Identification and validation a TGF-β-associated long non-coding RNA of head and neck squamous cell carcinoma by bioinformatics method

Teng Huang, Wei Huang, Hong Lu, Biyun Zhang, Jun Ma, Di Zhao, Yijun Wang, Dahai Yu, Xia HeJournal of Translational Medicine 2018 · DOI

TL;DR

This study identifies a novel long non-coding RNA (lncRNA), EPB41L4A-AS2, that is downregulated by TGF-β signaling in head and neck squamous cell carcinoma (HNSCC). EPB41L4A-AS2 appears to act as a tumor suppressor, inhibiting cell migration and invasion, and shows promise as a prognostic biomarker and therapeutic target for HNSCC.

What Problem Does This Paper Address?

Head and neck squamous cell carcinoma (HNSCC) is an aggressive cancer characterized by high rates of metastasis and recurrence. While the role of transforming growth factor-beta (TGF-β) in promoting tumor progression is well-established, its precise involvement in HNSCC metastasis, particularly through the action of long non-coding RNAs (lncRNAs), remains unclear. LncRNAs are emerging as critical regulators of various cellular processes, including cancer development, and understanding their interplay with signaling pathways like TGF-β is crucial for developing effective diagnostic and therapeutic strategies. This paper aims to identify and validate lncRNAs associated with TGF-β signaling in HNSCC and to elucidate their role in invasion and metastasis.

Methods

The researchers employed a multi-pronged bioinformatics and experimental approach:

  • Bioinformatics Analysis:
    • Gene Expression Profiling: They utilized publicly available datasets from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) to analyze lncRNA expression profiles in HNSCC.
  • TGF-β Pathway Analysis: They specifically looked for lncRNAs whose expression was altered by TGF-β signaling, using a dataset from hepatocellular carcinoma (GSE54800) as a proxy.
  • Invasion and Metastasis Signature: They identified lncRNAs differentially expressed in HNSCC patients with and without metastasis or recurrence.
  • Functional Enrichment Analysis: DAVID tools were used to predict the biological processes and pathways associated with the identified dysregulated lncRNAs.
  • ceRNA Network Construction: Tools like miRanda, miRTarBase, and TargetScan were used to predict lncRNA-miRNA-mRNA interactions, and Cytoscape was used to visualize the network.
  • RNA Binding Protein (RBP) Prediction: The RBP database was used to identify potential protein interactions of EPB41L4A-AS2.
  • Experimental Validation:
    • Quantitative Reverse Transcription-Polymerase Chain Reaction (qRT-PCR): Used to validate the expression levels of EPB41L4A-AS2 in HNSCC cell lines and patient samples.
  • Cell Culture and Transfection: HNSCC cell lines (CNE1) were cultured and transfected with EPB41L4A-AS2.
  • Epithelial-Mesenchymal Transition (EMT) Model: An in vitro EMT model was established by treating CNE1 cells with TGF-β1 to mimic the cellular changes associated with metastasis.
  • Wound Healing and Invasion Assays: These assays were performed to assess the effect of EPB41L4A-AS2 on cell migration and invasion capabilities.
  • Western Blot: Used to analyze protein expression levels of key EMT markers (Snail, vimentin, E-cadherin, N-cadherin) and components of the TGF-β signaling pathway (TGFBR1, Smad2, p-Smad2, Smad3, p-Smad3).

Datasets

The study utilized the following datasets:

  • TCGA (The Cancer Genome Atlas): Gene expression data and clinical data for HNSCC patients. Specifically, RNASeq datasets for 43 tumor-adjacent normal pairs and data for 127 HNSCC patients (39 with metastasis/recurrence, 88 without).
  • GEO (Gene Expression Omnibus):
    • GSE59652: RNASeq datasets for 7 tumor-adjacent normal pairs.
  • GSE54800: IncRNA expression profile related to TGF-β-induced invasion-metastasis in hepatocellular carcinoma (N=14).

No new benchmark datasets were introduced.

