-Cre ;Ai9(tdTomato) mice injected with Cre-dependent GCaMP6s across the z-axis. AP and NTS neurons are visible on all imaging planes due to the angle of the medulla for imaging (see Methods for additional details).

[0023] FIG. 7A-7I: In vivo responses of APGLP1Rand NTSGLP1Rneurons. (FIG. 7A) Top, simultaneous in vivo two-photon imaging of APGLP1Rand NTSGLP1Rneurons in mice. Bottom, images of surgical prep and cranial window for in vivo two-photon imaging of calcium dynamics in APGLP1Rand NTSGLP1Rneurons. (FIG. 7B) Proportion of APGLP1Rand NTSGLP1Rneurons activated by semaglutide (n=5, paired t-test, p=ns). (FIG. 7C) Heat maps depicting z-scores of responses of APGLP1Rneurons to vehicle (saline, ID) or Ensure (2 / 3 kCal per ml, 300 pl, ID) (n=375 neurons, 5 mice). Dashed lines indicate start of stimulus. (FIG. 7D) Heat maps depicting z-scores of responses of NTSGLP1Rneurons to vehicle (saline, ID) or Ensure (2 / 3 kCal per ml, 300 pl, ID) (n=266 neurons, 5 mice). Dashed white lines indicate start of stimulus. (FIG. 7E) Average activity responses (z- scores) in APGLP1Rneurons and NTSGLP1Rneurons to saline or Ensure administration (n=6, two-way ANOVA, AP: saline vs Ensure: p<0.01, NTS: saline vs. Ensure p<0.001). (FIG. 7F) Individual data points comparing APGLP1Rand NTSGLP1Rneuron activity responses (average z-scores) to vehicle and semaglutide. Dotted lines represent threshold for statistically significant neural activation (z=1.64, see Methods for additional details). (FIG. 7G) Individual heat maps for each mouse (n=7) depicting z-scores of responses of ApGLpiRan(jNTSGLpiR neurons to Ensure or cinacalcet. Dashed lines indicate start of stimulus. (FIG. 7H) Representative two-photon images of GCaMP6s across the z axis with neurons color-coded based on average responsivity to nutrients (Ensure) or the aversive stimulus cinacalcet, with the most nutrient-responsive neurons (index -10) and the most aversion-responsive neurons (index 10). Boundary between AP and NTS indicated by dashed line. (FIG. 71) Individual heat maps for each mouse (n=6) depicting z-scores of responses of APGLP1Rand NTSGLP1Rneurons to Ensure or LiCl. Dashed lines indicate start of stimulus.

[0024] FIG. 8A-8F: Distribution of and gene expression in APGLP1Rand NTSGLP1Rneurons. (FIG. 8A) Schematic and representative images of viral (AAV2.2-FLEX- tdTomato) separation of AP (FIG. 8A) and NTS (FIG. 8B) using 20 nl intracranial injections indicate that APGLP1Rneurons project to NTS but NTSGLP1Rneurons only project to the border of the AP. (FIG. 8C) Quantification of tdTomato-expressing GLP1R neurons in AP and NTS after viral injection into AP. (FIG. 8D) Quantification of tdTomato- expressing GLP1R neurons in AP and NTS after viral injection into NTS. (FIG. 8E) Uniform manifold approximation and projection (UMAP) plot showing AP and NTS neuron subtypes. Analysis was made after combining data sets from Zhang et al., (2021)14, Ludwig et al (2021)17, and Adriaenssens et al (2023)18, which all contained cells from both the AP and NTS. (FIG. 8F) Dot plots indicating normalized expression of genes in Glplr+ cells of the AP and NTS.

[0025] FIG. 9A-9L: APGLP1Rand NTSGLP1Rneurons independently drive nausea-like behaviors and satiety, respectively. (FIG. 9A) Schematic for recording orofacial taste reactivity responses to flavor stimuli delivered via intraoral cannula that were paired with chemogenetic neural activation, and example hedonic and aversive taste reactivity responses. See Methods and FIG. 10 for detailed description of taste reactivity behaviors. (FIG. 9B) Composite score of hedonic (left, n=6-7 / group, two-way ANOVA, control baseline vs. conditioned: p=ns, hM3Dq baseline vs, conditioned: p<0.001) and aversive (right, n=6-7 / group, two-way ANOVA, control baseline vs. conditioned: p=ns, hM3Dq baseline vs, conditioned: p<0.001) taste reactivity responses to a flavor paired with chemogenetic (hM3Dq) or control (EGFP) activation of DVCGLP1Rneurons. (FIG. 9C) Schematic for measuring avoidance of flavors previously paired with chemogenetic neural activation (conditioned flavor avoidance, CFA). (FIG. 9D) Intake of flavor paired with DVCGLP1Rneuron stimulation (CS+) before (baseline) and after conditioning (n=6-

[0026] 9 / group, two-way ANOVA, control baseline vs. conditioned: p=ns, hM3Dq baseline vs, conditioned: p<0.01). (FIG. 9E) Schematic (top) and representative images (bottom) of 20 nl infusions of AAV2.2-hM3Dq in the AP or the NTS to restrict viral expression to either subregion, cc, central canal; DMV, dorsal motor nucleus of the vagus. (FIG. 9F) Food intake over 24 h (left, n=l 1 / group, two-way repeated measures ANOVA, vehicle vs. (f) Food intake over 24 h (left, n=l 1 / group, two-way repeated measures ANOVA, vehicle vs. p<0.001) in fasted mice with chemogenetic (hM3Dq) activation of APGLP1Rneurons. (FIG. 9G) Food intake over 24 h (left, n=l 1 / group, two-way repeated measures ANOVA, vehicle vs. CNO: p<0.001) and in the dark period (right, n=l 1 / group, unpaired t-test, vehicle vs. CNO: p<0.001) in fasted mice with chemogenetic (hM3Dq) activation of NTSGLP1Rneurons. (FIG. 9H) Relative dark period (12-h) food intake with chemogenetic activation of APGLP1Ror NTSGLP1Rneurons (n=l 1 / group, unpaired t-test, APGLP1Rvs. NTSGLP1Rstimulation: p=ns). Box in boxplot indicates median and 1stto 3rdquartiles, bounded by the min and max (whiskers). (FIG. 91) Inter-meal interval in mice following chemogenetic activation of APGLP1Ror NTSGLP1Rneurons (n=10-l 1 / group, two-way ANOVA, APGLP1Rvehicle vs. CNO: p=ns, NTSGLP1Rvehicle vs. CNO: p<0.01). (FIG. 9J) Composite score of hedonic (left, n=5-8 / group, two-way ANOVA, APGLP1Rbaseline vs. conditioned: p<0.001, NTSGLP1Rbaseline vs. conditioned: p=ns) and aversive (right, n=5- 8 / group, two-way ANOVA, APGLP1Rbaseline vs. conditioned: p<0.001, NTSGLP1Rbaseline vs. conditioned: p=ns) taste reactivity responses to a flavor paired with chemogenetic (hM3Dq) activation of APGLP1Ror NTSGLP1Rneurons. (FIG. 9K) Intake of flavor paired with APGLP1Ror NTSGLP1Rneuron stimulation (CS+) before (baseline) and after conditioning (n=12-13 / group, two-way ANOVA, APGLP1Rbaseline vs. conditioned: p<0.001, NTSGLP1Rbaseline vs. conditioned: p=ns). (FIG. 9L) Body weight change across 42 days in chow-fed mice with chronic (NaChBac) or control (EGFP) activation of NTSGLP1Rneurons (n=4 / group, two-way repeated measures ANOVA, control vs.

[0027] NaChBah: p<0.001). Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0028] FIG. 10A-10N: Verification of real-time and conditioned orofacial taste reactivity taste reactivity in mice. (FIG. 10 A, 10B) Schematics of stereotyped hedonic (FIG. 10 A) and aversive (FIG. 10B) orofacial behavioral responses in mice. Modified from schematics in Grill and Norgren (1978)26. (FIG. 10C, 10D) Composite score (FIG. 10C, n=5- 10 / group, one-way ANOVA, all ps<0.05) and individual behaviors (FIG. 10D, n=5- 10 / group, two-way ANOVA, all ps<0.001) for hedonic taste reactivity responses to quinine concentrations (0, 2.4, 4 mM). (FIG. 10E, 10F) Composite score (FIG. 10E, n=5- 10 / group, one-way ANOVA, all ps<0.01) and individual behaviors (FIG. 10F, 5-10 / group, two-way ANOVA, all ps<0.05) for aversive taste reactivity responses to quinine concentrations (0, 2.4, 4 mM). (FIG. 10G, 10H) Composite scores (FIG. 10G, n=6 / group, paired t-tests, ps<0.05) and individual behaviors (FIG. 10H, 6 / group, two-way ANOVA, RMM: p<0.05, G: p<0.001) for real-time hedonic and aversive taste reactivity responses to a flavor paired with cinacalcet (15 pmol / kg, IP). (FIG. 101, 10J) Composite scores (FIG. 101, n=6 / group, paired t-tests, aversive: p<0.05) and individual behaviors (FIG. 10J, 6 / group, two-way ANOVA, G: p<0.001) for conditioned hedonic and aversive taste reactivity responses to a flavor paired with cinacalcet (15 pmol / kg, IP). (FIG. 10K, 10L) Composite scores (FIG. 10K, n=5 / group, paired t-tests, ps<0.01) and individual behaviors (FIG. 10L, 6 / group, two-way ANOVA, RMM: p<0.001, PL: p<0.05, G: p<0.001, CR: p<0.001) for real-time hedonic and aversive taste reactivity responses to a flavor paired with LiCl (6 mmol / kg, IP). (FIG. 10M, ION) Composite scores (FIG. 10M, n=5 / group, paired t-tests, ps<0.05) and individual behaviors (FIG. ION, 6 / group, two-way ANOVA, RMM: p<0.05, G: p<0.001, CR: p<0.05) for conditioned hedonic and aversive taste reactivity responses to a flavor paired with LiCl (6 mmol / kg, IP). Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0029] FIG. 11 A-l IE: Individual taste reactivity behaviors with chemogenetic activation of DVCGLP1Rneurons. (FIG. 11 A) Schematic for testing the effects of DVCGLP1Rneuron chemogenetic activation on taste reactivity. (FIG. 1 IB, 11C) Individual hedonic taste reactivity behaviors at baseline and after conditioning in control (FIG. 1 IB, n=6 / group, two-way ANOVA, all ps=ns) and experimental (FIG. 11C, n=7 / group, two-way ANOVA, RMM: p<0.001, LTP: p<0.05) mice. (FIG. 11D, HE) Individual aversive taste reactivity behaviors at baseline and after conditioning in control (FIG. 1 ID, n=6 / group, two-way ANOVA, all ps=ns) and experimental (FIG. 1 IE, n=7 / group, two-way ANOVA, G: p<0.01, CR: p<0.001) mice. Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0030] FIG. 12A-12B: Number of APGLP1R, but not NTSGLP1R, neurons expressing chemogenetic receptors correlates with avoidance behavior. (FIG. 12 A) Representative image of hM3Dq expression in the DVC (AP and NTS). (FIG. 12B) Correlation between the number of hM3Dq-expressing neurons (x-axis) and behavior [y-axis; avoidance (CS+ intake in CFA assay) or anorexia (food intake) elicited by DVCGLP1Rneuron stimulation] in the AP (left graphs) or NTS (right graphs) (n=9, Pearson correlation, p<0.05 for AP / avoidance, other ps=ns). Light grey color indicates statistical significance.

[0031] FIG. 13A-13K: Behavioral effects of APGLP1Rand NTSGLP1Rneuron chemogenetic activation. (FIG. 13 A) Strategy for viral injections to directly compare behavioral results of activation of DVCGLP1R, APGLP1R, and NTSGLP1Rneurons: we used the same viral serotype and injection protocol to activate neurons in each of these regions. (FIG. 13B) 12-h food intake in mice following activation of DVCGLP1R, APGLP1R, and NTSGLP1Rneurons (n=7-l 1 / group, two-way ANOVA, all psO.OOl). (FIG. 13C) Meal size in fasted mice following chemogenetic activation of APGLP1Ror NTSGLP1Rneurons (n=10-l 1 / group, two-way ANOVA, APGLPlR-hM3Dq vehicle vs. CNO: p<0.001, NTSGLP1R-hM3Dq vehicle vs. CNO: p=ns). (FIG. 13D, 13E) Energy expenditure in fasted mice with chemogenetic APGLP1R(FIG. 13D) or NTSGLP1R(FIG. 13E) neuron activation (n=l 1 / group). Shaded area represents dark period. (FIG. 13F) Dark period energy expenditure normalized to body mass by ANCOVA in mice with chemogenetic APGLP1Ror NTSGLP1Rneuron activation (n=l 1 / group, ANCOVA, p=ns). (FIG. 13G) Schematic describing protocol for taste reactivity experiments. (FIG. 13H, 131) Individual hedonic taste reactivity behaviors at baseline and after conditioning with chemogenetic APGLP1R(FIG. 13H, n=8, two-way ANOVA, RMM baseline vs. conditioned p<0.001) or NTSGLP1R(FIG. 131, n=5, two-way ANOVA, all ps=ns) neuron activation. (FIG. 13 J, 13K) Individual aversive taste reactivity behaviors at baseline and after conditioning with chemogenetic APGLP1R(FIG. 13 J, n=8, two-way ANOVA, CR baseline vs. conditioned: p<0.001) or NTSGLP1R(FIG. 13K, n=5, two-way ANOVA, all ps=ns) neuron activation. Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0032] FIG. 14A-14F: Obesity drugs reduce food intake even when aversion circuitry (ApGLpiRneurons) is blocked. (FIG. 14A) Mice with chemogenetic inhibition (hM4Di) of APGLP1RorNTSGLP1Rneurons (FIG. 14B), or deletion of GLP1R in AP or NTS (FIG. 14E), underwent a conditioned flavor avoidance assay to semaglutide (120pg / kg). (FIG, 14B) Intake of flavor paired with semaglutide (CS+) with chemogenetic inhibition of PGLpiRor NTSGLP1Rneurons before (baseline) and after conditioning (n=9-10 / group, two-way ANOVA, APGLP1Rbaseline vs. conditioned: p=ns, NTSGLP1Rbaseline vs. conditioned: p<0.05). (FIG. 14C) Food intake (4 h) was measured in response to exendin- 4 or semaglutide following APGLP1Ror NTSGLP1Rneuron inhibition (FIG. 14D) or GLP1R deletion (FIG. 14F). For exendin-4 studies, food was returned immediately after injection. For semaglutide studies, food was returned 4 h post-injection. (FIG. 14D) Food intake in response to exendin-4 (20 pg / kg) in mice with APGLP1Ror NTSGLP1Rneuron inhibition (n=9-10 / group, two-way ANOVA, all ps<0.05). (FIG. 14E) Intake of flavor paired with semaglutide (CS+) in mice with deletion of GLP1R in AP or NTS neurons before (baseline) and after conditioning (n=8 / group, two-way ANOVA, APGLP1Rbaseline vs. conditioned: p=ns, NTSGLP1Rbaseline vs. conditioned: p<0.001). (FIG. 14F) Food intake in response to exendin-4 (left) or semaglutide (right) in mice with AP or NTS GLP1R deletion (n=8 / group, two-way ANOVA, all ps<0.01). Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0033] FIG. 15A-15G: Deletion of GLP1R in DVC blocks anorexia and aversion by obesity drugs. (FIG. 15 A) Schematic for deletion of GLP1R in DVC neurons using Glplr^A mice and injection of AAV2.2-Cre. (FIG. 15B) Quantification of efficacy of GLP1R deletion: relative number of Glplr+ neurons in the DVC in WT and Glplr^A mice after DVC injection of AAV2.2-Cre (n=3 / group, unpaired t-test, p<0.001). (FIG. 15C) Mice underwent a conditioned flavor avoidance assay to semaglutide (120 pg / kg, IP). (FIG. 15D) Intake of flavor paired with semaglutide (CS+) with in control (WT) or DVC GLPIR-deleted (Glplr+1) mice before (baseline) and after conditioning (n=6-8 / group, two-way ANOVA, WT baseline vs. conditioned: p<0.001, Glplr+ baseline vs. conditioned: p=ns). (FIG. 15E) Food intake (4 h) was measured in response to exendin-4 (20 pg / kg, IP) or semaglutide (10 pg / kg, SQ). For exendin-4 studies, food was returned immediately after injection. For semaglutide studies, food was returned 4 h post-injection. (FIG. 15F) Food intake in response to exendin-4 in WT and Glplr+1mice (FIG. 15E, n=6- 7 / group, two-way ANOVA, WT: p<0.01, Glplr+1-. p=ns) (FIG. 15G) Food intake in response to semaglutide in WT and Glplr+ mice (FIG. 15E, n=7-8 / group, two-way ANOVA, WT: p<0.05, Glplr+1-. p=ns) Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0034] FIG. 16A-16K: APGLP1Rand NTSGLP1Rneurons form parallel projections to mediate functionally-dissociable behaviors. (FIG. 16A) Cre-dependent ChR2-EYFP was injected into the DVC of Glplr-ires-Cre mice (top left); representative images of the injection site (top middle); axons in the paraventricular hypothalamic nucleus (PVH, top right) and the lateral parabrachial nucleus (1PBN, bottom) (n=3). Scale bar, 50 pm. DMV, dorsal motor nucleus of the vagus; elPBN, external lateral PBN; clPBN, central lateral PBN; scp, superior cerebellar peduncle; 3 V, 3rdventricle. (FIG. 16B) Cre-dependent H129-DTK-TT was injected into the DVC of Glplr-ires-Cre mice (left); representative images of anterogradely labeled neurons in the 1PBN (middle) and PVH (right) 2 days post-injection (n=3). Scale bar, 50 pm. (FIG. 16C) The 1PBN or PVH was injected with PRV-263 which expresses tdTomato in all retrogradely-transported neurons and EYFP / mCerulean in all Cre-dependent retrogradely-transported neurons of Glplr-ires-Cre mice (left); representative images of NTS (top middle) or AP (bottom middle) EYFP / mCerulean-expressing (i.e., GLPIR-expressing) neurons that project to the PVH and IPBN, respectively. Scale bar, 100 pm. (Right) Quantification of retrogradely labeled EYFP / mCerulean neurons from the PVH and IPBN (n=3 / group). (FIG. 16D) Cre- dependent ChR2-EYFP was injected into the AP of Glplr-ires-Cre mice (left); representative image of axons in the IPBN (right) (n=3). Scale bar, 50 pm. (FIG. 16E) Cre-dependent ChR2-EYFP was injected into the NTS of Glplr-ires-Cre mice (left); representative image of axons in the PVH (right) (n=3). Scale bar, 50 pm. (FIG. 16F) Cre- dependent ChR2-EYFP was injected into the DVC of Glplr-ires-Cre mice and an optic fiber was implanted above the IPBN or PVH (left); representative images of fiber placement above IPBN (middle) and PVH (right). (FIG. 16G) Food intake in fasted mice with (+) and without (-) optogenetic stimulation of DVCGLP1Raxons in the IPBN (left, n=10, paired t-test, no stim vs. stim: p<0.001) or PVH (right, n=7, paired t-test, no stim vs. stim: p<0.01). (FIG. 16H) Schematic for real-time place avoidance assay to measure valence associated with optogenetic stimulation of DVCGLP1Rprojections to the IPBN or PVH. (FIG. 161) Fraction of time spent in the stimulation zone in mice with (+) and without (-) optogenetic stimulation of DVCGLP1Raxons in the IPBN (left, n=10, paired t- test, p<0.01) or the PVH (right, n=7, paired t-test, p=ns). (FIG. 16 J) Representative heat map of time spent in real-time place avoidance apparatus in mice with and without optogenetic axon stimulation. (FIG. 16K) Composite score of hedonic (left, n=6 / group, two-way ANOVA, IPBN no stim vs. stim: p<0.001, PVH no stim vs. stim: p=ns) and aversive (right, n=6 / group, two-way ANOVA, IPBN no stim vs. stim: p<0.001, PVH no stim vs. stim: p=ns) real-time taste reactivity responses to a flavor stimulus paired with optogenetic axon stimulation. Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0035] FIG. 17A-17I: DVCGLP1Rneurons project to the 1PBN and PVH. (FIG. 17A) Glplr-ires-Cre mice were injected in the DVC with AAV5-DIO-ChR2-EYFP (n=3). (FIG. 17B) Image of viral expression at the injection site (DVC: AP and NTS), cc, central canal; DMV, dorsal motor nucleus of the vagus. (FIG. 17C-17G) Representative images of DVCGLP1Raxons in brain regions involved in feeding behavior: 1PBN (FIG. 17C), PVH (FIG. 17D), ARC (FIG. 17E), BNST (FIG. 17F), CeA (FIG. 17G). Boxes indicate zoomed in regions. 3V, third ventricle; ac, anterior commissure; BNST, bed nucleus of the stria terminalis; CeA, central amygdala; LV, lateral ventricle; scp, superior cerebellar peduncle. (FIG. 17H) Cre-dependent H129-DTK-TT was injected into the DVC of Glplr-ires-Cre mice (n=3). (FIG. 171) Representative image of starter cells for anterograde H129-DTK- TT-mediated tracing from DVCGLP1Rneurons.

