AI-Guided Patient-Derived Endothelial Modeling Identifies a Reversible Moyamoya-Specific miRNA Deficiency Corrected...

Area of Science:

• Vascular Biology
• Regenerative Medicine
• Bioinformatics

Background:

• Moyamoya disease (MMD) is a progressive cerebrovascular disorder marked by abnormal blood vessel growth.
• Endothelial cells play a crucial role in angiogenesis, the formation of new blood vessels.

Purpose of the Study:

• To investigate molecular alterations in MMD using patient-derived cells and extracellular vesicles (EVs).
• To explore the therapeutic potential of mesenchymal stem cell-derived EVs (MSC-EVs) in reversing MMD-associated endothelial dysfunction.
• To develop an AI framework for identifying miRNA-mRNA interactions in vascular disorders.

Main Methods:

• Profiling of plasma EV-derived miRNAs in MMD patients and healthy controls.
• Generation of patient-specific induced pluripotent stem cell-derived endothelial cells (iPSC-ECs).
• Treatment of MMD iPSC-ECs with MSC-EVs and assessment of miRNA restoration and angiogenic function.
• Development of an AI framework integrating databases, literature mining, and large language models (LLMs) for miRNA-mRNA interaction prediction.

Main Results:

• MMD plasma EVs and iPSC-ECs showed depletion of angiogenesis-related miRNAs.
• MSC-EVs restored depleted miRNAs in MMD iPSC-ECs, normalized target mRNA expression, and rescued angiogenic function.
• The AI framework, enhanced by LLM-derived context, improved miRNA-mRNA interaction prediction accuracy.
• MSC-EV therapy demonstrated significant reversal of molecular and functional endothelial deficits.

Conclusions:

• MSC-EVs hold therapeutic promise for MMD by restoring endothelial function.
• The developed AI framework aids in prioritizing EV-based therapeutic targets for vascular disorders.
• Integrated profiling and AI-driven analysis provide a robust approach for understanding and treating MMD.