The role of ncRNA regulatory mechanisms in diseases - case on gestational diabetes

Dong Gao, Liping Ren, Yu-Duo Hao, Nalini Schaduangrat, Xiao-Wei Liu, Shi-Shi Yuan, Yu-He Yang, Yan Wang, Watshara Shoombuatong, Hui DingBriefings in Bioinformatics 2024 · DOI

TL;DR

This review explores the critical role of non-coding RNAs (ncRNAs) in gestational diabetes mellitus (GDM), detailing how long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and circular RNAs (circRNAs) contribute to GDM pathogenesis. It highlights their potential as diagnostic biomarkers and therapeutic targets, and discusses the promise of deep learning approaches for uncovering these complex regulatory mechanisms.

What Problem Does This Paper Address?

Gestational diabetes mellitus (GDM) is a common pregnancy complication with significant short- and long-term adverse effects on both mother and offspring, including increased risks of type 2 diabetes and cardiovascular disease. Despite its prevalence and impact, the exact pathogenesis of GDM remains unclear. This paper addresses the need for a deeper understanding of GDM's molecular underpinnings to facilitate early diagnosis, risk assessment, and targeted therapies. Specifically, it focuses on the emerging role of non-coding RNAs (ncRNAs) – RNA molecules that don't code for proteins but are crucial regulators of gene expression – in the development and progression of GDM.

Methods

This paper is a comprehensive review, not an experimental study. Therefore, it doesn't present novel methods or algorithms. Instead, it synthesizes existing research on the roles of lncRNAs, miRNAs, and circRNAs in GDM. It discusses how these ncRNAs function through various mechanisms, including acting as miRNA sponges (ceRNAs), regulating gene transcription, and influencing protein activity. The paper also highlights the potential of deep learning (DL) models, particularly those incorporating attention mechanisms, for identifying ncRNA biomarkers and predicting their pathogenic roles in GDM.

Datasets

Not discussed in the paper. The paper is a review of existing literature and does not present new experimental data or datasets.

Key Insights and Ideas

  • ncRNAs are key players in GDM: Beyond protein-coding genes, ncRNAs are now recognized as critical regulators in GDM, influencing everything from insulin sensitivity to placental development.
  • Diverse ncRNA roles:
    • lncRNAs: Can act as decoys for miRNAs, guide chromatin remodeling, and influence gene expression in cis or trans. Their dysregulation is linked to GDM complications.
  • miRNAs: Regulate gene expression post-transcriptionally by binding to target mRNAs, impacting processes like cell proliferation and metabolism. Aberrant miRNA levels are strongly associated with GDM.
  • circRNAs: Offer stability due to their circular structure and can also function as miRNA sponges, influencing gene expression and cellular processes relevant to GDM.
  • Biomarker potential: Many ncRNAs show altered expression levels in GDM patients (in plasma, serum, or placental tissues), making them promising candidates for early diagnostic biomarkers.
  • Deep learning for discovery: The complexity of ncRNA regulatory networks necessitates advanced computational approaches. Deep learning, especially with attention mechanisms, shows great promise for identifying novel ncRNA biomarkers and deciphering their intricate roles in GDM pathogenesis.

Key Results

The paper summarizes numerous studies identifying specific lncRNAs (e.g., HOTAIR, MEG8, SNHG17), miRNAs (e.g., miR-143-3p, miR-517a-3p, miR-16-5p), and circRNAs (e.g., circACTR2, circVEGFC, hsa_circ_0003218) that are dysregulated in GDM. These ncRNAs have been shown to:

  • Influence insulin resistance and glucose metabolism: For instance, lncRNA HCG27 enhances glucose uptake, while miR-152 inhibits hepatic insulin resistance.
  • Affect placental function: Several ncRNAs are implicated in trophoblast cell proliferation, migration, and invasion, crucial for a healthy pregnancy.
  • Serve as diagnostic indicators: Studies have shown that specific circulating ncRNAs can predict GDM risk or presence with notable accuracy. For example, a 3-miRNA signature (miR-16-5p, miR-29a-3p, and miR-134-5p) effectively predicted GDM in one study.
  • Deep learning models show promise: While not presenting new DL models, the review highlights how existing DL approaches have been used to predict ncRNA-disease associations and protein interactions, suggesting their utility for GDM biomarker discovery.

Limitations and Future Work

  • Experimental validation is key: While many ncRNAs show altered expression, their precise functional roles and therapeutic potential often require further experimental validation.
  • Complexity of networks: ncRNAs operate within complex regulatory networks, making it challenging to fully elucidate their mechanisms and interactions.
  • Need for cost-effective methods: Current experimental methods for analyzing ncRNAs are time-consuming and costly. There's an urgent need for more efficient bioinformatics tools, like those leveraging deep learning, to accelerate discovery.
  • Therapeutic potential: While ncRNAs show promise as biomarkers, their development into effective therapeutic agents for GDM requires further research.
  • Deep learning applications: The authors emphasize the need to further explore deep learning, particularly attention mechanisms, for identifying ncRNA biomarkers, predicting their pathogenic mechanisms, and constructing interaction networks with drugs for targeted therapies.

Summary updated

BibTeX
@article{gao2024role,
  author = {Dong Gao and Liping Ren and Yu-Duo Hao and Nalini Schaduangrat and Xiao-Wei Liu and Shi-Shi Yuan and Yu-He Yang and Yan Wang and Watshara Shoombuatong and Hui Ding},
  title = {{The role of ncRNA regulatory mechanisms in diseases - case on gestational diabetes}},
  journal = {Briefings in Bioinformatics},
  year = {2024},
  doi = {10.1093/bib/bbad489},
}
Figure