[0036] FIG. 18A-18D: Retrograde tracing from 1PBN and PVH to DVCGLP1Rneurons. (FIG. 18 A) Schematic of mechanism for fluorophore expression using PRV-263. (FIG. 18B) Glplr-ires-Cre mice (n=3 / group) were injected in the 1PBN or PVH with PRV-263 and DVC brain sections were imaged for retrogradely-transported Cre-positive (GLP1R+, mCerulean / EYFP) and Cre-negative (GLP1R-, tdTomato) neurons in the AP and NTS. (FIG. 18C) Representative images of PRV-263 in Cre-positive (GLP1R+, mCerulean / EYFP) and Cre-negative (GLP1R-, tdTomato) neurons of the AP and NTS after viral injection in the 1PBN. Distances indicated mm from bregma. (FIG. 18D) Representative images of PRV-263 in Cre-positive (GLP1R+, mCerulean / EYFP) and Cre- negative (GLP1R-, tdTomato) neurons of the AP and NTS after viral injection in the PVH. Distances indicated mm from bregma.

[0037] FIG. 19A-19F: DVCGLP1Rneurons project to 1PBN and PVH without collateralization. (FIG. 19 A) Cre-dependent ChR2-EYFP was injected into the AP of Glplr-ires-Cre mice (n=3). (FIG. 19B) Representative image of PVH in mouse with EY FP injection in APGLP1Rneurons. Scale bar, 50 pm. (FIG. 19C) Representative images of PBN projections across the rostral-caudal axis, distances indicate mm from bregma. (FIG. 19D) Cre-dependent ChR2-EYFP was injected into the NTS of Glplr-ires-Cre mice (n=3). (FIG. 19E) Representative image of 1PBN in mouse with EYFP injection in NTSGLP1Rneurons. Scale bar, 50 pm. (FIG. 19F) Representative images of PVH projections across the rostral-caudal axis, distances indicate mm from bregma.

[0038] FIG. 20A-20C: Neural activation in the 1PBN and PVH with DVCGLP1Rneuron stimulation. (FIG. 20 A) Schematic for RNA in situ hybridization for Fos and CGRP (Calca) in 1PBN or Fos and MC4R in PVH after chemogenetic stimulation of DVCGLP1Rneurons. (FIG. 20B) Representative images (left) and quantification (right) of colocalization of Fos and Calca in the 1PBN after control (vehicle) or chemogenetic (CNO) stimulation (n=3 / group, unpaired t-test, p<0.05). elPBN, external lateral PBN; clPBN, central lateral PBN. (FIG. 20C) Representative images (left) and quantification of colocalization (right) of Fos and Mc4r in the PVH after control (vehicle) or chemogenetic (CNO) chemogenetic stimulation (n=3 / group, unpaired t-test, p<0.01).

[0039] FIG. 21A-21N: Behavioral effects of APGLP1R^1PBN and NTSGLP1R^PVH neuron optogenetic activation. (FIG. 21A, 21B) Food intake (I h) in control mice expressing EYFP in APGLP1R— >1PBN (FIG. 21 A, n=6, paired t-test, p=ns) and NTSGLP1R— >PVH (FIG. 2 IB, n=5, paired t-test, p=ns) neurons. (FIG. 21C, 2 ID) Time spent in stimulation zone during real-time place avoidance assay in ad libitum-fed control mice expressing EYFP in APGLP1R— >1PBN (FIG. 21C, n=5-6 / group, two-way ANOVA, all ps=ns) or NTSGLP1R— >PVH (FIG. 21D, n=5, two-way ANOVA, all ps=ns) neurons. (FIG. 2 IE, 2 IF) Time spent in stimulation zone during real-time place avoidance assay in fasted control mice expressing EYFP in APGLP1R— >1PBN (FIG. 2 IE, n=5-6 / group, two-way ANOVA, all ps=ns) or NTSGLP1R— >PVH (FIG. 21F, n=5-6 / group, two-way ANOVA, all ps=ns). (FIG. 21G, 21H) Time spent in stimulation zone during real-time place avoidance assay in ad libitum-fed mice expressing ChR2 in APGLP1R— >1PBN (FIG. 21G, n=10- 11 / group, two-way ANOVA, Trials 2&3 no stim vs. stim: ps<0.05) or NTSGLP1R— >PVH (FIG. 21H, n=7, two-way ANOVA, all ps=ns). (FIG. 211, 21 J) Time spent in stimulation zone during real-time place avoidance assay in fasted mice expressing ChR2 in APGLP1R^1PBN (FIG. 211, n=l 1, two-way ANOVA, all ps=ns) or NTSGLP1R^PVH (FIG. 21 J, n=7, two-way ANOVA, all ps=ns) neurons. (FIG. 21K, 21L) Individual hedonic taste reactivity behaviors at baseline (pre-stimulation) and during optogenetic stimulation of APGLpiR— >1PBN (FIG. 21K, n=6, two-way ANOVA, RMM and LTP no stim vs. stim: ps<0.05) or NTSGLP1R— >PVH (FIG. 21L, n=6, two-way ANOVA, all ps=ns) neurons. (FIG. 2 IM, 2 IN) Individual aversive taste reactivity behaviors at baseline (pre- stimulation) and during optogenetic stimulation of APGLP1R— >1PBN (FIG. 2 IM, n=6, two- way ANOVA, CR no stim vs. stim: p<0.001) or NTSGLP1R— >PVH (FIG. 21N, n=6, two- way ANOVA, all ps=ns) neurons. Grey bars, no stimulation; colored bars, optogenetic stimulation. Data are presented as mean ± S.E.M. *p<0.05, **p<0.01, ***p<0.001.

[0040] FIG. 22A-22H: Volcano plots of genes that are differentially expressed in NTS and AP neurons.

[0041] DETAILED DESCRIPTION OF THE INVENTION

[0042] The present subject matter may be understood more readily by reference to the following detailed description which forms part of this disclosure. It is to be understood that this invention is not limited to the specific products, methods, conditions or parameters described and / or shown herein, and that the terminology used herein for the purpose of describing particular embodiments by way of example only and is not intended to be limiting of the claimed invention.

[0043] Dissociating neural circuits that mediate the therapeutic effects and side effects of brain-targeted drugs could dramatically improve our ability to treat disease. The most successful obesity therapeutics - glucagon-like peptide-1 receptor (GLP1R) agonists - cause aversive responses such as nausea and vomiting, effects which may contribute to their efficacy.

[0044] Here, we unexpectedly discovered that the neural circuits mediating the effects of satiety and aversion are anatomically, physiologically, and functionally separable. Systematic investigation of hindbrain, arcuate hypothalamic, and vagal GLP1R populations revealed that only hindbrain neurons are required for the suppression of food intake and body weight by GLP1 -based obesity drugs. Simultaneous in vivo two-photon imaging of hindbrain subregions demonstrated that while area postrema (AP) GLP1R neurons respond to both nutritive and aversive stimuli, nucleus tractus solitarius (NTS) GLP1R neurons are biased toward nutritive stimuli.

[0045] We therefore separately activated these two neuron populations to determine their effects on behavior. Strikingly, activation of NTSGLP1Rneurons triggered satiety in the absence of aversion, whereas activation of APGLP1Rneurons triggered strong aversion accompanied by anorexia. Anatomical and behavioral analyses demonstrated that NTSGLP1Rand APGLP1Rneurons send parallel projections to different downstream brain regions to drive satiety and aversion, respectively. Importantly, GLP1R agonists reduced food intake even when the aversion pathway was blocked. Overall, these findings reveal NTSGLP1Rneurons as a population that suppresses food intake and body weight without causing aversion. Refining obesity drugs to selectively target this population may promote weight loss while avoiding adverse side effects that limit treatment adherence.

[0046] This invention will be used, inter alia, in high-throughput cell screening assays using chemical libraries to identify therapeutic compounds that trigger satiety in the absence of aversion, including known agonists of GLP1R.

[0047] Unless otherwise defined herein, scientific and technical terms used in connection with the present application shall have the meanings that are commonly understood by those of ordinary skill in the art. In addition to definitions included in this sub-section, further definitions of terms are interspersed throughout the text.

[0048] In this invention, “a” or “an” means “at least one” or “one or more,” etc., unless clearly indicated otherwise by context. The term “or” means “and / or” unless stated otherwise. In the case of a multiple-dependent claim, however, use of the term “or” refers back to more than one preceding claim in the alternative only.

[0049] For purposes of the invention, “nucleic acid”, “nucleotide sequence” or a “nucleic acid molecule” as used herein refers to any DNA or RNA molecule, either single or double stranded and, if single stranded, the molecule of its complementary sequence in either linear or circular form. In discussing nucleic acid molecules, a sequence or structure of a particular nucleic acid molecule may be described herein according to the normal convention of providing the sequence in the 5’ to 3’ direction. With reference to nucleic acids of the invention, the term “isolated nucleic acid” is sometimes used. This term, when applied to DNA, refers to a DNA molecule that is separated from sequences with which it is immediately contiguous in the naturally occurring genome of the organism in which it originated. For example, an “isolated nucleic acid” may comprise a DNA molecule inserted into a vector, such as a plasmid or virus vector, or integrated into the genomic DNA of a prokaryotic or eukaryotic cell or host organism. Alternatively, this term may refer to a DNA that has been sufficiently separated from (e.g., substantially free of) other cellular components with which it would naturally be associated. “Isolated” is not meant to exclude artificial or synthetic mixtures with other compounds or materials, or the presence of impurities that do not interfere with the fundamental activity, and that may be present, for example, due to incomplete purification. When applied to RNA, the term “isolated nucleic acid” refers primarily to an RNA molecule encoded by an isolated DNA molecule as defined above. Alternatively, the term may refer to an RNA molecule that has been sufficiently separated from other nucleic acids with which it would be associated in its natural state (i.e., in cells or tissues). An isolated nucleic acid (either DNA or RNA) may further represent a molecule produced directly by biological or synthetic means and separated from other components present during its production.

[0050] The terms “polynucleotide”, “nucleotide”, “nucleotide sequence”, “nucleic acid” and “oligonucleotide” are used interchangeably. They refer to a polymeric form of nucleotides of any length, either deoxyribonucleotides or ribonucleotides, or analogs thereof. A polynucleotide may comprise one or more modified nucleotides, such as methylated nucleotides and nucleotide analogs. If present, modifications to the nucleotide structure may be imparted before or after assembly of the polymer. The sequence of nucleotides may be interrupted by non-nucleotide components. A polynucleotide may be further modified after polymerization, such as by conjugation with a labeling component.

[0051] As used herein, the terms “protein” and “polypeptide” are used interchangeably herein to designate a series of amino acid residues, connected to each other by peptide bonds between the alpha-amino and carboxy groups of adjacent residues. The terms “protein”, and “polypeptide” refer to a polymer of amino acids, including modified amino acids (e.g., phosphorylated, glycated, glycosylated, etc.) and amino acid analogs, regardless of its size or function. “Protein” and “polypeptide” are often used in reference to relatively large polypeptides, whereas the term “peptide” is often used in reference to small polypeptides, but usage of these terms in the art overlaps. The terms “protein” and “polypeptide” are used interchangeably herein when referring to a gene product and fragments thereof. Thus, exemplary polypeptides or proteins include gene products, naturally occurring proteins, homologs, orthologs, paralogs, fragments and other equivalents, variants, fragments, and analogs of the foregoing.

[0052] A “derivative” of a polypeptide, polynucleotide or fragments thereof means a sequence modified by varying the sequence of the construct, e.g., by manipulation of the nucleic acid encoding the protein or by altering the protein itself. “Derivatives” of a gene or nucleotide sequence refers to any isolated nucleic acid molecule that contains significant sequence similarity to the gene or nucleotide sequence or a part thereof. In addition, “derivatives” include such isolated nucleic acids containing modified nucleotides or mimetics of naturally-occurring nucleotides.

[0053] The term “functional” as used herein implies that the nucleic or amino acid sequence is functional for the recited assay or purpose.

[0054] According to the present invention, an isolated or biologically pure molecule or cell is a compound that has been removed from its natural milieu. As such, “isolated” and “biologically pure” do not necessarily reflect the extent to which the compound has been purified. An isolated compound of the present invention can be obtained from its natural source, can be produced using laboratory synthetic techniques or can be produced by any such chemical synthetic route.

[0055] Variants which are biologically active, refer to those, which maintains the native function of the protein. Preferably, naturally or non-naturally occurring polypeptide variants have amino acid sequences which are at least about 55, 60, 65, or 70, preferably about 75, 80, 85, 90, 91, 92, 93, 94, 95, 96, 97, or 98, 99, or 100% identical to the full- length amino acid sequence or a fragment thereof. Percent identity between a putative polypeptide variant and a full-length amino acid sequence is determined using the Blast2 alignment program (Blosum62, Expect 10, standard genetic codes). Variations in percent identity can be due, for example, to amino acid substitutions, insertions, or deletions. Amino acid substitutions are defined as one for one amino acid replacements. They are conservative in nature when the substituted amino acid has similar structural and / or chemical properties. Examples of conservative replacements are substitution of a leucine with an isoleucine or valine, an aspartate with a glutamate, or a threonine with a serine.

[0056] The term “identity” as used herein and as known in the art, is the relationship between two or more oligo sequences, and is determined by comparing the sequences. Identity also means the degree of sequence relatedness between oligo sequences, as determined by the match between strings of such sequences. Identity can be readily calculated (see, e.g., Computation Molecular Biology, Lesk, A. M., eds., Oxford University Press, New York (1998), and Biocomputing: Informatics and Genome Projects, Smith, D. W., ed., Academic Press, New York (1993), both of which are incorporated by reference herein). While a number of methods to measure identity between two polynucleotide sequences are available, the term is well known to skilled artisans (see, e.g., Sequence Analysis in Molecular Biology, von Heinje, G., Academic Press (1987); and Sequence Analysis Primer, Gribskovm, M. and Devereux, J., eds., M. Stockton Press, New York (1991)). Methods commonly employed to determine identity between oligo sequences include, for example, those disclosed in Carillo, H., and Lipman, D., Siam J. Applied Math. (1988) 48: 1073. In certain embodiments, the present invention may have 75, 80, 85, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, or 100% identity with the SEQ ID NOs disclosed herein.

[0057] Amino acid insertions or deletions are changes to or within an amino acid sequence. They typically fall in the range of about 1 to 5 amino acids. Guidance in determining which amino acid residues can be substituted, inserted, or deleted without abolishing biological or immunological activity of polypeptide can be found using computer programs well known in the art, such as DNASTAR software. Whether an amino acid change results in a biologically active therapeutic fusion polypeptide can readily be determined by assaying for native activity, as described for example, in the specific Examples, below.

[0058] Reference to genetic sequences herein refers to single- or double-stranded nucleic acid sequences and comprises a coding sequence or the complement of a coding sequence for polypeptide of interest. Degenerate nucleic acid sequences encoding polypeptides, as well as homologous nucleotide sequences which are at least about 50, 55, 60, 65, 60, preferably about 75, 80, 85, 90, 91, 92, 93, 94, 95, 96, 97, or 98, 99, or 100% identical to the DNA may be used in accordance with the teachings herein polynucleotides. Percent sequence identity between the sequences of two polynucleotides is determined using computer programs such as ALIGN which employ the FASTA algorithm, using an affine gap search with a gap open penalty of -12 and a gap extension penalty of -2. Complementary DNA (cDNA) molecules, species homologs, and variants of nucleic acid sequences which encode biologically active polypeptides also are useful polynucleotides.