Key Insights and Ideas

  • EPB41L4A-AS2 as a TGF-β Target: The study reveals that EPB41L4A-AS2 is downregulated by TGF-β signaling, suggesting it's a direct or indirect target of this pathway in HNSCC.
  • Tumor Suppressor Role: Overexpression of EPB41L4A-AS2 significantly inhibited cell migration and invasion in vitro, pointing to its role as a tumor suppressor.
  • Inhibition of TGF-β Signaling Axis: EPB41L4A-AS2 was found to suppress TGFBR1 expression, a key receptor in the TGF-β pathway, and consequently inhibit downstream signaling components like p-Smad2/3 and Snail.
  • Prognostic Biomarker Potential: Lower expression of EPB41L4A-AS2 correlated with more advanced tumor stages (T, N, and overall stage) and poorer patient survival, highlighting its potential as a prognostic biomarker.
  • EMT Regulation: EPB41L4A-AS2 appears to counteract the pro-EMT effects induced by TGF-β, promoting epithelial characteristics (e.g., increased E-cadherin) and suppressing mesenchymal traits (e.g., decreased vimentin).

Key Results

  • TGF-β Upregulation in HNSCC: TGF-β expression was found to be significantly higher in HNSCC tissues compared to normal tissues.
  • Worse Survival with High TGF-β: Patients with high TGF-β expression exhibited significantly worse overall survival.
  • Dysregulated TGF-β-Associated lncRNAs: Three lncRNAs were identified as dysregulated by TGF-β in HNSCC: EPB41L4A-AS2 and LINC00515 were downregulated, while MIR4435-2HG was upregulated.
  • EPB41L4A-AS2 Downregulation: EPB41L4A-AS2 was downregulated in HNSCC tissues and further decreased in metastatic or recurrent tissues.
  • Correlation with Clinicopathological Features: Low EPB41L4A-AS2 expression was significantly correlated with increased N classification, T classification, tumor stage, and perineural invasion.
  • Survival Impact: High EPB41L4A-AS2 expression was associated with better overall survival and relapse-free survival.
  • Inhibition of Migration and Invasion: Overexpression of EPB41L4A-AS2 in an EMT model significantly reduced cell migration and invasion.
  • Suppression of TGFBR1: EPB41L4A-AS2 overexpression suppressed TGFBR1 expression, and a negative correlation was observed between EPB41L4A-AS2 and TGFBR1 in HNSCC tissues.

Limitations and Future Work

  • In Vitro vs. In Vivo: While the study established EPB41L4A-AS2's role in vitro, further in vivo studies are needed to fully elucidate its mechanisms and therapeutic potential.
  • Mechanism of TGF-β Regulation: The precise mechanism by which TGF-β downregulates EPB41L4A-AS2 expression requires further investigation. The authors hypothesize it might be a direct target of the TGF-β/Smad pathway or involve promoter methylation.
  • ceRNA and RBP Interactions: While bioinformatics predicted potential ceRNA networks and RBP interactions for EPB41L4A-AS2, experimental validation of these interactions is needed.
  • Therapeutic Target Exploration: The authors suggest EPB41L4A-AS2 as a promising therapeutic target for anti-metastasis therapies, implying future work should focus on developing strategies to restore its expression or function.

Summary updated

Abstract

EPB41L4A-AS2 might serve as a negative regulator of TGF-β signaling and as an effective prognostic biomarker and important target in anti-metastasis therapies of HNSCC patients.
BibTeX
@article{huang2018identification,
  author = {Teng Huang and Wei Huang and Hong Lu and Biyun Zhang and Jun Ma and Di Zhao and Yijun Wang and Dahai Yu and Xia He},
  title = {{Identification and validation a TGF-β-associated long non-coding RNA of head and neck squamous cell carcinoma by bioinformatics method}},
  journal = {Journal of Translational Medicine},
  year = {2018},
  doi = {10.1186/s12967-018-1418-6},
}
Figure