Dulaglutide treatment reverses depression-like behavior and hippocampal metabolomic homeostasis in mice exposed to chronic mild stress

Abstract

Introduction: Treatment strategies for depression based on interventions for glucose and lipid metabolism disorders are receiving increasing attention. Investigating the mechanism of their antidepressant effect and exploring new diagnostic and therapeutic biomarkers have attracted increasing attention. Dulaglutide, a long-acting GLP-1 receptor agonist, has been reported to alleviate cognitive deficits and neuronal damage. However, the antidepressant effect of dulaglutide and, especially, the underlying mechanism are still poorly understood. In this study, we aimed to explore the underlying biomarkers of depression and potential modulatory targets of dulaglutide in chronic mild stress (CMS) mice.

Methods: Sixty mice were randomly divided into a control group (CON group), a CMS+Vehicle group (CMS+Veh group), a CMS+0.3 mg/kg dulaglutide group (Low Dula group), and a CMS+0.6 mg/kg dulaglutide group (High Dula group). Numerous behavioral tests, mainly the open field test, forced swimming test, and tail suspension test, were applied to evaluate the potential effect of dulaglutide treatment on anxiety- and depression-like behaviors in mice exposed to chronic stress. Furthermore, a liquid chromatography-tandem mass spectrometry-based metabolomics approach was utilized to investigate the associated mechanisms of dulaglutide treatment.

Results: Three weeks of dulaglutide treatment significantly reversed depressive-like but not anxiety-like behaviors in mice exposed to chronic stress for 4 weeks. The results from the metabolomics analysis showed that a total of 20 differentially expressed metabolites were identified between the CON and CMS+Veh groups, and 46 metabolites were selected between the CMS+Veh and High Dula groups in the hippocampus of the mice. Comprehensive analysis indicated that lipid metabolism, amino acid metabolism, energy metabolism, and tryptophan metabolism were disrupted in model mice that experienced depression and underwent dulaglutide therapy.

Conclusion: The antidepressant effects of dulaglutide in a CMS depression model were confirmed. We identified 64 different metabolites and four major pathways associated with metabolic pathophysiological processes. These primary data provide a new perspective for understanding the antidepressant-like effects of dulaglutide and may facilitate the use of dulaglutide as a potential therapeutic strategy for depression.

Keywords: depression; dulaglutide; liquid chromatography-tandem mass spectrometry; metabolomics.

FIGURE 1

Chronic mild stress (CMS) model was successfully established. (a) Schematic of our study design. (b) Changes in body weight during the entire experimental period (n = 13–15). (c,d) Time spent in the center (c) and total distance (d) in the OFT after CMS handling (n = 13–15). (e,f) Floating time (e) and latency to floating (f) in the FST after CMS handling (n = 13–15). Data are presented as the mean ± SEM, least significant difference (LSD) post hoc test. *p < .05, **p < .01, ***p < .001 versus the CON group.

FIGURE 2

Dulaglutide reverses depressive‐like behavior in stressed mice. (a,b) Time spent in the center (a) and total distance (b) in the open field test (OFT) after dulaglutide treatment (n = 13–15). (c,d) Floating time (c) and latency to floating (d) in the forced swimming test (FST) after dulaglutide treatment (n = 13–15). (e,f) Immobility time (e) and latency to immobility (f) in the tail suspension test (TST) after dulaglutide treatment (n = 13–15). Data were presented as the mean ± SEM, least significant difference (LSD) post hoc test, *p < .05, **p < .01, ***p < .001 versus the CON group. ^# p < .05, ^## p < .01, ^### p < .001 versus the CMS+Veh group.

FIGURE 3

Multivariate statistical analysis between CON, CMS + Veh, and High Dula groups. (a,b) Superimposed chromatograms of quality control (QC) samples in positive and negative modes (n = 6). (c) Partial least squares‐discriminate analysis (PLS‐DA) scores plot. (d,e) Orthogonal partial least‐squares‐discriminant analysis (OPLS‐DA) scores plot. (f) Statistical validation of the PLS‐DA model through 200× permutation testing. CMS, chronic mild stress; CON, control; CMS+Veh, CMS+Vehicle; High Dula, CMS+0.6 mg/kg dulaglutide.

FIGURE 4

Different metabolites in CON, CMS+Veh, and High Dula groups. (a,c) The different metabolites between CON/CMS+Veh group (a) and High Dula/CMS+Veh group (c) are shown by the volcano plot. Red indicates significantly up‐regulated metabolites, whereas blue indicates significantly down‐regulated metabolites after dulaglutide treatment. (b,d) Z‐Score analysis of significantly different metabolites by Top 20 VIP ranking for both CON/CMS+Veh group (b) and High Dula/CMS+Veh group (d). CMS, chronic mild stress; CON, control; CMS+Veh, CMS+Vehicle; High Dula, CMS+0.6 mg/kg dulaglutide; VIP, variable importance of projection.

FIGURE 5

Heatmap and metabolic pathway analysis in CON, CMS+Veh, and High Dula groups. (a) The different metabolites are shown by heatmap. Each column is a sample, each row is a metabolite, the color reflects the relative abundance of metabolites, and similar samples are clustered together. (b,c) Bubble chart of the metabolic pathway analysis of CON/CMS+Veh (b) and CMS+Veh/High Dula group (c). The p‐value and impact factor are represented by dot color and dot size, respectively. CMS, chronic mild stress; CON, control; CMS+Veh, CMS+Vehicle; High Dula, CMS+0.6 mg/kg dulaglutide.