[0059] Variants and homologs of the nucleic acid sequences described above also are useful nucleic acid sequences. Typically, homologous polynucleotide sequences can be identified by hybridization of candidate polynucleotides to known polynucleotides under stringent conditions, as is known in the art. For example, using the following wash conditions: 2*SSC (0.3 M NaCl, 0.03 M sodium citrate, pH 7.0), 0.1% SDS, room temperature twice, 30 minutes each; then 2*SSC, 0.1% SDS, 50° C. once, 30 minutes; then 2*SSC, room temperature twice, 10 minutes each, homologous sequences can be identified which contain at most about 25-30% basepair mismatches. More preferably, homologous nucleic acid strands contain 15-25% basepair mismatches, even more preferably 5-15% basepair mismatches.

[0060] As used herein the term “wild type” is a term of the art understood by skilled persons and means the typical form of an organism, strain, gene or characteristic as it occurs in nature as distinguished from mutant or variant forms. As used herein the term “variant” should be taken to mean the exhibition of qualities that have a pattern that deviates from the wild type or a comprises non naturally occurring components.

[0061] The terms “agent” and “test compound” refers to any purified molecule, substantially purified molecule, molecules that are one or more components of a mixture of compounds, or a mixture of a compound with any other material that can be analyzed using the methods of the present invention. Test compounds can be organic or inorganic chemicals, or biomolecules, and all fragments, analogs, homologs, conjugates, and derivatives thereof. Biomolecules include proteins, polypeptides, nucleic acids, lipids, monosaccharides, polysaccharides, and all fragments, analogs, homologs, conjugates, and derivatives thereof. Test compounds can be of natural or synthetic origin, and can be isolated or purified from their naturally occurring sources, or can be synthesized de novo. Test compounds can be defined in terms of structure or composition, or can be undefined. The compound can be an isolated product of unknown structure, a mixture of several known products, or an undefined composition comprising one or more compounds. Examples of undefined compositions include cell and tissue extracts, growth medium in which prokaryotic, eukaryotic, and archaebacterial cells have been cultured, fermentation broths, protein expression libraries, and the like.

[0062] It is also contemplated that the term “compound” or “compounds” refers to the compounds discussed herein and includes precursors and derivatives of the compounds, and pharmaceutically acceptable salts of the compounds, precursors, and derivatives.

[0063] In certain embodiments, the test compound can reduce an aversive response to an obesity therapeutic. The phrase obesity therapeutic refers to any drug used to treat obesity, a symptom of obesity, or any metabolic disorder / condition associated with body fat distribution. Exemplary metabolic disorders include, without limitation, type 2 diabetes, hyperlipidemia or dyslipidemia (high or altered circulating levels of low-density lipoprotein cholesterol (LDL-C), triglycerides, very low-density lipoprotein cholesterol (VLDL-C), apolipoprotein B or other lipid fractions), obesity (particularly abdominal obesity), lipodystrophy (such as an inability to deposit fat in adipose depots regionally (partial lipodystrophy) or in the whole body (lipoatrophy)), insulin resistance or higher or altered insulin levels at fasting or during a glucose or insulin challenge, liver fat deposition or fatty liver disease and their complications (such as, for example, cirrhosis, fibrosis, or inflammation of the liver), higher or elevated or altered liver enzyme levels or other markers of liver damage, inflammation or fat deposition, higher blood pressure and / or hypertension, higher blood sugar or glucose or hyperglycemia, metabolic syndrome, coronary artery disease, and other atherosclerotic conditions, and the complications of each of the aforementioned conditions.

[0064] Examples of obesity therapeutics include without limitation glucagon-like protein 1 receptor (GLP1R) agonists, (such as exenatide, liraglutide, and semaglutide), metformin, insulin, sulfonylureas (such as glyburide, glipizide, and glimepiride), meglitinides (such as repaglinide and nateglinde), thiazolidinediones (such as rosiglitazone and pioglitazone), DPP -4 inhibitors (such as sitagliptin-saxaghptin and linagliptin), SGLT2 inhibitors (such as canagliflozin, dapagliflozin, and empagliflozin), orlistat, phentermine, topiramate, bupropion, naltrexone, and liraglutide. Obesity therapeutics also include agents that treat or inhibit elevated triglycerides include, but are not limited to: statins (such as rosuvastatin, simvastatin, and atorvastatin), fibrates (such as fenofibrate, gemfibrozil, and fenofibric acid), nicotinic acid (such as niacin), and fatty acids (such as omega-3 fatty acids). Additional obesity therapeutics are known in the art including those identified in US Patent Application Publication No. US 2022 / 0184114 Al.

[0065] SCREENING METHODS

[0066] The present disclosure provides certain compounds and polypeptide sequences that are useful in simulating in vitro the response that certain human area postrema neurons that express glucagon -like peptide- 1 receptor (APGLP1Rneurons) and nucleus tractus solitarius neurons that express glucagon-like peptide 1 receptor (NTSGLP1Rneurons) exhibit when exposed to such compounds. In certain embodiments, these compounds and polypeptide sequences reduce an aversive response to an obesity therapeutic.

[0067] As used herein, the term “modulate” means any change, increase, or decrease in the amount, quality, or effect of a particular activity or protein. “Modulators” refer to any inhibitory or activating molecules identified using in vitro and in vivo assays for, e.g., agonists, antagonists, and their homologs, including fragments, variants, and mimetics, as defined herein, that exert substantially the same biological activity as the molecule. “Inhibitors” or “antagonists” are modulating compounds that reduce, decrease, block, prevent, delay activation, inactivate, desensitize, or downregulate the biological activity or expression of a molecule or pathway of interest. “Inducers,” “activators,” or “agonists” are modulating compounds that increase, induce, stimulate, open, activate, facilitate, enhance activation, sensitize, or upregulate a molecule or pathway of interest. In some preferred aspects of the invention, the level of inhibition or upregulation of the expression or biological activity of a molecule or pathway of interest refers to a decrease (inhibition or downregulation) or increase (upregulation) of greater than from about 50% to about 99%, and more specifically, about 50%, 51%, 52%, 53%, 54%, 55%, 56%, 57%, 58%, 59%,

[0068] 60%, 61%, 62%, 63%, 64%, 65%, 66%, 67%, 68%, 69% 70%, 71%, 72%, 73%, 74%,

[0069] 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%,

[0070] 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or more. The inhibition or upregulation may be direct, i.e., operate on the molecule or pathway of interest itself, or indirect, i.e., operate on a molecule or pathway that affects the molecule or pathway of interest.

[0071] In the present disclosure, “identification”, “separation”, “sorting”, “isolation”, “purification”, or “screening” has a meaning commonly used in the art, and typically refers to sorting of a target such as an organism, a cell, a substance, data, or the like that has certain targeted characteristics from a population containing a large number of cells by a specific manipulation / evaluation method.

[0072] In certain aspects, the disclosure provides methods of identifying compounds that inhibit aversive response to obesity therapeutics. An aversive response refers to any undesired effect of a drug or treatment. An “aversive response to an obesity therapeutic” or “obesity therapy-related aversive response” refers to any undesired effect of an obesity therapeutic. Exemplary obesity therapy-related aversive responses are any gastrointestinal symptoms including nausea and vomiting.

[0073] In certain aspects, the methods comprise introducing a test compound to area postrema neuron that expresses glucagon-like peptide- 1 receptor (APGLP1Rneuron), and measuring response of the APGLP1Rneuron to the test compound by comparing activity of the APGLP1Rneuron in the presence and the absence of the test compound. In certain embodiments, test compounds do not cause aversive responses when the activity of the APGLpiRneuronis not increased. In certain embodiments the activity of the APGLP1Rneuron is decreased. In certain embodiments, the activity of the APGLP1Rneuron is decreased by 100%, 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80%, 79%, 78%, 77%, 76%, 75%, 74%, 73%,

[0074] 72%, 71%, 70%, 69%, 68%, 67%, 66%, 65%, 64%, 63%, 62%, 61%, 60%, 59%, 58%,

[0075] 57%, 56%, 55%, 54%, 53%, 52%, 51%, 50%, 49%, 48%, 47%, 46%, 45%, 44%, 43%,

[0076] 42%, 41%, 40%, 39%, 38%, 37%, 36%, 35%, 34%, 33%, 32%, 31%, 30%, 29%, 28%,

[0077] 27%, 26%, 25%, 24%, 23%, 22%, 21%, 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%,

[0078] 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%.

[0079] In certain embodiments, the method further comprises contacting a nucleus tractus solitarius neuron that expresses glucagon-like peptide 1 receptor (NTSGLP1Rneuron), and measuring response of the NTSGLP1Rneuron to the test compound by comparing activity of the NTSGLP1Rneuron in the presence and the absence of the test compound. In certain embodiments, test compounds do not cause aversive responses when the activity of the NT SGLP1Rneuron is not decreased or is slightly decreased. In certain embodiments, the activity of the NTSGLP1Rneuron is decreased by less than 100%, 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%,

[0080] 80%, 79%, 78%, 77%, 76%, 75%, 74%, 73%, 72%, 71%, 70%, 69%, 68%, 67%, 66%,

[0081] 65%, 64%, 63%, 62%, 61%, 60%, 59%, 58%, 57%, 56%, 55%, 54%, 53%, 52%, 51%,

[0082] 50%, 49%, 48%, 47%, 46%, 45%, 44%, 43%, 42%, 41%, 40%, 39%, 38%, 37%, 36%,

[0083] 35%, 34%, 33%, 32%, 31%, 30%, 29%, 28%, 27%, 26%, 25%, 24%, 23%, 22%, 21%,

[0084] 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%.

[0085] In certain embodiments the methods further comprise measuring the responses of the APGLP1Rneurons and NTSGLP1Rneurons to a known aversive therapeutic compound and identifying the test compounds that do not cause aversive responses based on the measured responses to the test compound and the known aversive therapeutic compound.

[0086] The phrase “known aversive therapeutic compound” refers to any compound that is known to activate an APGLP1Rneuron and cause an aversive response when administered to a patient. In certain embodiments, the known aversive therapeutic compound is a GLP1R agonist. In certain embodiments the GLP1R agonist is exenatide, liraglutide, or semaglutide.

[0087] Also provided herein are methods of screening for therapeutic agents for treating a GLP1R agonist related aversive response in a patient, the method comprising expressing one or more area postrema neuron (AP) gene in a cell; contacting the cell with a test compound; determining the effect of the test compound on the expression level of the one or more AP genes in comparison to that in the absence of test compound; and selecting the test compounds that can reduce or inhibit the expression levels of the genes as the therapeutic agent.

[0088] The phrase “area postrema neuron gene” or “AP gene” refers to any gene is that has increased expression in AP neurons over NTS neurons. In certain embodiments the AP gene is selected from the genes identified in Figure 22. In certain embodiments, the AP gene is selected from Erbb4, Kcnabl, Unc5d, Csmd3, Grm7, Tenm3, Cdkl4, Robo2, Mgat4c, Kctd8, Slc24a3, Ntrk3, Luzp2, Ncam2, Rora, Slc2al3, Neill, Grin2b, Meisl, Spockl, Rmst, Lhfpl3, Epha5, Dpyd, Ccdc85a, Sgcz, Sema6d, Dabl, Gabra2, Pcdhl5, Cdhl2, Cntn4, Cntn6, Hs3st5, 01fm3, Nell2, Sorcs3, Nkain2, Grin2a, Grm8, Mpped2, Aff2, Sorcsl, Pcsk2, Asic2, Sytl, Brinp3, Kirrel3 Gabrbl, Rbfoxl, Slc8alTrpc5, Thsd7b, Ammercrl, Grik4, Zfpm2, Cdh4, Sema3e, Raplgap, Chnl, and Tmtc2.

[0089] In certain embodiments, administration of the test compound decreases expression of the AP gene by least 100%, 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80%, 79%, 78%, 77%, 76%, 75%,

[0090] 74%, 73%, 72%, 71%, 70%, 69%, 68%, 67%, 66%, 65%, 64%, 63%, 62%, 61%, 60%,

[0091] 59%, 58%, 57%, 56%, 55%, 54%, 53%, 52%, 51%, 50%, 49%, 48%, 47%, 46%, 45%,

[0092] 44%, 43%, 42%, 41%, 40%, 39%, 38%, 37%, 36%, 35%, 34%, 33%, 32%, 31%, 30%,

[0093] 29%, 28%, 27%, 26%, 25%, 24%, 23%, 22%, 21%, 20%, 19%, 18%, 17%, 16%, 15%,

[0094] 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%.

[0095] In certain embodiments, the method further comprises expressing one or more dorsal vagal complex (DVC) gene in a cell; contacting the cell with the test compound; determining the effect of the test compound on the expression level of the one or more DVC genes in comparison to that in the absence of test compound; and selecting the test compounds that do not significantly reduce the expression levels of the genes as the therapeutic agents. The phrase “dorsal vagal complex gene” or “DVC gene” refers to a gene that has increased expression in DVC neurons over other neurons. DVC genes can include AP genes and NTS genes. In certain embodiments the DVC gene is selected from the genes identified in Figure 22.

[0096] In certain embodiments, administration of the test compound increases expression of the DVC gene by least 100%, 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80%, 79%, 78%, 77%, 76%, 75%,

[0097] 74%, 73%, 72%, 71%, 70%, 69%, 68%, 67%, 66%, 65%, 64%, 63%, 62%, 61%, 60%,

[0098] 59%, 58%, 57%, 56%, 55%, 54%, 53%, 52%, 51%, 50%, 49%, 48%, 47%, 46%, 45%,

[0099] 44%, 43%, 42%, 41%, 40%, 39%, 38%, 37%, 36%, 35%, 34%, 33%, 32%, 31%, 30%,

[0100] 29%, 28%, 27%, 26%, 25%, 24%, 23%, 22%, 21%, 20%, 19%, 18%, 17%, 16%, 15%,

[0101] 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%.

[0102] In certain embodiments, the DVC gene is a nucleus tractus solitarius gene. The phrase “nucleus tractus solitarius gene” or “NTS gene” refers to any gene is that has increased expression in NTS neurons over AP neurons. As NTS genes are differentially expressed in NTS neutrons, they are responsible for the satiety response to obesity therapeutics. In certain embodiments the NTS gene is selected from the genes identified in Figure 22. In certain embodiments, the NTS gene is selected from Fstl4, Ephbl, Col25al, Synpr, Cadps2, Gria3, Ndst3, Klhll, Mania, Hs3st2, Pcdhl lx, Ptchd4, Ldb2, Glra3, Adcy8, Teadl, Rab27b, Fat3, Aik, Htr2c, Esrrg, Pdelc, Dgkg, Mei4, Cdh8, Hs3st4, Slcl7a6, Kcnhl, Nrpl, Esrl, Ak5, Fut9, Btbdl lNckap5, Calnl, Rbfox3, Cntnap4, Limal, Mrap2, and Cdh7.

[0103] In certain embodiments, administration of the test compound increases expression of the NTS gene by least 100%, 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80%, 79%, 78%, 77%, 76%, 75%,

[0104] 74%, 73%, 72%, 71%, 70%, 69%, 68%, 67%, 66%, 65%, 64%, 63%, 62%, 61%, 60%,

[0105] 59%, 58%, 57%, 56%, 55%, 54%, 53%, 52%, 51%, 50%, 49%, 48%, 47%, 46%, 45%,

[0106] 44%, 43%, 42%, 41%, 40%, 39%, 38%, 37%, 36%, 35%, 34%, 33%, 32%, 31%, 30%,

[0107] 29%, 28%, 27%, 26%, 25%, 24%, 23%, 22%, 21%, 20%, 19%, 18%, 17%, 16%, 15%,

[0108] 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%.

[0109] The effect of the test compound on the biological activity of the neurons can be determined by any means suitable in the art. The test compound can be assessed at multiple concentrations. In some aspects, the test compound is assessed for its ability to modulate at least one biological activity of the AP neurons or NTS neurons. Exemplary methods for assessing the biologically activity of the neurons include, without limitation, in vivo neural activity (calcium, voltage) measurements (using fiber photometry, 2-photon imaging; and ex vivo histological analysis of neural activity (e.g., Fos).

[0110] In another aspect of the invention, methods of treating an obesity-related aversive response are provided. The method comprises administering the compound identified above. This compound may be administered in combination with an obesity therapeutic. In certain embodiments, after administering the compound, the patient does not have any aversive response to the obesity therapeutic. In certain embodiments, after administering the compound, the patient has fewer aversive responses to the obesity therapeutic than a patient that received the obesity therapeutic alone. In certain embodiments, after administering the compound, the patient has a decreased aversive response to the obesity therapeutic than a patient that received the obesity therapeutic alone. In certain embodiments, the aversive responses are reduced by at least 100%, 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%,

[0111] 80%, 79%, 78%, 77%, 76%, 75%, 74%, 73%, 72%, 71%, 70%, 69%, 68%, 67%, 66%,

[0112] 65%, 64%, 63%, 62%, 61%, 60%, 59%, 58%, 57%, 56%, 55%, 54%, 53%, 52%, 51%,

[0113] 50%, 49%, 48%, 47%, 46%, 45%, 44%, 43%, 42%, 41%, 40%, 39%, 38%, 37%, 36%,

[0114] 35%, 34%, 33%, 32%, 31%, 30%, 29%, 28%, 27%, 26%, 25%, 24%, 23%, 22%, 21%,

[0115] 20%, 19%, 18%, 17%, 16%, 15%, 14%, 13%, 12%, 11%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, or 1%.

[0116] HIGH-THROUGHPUT SCREENING ASSAYS

[0117] The invention also includes within its scope high-throughput screening assays to identify compounds that modulate the biological activity that causes an aversive reaction to an anti-obesity therapeutic. High-throughput screening assays permit screening of large numbers of test compounds in an efficient manner. Each well of the microtiter plate can be used to run a separate assay against a candidate modulator. A microtiter plate permits screening of multiple concentrations of a test compound, multiple test compounds, alone or in combination with other test compounds. The assays may take place in the presence of additional agonists or antagonists. Data obtained for the test compounds are compared with measurements taken in the presence of known agonists or antagonists and / or to control samples (such as a non-stimulatory / non-inhibitory medium).

