Background
Combined treatment based on anthracyclines, such as doxorubicin, epirubicin, and pirarubicin, is usually the standard regimen for the first-line treatment of breast cancer. It has definite therapeutic effect and is indispensable. However, cardiotoxicity is the most serious side effect of anthracyclines. Both clinical research and practical observations have shown that cardiotoxicity induced by anthracyclines is often progressive and irreversible [
1,
2]. Cardiac injury might occur even when anthracyclines are used for the first time. Therefore, it is particularly important to actively prevent the cardiotoxicity induced by anthracyclines. Currently, the therapeutic effect on patients with anthracycline-induced cardiotoxicity is highly limited. The current diagnostic methods and treatment levels still fail to lower the incidence of anthracycline-induced cardiotoxicity effectively. This is because anthracycline-induced cardiotoxicity is a complicated, multifactorial, and multistep biological process. The existing research achievements are not sufficient to fully reveal the mechanism of its incidence and development. It is important to identify methods for the early specific diagnosis and effective prognosis.
During recent years, the use of high-throughput methods to detect gene expression has become a common practice. Microarray chip technology can quantify tens of thousands of gene transcript information simultaneously. The gene expression omnibus Gene Expression Omnibus (GEO) is currently the largest public high-throughput molecular abundance expression database globally. It mainly stores gene expression data. The GEO allows researchers to upload, download, save, and retrieve different types of genomic data.
In the present study, we aimed to (1) analyze the cardiotoxicity-related genes after breast cancer chemotherapy in gene expression database and (2) carry out bioinformatic analysis to identify cardiotoxicity-related abnormal expressions, the biomarkers of such abnormal expressions, and the key regulatory pathways after breast cancer chemotherapy.
Discussion
The mechanism of anthracycline-induced cardiotoxicity has not been fully understood. The existing evidences reveal that it is directly related to the free radicals produced by anthracyclines [
7,
8]. Unlike their mechanism of anti-tumor activity, the main mechanisms of anthracycline-induced cardiotoxicity involve iron-mediated production of reactive oxygen species (ROS) and the promotion of myocardial oxidative stress. Anthracyclines chelate iron ions and trigger the generation of oxygen free radicals, especially the hydroxyl free radicals, resulting in lipid peroxidation of myocardial cell membrane, myocardial mitochondrial DNA damage, etc. [
9,
10]. The other mechanisms include the formation of toxic drug metabolites, inhibition of nucleotide and protein synthesis, release of vasoactive amines, suppression of specific gene expression, impairment of mitochondrial membrane binding, aggregation of creatine kinase activity, induction of apoptosis, interference of intracellular calcium homeostasis, changes in respiratory chain proteins, induction of nitric oxide synthase, enhancement of mitochondrial cytochrome C release, etc. [
11,
12]. Other studies have shown that anthracyclines can lead to myocardial cell damage, inducing cardiac mitochondrial diseases and impairment of mitochondrial DNA and the respiratory chain in chronic cardiomyopathy [
13‐
15]. However, a solution to this problem is yet to be identified. One of the main reasons behind this is that the mechanism of pathogenesis of anthracycline-induced diseases is not fully understood. Modern medical studies have shown that it is difficult to fully explain the origin and development of a disease using the results regarding a single gene or signaling pathway. High-throughput biological detection technology, or gene chip detection technology, can obtain a large amount of gene expression information simultaneously. We can understand the biological molecules in living cells and their interactions by analyzing such gene expression information through bioinformatic methods. This way, the key factors involved in the pathogenesis of diseases can be identified, which can serve as references for target design in clinical treatment [
16].
Recent studies have shown that miRNA is involved in the processes of regulating the growth and development, mechanical remodeling, and electrical remodeling of the heart and is closely related to heart diseases [
17]. The overexpression of miRNA-1 and miRNA-133 can inhibit cardiomyocyte hypertrophy [
18]. Up-regulated miRNA-21 due to myocardial emergency can lead to fibroblast proliferation and myocardial interstitial fibrosis via [
19]. miRNA-29, miRNA-133, and miRNA-30 have also been proved to be involved in cardiac fibrosis [
20]. Terentyev et al. [
21] showed that miRNA-1 can also act This enhances the release of calcium ions and promotes cardiac arrhythmogenesis. Wijnen et al. [
22] evaluated the role of miRNA in myocardial fibrosis using a specific miRNA transgenic mouse model and methods such as the loss-of-function and gain-of-function.
