Multidimensional Single-Nuclei RNA-Seq Reconstruction of Adipose Tissue Reveals Adipocyte Plasticity Underlying Thermogenic Response

Abstract

Adipose tissue has been classified based on its morphology and function as white, brown, or beige / brite. It plays an essential role as a regulator of systemic metabolism through paracrine and endocrine signals. Recently, multiple adipocyte subtypes have been revealed using RNA sequencing technology, going beyond simply defined morphology but by their cellular origin, adaptation to metabolic stress, and plasticity. Here, we performed an in-depth analysis of publicly available single-nuclei RNAseq from adipose tissue and utilized a workflow template to characterize adipocyte plasticity, heterogeneity, and secretome profiles. The reanalyzed dataset led to the identification of different subtypes of adipocytes including three subpopulations of thermogenic adipocytes and provided a characterization of distinct transcriptional profiles along the adipocyte trajectory under thermogenic challenges. This study provides a useful resource for further investigations regarding mechanisms related to adipocyte plasticity and trans-differentiation.

Regulatory modules of human thermogenic adipocytes: functional genomics of large cohort and Meta-analysis derived marker-genes

“…By comparing the results of recent snRNA-seq. studies, all of them explored large tissue heterogeneity, but different factors have been highlighted to be determinants in the formation of the thermogenic phenotype [ 13 – 15 ]. When we investigated the network position based-scores of these identified factors to estimate their role in adipose thermogenicity, we found good concordances.…”

“…When we investigated the network position based-scores of these identified factors to estimate their role in adipose thermogenicity, we found good concordances. For example, in their single-nucleotide RNA-seq study, Biagi et al [ 13 ] identified 7 transcription factors (PPARG, ERG1, STAT3, BHLHE40, ESR1, CEBPD, PPARD) whose expression positively correlated with UCP1 expression and five (PPARG, AR, ESR1, GATA2, Trp63) that correlated with low UCP1. When compared to our results, AR ranks first in the number of bridges list among the 3705 proteins (expanded protein list) and our core proteins determined by network analysis, indicating that it may have potentially outstanding significance in the linkage of the proteins that make up the network.…”