[0118] Serial assays may be performed to narrow down the pool of test compounds that act as aversive. For example, the in vitro assays of the invention may be combined with cell-based assays as a secondary or confirming screening step. Such assays have been described, for example, in published U.S. Patent Application 2005 / 0059094 to Servant et al.

[0119] HOST CELLS

[0120] In certain aspects, the AP genes, NTS genes, and / or DVC genes can be expressed in a recombinant expression system such as, but not limited to, host cells. A “host cell” refers to a cell that may be or has been a recipient of a vector or isolated polynucleotide. Host cells may be prokaryotic cells or eukaryotic cells. Exemplary eukaryotic cells include mammalian cells, such as primate or non-primate animal cells; fungal cells, such as yeast; plant cells; and insect cells. Non-limiting exemplary mammalian cells include, but are not limited to, HEK cells, NSO cells, PER.C6® cells (Crucell), and 293 and CHO cells, and their derivatives, such as 293-6E and DG44 cells, respectively. In certain embodiments, the cell is a host cell is transfected with the vector described below. Methods for delivering vectors / constructs or other nucleic acids (such as in vitro transcribed RNA) into host cells can include electroporation, transformation, transfection or transduction.

[0121] VECTORS

[0122] Also described herein are vectors for directing expression the DVC genes, AP genes, and NTS genes described above.

[0123] The term “delivery” as used herein refers to the introduction of foreign molecule (i.e., AP gene nucleotide containing nanoparticle) into cells. The term “administration” as used herein means the introduction of a foreign molecule into a cell. The term is intended to be synonymous with the term “delivery”.

[0124] A “vector” as used herein is a biological or chemical moiety comprising a nucleic acid sequence which can be introduced into an appropriate host cell for replication or expression of said nucleic acid sequence. Common vectors include non-viral vectors and viral vectors. As used herein, a non-viral system might be selected from nanoparticles, electroporation systems and novel biomaterials, naked DNA, phage, transposon, plasmids, cosmids (Phillip McClean, www.ndsu.edu / pubweb / ~mcclean / - plsc731 / cloning / cloning4.htm) and artificial chromosomes (Gong, Shiaoching, et al. “A gene expression atlas of the central nervous system based on bacterial artificial chromosomes.” Nature 425.6961 (2003): 917-925).

[0125] As used herein, an “expression cassette” refers to a nucleic acid molecule which comprises a biologically useful nucleic acid sequence (e.g., a gene cDNA encoding a protein, enzyme or other useful gene product, such as the DVC genes, AP genes, and NTS genes described herein) and regulatory sequences operably linked thereto which direct or modulate transcription, translation, and / or expression of the nucleic acid sequence and its gene product.

[0126] As used herein, “operably linked” sequences include both regulatory sequences that are contiguous with the nucleic acid sequence and regulatory sequences that act in trans or at a distance to control the sequence. Such regulatory sequences typically include, e.g., one or more of a promoter, an enhancer, an intron, a Kozak sequence, a polyadenylation sequence, and a TATA signal. The expression cassette may contain regulatory sequences upstream (5’ to) of the gene sequence, e.g., one or more of a promoter, an enhancer, an intron, etc., and one or more of an enhancer, or regulatory sequences downstream (3’ to) a gene sequence, e.g., 3’ untranslated region comprising a polyadenylation site, among other elements. In other embodiments, the term “transgene” refers to one or more DNA sequences from an exogenous source which are inserted into a target cell. Typically, such an expression cassette for generating a viral vector contains the coding sequence for the gene product described herein flanked by packaging signals of the viral genome and other expression control sequences such as those described herein. In certain embodiments, a vector genome may contain two or more expression cassettes.

[0127] In addition to the coding sequence, in certain embodiments the vector includes regulatory sequences which direct expression in a host cell. In certain embodiments, the regulatory elements include a promoter.

[0128] In addition to a promoter, the vector may contain one or more appropriate “regulatory elements” or “regulatory sequences”, which comprise but are not limited to an enhancer; transcription factor; transcription terminator; efficient RNA processing signals such as splicing and polyadenylation signals (poly A); sequences that stabilize cytoplasmic mRNA, for example Woodchuck Hepatitis Virus (WHP) Posttranscriptional Regulatory Element (WPRE); sequences that enhance translation efficiency (i.e., Kozak consensus sequence); sequences that enhance protein stability; and when desired, sequences that enhance secretion of the encoded product. Examples of suitable polyA sequences include, e.g., SV40, bovine growth hormone (bGH), and TK polyA. Examples of suitable enhancers include, e.g., the alpha fetoprotein enhancer, the TTR minimal promoter / enhancer, LSP (TH-binding globulin promoter / alphal-microglobulin / bikunin enhancer), amongst others. These control sequences or the regulatory sequences are operably linked to the nuclease coding sequence or transgene coding sequence.

[0129] Conventional viral and non-viral based gene transfer methods can be used to introduce nucleic acids in cells or target tissues. Such methods can be used to administer nucleic acids encoding inhibitory compounds to cells in culture, or in a host organism. Non-viral vector delivery systems include DNA plasmids, RNA (e.g., a transcript of a vector described herein), naked nucleic acid, and nucleic acid complexed with a delivery vehicle, such as a liposome. Viral vector delivery systems include DNA and RNA viruses, which have either episomal or integrated genomes after delivery to the cell. For a review of gene therapy procedures, see Anderson, Science 256:808-813 (1992); Nabel & Feigner, TIBTECH 11 :211-217 (1993); Mitani & Caskey, TIBTECH 11 : 162-166 (1993); Dillon, TIBTECH 11 : 167-175 (1993); Miller, Nature 357:455-460 (1992); Van Brunt, Biotechnology 6(10): 1149-1154 (1988); Vigne, Restorative Neurology and Neuroscience 8:35-36 (1995); Kremer & Perricaudet, British Medical Bulletin 51(1):31-44 (1995); Haddada et al., in Current Topics in Microbiology and Immunology Doerfler and Bihm (eds) (1995); and Yu et al., Gene Therapy 1 : 13-26 (1994).

[0130] Methods of non-viral delivery of nucleic acids include lipofection, nucleofection, microinjection, biolistics, virosomes, liposomes, immunoliposomes, polycation or lipidmucleic acid conjugates, naked DNA, artificial virions, and agent-enhanced uptake of DNA. Lipofection is described in e.g., U.S. Pat. Nos. 5,049,386, 4,946,787; and 4,897,355) and lipofection reagents are sold commercially (e.g., TransfectamTM and LipofectinTM). Cationic and neutral lipids that are suitable for efficient receptorrecognition lipofection of polynucleotides include those of Feigner, WO 91 / 17424; WO 91 / 16024. Delivery can be to cells (e.g., in vitro or ex vivo administration) or target tissues (e.g., in vivo administration). The preparation of lipidmucleic acid complexes, including targeted liposomes such as immunolipid complexes, is well known to one of skill in the art (see, e.g., Crystal, Science 270:404-410 (1995); Blaese et al., Cancer Gene Ther. 2:291-297 (1995); Behr et al., Bioconjugate Chem. 5:382-389 (1994); Remy et al., Bioconjugate Chem. 5:647-654 (1994); Gao et al., Gene Therapy 2:710-722 (1995); Ahmad et al., Cancer Res. 52:4817- 4820 (1992); U.S. Pat. Nos. 4,186,183, 4,217,344, 4,235,871, 4,261,975, 4,485,054, 4,501,728, 4,774,085, 4,837,028, and 4,946,787).

[0131] Specific embodiments:

[0132] 1. A method of identifying a therapeutic compound that does not cause an aversive response, the method comprising:

[0133] (a) contacting a test compound and a postrema neuron that expresses glucagon-like peptide- 1 receptor (APGLP1Rneuron);

[0134] (b) contacting the test compound with a nucleus tractus solitarius neuron that expresses glucagon-like peptide 1 receptor (NTSGLP1Rneuron);

[0135] (c) measuring a response of the APGLP1Rneuron to the test compound by comparing activity of the APGLP1Rneuron in the presence and the absence of the test compound; and

[0136] (d) measuring a response of the NTSGLP1Rneuron to the test compound by comparing activity of the NTSGLP1Rneuron in the presence and the absence of the test compound; wherein said test compound does not cause an aversive response when the activity of the APGLP1Rneuron is not increased and the activity of the NTSGLP1Rneuron is not decreased in the presence of the test compound or is decreased by less than 50%.

[0137] 2. The method of embodiment 1, further comprising

[0138] (e) measuring the responses of the APGLP1Rneuron and the NTSGLP1Rneuron to a known aversive therapeutic compound; and

[0139] (f) identifying test compounds that do not cause aversive responses based on the measured responses to the test compound and known aversive therapeutic compound.

[0140] 3. The method of any one of embodiments 1 or 2, wherein the known aversive therapeutic compound is a glucagon-like peptide 1 receptor (GLP1R) agonist. 4. The method of any one of the preceding embodiments, wherein the activity of the APGLpiRneuron is decreased.

[0141] 5. The method of any one of the preceding embodiments, wherein the activity of the NTSGLP1Rneuron is the same.

[0142] 6. The method of any one of embodiments 1-4 wherein the activity of the NTSGLP1Rneuron is increased.

[0143] 7. The method of any one of embodiments 1-6, wherein the aversive response is nausea or vomiting.

[0144] 8. A method for screening therapeutic agents for treating GLP1R agonist related aversive response in a patient in need thereof, the method comprising

[0145] (a) expressing an area postrema neuron (AP) gene in a cell;

[0146] (b) contacting the cell with a test compound;

[0147] (c) determining the effect of the test compound on the expression level of the AP gene in comparison to that in the absence of the test compound; and

[0148] (d) selecting the agents that can reduce or inhibit the expression levels of the genes as the therapeutic agent.

[0149] 9. The method of embodiment 8, further comprising:

[0150] (e) expressing a dorsal vagal complex (DVC) gene in a cell;

[0151] (f) contacting the cell with the test compound;

[0152] (g) determining the effect of the test compound on the expression level of the DVC gene in comparison to that in the absence of the test compound; and

[0153] (h) selecting the test compounds that do not significantly reduce the expression levels of the gene as the therapeutic agent.

[0154] 10. The method of embodiment 9, wherein the DVC gene is a gene that expresses at the same level, or a greater level, in untreated NTS neurons when compared to AP neurons. 11. The method of any one of embodiments 8-10, wherein the cell expressing the AP gene is an APGLP1Rneuron.

[0155] 12. The method of any one of embodiments 9-11, wherein the cell expressing the DVC gene is an NTSGLP1Rneuron.

[0156] 13. The method of any one of embodiments 8-12, wherein the AP gene is Erbb4, Kcnabl, Unc5d, Csmd3, Grm7, Tenm3, Cdkl4, Robo2, Mgat4c, Kctd8, Slc24a3, Ntrk3, Luzp2, Ncam2, Rora, Slc2al3, Neill, Grin2b, Meisl, Spockl, Rmst, Lhfpl3, Epha5, Dpyd, Ccdc85a, Sgcz, Sema6d, Dabl, Gabra2, Pcdhl5, Cdhl2, Cntn4, Cntn6, Hs3st5, Olfm3, Nell2, Sorcs3, Nkain2, Grin2a, Grm8, Mpped2, Aff2, Sorcsl, Pcsk2, Asic2, Sytl, Brinp3, Kirrel3 Gabrbl, Rbfoxl, Slc8alTrpc5, Thsd7b, Ammercrl, Grik4, Zfpm2, Cdh4, Sema3e, Raplgap, Chnl, or Tmtc2.

[0157] 14. The method of any one of embodiments 9-15, wherein the NTS gene is Fstl4, Ephbl, Col25al, Synpr, Cadps2, Gria3, Ndst3, Klhll, Mania, Hs3st2, Pcdhl lx, Ptchd4, Ldb2, Glra3, Adcy8, Teadl, Rab27b, Fat3, Aik, Htr2c, Esrrg, Pdelc, Dgkg, Mei4, Cdh8, Hs3st4, Slcl7a6, Kcnhl, Nrpl, Esrl, Ak5, Fut9, Btbdl lNckap5, Calnl, Rbfox3, Cntnap4, Limal, Mrap2, or Cdh7.

[0158] 15. A therapeutic compound identified by the method of any one of the preceding embodiments.

[0159] 16. A method for treating an obesity-related aversive response in a patient in need thereof, the method comprising administering to the patient the therapeutic compound of embodiment 15.

[0160] 17. The method of embodiment 16, wherein said patient does not have an aversive response to said therapeutic compound.

[0161] 18. The method of embodiment 16, wherein the patient has fewer aversive responses when compared to a patient treated with the obesity therapeutic alone. EXAMPLES

[0162] The invention is now described with reference to the following examples. These examples are provided for the purpose of illustration only and the invention should in no way be construed as being limited to these examples but rather should be construed to encompass any and all variations that become evident as a result of the teaching provided herein.

[0163] EXAMPLE 1 : MATERIALS AND METHODS

[0164] The materials and methods below are provided to facilitate the practice of the present invention.

[0165] Drugs and reagents

[0166] Paraformaldehyde (PF A, 441244), bovine serum albumin (BSA, A7906), Triton X- 100 (T8787), dimethyl sulfoxide (DMSO, D8418), 2,2-thiodiethanol (TDE, 166782), cinacalcet (SML2012), lithium chloride (LiCl, L9650), saccharin (240931) and fast green FCF (68724) were purchased from Millipore Sigma. Phosphate buffered saline (SH30013.04) was purchased from Cytiva. Fluoro-Gel (17985) was purchased from Electron Microscopy Science. Clozapine N-oxide (CNO, 4936), RNAscope Multiplex Fluorescent V2 Assay (323100), and RNAscope probes were purchased from Bio-Techne. Exendin-4 (Ex-4, H8730) was purchased from Bachem. Semaglutide (29969) was purchased from Cayman.

[0167] Adeno-associated viral vectors

[0168] AAV5-EFla-DIO-taCasp3-TEVp (45580-AAV5), AAV5-syn-DIO-hM3Dq- mCherry (44361-AAV5), AAV5-syn-DIO-EGFP (50457-AAV5), AAV2-syn-DIO- hM3Dq-mCherry (44361-AAV2), AAV2-syn-DIO-hM4Di-mCherry (44362-AAV2), AAV5-EFla-DIO-hChR2(H134R)-EYFP (20298-AAV5), AAV5-EFla-DIO-EYFP (27056-AAV5), AAVl-syn-DIO-GCamP6s (100845-AAV1), AAV2-syn-Cre-P2A- dTomato (107738-AAV2), and AAV5-EFla-DIO-EGFPL10a (98747) were purchased from Addgene. AAV9-EFla-DIO-DTA (v62-9) was purchased from University of Zurich viral vector facility. AAVDJ8-EFla-DIO-NaChBac-EGFP was a gift from Benjamin Arenkiel (Baylor College of Medicine). Herpes simplex virus type 1 (HSV, H129-DTK- TT) and pseudorabies virus (PRV-263) were purchased from the Center for Neuroanatomy with Neurotropic Viruses (CNNV funded by NIH P40 ODO 1096).

[0169] Surgery

[0170] Mice were anesthetized with inhaled isoflurane (1-2%) and received bupivacaine (2 mg / kg, subcutaneous) at the site of surgical incision and meloxicam (5 mg / kg, subcutaneous) analgesia once daily for 3 days post-surgery. All mice were given at least two weeks for recovery prior to experimentation unless otherwise noted.

[0171] Viral injections and optical fiber implantation

[0172] All mice that received surgeries were Glplr-ires-Cre mice except where noted.

[0173] Viral expression specifically in GLP1R neurons was validated using in situ hybridization (Fig. 2 A). Mice received viral injections with or without optical fiber implantation with methods similar to those we have previously published71,72. Briefly, mice were anesthetized and placed in a stereotaxic frame. For central GLP1R neuron ablation, 100 nl of AAV5-EFla-DIO-taCasp3-TEVp was injected into the DVC (0.45 mm AP, ±0.15 mm ML, -0.18 mm DV from the obex) or the ARC (-1.35 mm AP, ±0.25 mm ML, -6.15 mm DV from bregma). For ablation of GLP1R neurons in the nodose ganglion (NG), we injected AAV9-EFla-DIO-DTA with 0.05% Fast Green FCF. Genes carried by the AAV5 serotype do not transfect in NG neurons. Because there is no available packaged Caspase with AAV9 serotype (which works well in NG neurons), and because DTA has been previously used to ablate NG neurons73, we used a different virus (AAV9-EFla-DIO- DTA) than that used in the ARC and DVC to ablate NGGLP1Rneurons.

[0174] For NG injection, an incision was made in the neck and the sternohyoid and stemomastoid muscle connective tissue was blunt dissected and moved aside to expose the carotid artery. The vagus nerve was gently separated and followeACd to reach the nodose ganglion as it enters the foramen. A glass micropipette was advanced into the ganglion, and the virus was injected. The procedure was repeated on the contralateral ganglion. We verified that both techniques (Casp3 and DTA) effectively ablated our neurons of interest (Fig. 2B, 2C).

[0175] For chemogenetic activation studies, 100 nl of AAV5-syn-DIO-hM3D-mCherry or AAV5-syn-DIO-EGFP was injected in the DVC, 20 nl of AAV2-syn-DIO-hM3D- mCherry was injected in the AP (0.45 mm AP, 0 mm ML, -0.16 mm DV from the obex) and the NTS (0.45 mm AP, ±0.30 mm ML, -0.2 mm 0 DV from the obex). For chronic neural activation studies, 100 nl of AAVDJ8-EFla-DIO-NaChBac-EGFP or AAV5-syn- DIO-EGFP was bilaterally injected in the DVC. To knock out the GLP1R, 100 nl of AAV2-syn-Cre-P2A-dTomato was bilaterally injected in the DVC, or 20 nl of the vector was injected in the AP or the NTS, of Glplr^fl or control C57BL / 6J mice.

