Prediction of a potential drug target based on protein druggability for thyroid cancer

Yifei Yang, Bin Yu, Xiuxia Zhang, Yunhua ZhuResearch Square 2020 · DOI

Abstract

Thyroid cancer is a prevalent endocrine malignancy often requiring surgical intervention. This study aimed to identify novel drug targets and potential therapeutic agents through a multi-omics bioinformatics pipeline. By integrating gene co-expression networks, DNA methylation data, and protein druggability simulations, researchers identified TNIK as a high-potential target. Pathway analysis suggests that the MEK1/2 inhibitor Binimetinib could be repositioned for thyroid cancer treatment.

Introduction

Thyroid cancer (TC) incidence is rising globally. While many patients respond well to standard care, advanced or refractory cases necessitate the development of targeted therapies. Traditional drug discovery is resource-intensive; thus, bioinformatics-driven target identification and drug repositioning offer efficient alternatives. This study focuses on identifying genes that are not only differentially expressed and methylated but also "druggable"—possessing structural pockets suitable for small-molecule binding.

Methods

The study utilized RNA-seq and DNA methylation data from The Cancer Genome Atlas (TCGA), comparing 568 thyroid cancer samples against normal tissues.

  1. WGCNA: Weighted Gene Co-expression Network Analysis identified modules of co-regulated genes associated with the disease.
  2. Methylation Analysis: Differentially methylated genes (DMGs) were screened to identify epigenetic drivers, specifically looking for hypomethylation which often leads to gene overexpression.
  3. Druggability Prediction: The PockDrug-Server analyzed the 3D structures of candidate proteins to predict the probability of drug-binding pockets.
  4. Network Analysis: Protein-Protein Interaction (PPI) networks were constructed using STRING and Cytoscape to identify influential "hub" genes.
  5. Functional Enrichment: Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses identified the biological pathways involved.

Results

The WGCNA "red module" showed the strongest correlation with thyroid cancer (r = 0.51, p = 3e-38). By intersecting this module with DMGs, three genes were prioritized: HEY2, TNIK, and LRP4. All three were found to be hypomethylated in tumor tissues, suggesting epigenetic activation.

Structural analysis via PockDrug-Server determined that only TNIK (Traf2- and Nck-interacting kinase) was druggable. TNIK contained eight predicted pockets, with pocket P6 showing a high druggability probability of 0.92.

A PPI network centered on TNIK revealed interactions with 10 key genes. Functional enrichment showed these genes are primarily involved in the MAPK and Wnt signaling pathways, both critical for tumor proliferation and development. A search of the DrugBank database identified Binimetinib, an established MEK1/2 inhibitor, as a potential therapeutic agent capable of disrupting this signaling network.

Discussion and Conclusion

This in silico study identifies TNIK as a novel, druggable target in thyroid cancer. The association of TNIK with the MAPK and Wnt pathways underscores its role in oncogenesis. Furthermore, the identification of Binimetinib suggests a viable path for drug repositioning. While these bioinformatics findings are robust, experimental validation through in vitro and in vivo models is necessary to confirm the clinical efficacy of targeting TNIK with Binimetinib in thyroid cancer patients.

Summary updated

Abstract

<title>Abstract</title> <bold>Background:</bold> Thyroid cancer is a common endocrine malignancy; however, its treatment is still surgical. With the development and application of targeted therapy in cancer treatment, there are great development prospects in researching targeted drugs for thyroid cancer.<bold> </bold><bold>Methods:</bold> Differentially expressed mRNAs between thyroid cancerous tissue and normal thyroid tissues were screened from The Cancer Genome Atlas (TCGA) database. Using weighted gene coexpression network analysis (WGCNA) to build co-expression modules and combined with differentially methylated gene (DMG) analysis. The druggability was analyzed by PockDrug-Server. Due to drug repositioning to seek targeted drugs to treat thyroid cancer we constructed a protein-protein interaction (PPI) network, and screened out a drug target of thyroid cancer. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes pathway (KEGG) were used to analysis the protein enrichment of PPI network. <bold>Results:</bold> In the present study, the red module was significantly correlated with thyroid cancer. With DMG analysis, we screened out three genes: HEY2 , TNIK and LRP4 . These three genes were hypomethylation in tumors. The druggability based on PockDrug-Server predicted that only TNIK had protein pocket druggability. With PPI model for TNIK, there were ten genes interacted with TNIK. These genes were enriched in the MAPK and Wnt pathways, which are correlated with tumor proliferation, differentiation, and development. Upon searching for drugs against these 10 genes in Drugbank, it was determined that the targeted drug Binimetinib which is MEK1/2 inhibition. Therefore, we hypothesized that Binimetinib can be used as a targeted drug and TNIK can be regard as drug target for thyroid cancer therapy. <bold>Conclusion: </bold>Our research provides a bioinformatics method for screening drugs target and provides a theoretical basis for targeted therapy for thyroid cancer.
BibTeX
@inproceedings{yang2020prediction,
  author = {Yifei Yang and Bin Yu and Xiuxia Zhang and Yunhua Zhu},
  title = {{Prediction of a potential drug target based on protein druggability for thyroid cancer}},
  booktitle = {Research Square},
  year = {2020},
  doi = {10.21203/rs.3.rs-30875/v1},
}
Figure