Transcriptome analysis of changes associated with dexrazoxane or GDF15 pretreatment to protect against doxorubicin in human embryonic stem cell-derived cardiomyocytes

We recently showed in human embryonic stem cell-derived cardiomyocytes (hESC-CMs) that growth differentiation factor 15 (GDF15) attenuated doxorubicin (Dox)-induced toxicity and mitochondrial dysfunction with a similar efficacy to dexrazoxane, an approved cardioprotective agent.

We aim to perform RNA sequencing (RNA-Seq) to reveal gene expression changes associated with GDF15 or dexrazoxane pretreatment preceding Dox exposure in hESC-CMs.

hESC-CMs were cultured for 60–120 days, exhibited spontaneous beating, and expressed cardiac troponin protein in >90% cells. Dexrazoxane (Dex, 200µM, 1h) or GDF15 (300ng/ml, 24h) were applied to hESC-CMs as pretreatment before incubation with doxorubicin (Dox, 1.5µM) for 8h. Thereafter, total RNA was isolated, checked for integrity, and sequenced. RNA library was constructed using 1μg total RNA. Illumina NextSeq 500 was used for sequencing and 100bp paired-end reads were generated. Quality trimmed reads were mapped to the human genome (mm10) by Tophat2, and gene expression mapped reads of genes overlapping with gene annotations in Ensemble release 96 were calculated by Cufflinks. DESEQ2 was used for detecting differentially expressed gene (DEG) transcripts at 8h. Three biological replicates were used for each condition. Significant DEGs were defined as log2 fold change >1 or <-1, and p<0.05. Gene set enrichment was done using Gene Ontology analysis with FDR q≤0.05. Experimental groups were comparatively analysed: 1) Dox vs. Control, 2) GDF15 + Dox vs. Dox, and 3) Dexrazoxane + Dox vs. Dox. Select genes were validated by qPCR.

In Group 1 analysis, Dox significantly upregulated 796 genes and downregulated 1,561 genes relative to Control (Fig A). Genes involved in negative regulation of cell proliferation were upregulated, whereas those related to transcription were suppressed (Fig B). Proapoptotic genes, including TP53 and FAS, were increased by Dox, whereas antiapoptotic genes (MMP11, CXCL7, PIK3R1) and genes related to heart development (HAND2, IRX4, NKX2-5) and cellular response to fatty acid (CREB1) and insulin (IRS1) were decreased, suggesting compromised cardiac metabolism and repair. In Group 2 analysis, GDF15 upregulated 187 and downregulated 24 genes. Significantly increased DEGs were related to SMAD and prosurvival (FZD4, GDF6, TCF7, FZD2, GDF7, WNT4) signalling pathways (Fig C). In Group 3 analysis, 52 and 125 genes were upregulated and downregulated, respectively. Increased DEGs were those associated with proliferation (IL6R, NTRK2), whereas others related to growth suppression and apoptosis (MT2A, MT1E, MT1F) were decreased (Fig D). Prosurvival genes upregulated by GDF15 against Dox were validated by qPCR (Fig E).

Transcriptome analysis revealed known genes associated with Dox cytotoxicity and Dex-induced cytoprotection, and identified GDF15-upregulated downstream prosurvival genes that warrant future mechanistic studies.