[0176] For quantification of DVCGLP1Rneurons, and for whole-mount tissue confocal imaging, 100 nl of AAV5-EFla-DIO-EGFPL10a was bilaterally injected in the DVC. For optogenetic activation and optical fiber implantation, 100 nl of AAV5-EFla-DIO- hChR2(H134R)-EYFP or AAV5-EFla-DIO-EYFP was injected in the DVC. A 200 pm optical fiber (FT200UMT, Thorlabs) was placed above either the 1PBN (-1.20 mm AP, 1.60 mm ML, -3.40 mm DV from lambda) or PVH (-0.50 mm AP, 0.20 mm ML, -4.8 mm DV from bregma) and secured to the skull with dental cement.

[0177] For in vivo calcium imaging, 150 nl of AAVl-syn-DIO-GCamP6s was injected in the DVC (0.25 mm AP, ±0.15 mm ML, -4.15 mm DV from the occipital suture) of Glplr- ires-Cre, Ai9 mice. For anterograde tracing, 100 nl of HSV H129-ATK-TT (5 x 108vg / ml) was bilaterally injected in the DVC, and brains were collected 24 and 48 h after injection. For retrograde tracing, 150 nl of PRV-263 (lx 109vg / ml) was injected in either the 1PBN (-1.40 mm AP, ±0.16 mm ML, -3.60 mm DV from lambda) or PVH (-0.80 mm AP ±0.30 mm ML, -4.95mm DV from bregma), and brains were collected 36 h after injection.

[0178] Intraoral Cannula Surgery

[0179] Mice were implanted with a unilateral intraoral (IO) cannula consisting of polyethylene tubing (PE50CL100, Braintree Scientific), heat-flared to hold a Teflon washer (5612-120-25, Seastrom) on the proximal end and press fit with 23G stainless steel tubing on the distal end. The cannulas were implanted inside of the cheek, just lateral to the first maxillary molar, as we previously described in rats74. The distal end of the catheter was secured to the skull with dental cement. Mice received moistened chow for 2 days post-surgery before returning to maintenance on regular chow.

[0180] Immunohistochemistry, in situ hybridization, and imaging

[0181] Mice were transcardially perfused with PBS followed by 4% PF A, and brains were collected and post-fixed overnight in PF A. Brains were coronally sectioned (16-150 pm) on a vibratome or cryostat. For immunohistochemistry, brain sections were incubated with primary antibodies in 1% BSA and 0.1% Triton X-100 in PBS overnight at 4°C. On the second day, brain sections were washed with PBS 3 times, followed by incubation with secondary antibodies for 1 h at room temperature. Brain sections were washed with PBS 3 times, mounted on slides, and cover slipped with Fluoro-Gel. Antibodies used were sheep anti-EGFP (1 : 1000, 4745-1051, Bio-Rad), rabbit anti-RFP (1 : 1000, 600-401-379, Rockland), goat anti-sheep IgG Alex488 (1 :500, 713-545-147, Jackson ImmunoResearch), and goat anti-rabbit IgG Cy3 (1 :500, 711-165-152, Jackson ImmunoResearch).

[0182] For in situ hybridization, RNAscope Multiplex Fluorescent V2 Assay was performed based on the manufacturer’s protocol. Briefly, brain sections were pretreated with 4% PF A, an ethanol gradient, and protease III. Nodose ganglia were pretreated with an ethanol gradient and protease IV (20 min at room temperature). Treated sections were hybridized with Glplr, Calca, Mc4r, and / or Fos probes (418851-C3, 578771-C1, 319181- C3, and 316921-C2, respectively), followed by amplification and detection reagents. Lastly, sections were counterstained with DAPI and mounted on slides with Fluoro-Gel. Epifluorescence images were acquired with a slide scanner (BZ-X800, Keyence) with a 10X objective lens and confocal microscope (Stellaris 5, Leica, purchased with S100D030354, PI: Alhadeff) with a 20X objective lens. Images were analyzed (3 sections / mouse / experiment) using the spots function in Imaris 9.3 software (Oxford Instruments) to detect positive neurons based on the size / shape of neurons and the image background. We manually marked the subregions of the DVC based on anatomical landmarks using the surface function in Imaris. Colocalization was calculated using the filter function in Imaris.

[0183] Whole-mount tissue imaging

[0184] Brains were collected and post-fixed overnight in PF A. Brains were washed with PBS and a portion of the cerebellum was removed to expose the medulla. The tissue was incubated in a TDE gradient (10, 25, 47% in water) until equilibrium was reached. Cleared tissue was imaged from dorsal to ventral (480 pm) with a two-photon microscope (Ultima 2Pplus, Bruker) with a water-immersion objective lens (XLSLPLN25XSVMP2, Evident Scientific). Fluorophore-expressing neurons were detected with the spots function, and 3D structure was constructed using Imaris 9.3 software.

[0185] Experiments using GLP1R agonists

[0186] Effects of GLP1R neuron ablation on Exendin-4 (Ex-4) induced anorexia: Mice were fasted overnight and IP injected with saline, 5, 10, or 20 pg / kg of Ex-4. 15 min post-injection, mice were given access to food, and intake was measured at 0.5, 1, 2, 4, and 8h.

[0187] Effects of GLP1R neuron ablation on semaglutide-induced anorexia:

[0188] Mice were fasted overnight and SQ injected with vehicle (0.25% of DMSO in saline) or 10 pg / kg of semaglutide. 4 h post-injection, mice were given to access to food, and food intake was measured 4 h later.

[0189] Effects of GLP1R neuron inhibition on Ex-4-induced anorexia:

[0190] Mice were fasted overnight and IP injected with vehicle or 1 mg / kg CNO. 15 min after the first injection, mice were IP injected with saline or 20 pg / kg of Ex -4. 15 min after the second injection, mice were given access to food, and food intake was measured at 0.5, 1, 2, 4, and 8h.

[0191] Effects of GLP1R neuron ablation on semaglutide-induced prevention of weight gain:

[0192] Mice were maintained on 60% high-fat diet and received biweekly SQ injections of 12 pg / kg semaglutide. Weekly body mass, food intake, and body composition (measured by magnetic resonance, minispec, Bruker) were measured.

[0193] Effects of GLP1R neuron ablation on semaglutide-induced weight loss:

[0194] Mice were maintained on 60% high-fat diet (DI 2492, Research Diets) for 8 weeks, followed by SQ injection of 40 pg / kg semaglutide every 2 days for the one week and daily for the next 2 weeks. Body mass was measured daily.

[0195] Taste Reactivity Assay

[0196] Orofacial taste reactivity tests26were modified based on our protocol in rats74. A mirror was mounted under a chamber with a clear plexiglass floor at a -45° angle, and a digital video camera (Panasonic XI 500 4K) was positioned facing the mirror on a tripod ~35 cm away to capture food-evoked taste reactivity behaviors. Low-sodium chicken broth (50%, Pacific food) or saccharin (0.05%) was used as novel flavor and was infused (50 pl) via IO cannula at a rate of 0.1 ml / min. Each infusion was separated by 10-30 s. Mice were naive to the flavor on the first day. Taste reactivity habituation and testing were conducted in a cylindrical chamber with clear Plexiglas walls and floor. Video files were viewed offline using iMovie v. 10.2.2 and were slowed to 25% of the original speed for analysis. Responses were categorized into two affective reaction patterns - hedonic reactions and disgust-aversive reactions (Fig. 10). Hedonic reactions included rhythmic mouth movements (including rhythmic tongue protrusions), lateral tongue protrusions, and paw licks. Aversive reactions included gapes, chin rubs and facilitated fluid rejection (including face washing and forelimb flails with ejection of fluid). To give approximately equal weight to all taste reactivity responses, we counted the numbers of gapes, chin rubs, and lateral tongue protrusions, while the remaining continuous responses were quantified in time bins. Specifically, we measured the time spent (in seconds) engaging in rhythmic mouth movements, paw licks and facilitated fluid rejection, and to weight these equally with the other behaviors, we normalized each by a factor of 0.8. The composite score was calculated by summing all of the resulting measures.

[0197] Quinine-induced taste reactivity:

[0198] Mice received an IO infusion of water, 2.4 mM quinine, or 4 mM quinine in a counterbalanced order. Mice received 4 IO infusions of each stimuli, and video recordings were analyzed.

[0199] Cinacalcet or LiCl-induced real time taste reactivity:

[0200] To determine whether systemic cinacalcet or LiCl administration causes real-time taste aversion, 5 min prior to first IO infusion, mice received an IP injection of LiCl (6 mmol / kg in saline) or cinacalcet (15 pmol / kg in saline with 5% DMSO). Mice received an IO infusion every 5 min for the following 30 min. The experiment was run in a within- subject manner where each mouse received the control and experimental condition paired with either chicken stock or saccharin.

[0201] Cinacalcet-induced conditioned taste reactivity:

[0202] To determine whether cinacalcet produces a conditioned taste aversion, experimental and control mice received 8 IO infusions of a flavor, followed by IP injection of cinacalcet (15 pmol / kg in saline with 5% DMSO). Three conditioning sessions and one testing session were performed. In between each conditioning session, mice had a 24-h washout time, and the testing session was performed 48 h after the last conditioning session. Taste reactivity to the flavor stimulus was recorded and data from the first 4 infusions from testing (conditioned) sessions were analyzed. LiCl-induced taste reactivity:

[0203] To determine whether LiCl produces a conditioned taste aversion, experimental and control mice received 8 IO infusions of a flavor, followed by IP injection of LiCl (6 mmol / kg in saline). Two conditioning sessions and one testing session were performed, with a 24-h washout time in between each session. Taste reactivity to the flavor stimulus was recorded and data from the first 4 infusions from testing (conditioned) sessions were analyzed.

[0204] Chemogenetic GLP1R neuron activation-induced conditioned taste reactivity:

[0205] To determine whether DVCGLP1R, APGLP1R, or NTSGLP1Rneuron activation conditions taste aversion, experimental and control mice received 8 IO infusions of flavor, followed by IP injection of 1 mg / kg CNO. We waited 24 h between taste reactivity sessions to wash out effects of CNO. For each experiment, 2-3 conditioning sessions and 1 testing session were performed. Taste reactivity to the flavor stimulus was recorded and data from the first 4 infusions from the baseline and testing (conditioned) sessions were analyzed.

[0206] Real-time taste reactivity with GLP1R neuron optogenetic stimulation:

[0207] The axon terminals of DVCGLP1Rneurons at the 1PBN or the PVH were unilaterally optogenetically stimulated in experimental and control mice with 10 ms pulses (20 mW, 30 Hz, l / 5s on, 0.5s off)54while mice received 8 IO infusions of the novel flavor. Data from the last 4 IO infusions (during stimulation) were analyzed. See below for detailed methods for optogenetic stimulation.

[0208] Conditioned flavor avoidance (CFA) assay

[0209] CFA by chemogenetic activation of GLP1R neurons:

[0210] Mice with excitatory chemogenetic receptors in DVCGLP1R, APGLP1R, or NTSGLP1Rneurons, or control mice, were singly housed and acclimated to the testing chamber with 19-h water deprivation for 3 days. To condition the flavor to GLP1R neuron activation, water-deprived mice were allowed to drink 50% chicken broth for 20 min, followed by IP injection of 1 mg / kg CNO, for two conditioning days. On the test day, 20-min chicken broth intake was measured.

[0211] Effect of chemogenetic inhibition on semaglutide-induced CFA: Mice were singly housed and acclimated to the testing chamber with 19-h water deprivation for 3 days. Water-deprived mice were allowed to drink 50% chicken broth, followed by IP injection of 1 mg / kg CNO. 15 min post-injection of CNO, mice were SQ injected with 120 pg / kg semaglutide. To maintain the suppression of GLP1R neurons for the duration of semaglutide action, mice were IP injected with 1 mg / kg CNO every 12 h for 2 days. After two rounds of conditioning and an extra day for drug washout (where we verified that mice were no longer losing weight), 20-min chicken broth intake (CFA test) was measured.

[0212] Effect of GLP1R deletion on semaglutide-induced CFA:

[0213] Mice were singly housed and acclimated to the testing chamber with 19-h water deprivation for 3 days. Water-deprived mice were allowed to drink 50% chicken broth. After 20 min, mice were SQ injected with 120 pg / kg semaglutide. After two rounds of conditioning and an extra day for drug washout (where we verified that mice were no longer losing weight), 20-min chicken broth intake (CFA test) was measured.

[0214] In vivo optogenetic stimulation experiments

[0215] The output beam from a diode laser (450 nm, Lasever) was controlled by a microcontroller (Arduino Uno) running a pulse generation script. The laser was coupled to a multimode optical fiber (200 mm core, NA 0.37, Doric) with a 1.25 mm OD zirconium ferrule (Kientech) and mating sleeve that allowed delivery of light to the brain by coupling to the implanted ferrule-capped optical fiber in the mouse.

[0216] Effects of optogenetic stimulation of DVCGLPIRprojections on food intake:

[0217] Mice were fasted overnight and axon terminals of DVCGLPIR neurons in the 1PBN or PVH (or controls) were optogenetically activated with 10-ms pluses (20 mW, 30Hz, Is on, 4 s off)53for Ih, after which food intake was measured.

[0218] Effects of optogenetic stimulation of DVCGLPIRprojections on real-time avoidance:

[0219] Real-time place avoidance was performed in a two-chambered apparatus (ENV- 3013, Med Associates) in which one chamber was paired with optogenetic activation (10 ms pulses, 20 mW, 30 Hz, 1.5s on, 0.5 s off)54of DVCGLP1Rneurons in the 1PBN or PVH (or controls). Mouse position was tracked by Etho Vision XT system (Noldus), which triggered optogenetic stimulation when mice entered the appropriate side of the apparatus. Mice were tested both in the ab libitum and fasted states, with 10-min of no stimulation, followed by 20 min with stimulation, repeated for 3 days53.

[0220] Energy Expenditure and Meal Pattern Measurements

[0221] Mice were individually housed and acclimated to habituation monitoring chambers for 2 days prior to data collection. For chemogenetic activation of DVCGLP1R, APGLP1R, or NTSGLP1Rneurons, mice were fasted for 24 h and IP injected with 1 mg / kg CNO or vehicle (2% DMSO in saline) at the onset of dark period. For experiments with chronic DVCGLP1Ractivation (via NaChBac), mice were fed ad libitum. Food intake, physical activity, and O2 / CO2 exchange were monitored in metabolic chambers (PhenoMaster, TSE systems or Promethion, Sable Systems) and data were analyzed as we have previously published70. For meal pattern analyses, a meal was considered a bout of food intake >0.02 g. The termination of a meal was determined as >10 min with no measurable feeding70. Mice with excessive food grinding behaviors (<5% of all mice tested) were excluded from all analyses.

[0222] In vivo calcium imaging

[0223] For in vivo calcium imaging, 150 nl of AAVl-syn-DIO-GCamP6s was injected in the DVC (0.25 mm AP, ±0.15 mm ML, -4.15 mm DV from the occipital suture) of Glplr- ires-Cre ;Ai9(tdTomato) mice. After 4 weeks of recovery from stereotaxic surgery, mice were anesthetized with isoflurane (1-2%) to receive an intravenous (IV) catheter, an intraduodenal (ID) catheter, an ID exit port, and a cranial window above the DVC. An incision was made in the neck area, and the sternohyoid muscle and fatty tissue were moved aside to expose jugular vein. The superior end of jugular vein was ligated, and the jugular vein was nicked with 22G needle. A catheter (C20PU-MJV1458, Instech Laboratories) was inserted into the vein and tied with silk thread. An abdominal midline incision was performed to expose duodenum and stomach, and a duodenal catheter (MRE- 033, Braintree Scientific) was inserted through a puncture hole right below the pyloric sphincter and secured to the tissue with glue. For the intestinal exit port, the duodenum was truncated ~3 cm below the pyloric sphincter, and the intestinal exit port was left outside of the abdominal cavity.

[0224] Mice were placed on a custom-made platform to bend the head approximately 45 degrees downward. To expose the obex, a midline incision was made between the ears. The splenius capitis muscle was retracted and the cranial meninges were removed. The skull above the cerebellar lobules was carefully removed by Friedman-Pearson rongeurs (16220-14, Fine Science Tools). To avoid excess bleeding from bone cutting, hemostatic sponges (HY-80208, Hygitech) were applied at the bleeding site as needed. The lobules were pushed rostrally and anchored with a rolled Kimwipe. Artificial cerebrospinal fluid (59-7316, Harvard Apparatus) was applied to avoid drying. We used a stainless-steel column (3 mm in length, 50415K15, McMaster-Carr) attached to a #0 round coverslip (64- 0726, Warner Instruments) as a window that we placed on the top of the medulla, which we tilted 5-10 degrees downward along the rostral-caudal axis. This angle enabled simultaneous imaging of the AP and NTS across all z planes. To stabilize the window, a custom-made holder attached to a micromanipulator (MX10R, Siskiyou) was attached to the stainless-steel column and gently pressed downward against the surface of medulla, similar to the approach in Ran et al., 202256. To prevent respiratory arrest, blood flow in the vein of inferior cerebellar peduncle was monitored. The gap between the skull, the window, and the custom-made holder was filled with silicone elastomer (KWIK-SIL, World Precision Instruments), and the window was filled with water. The vein of inferior cerebellar peduncle was used as a landmark to locate the DVC under microscope (see Fig. 7A).

[0225] In vivo calcium images from anesthetized mice were acquired with a two-photon microscope (Ultima 2Pplus, Bruker) with a water immersion objective lens (XLSLPLN25XSVMP2, Evident Scientific). The laser (Insight X3 Dual, Spectra-Physics) was tuned to 860 nm (for GCaMP6s) and 1040 nm (for tdTomato, serving as structural channel for data analysis) with 30-60-mW laser power depending on the depth of imaging.

