A comparative transcriptomic analysis of glucagon-like peptide-1 receptor- and glucose-dependent insulinotropic polypeptide receptor-expressing cells in the hypothalamus

OBJECTIVE: The hypothalamus is a key region of the brain implicated in homeostatic regulation, and is an integral centre for the control of feeding behaviour. Glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are incretin hormones with potent glucoregulatory function through engagement of their respective cognate receptors, GLP-1R and GIPR. Recent evidence indicates that there is a synergistic effect of combining GIP- and GLP-1-based pharmacology on appetite and body weight. The mechanisms underlying the enhanced weight loss exhibited by GIPR/GLP-1R co-agonism are unknown. Gipr and Glp1r are expressed in the hypothalamus in both rodents and humans. To better understand incretin receptor-expressing cell populations, we compared the cell types and expression profiles of Gipr- and Glp1r-expressing hypothalamic cells using single-cell RNA sequencing. METHODS: Using Glp1r-Cre or Gipr-Cre transgenic mouse lines, fluorescent reporters were introduced into either Glp1r- or Gipr-expressing cells, respectively, upon crossing with a ROSA26-EYFP reporter strain. From the hypothalami of these mice, fluorescent Glp1rEYFP+ or GiprEYFP+ cells were FACS-purified and sequenced using single-cell RNA sequencing. Transcriptomic analysis provided a survey of both non-neuronal and neuronal cells, and comparisons between Glp1rEYFP+ and GiprEYFP + populations were made. RESULTS: A total of 14,091 Glp1rEYFP+ and GiprEYFP+ cells were isolated, sequenced and taken forward for bioinformatic analysis. Both Glp1rEYFP+ and GiprEYFP+ hypothalamic populations were transcriptomically highly heterogeneous, representing vascular cell types, oligodendrocytes, astrocytes, microglia, and neurons. The majority of GiprEYFP+ cells were non-neuronal, whereas the Glp1rEYFP+ population was evenly split between neuronal and non-neuronal cell types. Both Glp1rEYFP+ and GiprEYFP+ oligodendrocytes express markers for mature, myelin-forming oligodendrocytes. While mural cells are represented in both Glp1rEYFP+ and GiprEYFP+ populations, Glp1rEYFP+ mural cells are largely smooth muscle cells, while the majority of GiprEYFP+ mural cells are pericytes. The co-expression of regional markers indicate that clusters of Glp1rEYFP+ and GiprEYFP+ neurons have been isolated from the arcuate, ventromedial, lateral, tuberal, suprachiasmatic, and premammillary nuclei of the hypothalamus. CONCLUSIONS: We have provided a detailed comparison of Glp1r and Gipr cells of the hypothalamus with single-cell resolution. This resource will provide mechanistic insight into how engaging Gipr- and Glp1r-expressing cells of the hypothalamus may result in changes in feeding behaviour and energy balance.

Selecting single cell clustering parameter values using subsampling-based robustness metrics.

BACKGROUND: Generating and analysing single-cell data has become a widespread approach to examine tissue heterogeneity, and numerous algorithms exist for clustering these datasets to identify putative cell types with shared transcriptomic signatures. However, many of these clustering workflows rely on user-tuned parameter values, tailored to each dataset, to identify a set of biologically relevant clusters. Whereas users often develop their own intuition as to the optimal range of parameters for clustering on each data set, the lack of systematic approaches to identify this range can be daunting to new users of any given workflow. In addition, an optimal parameter set does not guarantee that all clusters are equally well-resolved, given the heterogeneity in transcriptomic signatures in most biological systems. RESULTS: Here, we illustrate a subsampling-based approach (chooseR) that simultaneously guides parameter selection and characterizes cluster robustness. Through bootstrapped iterative clustering across a range of parameters, chooseR was used to select parameter values for two distinct clustering workflows (Seurat and scVI). In each case, chooseR identified parameters that produced biologically relevant clusters from both well-characterized (human PBMC) and complex (mouse spinal cord) datasets. Moreover, it provided a simple "robustness score" for each of these clusters, facilitating the assessment of cluster quality. CONCLUSION: chooseR is a simple, conceptually understandable tool that can be used flexibly across clustering algorithms, workflows, and datasets to guide clustering parameter selection and characterize cluster robustness.