Modelling the gene expression and the DNA-binding in the 3T3-L1 differentiating adipocytes

Maximizing the utility of public data

The human genome project galvanized the scientific community around an ambitious goal. Upon completion, the project delivered several discoveries, and a new era of research commenced. More importantly, novel technologies and analysis methods materialized during the project period. The cost reduction allowed many more labs to generate high-throughput datasets. The project also served as a model for other extensive collaborations that generated large datasets. These datasets were made public and continue to accumulate in repositories. As a result, the scientific community should consider how these data can be utilized effectively for the purposes of research and the public good. A dataset can be re-analyzed, curated, or integrated with other forms of data to enhance its utility. We highlight three important areas to achieve this goal in this brief perspective. We also emphasize the critical requirements for these strategies to be successful. We draw on our own experience and others in using publicly available datasets to support, develop, and extend our research interest. Finally, we underline the beneficiaries and discuss some risks involved in data reuse.

Mouse Embryonic Fibroblast Adipogenic Potential: A Comprehensive Transcriptome Analysis

Our understanding of adipose tissue has progressed from an inert tissue for energy storage to be one of the largest endocrine organs regulating metabolic homoeostasis through its ability to synthesize and release various adipokines that regulate a myriad of pathways. The field of adipose tissue biology is growing due to this association with various chronic metabolic diseases. An important process in the regulation of adipose tissue biology is adipogenesis, which is the formation of new adipocytes. Investigating adipogenesis in vitro is currently a focus for identifying factors that might be utilized in clinically. A powerful tool for such work is high-throughput sequencing which can rapidly identify changes at gene expression level. Various cell models exist for studying adipogenesis and has been used in high-throughput studies, yet little is known about transcriptome profile that underlies adipogenesis in mouse embryonic fibroblasts. This study utilizes RNA-sequencing and computational analysis with DESeq2, gene ontology, protein–protein networks, and robust rank analysis to understand adipogenesis in mouse embryonic fibroblasts in-depth. Our analyses confirmed the requirement of mitotic clonal expansion prior to adipogenesis in this cell model and highlight the role of Cebpa and Cebpb in regulating adipogenesis through interactions of large numbers of genes.

A Small Fraction of Progenitors Differentiate Into Mature Adipocytes by Escaping the Constraints on the Cell Structure

Differentiating 3T3-L1 pre-adipocytes are a mixture of non-identical culture cells. It is vital to identify the cell types that respond to the induction stimulus to understand the pre-adipocyte potential and the mature adipocyte behavior. To test this hypothesis, we deconvoluted the gene expression profiles of the cell culture of MDI-induced 3T3-L1 cells. Then we estimated the fractions of the sub-populations and their changes in time. We characterized the sub-populations based on their specific expression profiles. Initial cell cultures comprised three distinct phenotypes. A small fraction of the starting cells responded to the induction and developed into mature adipocytes. Unresponsive cells were probably under structural constraints or were committed to differentiating into alternative phenotypes. Using the same population gene markers, similar proportions were found in induced human primary adipocyte cell cultures. The three sub-populations had diverse responses to treatment with various drugs and compounds. Only the response of the maturating sub-population resembled that estimated from the profiles of the mixture. We then showed that even at a low division rate, a small fraction of cells could increase its share in a dynamic two-populations model. Finally, we used a cell cycle expression index to validate that model. To sum, pre-adipocytes are a mixture of different cells of which a limited fraction become mature adipocytes.

Curated gene expression dataset of differentiating 3T3-L1 adipocytes under pharmacological and genetic perturbations

The 3T3-L1 cell line is used as an adipocyte differentiation model for the analysis of genes specifically expressed during the differentiation course. This cell model has several applications in obesity and insulin resistance research. We built a data resource to model gene expression of differentiating and mature adipocytes in response to several drugs and gene manipulations. We surveyed the literature survey for microarray datasets of differentiating 3T3-L1 cell line sampled at one or more time points under genetic or pharmacological perturbations. Data and metadata were obtained from the gene expression omnibus. The metadata were manually curated using unified language across the studies. Probe intensities were mapped and collapsed to genes using a reproducible pipeline. Samples were classified into none, genetically or pharmacologically modified. In addition to the clean datasets, two aggregated sets were further homogenized for illustration purposes. The curated datasets are available as an R/Bioconductor experimental data package curatedAdipoArray. The package documents the source code of the data collection, curation and processing. Finally, we used a subset of the data to effectively remove batch effects and reproduce biological observations.

Database URL

https://bioconductor.org/packages/curatedAdipoArray