Integrative Bioinformatic Analyses of Global Transcriptome Data Decipher Novel Molecular Insights into Cardiac Anti-Fibrotic Therapies.

Maximilian Fuchs,Thomas Thum,Robert Geffers,Annette Just,Lorenz A Kapsner,Meik Kunz,Fabian Philipp Kreutzer,Ke Xiao,Jan Fiedler,Christian Bogdan,Saskia Mitzka,Cesare M Terracciano,Filippo Perbellini,Hans-Ulrich Prokosch

doi:10.3390/ijms21134727

Abstract

Integrative bioinformatics is an emerging field in the big data era, offering a steadily increasing number of algorithms and analysis tools. However, for researchers in experimental life sciences it is often difficult to follow and properly apply the bioinformatical methods in order to unravel the complexity and systemic effects of omics data. Here, we present an integrative bioinformatics pipeline to decipher crucial biological insights from global transcriptome profiling data to validate innovative therapeutics. It is available as a web application for an interactive and simplified analysis without the need for programming skills or deep bioinformatics background. The approach was applied to an ex vivo cardiac model treated with natural anti-fibrotic compounds and we obtained new mechanistic insights into their anti-fibrotic action and molecular interplay with miRNAs in cardiac fibrosis. Several gene pathways associated with proliferation, extracellular matrix processes and wound healing were altered, and we could identify micro (mi) RNA-21-5p and miRNA-223-3p as key molecular components related to the anti-fibrotic treatment. Importantly, our pipeline is not restricted to a specific cell type or disease and can be broadly applied to better understand the unprecedented level of complexity in big data research.

Highlights

Bioinformatics is an emerging field in the big data era, offering a steadily increasing number of algorithms and analysis tools
We previously showed that therapeutic antagonism of miRNA-21 inhibited cardiac fibrosis in various cardiac disease models via modulating fibroblast ERK–MAP kinase activity [10,11]
Besides the use of non-coding RNA inhibitors, we recently demonstrated that the selected natural compounds bufalin and lycorine were effective in counteracting detrimental cardiac fibrosis in various animal models [14]

Summary

Introduction

Bioinformatics is an emerging field in the big data era, offering a steadily increasing number of algorithms and analysis tools. Database tools such as Expression Atlas [5], RNASeq-er API [6], and The Cancer Genome Atlas (TCGA) offer analyses including annotation, normalization, and differential expression analysis of public RNA-Seq data Such bioinformatics pipelines tend to be quite complicated as advanced statistics and programming tasks need to be combined, which remains challenging for researchers in experimental life sciences without background in bioinformatics. We observed an anti-fibrotic effect for a small molecular similar of the natural compound lycorine compared to DMSO in our ex vivo cardiac fibrosis model, whereas a similar of bufalin turned out to be ineffective. These fibroblasts differentiate into alpha-smooth muscle actin (αSMA)-expressing myofibroblasts (logFC = 1.3 after lyco-s) to stabilize the injured tissue [17] Both natural compounds, lycorine and bufalin, were previously selected due to their anti-proliferative and migration-inhibitory effects on cardiac fibroblasts [14]. Our findings underline the importance of in silico omics data analysis in preclinical experimental model systems to validate innovative therapeutics

Animal Experiments

Bioinformatics Transcriptome Analysis Approach

Luciferase Reporter Assay