A Lipopolysaccharide Mouse Model Mirrors Neuroinflammatory Transcriptional Signatures of Human AD, and the Glucagon-Like Peptide-1 Receptor Agonist Semaglutide Attenuates Neuroinflammation in this Model
Mette Q Ludwig1, Dylan M Belmont-Rausch1, Anna Secher2, Marie A Bentsen2, Stine Normann Hansen2, Anne-Mette Bjerregaard2, Kristoffer Niss2, Kristoffer L Egerod1, Charlotte T Hansen3, Kevin D Dalgaard2, Myrte Merkestein2, Dorte Holst2, Charles Pyke2, Franziska Wichern4, Joseph Polex-Wolf2, Tune H Pers1, Lotte B Knudsen2
1Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, 2Global Drug Discovery, Novo Nordisk A/S, 3Global Development, Novo Nordisk A/S, 4Gubra
Objective:
We aimed to investigate the molecular effects of semaglutide on lipopolysaccharide (LPS) induced hippocampal neuroinflammation in mice and further to compare human gene expression signatures in Alzheimer’s disease (AD) brain tissue with those induced by LPS.
Background:
Neuroinflammation is part of the pathophysiology in AD. AD neuroimaging studies highlight increased inflammation, while genome wide association studies indicate that many AD-associated genes are expressed in glial cells. The glucagon-like peptide-1 receptor agonist semaglutide is currently being investigated in two phase 3a randomized, placebo-controlled trials in people with early Alzheimer’s disease (evoke/evoke+).
Design/Methods:
Mice were treated daily with subcutaneous semaglutide (30nmol/kg) or vehicle for 28 days, and LPS or vehicle was administered on days 15-17. RNA-sequencing and immunohistochemistry were performed to assess transcriptional and morphological changes in the hippocampus at Day 19 and Day 28 (Day 2 and 11 post-LPS, respectively). RNA-sequencing analyses included differentially expressed gene analysis, weighted gene co-regulated network analysis, pathway analysis and gene set enrichment analysis on published human AD datasets.
Results:
Semaglutide treatment significantly decreased the area of ionized calcium binding adaptor molecule 1 (Iba1) positive microglia (quantitative assessment) and attenuated co-expressed inflammatory genes vs vehicle in LPS-treated mice (p<0.01) on Day 28. Pathway analysis was consistent with an overall dampening of inflammatory processes in mice treated with LPS and semaglutide. The mouse LPS model mirrored the transcriptional signatures identified in people with AD in data of human hippocampal subtype transcriptional signatures, suggesting that semaglutide could potentially alleviate homologous inflammatory molecular processes in humans.
Conclusions:
In an LPS-induced neuroinflammation mouse model, semaglutide reduced hippocampal neuroinflammation as measured by transcriptional changes and microglial area (Iba1). This could represent a novel mechanism whereby semaglutide may affect neuronal integrity and function, and thereby AD pathophysiology.