Long non-coding RNAs in biomarking COVID-19: a machine learning-based approach.

Area of Science:

• Molecular Biology
• Immunology
• Virology

Background:

• The COVID-19 pandemic has caused significant global mortality and morbidity.
• Limited data exists on the role of long non-coding RNAs (lncRNAs) in COVID-19 pathogenesis.
• This study investigates three specific lncRNAs (H19, TUG1, CRNDE) in relation to inflammation and fibrosis in COVID-19 patients.

Purpose of the Study:

• To elucidate the relationship between H19, TUG1, and CRNDE lncRNAs and key inflammatory/fibrotic factors (STAT3, α-SMA, TNF-α, IL-6) in COVID-19 patients.
• To assess the diagnostic potential of these lncRNAs in distinguishing COVID-19 patients from healthy individuals.

Main Methods:

• Collected peripheral blood mononuclear cells from 28 COVID-19 patients and 17 healthy controls.
• Utilized real-time quantitative PCR (RT-qPCR) and Western blotting to measure RNA and protein expression.
• Employed machine learning and ROC curve analysis to evaluate diagnostic capabilities.

Main Results:

• H19, TUG1, and CRNDE were significantly overexpressed in COVID-19 patients compared to controls.
• STAT3 and α-SMA expression increased at transcript and protein levels in patients and correlated with the three lncRNAs.
• IL-6 and TNF-α were also upregulated in COVID-19 patients.
• A CRNDE-H19 panel demonstrated significant ability to distinguish COVID-19 patients (AUC=0.86).

Conclusions:

• Overexpressed lncRNAs in COVID-19 patients may contribute to disease manifestations.
• Co-expression with STAT3 and α-SMA suggests lncRNAs' role in COVID-19-associated inflammation and fibrosis.
• The CRNDE-H19 combination serves as a promising biomarker for COVID-19 screening and diagnosis.
• Further research on lncRNAs could identify new therapeutic targets for viral infections.