Artificial Intelligence Advances Drug Delivery for Blinding Eye Diseases
Eye Diseases
Introduction
A team of researchers from the Wilmer Eye Institute at Johns Hopkins Medicine and the University of Maryland has made significant progress in the field of drug delivery for chronic blinding eye diseases using artificial intelligence (AI) models and machine-learning algorithms. By leveraging these technologies, the researchers successfully predicted which components of amino acids, known as peptides, would safely deliver therapeutic drugs to eye cells in animal models. This breakthrough holds immense promise for developing new and more tolerable treatments for prevalent eye conditions such as glaucoma and macular degeneration.
The Challenges of Current Eye Disease Treatments
Current drug therapies for chronic blinding eye diseases often involve multiple daily eye drops or frequent injections, which can be burdensome and challenging for patients to sustain over time. These treatments have prompted scientific efforts to develop delivery systems that can effectively bind to eye cells and extend the therapeutic impact of medications. While an implantable device was approved by the FDA in 2020 for treating glaucoma, its prolonged use has been associated with eye cell death, necessitating a return to traditional eye drops and injections.
AI-Powered Prediction of Effective Peptide Sequences
In their research, the team utilized AI-designed models to accurately predict peptide sequences that would bind to a specific chemical in rabbit eye cells, allowing for the safe and sustained release of medications over several weeks. The focus was on peptides that could bind to melanin, a compound widely present in specialized eye cell structures. Although previous studies had demonstrated the effectiveness of peptide-based drug delivery systems, this study aimed to identify peptides with strong binding capabilities to a widespread eye compound.
The AI model was trained on thousands of data points, including amino acid characteristics and peptide sequences, enabling it to learn the chemical and binding properties of different amino acid combinations. As a result, the AI model generated 127 peptides with varying penetration abilities, melanin binding affinity, and non-toxicity to eye cells.
Identification of HR97 as a Promising Peptide
Among the 127 peptides generated by the AI model, one peptide named HR97 stood out with the highest predicted success rate of binding to melanin. The research team verified the properties of these peptides, confirming that HR97 exhibited better uptake and binding within cells without causing any cell death. To assess HR97's performance in drug delivery, the researchers attached it to the glaucoma medication brimonidine and injected it into adult rabbit eyes.
Extended Drug Release and Favorable Results
The researchers found that HR97 enabled sustained release of brimonidine, with high drug concentrations observed in the eye cells for up to one month. This indicated that HR97 successfully penetrated the cells, bound to melanin, and released the drug over an extended period. Furthermore, the eye pressure-lowering effect of brimonidine lasted up to 18 days when bound to HR97, and no signs of irritation were detected in the rabbits' eyes.
Implications and Future Research
The application of AI to predict peptides for drug delivery holds significant implications for treating conditions involving melanin and targeting other specialized eye structures. The success of this study paves the way for improving patient care and quality of life through advanced drug delivery systems. However, further research is required to extend the duration of action, test the AI model's predictions with other drugs, and evaluate safety in human subjects.
The collaborative research effort between the Wilmer Eye Institute at Johns Hopkins Medicine and the University of Maryland has harnessed the power of artificial intelligence to revolutionize drug delivery for blinding eye diseases. By leveraging machine-learning algorithms, the researchers successfully predicted peptide sequences that effectively bind to eye cells, enabling the sustained release of therapeutic drugs
Reference
1. Hsueh, H.T., Chou, R.T., Rai, U. et al. Machine learning-driven multifunctional peptide engineering for sustained ocular drug delivery. Nat Commun 14, 2509 (2023). https://doi.org/10.1038/s41467-023-38056-w
2. Ophthalmologytimes.com: https://www.ophthalmologytimes.com/view/ai-used-to-advance-drug-delivery-system-for-glaucoma-and-other-chronic-diseases
Comments
There are no comments for this story
Be the first to respond and start the conversation.