[0226] Volumetric imaging was performed by using Optotune ETL (6 focal planes, 30 pm apart, 1.71 Hz), and this enabled simultaneous recording of up to hundreds of DVCGLP1Rneurons (215 ± 29 neurons on average per mouse, ranging from 36-208 APGLP1Rneurons and 31-100 NTSGLP1Rneurons per mouse) per imaging session in a field of view 746.5 x 746.5 pm. To test the effect of semaglutide on DVCGLP1Rneuron activity, IV infusions were performed in two imaging sessions. First, 100 pl of vehicle (0.25% DMSO in saline, 0.05 ml / min IV) was infused. Second, 200 pl of semaglutide (60 pg / kg, 0.05 ml / min IV) was infused. Baseline recordings were taken for 30 s prior to each infusion. To determine whether the same cells were responsive to nutritive and aversive stimuli, we imaged neural activity responses to: 300 pl of Ensure (66.7%, 0.1 ml / min ID, Abbott), 100 pl of cinacalcet (0.5mg / kg, 0.05 ml / min IV, 0.25% DMSO in saline), and 100 pl of LiCl (150 mg / kg, 0.05 ml / min IV). Controls for these solutions were: saline infused 0.1 ml / min ID, and 0.25% DMSO in saline infused 0.05 ml / min IV. Baseline recordings were taken for 60 s prior to each infusion. Across experiments, imaging was paused in between sessions with multiple infusions in the same mice to prevent bleaching artifacts.

[0227] Two-photon imaging data were analyzed using the CalmAn package in Python77. The data were first motion-corrected using the rigid implementation of the NoRMCorre algorithm78. The algorithm was run for two iterations on data from a structural channel (Glplr-tdTomato) and the computed shifts were subsequently applied to the functional channel (GCaMP6s signal). Cellpose, a neural network-based algorithm trained to perform cell segmentation, was then used to identify regions of interest (ROIs) given a maximum projection of the functional channel data along the time axis79. Any non-neuronal ROIs were manually discarded. The activity of neurons associated with these ROIs was subsequently extracted using constrained non-negative matrix factorization (CNMF), wherein the ROIs acquired with Cellpose served as seeds for the neuronal spatial components of the model80. Calcium transients were modeled with a second order autoregressive process with a decay time of approximately 1.8 s to capture GCaMP6s kinetics. Neuropil activity generally manifests in two-photon calcium imaging as less localized fluctuations in background fluorescence. If unaccounted for, these background signals could contaminate signals originating from neurons of interest. Therefore, when fitting the CNMF, two additional spatial components were used to capture any background neuropil activity (the ‘nb’ parameter for the CNMF in CalmAn). These components generally have spatial footprints that cover a large portion of the FOV because, as previously mentioned, the activity they represent is less localized. The background-subtracted traces for each ROI were normalized by computing z-scores relative to the mean and standard deviation of data during a 30-s baseline period prior to stimulus onset. The ROIs with the median of z- score across stimulation period larger than 1.64 (which is statistically significant at a=0.05 for a one-tailed z-test) were considered responders81, and all stimulus-responsive neurons were included in the analyses. Prior to analysis of calcium activity (to prevent any bias), boundaries between the AP and NTS were drawn for each z plane using (1) the differential density of GLP1R neurons in AP vs. NTS which creates a natural boundary (density is much higher in AP compared to NTS, see FIG. 4B and FIG. 5 A and 6A) and (2) the red halo of tdTomato-expressing GLP1R processes that project into the NTS but not into the AP (see FIG. 6). To display average neural activity responses for each mouse, the previously-calculated median z-scores for each neuron were averaged together and graphed separately for APGLP1Rand NTSGLP1Rresponses. To visualize differential stimulus preferences for responsive ROIs, we subtracted the median z-scores for nutrient responses from the median z-scores for aversive stimuli to produce a difference index. These index values were mapped to dots at the location of each responsive ROI in FIG. 4L, 4P and FIG. 7H, with the neurons more responsive to aversive stimuli having a score of 10 and the neurons most responsive to nutritive stimuli having a score of -10.

[0228] Transcriptomics

[0229] Publicly-available transcriptomic data of single cells / nuclei from the mouse dorsal vagal complex produced by Zhang et al14, Ludwig et al17, and Adriaenssens et al18were combined using Seurat (v.4.1.3) in R (v.4.3.0)82’83. Count matrices of unique transcripts for each library were normalized individually prior to dimensionality reduction analysis. Neuronal clusters from each dataset were identified and subsetted based on canonical marker gene expression (Snap25, Syp, Rbfox3, Mapt, Map2, Nefl, Neftn, Nefh, Dlg4, Sy 11). The resulting neuronal datasets were re-normalized individually prior to merging and integration using canonical correlation analysis followed by mutual nearest-neighbor detection. The integrated dataset was scaled and uniform manifold approximation and projection (UMAP) was generated using the top 30 principal components from principal component analysis (PCA). Clustering was performed using the Louvain algorithm with resolution 1.4. Neuronal clusters were assigned to the AP or NTS based on the expression patterns of signature genes previously identified84and confirmed with RNA in situ hybridization data from the Allen Brain Atlas. Co-expression of select targets in Glplr neurons was analyzed and grouped by region ID using the Seurat DotPlot function. Data were visualized using ggplot2.

[0230] Statistical analyses

[0231] All data were expressed as mean ± S.E.M. unless otherwise noted. Paired or unpaired two-tailed t-tests; one-way, two-way, and repeated-measures ANOVA (with post hoc Bonferroni comparisons); Fisher’s exact test; and Pearson regression were performed where appropriate using Prism software. EXAMPLE 2: ANALYSIS OF HINDBRAND GLP1R NEURONS

[0232] Although there are GLPIR-expressing cells throughout the periphery (e.g., pancreas, kidney, gastrointestinal tract, vagus nerve, etc.) as well as across the brain6,7, peripherally-delivered GLP1 -based obesity drugs directly target neurons in the hindbrain dorsal vagal complex (DVC), arcuate nucleus of the hypothalamus (ARC), vagal afferents (nodose complex (DVC), arcuate nucleus of the hypothalamus (ARC), and / or vagal afferents (no dose ganglion, NG). It is thought that action on GLP1R in these regions contributes to food intake reduction8'10. However, the necessity of each of these neural populations (FIG. 1 A) in the anorexic and weight loss effects of obesity therapeutics has not been systematically tested.

[0233] To determine the contribution of these populations to GLP1R agonist-induced food intake suppression, we ablated each population by performing viral injections of Cre- dependent Caspase 3 (Casp3, for DVC and ARC neurons), Diphtheria toxin subunit A (DTA, for NG neurons), or a control virus (EGFP) in Glplr-ires-Cre mice (FIG. IB). These manipulations selectively ablated DVC, ARC, or NG GLP1R neurons, respectively, without influencing the other GLP1R populations (FIGs. 2A-2C). Ablation of only DVCGLP1Rneurons, but not the populations in the ARC or NG, completely blocked food intake suppression by both exendin-4 (exenatide, the first FDA-approved GLP1 -based drug; FIG. 1C, FIG. 2D) and semaglutide (the most effective and recently-approved GLP1 -based obesity drug; FIG. ID). Ablation of DVCGLP1Rneurons also blunted the longterm prevention of weight gain (FIG. 2E-2G) by semaglutide in mice maintained on a high-fat diet, an effect that was mediated by changes in food intake (FIG. 2h). Most importantly, DVCGLP1Rneuron ablation, but not ablation of ARCGLP1Ror NGGLP1Rneurons, blunted weight loss by semaglutide in obese mice (FIG. IE), demonstrating the necessity of DVCGLP1Rneurons for the efficacy of this obesity drug. Together, these data highlight the DVC as a critical site of action for GLPIR-mediated weight loss therapeutics.

[0234] We next tested how activation of DVCGLP1Rneurons influences feeding behaviour. Acute chemogenetic (hM3Dq) activation of DVCGLP1Rneurons potently suppressed food intake in food deprived mice (FIG. IF). To determine whether chronic activation of DVCGLP1Rneurons influences long-term energy balance, we injected Glplr-ires-Cre mice in the DVC with a Cre-dependent virus encoding NaChBac, a modified bacterial sodium channel which chronically upregulates neural activity11. Chronic activation of DVCGLP1Rneurons not only reduced body weight in lean, chow fed mice (FIG. 1G) but also prevented weight gain and fat mass accumulation in mice maintained on a high-fat diet (FIG. 1H). These effects were mediated by the suppression of food intake due to increased satiety - indicated by increased inter-meal interval and no change in meal size (FIG. 3 A- D) - and not by changes in energy expenditure, as acute or chronic activation of DVCGLP1Rneurons decreased total energy expenditure (FIG. 3E-3H). Together, these data demonstrate the power of hindbrain GLP1R neuron activity in reducing food intake and preventing weight gain in addition to being the critical set of drug-accessible neurons responsible for the efficacy of GLP1 -based obesity therapeutics.

[0235] EXAMPLE 3: ANALYSIS OF IN VIVO RESPONSE PROFILES OF APGLP1RAND NTSGLP1RNEURONS

[0236] Within the DVC, the area postrema (AP) and nucleus tractus solitarius (NTS) both express GLP1R12 13(FIG. 4A, 4B, and FIG. 5). These regions are both implicated in food intake control and nausea-like behaviours14'16. We therefore sought to determine potential differences in neural activity patterns of APGLP1Rand NTSGLP1Rneurons. We first created a 3D reconstruction of the neural populations after injecting Glplr-ires-Cre mice with a Cre-dependent virus expressing a soma-restricted fluorophore and imaging cleared brain tissue. This approach enabled clear visualization of the two major populations of GLP1R neurons within the AP and the NTS, as well as a much smaller, caudal -lateral population in the cuneate nucleus (FIG. 4B), indicating that we can anatomically distinguish subregions of the DVC in intact tissue. Therefore, we set out to monitor in vivo neural activity in APGLP1Rand NTSGLP1Rneurons in anesthetized mice.

[0237] We injected Glplr-ires-Cre ;Ai9 mice with a virus expressing a Cre-dependent genetically-encoded calcium indicator, GCaMP6s, and performed simultaneous two- photon imaging of the AP and NTS through a cranial window (FIG. 4C, 4D, FIG. 6, and FIG. 7A). We first monitored neural activity in response to semaglutide. Both the APGLP1Rand NTSGLP1Rneurons were significantly activated by semaglutide in comparison to vehicle administration (FIG. 4E-4H), and similar proportions of neurons were activated in the AP and NTS (FIG. 7B). Because GLP1R signaling causes both satiety and nausea / aversion, we next analysed responsivity to nutrients [Ensure, administered intraduodenally (ID)] or nauseogenic / aversive stimuli [cinacalcet or lithium chloride (LiCl), both administered IV], Cinacalcet activates the calcium sensing receptor, a GPCR that is expressed in the AP and NTS (FIG. 8E, 8F)14,17’18and causes nausea and aversion in humans and mice14,19'21.

[0238] The proportions and neural responses of GLP1R neurons that were activated by the nutritive or aversive stimuli were significantly different between the AP and NTS (FIG. 4I-4P, FIG. 7C-7I). When comparing calcium responses to nutrients with those to cinacalcet, 29% of responsive APGLP1Rneurons were activated by only nutrients, 65% were activated by only cinacalcet and 6% were activated by both stimuli (FIG. 41). Of responsive NTSGLP1Rneurons, 59% were activated by only nutrients, 35% were activated by only cinacalcet and 6% were activated by both stimuli (FIG. 41). These data indicate that APGLP1Rneurons are biased toward aversive stimuli while NTSGLP1Rneurons are biased toward nutritive stimuli. When comparing calcium responses to nutrients with those to LiCl, there was also a bias but only in the NTS. Of responsive APGLP1Rneurons, 47% were activated by only nutrients, 41% were activated by only LiCl and 12% were activated by both stimuli (FIG. 4M). Of responsive NTSGLP1Rneurons, 69% were activated by only nutrients, 19% were activated by only LiCl and 12% were activated by both stimuli (FIG. 4M). Average neural activity responses to nutritive and aversive stimuli were generally consistent across mice, with greater responses to nutrients in NTSGLP1Rneurons and greater responses to aversive stimuli in APGLP1Rneurons (FIG. 4J, 4K, 4N, and 40). Interestingly, across APGLP1Rand NTSGLP1Rneurons, there were significant subpopulations of nutrient-responsive and aversion-responsive neurons that were inhibited by the other stimulus (FIG. 7G, 71). In contrast, very few neurons were inhibited by control solutions (FIG. 4E, 4F and FIG. 7C, 7D) or semaglutide (FIG. 4E, 4F). Overall, these findings reveal that DVCGLP1Rneurons are broadly tuned to GLP1 -based obesity drugs, but that there are differences in subregion responses to nutritive and aversive stimuli, where NTSGLP1Rneurons are biased toward responsivity to nutritive over aversive / nauseogenic stimuli.

[0239] EXAMPLE 4: ANALYSIS OF HINDBRAIN ACTIVITY OF APGLP1RAND NTSGLP1RNEURONS GLP1 drug-induced nausea and anorexia co-occur, but it was previously unknown whether these features are intrinsically linked as a mechanism for food intake suppression and weight loss. Given the differences in APGLP1Rand NTSGLP1Rneuron responses to nutritive and aversive stimuli, we next tested whether neural activity in these populations differentially contributes to aversion. To accurately examine aversion responses, we designed an approach to measure hedonic and aversive orofacial taste reactivity while activating GLP1R neurons (FIG. 9A). As a positive control, mice were implanted with chronic intraoral cannulas and infused with quinine or solutions paired with cinacalcet or LiCl to characterize behavioural responses (FIG. 10A, 10B). Quinine-infused mice displayed fewer hedonic responses (i.e., rhythmic mouth movements, lateral tongue protrusions, and paw licks) and greater aversive responses (i.e., gaping, chin rubs, and facilitated fluid rejection) (FIG. 10C-10F). Similarly, pairing cinacalcet or LiCl to a novel hedonic flavour stimulus caused both real-time and conditioned aversive taste reactivity (FIG. 10G-10N).

[0240] To determine whether activation of DVCGLP1Rneurons triggers aversion, we injected Glplr-ires-Cre mice in the DVC with Cre-dependent hM3Dq and paired chemogenetic activation of DVCGLP1Rneurons with intraoral delivery of a novel flavour (FIG. 9A). After pairing, mice displayed fewer conditioned hedonic responses and increased conditioned aversive responses compared to control mice (FIG. 9B, FIG. 11 A- 1 IE). We complemented this result with data from the more commonly used conditioned flavour avoidance (CFA) assay (FIG. 9C). Activation of DVCGLP1Rneurons conditioned a flavour avoidance (FIG. 9D). We quantified the number of cells expressing hM3Dq in the DVC and found a negative correlation between the number of hM3Dq+ neurons in the AP, but not the NTS, and CS+ intake (measure of conditioned avoidance, FIG. 12A-12B). This finding supports our in vivo calcium imaging data, which together indicates that APGLP1Rneurons drive anorexia through inducing nausea / aversion and NTSGLP1Rneurons drive food intake inhibition through aversion- independent mechanisms.

[0241] To directly test this hypothesis, we developed a strategy to restrict viral expression to either the AP or the NTS neurons. We administered 20 nl injections of an AAV2.2 (which has limited diffusion compared to other serotypes) virus encoding Cre-dependent hM3Dq to either region. This enabled successful labelling of AP or NTS neurons without spread into neighbouring regions (FIG. 9E, FIG. 8A-8D). Activation of either APGLP1R (FIG. 9F) or NTSGLP1R(FIG. 9G) neurons inhibited food intake to the same magnitude (FIG. 8H), demonstrating that individual activation of either population is sufficient to drive food intake reduction. Further, the magnitude of food intake inhibition by either PGLpiRor NTSGLP1Rneurons was comparable to the effect of DVCGLP1Rstimulation (i.e., activating both APGLP1Ror NTSGLP1Rneurons, FIG. 13A-13B). Activation of NTSGLP1Rneurons, but not APGLP1Rneurons, increased inter-meal interval (FIG. 91) but had no effect on meal size (FIG. 13C), indicating that NTSGLP1Rneurons increase satiety. Of note, the effects of GLP1R neuron signalling on meal patterns also varied with acute vs. chronic stimulation. There were no differences between the effects of APGLP1Ror NTSGLP1Rneuron activation on energy expenditure (FIG. 13D-13F).

[0242] We next analysed the effects of APGLP1Rand NTSGLP1Rneuron signalling on aversion. Strikingly, only activation of APGLP1Rneurons, and not NTSGLP1Rneurons, reduced hedonic and increased aversive taste reactivity responses (FIG. 9 J, FIG. 13G- 13K). Similarly, APGLP1Rneuron activation, but not NTSGLP1Rneuron activation, was sufficient to condition a CFA (FIG. 9K). These data highlight NTSGLP1Rneurons as a population that robustly inhibits food intake without inducing aversion. Importantly, chronic activation of NTSGLP1Rneurons is sufficient to reduce body weight (FIG. 9L).

[0243] Finally, we combined and re-analysed three existing single cell / nucleus RNA sequencing datasets14,17,18to determine that both APGLP1Rand NTSGLP1Rneurons are mixed excitatory and inhibitory populations, with a greater proportion of NTSGLP1Rneurons being excitatory (FIG. 8E, 8F). This analysis also indicates that APGLP1Rand NTSGLP1Rneurons are largely distinct from other DVC cell types (e.g., CCK, Adacyapl, Tael, Gcg, Calcr, FIG. 8E, 8F). Together with physiological and behavioural results, these data indicate that APGLP1Rand NTSGLP1Rneurons are generally distinct from other hindbrain cell types and mediate aversion and satiety responses, respectively.

[0244] EXAMPLE 5: EFFECT OF OBESITY DRUGS IN THE ABSENCE OF APGLP1RNEURONS

[0245] To determine whether obesity drugs are effective at supressing food intake when the aversion pathway is inhibited, we treated mice with GLP1 -based obesity drugs and inhibited neutral activity in APGLP1Ror NTSGLP1Rneurons while measuring food intake and flavor avoidance. Chemogenetic inhibition (hM4Di) of APGLP1Rneurons, but not NTSGLP1Rneurons, blocked the CFA to a semaglutide-paired flavour (FIG. 14A, 14B). Importantly, however, inhibiting neural activity in either population did not block the food intake suppression by GLP1 drugs (FIG. 14C, 14D). These data demonstrate that the anorexic effect of GLP1 based-obesity drugs remains intact even when the aversion pathway is inhibited.

