Integrated Bioinformatics Analysis for the Screening of Hub Genes and Therapeutic Drugs in Androgen Receptor-Positive TNBC

Qiao-Nan Guo, Pengjun Qiu, Qingzhi Yao, Jianpeng Chen, Jianqing LinDisease 2022 · DOI

TL;DR

This study leverages integrated bioinformatics approaches to identify key genes and potential therapeutic targets for Androgen Receptor-positive Triple-Negative Breast Cancer (AR+ TNBC). They pinpoint TFF1 as a crucial hub gene associated with poorer prognosis and identify two microRNAs (hsa-miR-520g-3p and hsa-miR-520h) that regulate it, suggesting their potential as diagnostic and therapeutic tools.

What Problem Does This Paper Address?

Triple-Negative Breast Cancer (TNBC) is a particularly aggressive and lethal subtype of breast cancer, often diagnosed late and with limited treatment options. While the role of the Androgen Receptor (AR) in TNBC is still being elucidated, some evidence suggests AR-positive TNBC may have a distinct clinical profile. The challenge lies in identifying reliable biomarkers for early detection, prognosis prediction, and novel therapeutic targets specifically for this AR+ TNBC subtype, which could lead to improved patient outcomes.

Methods

The researchers employed a multi-pronged bioinformatics strategy:

  • Weighted Gene Coexpression Network Analysis (WGCNA): This method was used to identify modules of genes that are highly correlated in their expression patterns. They specifically looked for modules associated with AR-positive TNBC.
  • Differential Gene Expression (DEG) Analysis: Using the limma package, they identified genes that were significantly up- or down-regulated between AR-positive TNBC and other TNBC subtypes.
  • Module-Trait Relationship Analysis: This step helped to connect the identified gene modules with clinical characteristics, such as AR status.
  • Protein-Protein Interaction (PPI) Network Analysis: Using the STRING database and Cytoscape, they constructed a network of interactions among the differentially expressed genes to identify "hub genes" – genes that are central to the network. The MCC algorithm was used for this selection.
  • Functional Enrichment Analysis (GO and KEGG): These analyses were performed on the identified DEGs to understand their biological processes, cellular components, molecular functions, and associated pathways.
  • Survival Analysis (Kaplan-Meier plotter): This was used to assess the prognostic value of the identified hub genes and candidate miRNAs.
  • miRNA Target Prediction (TargetScan and ENCORI): These databases were used to predict microRNAs that might regulate the identified hub genes.
  • Drug-Gene Interaction Database (DGIdb): This was used to identify potential therapeutic drugs that target the identified hub genes.

Datasets

  • GSE76124: A gene expression microarray dataset from the Gene Expression Omnibus (GEO) database, used for the primary analysis. This dataset included TNBC samples categorized by molecular subtypes, including AR-positive (LAR).
  • GSE167213: Another GEO dataset used for validation of TFF1 expression.
  • TCGA and GEO databases: Used for survival analysis.
  • starBase v3.0 project: Used for expression analysis of candidate miRNAs.

Key Insights and Ideas

  • AR+ TNBC as a distinct entity: The study reinforces the idea that AR-positive TNBC might represent a biologically and clinically distinct subtype, warranting specific investigation.
  • TFF1 as a critical player: The identification of TFF1 as a significantly upregulated hub gene associated with poorer survival in AR+ TNBC is a key finding, suggesting its potential as a prognostic biomarker.
  • miRNA-mediated regulation: The discovery that specific microRNAs (hsa-miR-520g-3p and hsa-miR-520h) are downregulated in breast cancer and regulate TFF1 opens avenues for understanding disease progression and developing targeted therapies.
  • Integrated approach power: The successful integration of WGCNA, DEG analysis, PPI networks, and survival analysis highlights the power of multi-omics and bioinformatics approaches for complex disease research.

Key Results

  • 88 DEGs identified: From the intersection of WGCNA-identified purple module genes and DEGs, 88 genes were selected for further analysis.
  • 10 hub genes identified: TFF1, FOXA1, ESR1, AGR2, TFF3, AGR3, GATA3, XBP1, SPDEF, and TOX3 were identified as hub genes, all found to be upregulated.
  • TFF1 linked to poor prognosis: Survival analysis revealed that TFF1 was the only hub gene significantly associated with lower survival rates in TNBC.
  • Two candidate miRNAs: hsa-miR-520g-3p and hsa-miR-520h were identified as key regulators of TFF1 and were found to be downregulated in breast cancer.
  • Two potential drugs: The Drug-Gene Interaction Database identified two small molecules (Afimoxifene and Raloxifene) as potential therapeutic targets for AR-positive TNBC.

Limitations and Future Work

The authors acknowledge several limitations:

  • Reliance on public datasets: The study primarily used existing datasets and did not involve in vivo or in vitro experiments.
  • Limited mechanistic insight: While they identified key genes and miRNAs, the detailed signaling pathways and regulatory mechanisms were not fully elucidated.
  • Need for validation: Further prospective research is needed to validate the diagnostic and prognostic value of TFF1, hsa-miR-520g-3p, and hsa-miR-520h, as well as to explore their precise roles and therapeutic potential.

Future work should focus on deeper molecular mechanism studies of the identified core genes and the establishment of experimental models to validate their roles in early detection, risk estimation, prognosis determination, and targeted treatment of AR-positive TNBC.

Summary updated

Abstract

As the most invasive and lethal subtype of breast cancer (BC), triple-negative breast carcinoma (TNBC) is of increasing interest. However, the androgen receptor (AR) still has an unclear role in TNBC. The current study is aimed at testing the diagnostic and therapeutic performance of novel biomarkers for AR-positive TNBC. The GSE76124 dataset was analyzed by combining WGCNA and other bioinformatics methods. Subsequently, function enrichment analysis was applied to identify the relationships between these differential expression genes (DEGs). Subsequently, the protein-protein interaction network was established, and the hub genes were identified by Cytoscape software. Eventually, the miRNA-hub gene modulate network was developed and the Drug-Gene Interaction Database (DGIdb) was applied to verify the potential drugs for AR-positive TNBC. In the current research, 88 DEGs in total were selected from the intersection of the purple module genes identified by WGCNA and limma package. TFF1, FOXA1, ESR1, AGR2, TFF3, AGR3, GATA3, XBP1, SPDEF, and TOX3 were selected as hub genes by the MCC method, which were all upregulated. The survival analysis suggested that TFF1 was the only one related to significant lower survival rate in TNBC. Ultimately, hsa-miR-520g-3p and hsa-miR-520h were found taking part in the regulation of TFF1, and 2 small molecules were identified as the potential targets for AR-positive TNBC treatment. As a result, our study suggested that hsa-miR-520g-3p, hsa-miR-520h, and TFF1 might have significant potential values for AR-positive TNBC diagnosis and prognosis prediction. TFF1, hsa-miR-520g-3, and hsa-miR-520h may serve as the novel therapeutic targets, and our findings offer further insights into the therapy of AR-positive TNBC.
BibTeX
@inproceedings{guo2022integrated,
  author = {Qiao-Nan Guo and Pengjun Qiu and Qingzhi Yao and Jianpeng Chen and Jianqing Lin},
  title = {{Integrated Bioinformatics Analysis for the Screening of Hub Genes and Therapeutic Drugs in Androgen Receptor-Positive TNBC}},
  booktitle = {Disease},
  year = {2022},
  doi = {10.1155/2022/4964793},
}
Figure