Repository logo

Mapping the Feeding Circuitry in the Mouse and Human Brain

Change log


Dowsett, Georgina KC 


The prevalence of obesity and its associated comorbidities has significantly grown over the last 50 years, imposing a great burden on people’s health. Consequently, understanding the mechanisms by which energy intake and bodyweight are controlled is extremely important to help us to further improve prevention and treatment of obesity. Genetic studies highlight the brain as a critical centre for food intake control. In particular, the hypothalamus and the hindbrain play key roles in the integration of signals regarding the body’s energy status and coordinating a response to alter changes in food intake. In particular, the hypothalamic leptin-melanocortin pathway is a well characterised signalling pathway implicated in food intake control, with genetic disruption resulting in severe obesity. Further, glucagon-like peptide 1 (GLP-1) is a hormone secreted from the gut postprandially and acts to potentiate insulin secretion, as well as acting as a satiation signal via activation of GLP-1R in the brain. Current anti-obesity therapeutics include analogues of GLP-1 and have been shown to activate GLP-1R in the hypothalamus and hindbrain to mediate their weight loss effects, through a reduction of food intake. However, most information about the neural circuitry involved in food intake control comes from murine studies, with little information about the hypothalamic cellular architecture in humans. Therefore, this thesis aims to map the feeding circuitry in the mouse and human brain, using transcriptomic approaches.

To map the mouse hindbrain, we performed single nucleus RNA sequencing on hindbrain samples from mice fed ad libitum and mice subjected to an overnight fast. Here, we profile 16,034 single nuclei, and demonstrate the oligodendrocytes are transcriptionally sensitive to an overnight fast. We characterise cells expressing druggable targets for obesity treatments and highlight their neurotransmitter and neuropeptide properties. Additionally, we discuss the HypoMap, a single cell gene expression atlas of the mouse hypothalamus, created through the integration of 18 single cell, and single nucleus RNA sequencing datasets. Here, we demonstrate the heterogeneity of neurons and non-neuronal cells in the hypothalamus, and profile cells involved in the leptin melanocortin pathway.

Through the integration of single nucleus RNA sequencing and spatial transcriptomics, we generate a spatially resolved single cell gene expression atlas of the human hypothalamus, named the HYPOMAP. Here we review non-neuronal cells, and neurons involved in the leptin-melanocortin pathway and identify transcriptionally and spatially distinct populations of POMC neurons, and neurons expressing the MC3R and MC4R.

Using the mouse and human hypothalamic atlases, we identify and characterise GLP-1R expressing populations, describing similarities and differences between the mouse and human. We confirm co-expression of GLP1R with key transcripts of interest in the human hypothalamus using in situ hybridization.

Collectively, this project has characterised the cellular architecture of the mouse and human hypothalamus (and mouse hindbrain) using transcriptomic approaches, profiling cells involved in the neurocircuitry of food intake. We describe similarities and differences between mouse and human appetitive pathways and profile neurons which could be directly modulated by current anti-obesity therapeutics. These datasets will serve as key resources for the identification of novel therapeutic targets for treating obesity.





Yeo, Giles


hypothalamus, Metabolism, Obesity, single cell RNA-sequencing


Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
Biotechnology and Biological Sciences Research Council (2292908)