Single-cell transcriptional networks in differentiating preadipocytes suggest drivers associated with tissue heterogeneity

Alfred K. Ramirez, Simon N. Dankel, Bashir Rastegarpanah, Weikang Cai, Ruidan Xue, Mark Crovella, Yu‐Hua Tseng, C. Ronald Kahn, Simon KasifNature Communications 2020 · DOI

TL;DR

This paper uses single-cell RNA sequencing and a novel network analysis approach to reveal that human white adipocytes are not a uniform cell type, but rather consist of at least two distinct classes that emerge during differentiation. These differences are linked to metabolic phenotypes and are driven by a network of zinc-finger proteins, suggesting a deeper understanding of adipose tissue heterogeneity and its implications for metabolic diseases.

What Problem Does This Paper Address?

White adipose tissue is crucial for metabolic health, but its different depots have varying associations with disease risk. It's unclear if these differences are inherent to the precursor cells (preadipocytes) or arise later. Existing methods for studying adipose tissue heterogeneity are limited, especially in humans. This paper aims to dissect the intrinsic cellular heterogeneity of human white preadipocytes during their differentiation into mature adipocytes, seeking to identify the underlying molecular drivers of these differences and their link to metabolic function.

Methods

The study employs a multi-pronged approach:

  • Single-cell RNA Sequencing (scRNA-seq): This is the core technology used to profile gene expression at the individual cell level during adipogenesis.
  • PAGODA (Pathway and Gene Set Overdispersion Analysis): A method to identify potential cellular heterogeneity by analyzing gene expression patterns.
  • t-SNE (t-Distributed Stochastic Neighbor Embedding): A dimensionality reduction technique used to visualize the scRNA-seq data and identify distinct cell clusters.
  • Differential Gene Expression Analysis: Used to pinpoint genes that are significantly different between the identified cell clusters.
  • Systems Biology Network Approach: A novel algorithm is developed to integrate protein-protein interaction (PPI) networks with single-cell gene expression data. This approach models gene expression as a snapshot of underlying biological networks, aiming to identify modules of co-regulated genes.
  • Clonal Expansion and Phenotypic Profiling: Human preadipocyte cell lines were clonally expanded and their metabolic phenotypes (e.g., glucose uptake, oxygen consumption) were measured. This allowed for correlation of gene expression signatures with functional outcomes.

Datasets

  • Human abdominal subcutaneous preadipocytes: scRNA-seq data from primary human abdominal subcutaneous preadipocytes undergoing in vitro adipogenesis (obtained from Soumillon et al., 2014, deposited in GEO: GSE53638).
  • Clonally expanded human neck white subcutaneous preadipocytes: RNA-seq data from 35 clonally expanded cell lines, with phenotypic measurements (deposited in GEO: GSE128253).
  • Fibroblast differentiation dataset: scRNA-seq data from fibroblasts differentiating into adipocytes, chondrocytes, and osteocytes (GSE37521) was used for analysis of ZNF expression across different lineages.

Key Insights and Ideas

  • Adipocyte Heterogeneity is Intrinsic: The study strongly suggests that distinct classes of white adipocytes emerge during the differentiation process from preadipocytes, indicating that heterogeneity is not solely due to external factors or depot-specific influences.
  • Metabolic Signatures Differentiate Cell Classes: The identified cell clusters exhibit distinct gene expression profiles related to metabolic pathways, particularly those involved in glycolysis and protein metabolism, suggesting functional differences in energy utilization.
  • A Novel Zinc-Finger Protein Network: A key finding is the identification of a specific network of zinc-finger proteins (ZNFs) that is highly expressed in one of the preadipocyte subpopulations. This network appears to be inversely correlated with adipogenic markers, suggesting it might be involved in maintaining a pool of adipocyte precursors that are resistant to differentiation.
  • Systems Biology for Uncovering Hidden Networks: The integration of scRNA-seq with PPI networks proved crucial for uncovering the ZNF network, which might be missed by traditional bulk RNA-seq or simpler analysis methods. This highlights the power of network-based approaches for understanding complex biological systems.

Key Results

  • scRNA-seq analysis of differentiating human preadipocytes revealed at least two distinct cell clusters at various stages of differentiation.
  • These clusters showed significant differences in gene expression related to metabolic pathways, with one cluster exhibiting higher activity in genes related to oxygen consumption and protein metabolism.
  • The study identified a network of 30 genes enriched for KRAB domains and C2H2 zinc-finger regions, with a tightly connected subnetwork of ZNFs (ZNF264, ZNF490, ZNF587, ZNF714).
  • The expression of these ZNFs was inversely correlated with the adipogenic marker FABP4 during differentiation, suggesting a role in maintaining a pre-adipocyte state.
  • Phenotypic profiling of clonally expanded cell lines confirmed differences in glucose uptake and differentiation capacity between cell populations that mirrored the scRNA-seq clusters.

Limitations and Future Work

  • Human Tissue Focus: While the study uses human preadipocytes, the findings are primarily from in vitro differentiation. Future work could explore these heterogeneities directly in human adipose tissue depots.
  • ZNF Network Function: The precise functional role of the identified ZNF network in regulating adipogenesis and maintaining preadipocyte populations needs further experimental validation.
  • Disease Context: The study focuses on normal differentiation. Investigating how these ZNF networks and cell subpopulations are altered in disease states like diabetes and obesity is a critical next step.
  • Therapeutic Potential: The authors speculate that understanding these differentiation trajectories and the ZNF network could lead to therapeutic strategies for reprogramming specific differentiation paths to combat metabolic diseases.

Summary updated

Abstract

White adipose tissue plays an important role in physiological homeostasis and metabolic disease. Different fat depots have distinct metabolic and inflammatory profiles and are differentially associated with disease risk. It is unclear whether these differences are intrinsic to the pre-differentiated stage. Using single-cell RNA sequencing, a unique network methodology and a data integration technique, we predict metabolic phenotypes in differentiating cells. Single-cell RNA-seq profiles of human preadipocytes during adipogenesis in vitro identifies at least two distinct classes of subcutaneous white adipocytes. These differences in gene expression are separate from the process of browning and beiging. Using a systems biology approach, we identify a new network of zinc-finger proteins that are expressed in one class of preadipocytes and is potentially involved in regulating adipogenesis. Our findings gain a deeper understanding of both the heterogeneity of white adipocytes and their link to normal metabolism and disease.
BibTeX
@inproceedings{ramirez2020single,
  author = {Alfred K. Ramirez and Simon N. Dankel and Bashir Rastegarpanah and Weikang Cai and Ruidan Xue and Mark Crovella and Yu‐Hua Tseng and C. Ronald Kahn and Simon Kasif},
  title = {{Single-cell transcriptional networks in differentiating preadipocytes suggest drivers associated with tissue heterogeneity}},
  booktitle = {Nature Communications},
  year = {2020},
  doi = {10.1038/s41467-020-16019-9},
}

Loading related content...

Figure