In the present study, we used the raw data of GSE40447 gene chip in the GEO database and adopted software packages to analyze differentially expressed miRNAs. The DGE analysis between group A and groups B and C showed that there were 30 intersection genes. With further analysis of the biological functions and signaling pathways, it was found that the main functions of the differentially expressed miRNA-mRNA included cell-cell adhesion, response to toxic substance, and lipid catabolic process. Their cellular components mainly included cytosol, nuclear speck, and cell-cell adhesion junction. The genes
VPS53, ITGB8, ZNF736, and
TTC26 play important roles in cholesterol transport in cardiac cells, mediating the interaction between cell-cell and cell-extracellular matrix, regulate transcription, and transport proteins [
23‐
25]. Among them
VPS53 and
PHLDB2, GPATCH2, ITGB8, ZNF736, TTC26 exhibited opposite expressions in groups A, B, and C. This suggests that
VPS53 and
PHLDB2, GPATCH2, ITGB8, ZNF736, TTC26 might have opposite effects on some processes of the disease. The molecular functions included cadherin binding involved in cell-cell adhesion and phospholipid binding. The signaling pathways found in the KEGG database showed that these target genes were mainly distributed in the hippocampus-dependent long-term potentiation and long-term depression pathways, intercellular gap junction pathway, gonadotropin-releasing hormone signal transduction pathway, and circadian entrainment pathway. The biological functions of miRNA were mainly realized through the regulation and control of their target genes. The analysis of the functions of these differentially expressed miRNAs revealed that hsa-miR-4638-3p, hsa-miR-5096, hsa-miR-4763-5p, hsa-miR-1273 g-3p, hsa-miR6192, hsa-miR-4726-5p, and hsa-miR-1273a have higher centrality. This indicates that these miRNA targets are extensive, including oncogenes, tumor suppressor genes, signal transduction genes, and cell cycle regulation genes. Therefore, target gene prediction can provide a theoretical basis for further target gene verification experiments and avoid blind obedience. Further analysis of the miRNA-target gene regulatory network also showed that the miRNAs such as hsa-miR-4638-3p and hsa-miR-1273 g-3p had the highest centrality to the surrounding target genes, suggesting that they are the core regulatory miRNAs. The PCR results were consistent with the predicted results. The results of clinical sample validation showed that there was a significant difference between hsa-miR4638-3p and hsa-1273 g-3p before and after anthracycline chemotherapy. According to GO annotation and signal pathway analysis, Rho Associated Coiled-coil Containing Protein Kinase 1(ROCK1) and Mitogen-Activated Protein Kinase 1(MAPK1) are two key genes in the pathway, which may regulate the cardiac toxicity of anthracycline drugs through TGF-β signaling pathway and adhesion signaling pathway. Rho-Associated Coiled-Protein Kinase (ROCK) has serine/threonine protein kinase activity. It is a Rho-binding protein associated with apoptosis, which is also the main molecule of the Rho family. The Ras Homolog C/Rho-Associated Coiled-Protein Kinase (RHO/ROCK) pathway plays an important role in mediating various cellular functions, including contraction, actin cytoskeleton, cell adhesion and movement, proliferation, cytokinesis, and gene expression, all of which are involved in the pathogenesis of cardiovascular disease [
26,
27]. RHO/ROCK specific inhibitors show promise in the prevention of heart disease. ROCK1 is activated after cardiomyocytes are treated with doxorubicin in vitro, and RHO/ROCK inhibitors can prevent doxorubicin cardiotoxicity [
28]. ROCK1 mediates autophagy dysregulation and apoptosis in adriamycin cardiotoxicity, and promotes cardiac remodeling and reverse dysfunction [
29]. MAPK1 activation provides a possible mechanism to prevent doxorubicin-induced cardiomyocyte injury [
30‐
32]. These findings suggested that whether specific miRNAs inhibitors or activators, specific pathway inhibitors or activators can be used to alleviate cardiotoxicity caused by anthracyclines; whether certain natural compounds can achieve heart protection by regulating specific miRNAs and target gene expression; Whether the detection of heart-specific miRNA in plasma can be used to identify patients with subclinical cardiotoxicity and prevent severe complications of anthracycline drugs. Therefore, miRNAs have considerable research value in cardiotoxicity caused by anthracyclines.
The present study had some limitations. Based on the results of bioinformatic analysis, we can predict numerous target genes. However, there are also some false positives; therefore, its accuracy needs further validation by molecular biology experiment, in order to identify the real target genes of each miRNA and to verify the exact role played by miRNA in the development and incidence of anthracycline-induced cardiac toxicity. In the verification experiment, The number of cases collected in this experiment is small, and the experimental results have certain limitations. The reliability of the data needs to be further expanded to confirm the sample size.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.