[0246] We next tested whether the expression of GLP1R in AP or NTS neurons is necessary for the effects of GLP1 -based obesity drugs on food intake and aversion. To achieve this, we deleted the GLP1R from the DVC, AP, or NTS by injecting Glplr^A mice with an AAV encoding Cre (FIG. 15A-15B). DVCGLP1Rdeletion attenuated both the anorexic and aversive effects of semaglutide (FIG. 15C-15G). Similar to our chemogenetic inhibition results, deleting GLP1R from the AP, but not the NTS, blocked the aversive responses to semaglutide (FIG. 14E). Neither APGLP1Rnor NTSGLP1Rdeletion affected exendin-4 or semaglutide-induced anorexia (FIG. 14F). These findings reveal that obesity drugs are still effective at suppressing food intake in the absence of nausea / aversion, suggesting that future drugs could be better targeted to improve satiety without aversion.

[0247] EXAMPLE 6: ANALYSIS OF INDEPENDENT MEDIATION OF SATIETY AND AVERSION

[0248] Our data demonstrate that APGLP1Rand NTSGLP1Rneurons independently drive aversion and satiety, respectively. To determine how this occurs, we first performed whole brain histological analysis after fluorophore (EYFP) labeling (FIG. 16A, FIG. 17A-17G) or H129-DTK-TT (tdTomato) anterograde tracing (FIG. 16B, FIG. 17H, 171) of DVCGLP1Rneurons. These data revealed ascending projections to only two regions - the lateral parabrachial nucleus (1PBN) and paraventricular hypothalamus (PVH) - in a remarkably distinct projection pattern compared to the more widespread projections of some other DVC subpopulations (e.g., TH29, Gcg / GLPl30, and POMC31).

[0249] To test whether APGLP1Rand NTSGLP1Rneurons have similar or divergent projections, we next injected a retrograde virus (PRV-263) into either the 1PBN or the PVH (FIG. 16C). This virus is taken up by axons and retrogradely transported by all cells to express tdTomato, and also expresses EYFP or mCerulean in a Cre-dependent manner. Analysis of DVC sections indicated that APGLP1Rand NTSGLP1Rneurons send largely separate projections to the 1PBN and PVH, respectively (FIG. 16C, FIG. 18), providing an anatomical basis for the functional differences between these GLPIR-expressing populations. To confirm that APGLP1Rand NTSGLP1Rneurons have distinct axon projections, we injected 20 nl of AAV2.2 virus encoding EYFP into the AP or NTS of Glplr-ires-Cre mice. This analysis revealed axon terminals in the 1PBN of APGLP1R- injected mice, and axon terminals in the PVH of NTSGLP1R-injected mice, with virtually no collateralization (FIG. 16D, 16E, FIG. 19). Chemogenetic stimulation of APGLP1Rand NTSGLP1Rneurons activated neurons in their respective downstream target regions: within the external 1PBN, the majority of activated (Fos+, 71%) neurons expressed calcitonin gene-related peptide [CGRP (Calca); FIG. 20A-20B]; and within the PVH, Fos+ cells (42%) expressed the melanocortin 4 receptor (MC4R; FIG. 20C ). 1PBNCGRPneurons suppress food intake and drive aversion, and PVHMC4Rneurons are known to mediate satiety. Therefore, APGLP1Rand NTSGLP1Rneurons send parallel projections to different downstream brain regions and cell types to mediate distinct anorexic behaviours.

[0250] To directly test the functional relevance of these divergent GLP1R projections, we injected Glplr-ires-Cre mice in the DVC with a Cre-dependent virus expressing channelrhodopsin-2 (ChR2) or control (EYFP) and implanted optic fibers above either the 1PBN or the PVH (FIG. 16F). Activation of either APGLP1R^1PBN or NTSGLP1R^PVH neurons significantly inhibited food intake in ChR2-expressing (FIG. 16G) but not EYFP- expressing (FIG. 21A-21B) mice. Taking advantage of the ability to rapidly and reversibly activate neural activity in APGLP1R— >1PBN or NTSGLP1R— >PVH neurons, we used a realtime place avoidance assay to test whether these projections mediate valence (FIG. 16H). Optogenetic activation of APGLP1R— >1PBN neurons, but not NTSGLP1R— >PVH neurons, significantly reduced time spent in the environment where mice received stimulation (FIG. 16H-16J, FIG. 21C-21J). Consistently, optogenetic activation of APGLP1R— >1PBN neurons, but not NTSGLP1R— >PVH neurons, reduced hedonic and increased aversive realtime taste reactivity to intraoral delivery of a flavour stimulus (FIG. 16K, FIG. 10G-10H, 10K-10L and FIG. 21K-21N). Overall, these data demonstrate the differential behavioural effects of APGLP1R— >PBN and NTSGLP1R— >PVH projections, and uncover NTSGLP1R— >PVH neurons as a neural target for obesity drugs to drive satiety without aversion.

[0251] The most promising obesity therapeutics - agonists of the GLP1R - cause nausea and vomiting, and it remains an open question whether the therapeutic and adverse effects of drugs like these can be functionally dissociated. Here, we demonstrate that hindbrain GLP1R neurons are the primary site of action for GLP1 -based obesity therapeutics, and reveal two anatomically and functionally distinct GLP1R projections - arising from the AP and the NTS - that drive aversion and satiety, respectively.

[0252] Previous work had identified GLPIR-expressing populations within the AP and the NTS, with recent single cell transcriptomics studies14,17,18’40confirming earlier histological reports12,13,41,42. We functionally dissociated APGLP1Rand NTSGLP1Rneurons in mice by developing strategies to independently manipulate and monitor in vivo neural activity in these populations. The current findings add to previous studies on hindbrain GLP1R signalling from our group and others43'47by highlighting a NTSGLP1R— >PVH circuit that is activated by nutrients and suppresses feeding without inducing aversion. This is especially significant given recent findings highlighting other NTS cell types (i.e., Tael15and ADCYAP148), which are largely distinct from NTSGLP1Rneurons, in mediating nausea-like or sickness behaviours. Importantly, we demonstrate that obesity drugs are fully effective at inducing satiety even when aversion circuitry is inhibited. This finding aligns with previous work showing that AP lesion does not block the satiating effects of GLP1R ligands44,49. In contrast, our findings challenge previous studies suggesting that GLP1- based therapeutics suppress food intake and body weight through action on distributed neural circuits including the hypothalamus10,49. Rather, our data indicates that drug- accessible GLP1R populations outside of the hindbrain (including the hypothalamus50) are dispensable for their therapeutic effects, a notion that is supported by earlier work in decerebrate rats51. Similarly, NTS serotonin 2C receptors (5HT2cR) are both necessary and sufficient for the anorexic effects of a previously-used obesity drug, lorcaserin52. These data emphasize the primacy of NTS signaling in the neural control of food intake and body weight, highlighting it as a target region for future development of weight loss pharmacotherapies.

[0253] We demonstrate that APGLP1Rneurons project to the 1PBN to transmit aversion, while NTSGLP1Rneurons project to the PVH without evoking aversion. Interestingly, NTSCCKneurons also project to the 1PBN and PVH, and NTSCCK— >1PBN projections cause negative valence but NTSCCK— >PVH projections do not53'55. Given that a minority of NTSGLP1Rneurons express CCK (FIG. 8E-8F), this pattern of parallel, functionally dissociable projections within the same cell type may reflect a more general organizational principle across NTS cell types that mediate satiety.

[0254] We developed a platform to monitor in vivo calcium dynamics simultaneously in APGLpiRand NTSGLP1Rneurons, and monitored responses to semaglutide as well as nutritive and nauseogenic / aversive stimuli. These data complement results from a recent paper demonstrating that gut signaling broadly activates NTS neurons, and importantly, provide in vivo evidence of hindbrain activation by GLPl-based obesity drugs. We also demonstrate that APGLP1Rand NTSGLP1Rneurons are differentially activated by nutritive and aversive stimuli. However, while the behavioural effects of APGLP1Rand NTSGLP1Rneuron activation were distinct, the physiological response properties of these populations were mixed in nature. Indeed, while APGLP1Rneurons generally had greater responses to aversive stimuli and NTSGLP1Rneurons had greater responses to nutritive stimuli, the neural tuning of these subregion responses was not absolute. The consistency in proportional APGLP1Rand NTSGLP1Rneural responses to these stimuli across mice point to the biased tuning of these regions, especially within the NTS. Moving forward, this imaging approach can be utilized to screen future potential obesity drugs for preferential activation of hindbrain satiety but not aversion circuits.

[0255] Our dissection of the neural circuitry mediating the satiety and aversion effects of GLP1- based obesity drugs demonstrates that distinct projections - APGLP1R— >1PBN and NTSGLP1R— >PVH neurons - mediate behaviourally-distinct anorexic effects. However, the finding that obesity drugs are still fully effective for anorexia, even in the absence of the aversion pathway emerging from APGLP1Rneurons, suggests that the intake-suppressive effects of these projections are not additive. In fact, activation of APGLP1Ror NTSGLP1Rneurons leads to the same magnitude of food intake suppression as activating both (i.e., activating all DVCGLP1Rneurons).

[0256] Furthermore, an extensive search of the literature has identified 192 differentially express genes in APGLP1Ror NTSGLP1Rneurons (FIG. 22). Monitoring expression levels of these genes in response to different compound would allow for the identification of additional therapeutics that prevent the aversive response to obesity therapeutics.

[0257] More broadly, this study demonstrates that anatomically and functionally dissociable neural circuits mediate different effects of the same drug. Because nausea and vomiting are side effects of thousands of treatments for human disease66, this concept could be applied to investigate other therapeutics - beyond obesity drugs - with the goal of developing drugs that better target mechanisms that eliminate pathologies and alleviate disease symptoms while avoiding negative side effects.

[0258] REFERENCES

[0259] 1. Wadden, T. A. et al. Effect of Subcutaneous Semaglutide vs Placebo as an Adjunct to Intensive Behavioral Therapy on Body Weight in Adults With Overweight or Obesity: The STEP 3 Randomized Clinical Trial. JAMA 325, 1403-1413 (2021). https: / / doi.org: 10.1001 / jama.2021.1831

[0260] 2. Wilding, J. P. H. et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. N Engl J Med 384, 989-1002 (2021). https: / / doi.org: 10.1056 / NEJMoa2032183

[0261] 3. Federation, W. O. World Obesity Atlas 2023.

[0262] 4. Jastreboff, A. M. et al. Triple-Hormone-Receptor Agonist Retatrutide for Obesity - A Phase 2 Trial. N Engl J Med 389, 514-526 (2023). https: / / doi.org: 10.1056 / NEJMoa2301972

[0263] 5. Jastreboff, A. M. et al. Tirzepatide Once Weekly for the Treatment of Obesity. N Engl J Med 387, 205-216 (2022). https: / / doi.org: 10.1056 / NEJMoa2206038

[0264] 6. Muller, T. D. et al. Glucagon-like peptide 1 (GLP-1). Mol Metab 30, 72- 130 (2019). https: / / doi.org: 10.1016 / j.molmet.2019.09.010

[0265] 7. Hayes, M. R., De Jonghe, B. C. & Kanoski, S. E. Role of the glucagon- like-peptide-1 receptor in the control of energy balance. Physiol Behav 100, 503-510 (2010). https: / / doi.org: 10.1016 / j.physbeh.2010.02.029

[0266] 8. Krieger, J. P. et al. Knockdown of GLP-1 Receptors in Vagal Afferents Affects Normal Food Intake and Glycemia. Diabetes 65, 34-43 (2016). https: / / doi.org: 10.2337 / dbl5-0973

[0267] 9. Kanoski, S. E., Fortin, S. M., Arnold, M., Grill, H. J. & Hayes, M. R. Peripheral and central GLP-1 receptor populations mediate the anorectic effects of peripherally administered GLP-1 receptor agonists, liraglutide and exendin-4. Endocrinology 152, 3103-3112 (2011). https: / / doi.org: 10.1210 / en.2011-0174

[0268] 10. Gabery, S. et al. Semaglutide lowers body weight in rodents via distributed neural pathways. JCI Insight 5 (2020). https: / / doi.org: 10.1172 / jci. insight.133429 11. Ren, D. et al. A prokaryotic voltage-gated sodium channel. Science 294, 2372-2375 (2001). https: / / doi.org: 10.1126 / science.1065635

[0269] 12. Jensen, C. B. et al. Characterization of the Glucagonlike Peptide-1 Receptor in Male Mouse Brain Using a Novel Antibody and In Situ Hybridization. Endocrinology 159, 665-675 (2018). https: / / doi.org: 10.1210 / en.2017-00812

[0270] 13. Merchenthaler, I., Lane, M. & Shughrue, P. Distribution of pre-pro- glucagon and glucagon-like peptide-1 receptor messenger RNAs in the rat central nervous system. J Comp Neurol 403, 261-280 (1999). https: / / doi.org: 10.1002 / (sici)1096- 9861(19990111 )403 :2<261 : : aid-cne8>3.0. co;2-5

[0271] 14. Zhang, C. et al. Area Postrema Cell Types that Mediate Nausea-Associated Behaviors. Neuron 109, 461-472 e465 (2021). https: / / doi. org: 10.1016 / j. neuron.2020.11.010

[0272] 15. Xie, Z. et al. The gut-to-brain axis for toxin-induced defensive responses. Cell 185, 4298-4316 e4221 (2022). https: / / doi.org: 10.1016 / j. cell.2022.10.001

[0273] 16. Grill, H. J. & Hayes, M. R. Hindbrain neurons as an essential hub in the neuroanatomically distributed control of energy balance. Cell Metab 16, 296-309 (2012). https: / / doi.org: 10.1016 / j. cmet.2012.06.015

[0274] 17. Ludwig, M. Q. et al. A genetic map of the mouse dorsal vagal complex and its role in obesity. Nat Metab 3, 530-545 (2021). https: / / doi.org: 10.1038 / s42255-021- 00363-1

[0275] 18. Adriaenssens, A. et al. Hypothalamic and brainstem glucose-dependent insulinotropic polypeptide receptor neurons employ distinct mechanisms to affect feeding. JCI Insight 8 (2023). https: / / doi.org: 10.1172 / jci. insight.164921

[0276] 19. Lindberg, J. S. et al. Cinacalcet HC1, an oral calcimimetic agent for the treatment of secondary hyperparathyroidism in hemodialysis and peritoneal dialysis: a randomized, double-blind, multicenter study. J Am Soc Nephrol 16, 800-807 (2005). https: / / doi.org: 10.1681 / ASN.2004060512

[0277] 20. Wuthrich, R. P., Martin, D. & Bilezikian, J. P. The role of calcimimetics in the treatment of hyperparathyroidism. Eur J Clin Invest 37, 915-922 (2007). https: / / doi.org: 10.1111 / j .1365-2362.2007.01874.x 21. Harrington, P. E. & Fotsch, C. Calcium sensing receptor activators: calcimimetics. Curr Med Chem 14, 3027-3034 (2007). https: / / doi.org: 10.2174 / 092986707782794096

[0278] 22. Kanoski, S. E., Rupprecht, L. E., Fortin, S. M., De Jonghe, B. C. & Hayes, M. R. The role of nausea in food intake and body weight suppression by peripheral GLP-1 receptor agonists, exendin-4 and liraglutide. Neuropharmacology 62, 1916-1927 (2012). http s : / / doi . org : 10.1016 / j . neuropharm .2011.12.022

[0279] 23. Pelchat, M. L., Grill, H. J., Rozin, P. & Jacobs, J. Quality of acquired responses to tastes by Rattus norvegicus depends on type of associated discomfort. J Comp Psychol 97, 140-153 (1983).

[0280] 24. Parker, L. A., Rana, S. A. & Limebeer, C. L. Conditioned nausea in rats: assessment by conditioned disgust reactions, rather than conditioned taste avoidance. Can J Exp Psychol 62, 198-209 (2008). https: / / doi.org: 10.1037 / a0012531

[0281] 25. Grill, H. J. Introduction: physiological mechanisms in conditioned taste aversions. Ann N Y Acad Sci 443, 67-88 (1985). https: / / doi.org: 10.1111 / j .1749- 6632.1985.tb27064.x

[0282] 26. Grill, H. J. & Norgren, R. The taste reactivity test. I. Mimetic responses to gustatory stimuli in neurologically normal rats. Brain Res 143, 263-279 (1978). https: / / doi.org: 10.1016 / 0006-8993(78)90568-1

[0283] 27. Watakabe, A. et al. Comparative analyses of adeno-associated viral vector serotypes 1, 2, 5, 8 and 9 in marmoset, mouse and macaque cerebral cortex. Neurosci Res 93, 144-157 (2015). https: / / doi.org: 10.1016 / j.neures.2014.09.002

[0284] 28. Wilson-Perez, H. E. et al. Vertical sleeve gastrectomy is effective in two genetic mouse models of glucagon-like Peptide 1 receptor deficiency. Diabetes 62, 23 SO- 2385 (2013). https: / / doi.org: 10.2337 / dbl2-1498

[0285] 29. Murphy, S. et al. Nucleus of the solitary tract A2 neurons control feeding behaviors via projections to the paraventricular hypothalamus. Neuropsychopharmacology (2022). https: / / doi . org: 10.1038 / s41386-022-01448-5

[0286] 30. Rinaman, L. Ascending projections from the caudal visceral nucleus of the solitary tract to brain regions involved in food intake and energy expenditure. Brain Res 1350, 18-34 (2010). https: / / doi.org: 10.1016 / j.brainres.2010.03.059 31. Wang, D. et al. Whole-brain mapping of the direct inputs and axonal projections of POMC and AgRP neurons. Front Neuroanat 9, 40 (2015). https: / / doi.org: 10.3389 / fnana.2015.00040

[0287] 32. Carter, M. E., Han, S. & Palmiter, R. D. Parabrachial calcitonin gene- related peptide neurons mediate conditioned taste aversion. J Neurosci 35, 4582-4586 (2015). https: / / doi.org:10.1523 / JNEUROSCI.3729-14.2015

[0288] 33. Campos, C. A., Bowen, A. J., Schwartz, M. W. & Palmiter, R. D. Parabrachial CGRP Neurons Control Meal Termination. Cell Metab 23, 811-820 (2016). https: / / doi.org: 10.1016 / j.cmet.2016.04.006

[0289] 34. Carter, M. E., Soden, M. E., Zweifel, L. S. & Palmiter, R. D. Genetic identification of a neural circuit that suppresses appetite. Nature 503, 111-114 (2013). https: / / doi.org: 10.1038 / naturel2596

