Harnessing the Potential of Carotenoids for Cancer Therapy: An Integrated Machine Learning and MST Based Approach

Area of Science:

• Molecular Biology
• Pharmacology
• Computational Chemistry

Background:

• Receptor tyrosine kinases (RTKs) are crucial for cell signaling but their dysregulation drives cancer.
• Carotenoids, plant metabolites, exhibit anti-cancer properties by targeting molecular pathways.
• Targeting RTKs is a key strategy in cancer therapy.

Purpose of the Study:

• To investigate the potential of carotenoids as inhibitors of receptor tyrosine kinases (RTKs).
• To identify specific carotenoid-RTK interactions using computational and in vitro methods.
• To assess the drug-likeness of potential carotenoid inhibitors.

Main Methods:

• In silico molecular docking of nine RTKs with various carotenoids.
• Molecular dynamics simulations to validate interactions.
• Machine learning algorithms for drug-likeness assessment.
• In vitro microscale thermophoresis (MST) to confirm binding affinity.

Main Results:

• Identified five potential RTK-carotenoid interactions: EGFR-fucoxanthin, FGFR2-peridinin, VEGFR2-canthaxanthin, PDGFRA-canthaxanthin, and ALK-crocin.
• Canthaxanthin demonstrated high binding affinity with PDGFRA and VEGFR2 via MST.
• Carotenoids showed favorable drug-likeness compared to established drugs.

Conclusions:

• Carotenoids are promising candidates for RTK-targeted cancer therapy, offering plant-based alternatives to conventional drugs.
• This study provides a novel ML-assisted approach for discovering phytochemicals as anti-cancer agents.
• Findings offer new insights into the development of phytochemicals for cancer treatment.