[0290] 35. Campos, C. A., Bowen, A. J., Roman, C. W. & Palmiter, R. D. Encoding of danger by parabrachial CGRP neurons. Nature 555, 617-622 (2018). http s : / / doi . org : 10.1038 / nature25511

[0291] 36. Li, M. M. et al. The Paraventricular Hypothalamus Regulates Satiety and Prevents Obesity via Two Genetically Distinct Circuits. Neuron 102, 653-667 e656 (2019). https: / / doi.org: 10.1016 / j. neuron.2019.02.028

[0292] 37. Zhang, S. X. et al. Competition between stochastic neuropeptide signals calibrates the rate of satiation. bioRxiv (2023). https: / / doi.org: 10.1101 / 2023.07.11.548551

[0293] 38. Fenselau, H. et al. A rapidly acting glutamatergic ARC— >PVH satiety circuit postsynaptically regulated by alpha-MSH. Nat Neurosci 20, 42-51 (2017). https: / / doi.org: 10.1038 / nn.4442

[0294] 39. Spector, A. C., Breslin, P. & Grill, H. J. Taste reactivity as a dependent measure of the rapid formation of conditioned taste aversion: a tool for the neural analysis of taste- visceral associations. Behav Neurosci 102, 942-952 (1988). https: / / doi.org: 10.1037 / / 0735-7044.102.6.942

[0295] 40. Dowsett, G. K. C. et al. A survey of the mouse hindbrain in the fed and fasted states using single-nucleus RNA sequencing. Mol Metab 53, 101240 (2021). https: / / doi.org: 10.1016 / j .molmet.2021.101240 41. Cork, S. C. et al. Distribution and characterisation of Glucagon-like peptide-1 receptor expressing cells in the mouse brain. Mol Metab 4, 718-731 (2015). https: / / doi.org: 10.1016 / j.molmet.2015.07.008

[0296] 42. Goke, R., Larsen, P. J., Mikkelsen, J. D. & Sheikh, S. P. Distribution of GLP-1 binding sites in the rat brain: evidence that exendin-4 is a ligand of brain GLP-1 binding sites. Eur J Neurosci 7, 2294-2300 (1995). https: / / doi.org: 10.1111 / j .1460- 9568.1995.tb00650.x

[0297] 43. Alhadeff, A. L. & Grill, H. J. Hindbrain nucleus tractus solitarius glucagon- like peptide-1 receptor signaling reduces appetitive and motivational aspects of feeding. Am J Physiol Regul Integr Comp Physiol 307, R465-470 (2014). https: / / doi. org: 10.1152 / ajpregu.00179.2014

[0298] 44. Fortin, S. M. et al. GABA neurons in the nucleus tractus solitarius express GLP-1 receptors and mediate anorectic effects of liraglutide in rats. Sci Transl Med 12 (2020). https: / / doi.org: 10.1126 / scitranslmed.aay8071

[0299] 45. Hayes, M. R. et al. Intracellular signals mediating the food intakesuppressive effects of hindbrain glucagon-like peptide-1 receptor activation. Cell Metab 13, 320-330 (2011). https: / / doi.org: 10.1016 / j.cmet.2011.02.001

[0300] 46. Alhadeff, A. L. et al. Endogenous Glucagon-like Peptide-1 Receptor Signaling in the Nucleus Tractus Solitarius is Required for Food Intake Control. Neuropsychopharmacology 42, 1471-1479 (2017). https: / / doi.org: 10.1038 / npp.2016.246

[0301] 47. Costa, A. et al. Anorectic and aversive effects of GLP-1 receptor agonism are mediated by brainstem cholecystokinin neurons, and modulated by GIP receptor activation. Mol Metab 55, 101407 (2022). https: / / doi.org: 10.1016 / j.molmet.202L 101407

[0302] 48. Ilanges, A. et al. Brainstem ADCYAP1(+) neurons control multiple aspects of sickness behaviour. Nature 609, 761-771 (2022). https: / / doi.org: 10.1038 / s41586-022- 05161-7

[0303] 49. Secher, A. et al. The arcuate nucleus mediates GLP-1 receptor agonist liraglutide-dependent weight loss. J Clin Invest 124, 4473-4488 (2014). https: / / doi.org: 10.1172 / JCI75276

[0304] 50. Burmeister, M. A. et al. The Hypothalamic Glucagon-Like Peptide 1 Receptor Is Sufficient but Not Necessary for the Regulation of Energy Balance and Glucose Homeostasis in Mice. Diabetes 66, 372-384 (2017). https: / / doi.org: 10.2337 / dbl6- 1102

[0305] 51. Hayes, M. R., Skibicka, K. P. & Grill, H. J. Caudal brainstem processing is sufficient for behavioral, sympathetic, and parasympathetic responses driven by peripheral and hindbrain glucagon-like-peptide-1 receptor stimulation. Endocrinology 149, 4059- 4068 (2008). https: / / doi.org: 10.1210 / en.2007-1743

[0306] 52. D'Agostino, G. et al. Nucleus of the Solitary Tract Serotonin 5-HT(2C) Receptors Modulate Food Intake. Cell Metab 28, 619-630 e615 (2018). https: / / doi.org: 10.1016 / j.cmet.2018.07.017

[0307] 53. D'Agostino, G. et al. Appetite controlled by a cholecystokinin nucleus of the solitary tract to hypothalamus neurocircuit. Elife 5 (2016). https: / / doi.org: 10.7554 / eLife.12225

[0308] 54. Roman, C. W., Sloat, S. R. & Palmiter, R. D. A tale of two circuits: CCK(NTS) neuron stimulation controls appetite and induces opposing motivational states by projections to distinct brain regions. Neuroscience 358, 316-324 (2017). https: / / doi.org: 10.1016 / j. neuroscience.2017.06.049

[0309] 55. Cheng, W. et al. Calcitonin Receptor Neurons in the Mouse Nucleus Tractus Solitarius Control Energy Balance via the Non-aversive Suppression of Feeding. Cell Metab 31, 301-312 e305 (2020). https: / / doi.org: 10.1016 / j. cmet.2019.12.012

[0310] 56. Ran, C., Boettcher, J. C., Kaye, J. A., Gallori, C. E. & Liberies, S. D. A brainstem map for visceral sensations. Nature 609, 320-326 (2022). https: / / doi.org: 10.1038 / s41586-022-05139-5

[0311] 57. Betley, J. N., Cao, Z. F., Ritola, K. D. & Sternson, S. M. Parallel, redundant circuit organization for homeostatic control of feeding behavior. Cell 155, 1337-1350 (2013). https: / / doi.org: 10.1016 / j.cell.2013.11.002

[0312] 58. Atasoy, D., Betley, J. N., Su, H. H. & Sternson, S. M. Deconstruction of a neural circuit for hunger. Nature 488, 172-177 (2012). https: / / doi.org: 10.1038 / naturel 1270

[0313] 59. Wu, Q., Boyle, M. P. & Palmiter, R. D. Loss of GABAergic signaling by AgRP neurons to the parabrachial nucleus leads to starvation. Cell 137, 1225-1234 (2009). https: / / doi.org: 10.1016 / j. cell.2009.04.022

[0314] 60. Wu, Q., Clark, M. S. & Palmiter, R. D. Deciphering a neuronal circuit that mediates appetite. Nature 483, 594-597 (2012). https: / / doi.org: 10.1038 / naturel 0899 61. Dickson, S. L. et al. The glucagon-like peptide 1 (GLP-1) analogue, exendin-4, decreases the rewarding value of food: a new role for mesolimbic GLP-1 receptors. J Neurosci 32, 4812-4820 (2012). https: / / doi.org: 10.1523 / JNEUROSCI.6326- 11.2012

[0315] 62. Egecioglu, E. et al. The glucagon-like peptide 1 analogue Exendin-4 attenuates alcohol mediated behaviors in rodents. Psychoneuroendocrinology 38, 1259- 1270 (2013). https: / / doi.org: 10.1016 / j.psyneuen.2012.11.009

[0316] 63. Sorensen, G. et al. The glucagon-like peptide 1 (GLP-1) receptor agonist exendin-4 reduces cocaine self-administration in mice. Physiol Behav 149, 262-268 (2015). https: / / doi.org: 10.1016 / j.physbeh.2015.06.013

[0317] 64. Tuesta, L. M. et al. GLP-1 acts on habenular avoidance circuits to control nicotine intake. Nat Neurosci 20, 708-716 (2017). https: / / doi.org: 10.1038 / nn.4540

[0318] 65. Schmidt, H. D. et al. Glucagon-Like Peptide-1 Receptor Activation in the Ventral Tegmental Area Decreases the Reinforcing Efficacy of Cocaine. Neuropsychopharmacology 41, 1917-1928 (2016). https: / / doi.org: 10.1038 / npp.2015.362

[0319] 66. Wickham, R. J. Nausea and Vomiting: a Palliative Care Imperative. Curr Oncol Rep 22, 1 (2020). https: / / doi.org: 10.1007 / sl 1912-020-0871-6

[0320] 67. Williams, E. K. et al. Sensory Neurons that Detect Stretch and Nutrients in the Digestive System. Cell 166, 209-221 (2016). https: / / doi.org: 10.1016 / j. cell.2016.05.011

[0321] 68. Madisen, L. et al. A toolbox of Cre-dependent optogenetic transgenic mice for light-induced activation and silencing. Nat Neurosci 15, 793-802 (2012). https: / / doi.org: 10.1038 / nn.3078

[0322] 69. Sisley, S. et al. Neuronal GLP1R mediates liraglutide's anorectic but not glucose-lowering effect. J Clin Invest 124, 2456-2463 (2014). https: / / doi.org: 10.1172 / JCI72434

[0323] 70. Huang, K. P. et al. Sex differences in response to short-term high fat diet in mice. Physiol Behav 221, 112894 (2020). https: / / doi.org: 10.1016 / j .physbeh.2020.112894

[0324] 71. Alhadeff, A. L. et al. A Neural Circuit for the Suppression of Pain by a Competing Need State. Cell 173, 140-152 el 15 (2018). https: / / doi.org: 10.1016 / j.cell.2018.02.057 72. Goldstein, N. et al. Hypothalamic detection of macronutrients via multiple gut-brain pathways. Cell Metab 33, 676-687 e675 (2021). https: / / doi.org: 10.1016 / j.cmet.2020.12.018

[0325] 73. Min, S. et al. Arterial Baroreceptors Sense Blood Pressure through Decorated Aortic Claws. Cell Rep 29, 2192-2201 e2193 (2019). https: / / doi.org: 10.1016 / j.celrep.2019.10.040

[0326] 74. Ghidewon, M. et al. Growth differentiation factor 15 (GDF15) and semaglutide inhibit food intake and body weight through largely distinct, additive mechanisms. Diabetes Obes Metab 24, 1010-1020 (2022). https: / / doi.org: 10. I l l 1 / dom.14663

[0327] 75. Wyvell, C. L. & Berridge, K. C. Intra-accumbens amphetamine increases the conditioned incentive salience of sucrose reward: enhancement of reward "wanting" without enhanced "liking" or response reinforcement. J Neurosci 20, 8122-8130 (2000). https: / / doi.org:10.1523 / JNEUROSCI.20-21-08122.2000

[0328] 76. Berridge, K. C. & Pecina, S. Benzodiazepines, appetite, and taste palatability. Neurosci Biobehav Rev 19, 121-131 (1995). https: / / doi.org: 10.1016 / 0149- 7634(94)00026-w

[0329] 77. Giovannucci, A. et al. CalmAn an open source tool for scalable calcium imaging data analysis. Elife 8 (2019). https: / / doi.org: 10.7554 / eLife.38173

[0330] 78. Pnevmatikakis, E. A. & Giovannucci, A. NoRMCorre: An online algorithm for piecewise rigid motion correction of calcium imaging data. J Neurosci Methods 291, 83-94 (2017). https: / / doi.org: 10.1016 / j.jneumeth.2017.07.031

[0331] 79. Stringer, C., Wang, T., Michaelos, M. & Pachitariu, M. Cellpose: a generalist algorithm for cellular segmentation. Nat Methods 18, 100-106 (2021). https: / / doi.org: 10.1038 / s41592-020-01018-x

[0332] 80. Pnevmatikakis, E. A. et al. Simultaneous Denoising, Deconvolution, and Demixing of Calcium Imaging Data. Neuron 89, 285-299 (2016). https: / / doi. org: 10.1016 / j. neuron.2015.11.037

[0333] 81. Tan, H. E. et al. The gut-brain axis mediates sugar preference. Nature 580, 511-516 (2020). https: / / doi.org: 10.1038 / s41586-020-2199-7 82. Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol 36, 411-420 (2018). https: / / doi.org: 10.1038 / nbt.4096

[0334] 83. Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573-3587. e3529 (2021). https: / / doi.org: 10.1016 / j. cell.2021.04.048

[0335] 84. Ludwig, M. Q., Todorov, P. V., Egerod, K. L., Olson, D. P. & Pers, T. H. Single-Cell Mapping of GLP-1 and GIP Receptor Expression in the Dorsal Vagal Complex. Diabetes 70, 1945-1955 (2021). https: / / doi.org: 10.2337 / dbi21-0003 Each and every patent, patent application, and publication, including publications listed herein and publicly available nucleic acid and amino acid sequences cited throughout the disclosure, is expressly incorporated herein by reference in its entirety. While this invention has been disclosed with reference to specific embodiments, it is apparent that other embodiments and variations of this invention are devised by others skilled in the art without departing from the true spirit and scope of the invention. The appended claims include such embodiments and equivalent variations.

Claims

WHAT IS CLAIMED IS:

1. A method of identifying a therapeutic compound that does not cause an aversive response, the method comprising:(a) contacting a test compound and a postrema neuron that expresses glucagon-like peptide- 1 receptor (APGLP1Rneuron);(b) contacting the test compound with a nucleus tractus solitarius neuron that expresses glucagon-like peptide 1 receptor (NTSGLP1Rneuron);(c) measuring a response of the APGLP1Rneuron to the test compound by comparing activity of the APGLP1Rneuron in the presence and the absence of the test compound; and(d) measuring a response of the NTSGLP1Rneuron to the test compound by comparing activity of the NTSGLP1Rneuron in the presence and the absence of the test compound; wherein said test compound does not cause an aversive response when the activity of the APGLP1Rneuron is not increased and the activity of the NTSGLP1Rneuron is not decreased in the presence of the test compound or is decreased by less than 50%.

2. The method of claim 1, further comprising(e) measuring the responses of the APGLP1Rneuron and the NTSGLP1Rneuron to a known aversive therapeutic compound; and(f) identifying test compounds that do not cause aversive responses based on the measured responses to the test compound and known aversive therapeutic compound.

3. The method of any one of claims 1 or 2, wherein the known aversive therapeutic compound is a glucagon-like peptide 1 receptor (GLP1R) agonist.

4. The method of any one of claims 1 or 2, wherein the activity of the APGLP1Rneuron is decreased.

5. The method of any one of claims 1 or 2, wherein the activity of the NTSGLP1Rneuron is the same.

6. The method of any one of claims 1 or 2 wherein the activity of the NTSGLP1Rneuron is increased.

7. The method of any one of claims 1 or 2, wherein the aversive response is nausea or vomiting.

8. A method for screening therapeutic agents for treating GLP1R agonist related aversive response in a patient in need thereof, the method comprising(a) expressing an area postrema neuron (AP) gene in a cell;(b) contacting the cell with a test compound;(c) determining the effect of the test compound on the expression level of the AP gene in comparison to that in the absence of the test compound; and(d) selecting the agents that can reduce or inhibit the expression levels of the genes as the therapeutic agent.

9. The method of claim 8, further comprising:(e) expressing a dorsal vagal complex (DVC) gene in a cell;(f) contacting the cell with the test compound;(g) determining the effect of the test compound on the expression level of the DVC gene in comparison to that in the absence of the test compound; and(h) selecting the test compounds that do not significantly reduce the expression levels of the gene as the therapeutic agent.

10. The method of claim 9, wherein the DVC gene is a gene that expresses at the same level, or a greater level, in untreated NTS neurons when compared to AP neurons.

11. The method of claim 8, wherein the cell expressing the AP gene is an APGLP1Rneuron.

12. The method of claim 91, wherein the cell expressing the DVC gene is an NTSGLP1Rneuron.

13. The method claim 8, wherein the AP gene is Erbb4, Kcnabl, Unc5d, Csmd3, Grm7, Tenm3, Cdkl4, Robo2, Mgat4c, Kctd8, Slc24a3, Ntrk3, Luzp2, Ncam2, Rora, Slc2al3, Neill, Grin2b, Meisl, Spockl, Rmst, Lhfpl3, Epha5, Dpyd, Ccdc85a, Sgcz,Sema6d, Dabl, Gabra2, Pcdhl5, Cdhl2, Cntn4, Cntn6, Hs3st5, 01fm3, Nell2, Sorcs3, Nkain2, Grin2a, Grm8, Mpped2, Aff2, Sorcsl, Pcsk2, Asic2, Sytl, Brinp3, Kirrel3 Gabrbl, Rbfoxl, Slc8alTrpc5, Thsd7b, Ammercrl, Grik4, Zfpm2, Cdh4, Sema3e, Rap 1 gap, Chnl, or Tmtc2.

14. The method claim 9, wherein the NTS gene is Fstl4, Ephbl, Col25al, Synpr, Cadps2, Gria3, Ndst3, Klhll, Mania, Hs3st2, Pcdhl lx, Ptchd4, Ldb2, Glra3, Adcy8, Teadl, Rab27b, Fat3, Aik, Htr2c, Esrrg, Pdelc, Dgkg, Mei4, Cdh8, Hs3st4, Slcl7a6, Kcnhl, Nrpl, Esrl, Ak5, Fut9, Btbdl INckapS, Calnl, Rbfox3, Cntnap4, Limal, Mrap2, or Cdh7.

15. A therapeutic compound identified by the method of any one of the preceding claims.

16. A method for treating an obesity-related aversive response in a patient in need thereof, the method comprising administering to the patient the therapeutic compound of claim 15.

17. The method of claim 16, wherein said patient does not have an aversive response to said therapeutic compound.

18. The method of claim 16, wherein the patient has fewer aversive responses when compared to a patient treated with the obesity therapeutic alone.

Images