TL;DR:
- Researchers at Wilmer Eye Institute, Johns Hopkins Medicine, have used artificial intelligence models and machine-learning algorithms to predict safe drug delivery to animal eye cells.
- The collaboration with the University of Maryland aims to develop better drug treatments for chronic blinding eye diseases like glaucoma and macular degeneration.
- Current drug therapies for these diseases may be challenging to sustain and tolerate over time, leading to the exploration of delivery systems that can extend the therapeutic impact.
- An implantable device approved by the FDA in 2020 showed promising results but also caused eye cell death in some cases.
- The research published in Nature Communications demonstrates the AI-designed models’ accuracy in predicting effective amino acid sequences (peptides) for drug delivery.
- The focus was on peptides that bind to melanin, a compound widely present in eye cells.
- Machine learning aided in predicting peptide sequences that could penetrate cells, bind to melanin, and exhibit non-toxicity.
- HR97 was identified as the peptide with the highest success rate of binding.
- HR97, when attached to the drug brimonidine, successfully delivered the drug to rabbit eye cells over an extended period, showing no signs of irritation.
- Future studies will explore extending the duration of action, testing the model’s predictions with other drugs, and ensuring safety in humans.
Main AI News:
Cutting-edge research conducted by the Wilmer Eye Institute, Johns Hopkins Medicine, in collaboration with the University of Maryland, showcases a breakthrough in drug delivery for chronic blinding eye diseases. By harnessing the power of artificial intelligence (AI) models and machine-learning algorithms, scientists have successfully predicted the ideal components of amino acids within therapeutic proteins that can safely transport drugs to animal eye cells. This innovative approach holds tremendous promise for developing new and more tolerable treatments for prevalent eye conditions like glaucoma and macular degeneration, which collectively affect millions of Americans.
Currently, drug therapies for these debilitating diseases necessitate multiple daily eyedrops or frequent eye injections. While effective, these treatment methods pose challenges in terms of long-term sustainability and patient tolerance. Consequently, scientists have been fervently exploring novel delivery systems that can bind to specific components of eye cells, thereby enhancing the therapeutic impact of the medications they carry. In 2020, the Food and Drug Administration approved an implantable device designed to release drugs for glaucoma treatment. Although it offered longer-lasting effects than traditional methods, extended usage of the device was found to cause eye cell death in some cases, compelling patients to revert to eye drops and injections.
The findings of the research, published in Nature Communications on May 2, demonstrate that AI-designed models accurately forecasted an effective sequence of amino acids, known as peptides or small proteins, capable of binding to a specific chemical in rabbit eye cells. This binding allowed for safe and sustained drug delivery over several weeks, reducing the frequency and rigidity of treatment schedules. The focus of the investigation revolved around peptides that bind to melanin—a pigment widely present in specialized eye cell structures and responsible for imparting color to the eyes.
Previous studies on peptide-based drug delivery systems have demonstrated their effectiveness. However, this research sought to identify peptides that could strongly bind with a widespread eye compound like melanin. To achieve this objective, the team of scientists leveraged the power of rapid machine learning and AI methods. By feeding a machine learning model thousands of data points, including amino acid characteristics and peptide sequences, the computer model “learned” the chemical and binding properties of specific amino acid combinations. Over time, it acquired the ability to predict peptide sequences with high potential for drug delivery using melanin.
The AI model generated 127 peptides predicted to possess varying capacities for penetrating melanin-containing cells, binding to melanin, and exhibiting non-toxicity toward the cells. Among these peptides, the model identified HR97 as having the highest success rate of binding. The properties of these peptides were subsequently confirmed, revealing superior uptake and binding within cells, with no indication of cell death.
To validate the model’s prediction, researchers attached HR97 to the drug brimonidine, commonly employed to treat glaucoma by reducing inner eye pressure, and administered it through injections into adult rabbit eyes. The team measured the levels of brimonidine in the eye cells by assessing the drug’s concentrations after administering the experimental drug delivery system. Results indicated that significant quantities of brimonidine were present for up to one month, indicating the successful penetration of cells by HR97, its binding to melanin, and the sustained release of the drug. Furthermore, the eye pressure-lowering effect of brimonidine, when bound to HR97, persisted for up to 18 days, with no signs of eye irritation in the rabbits.
Dr. Laura Ensign, the Marcella E. Woll professor of ophthalmology at the Johns Hopkins University School of Medicine and co-corresponding author of the research paper, emphasizes the far-reaching implications of utilizing AI to predict peptides for drug delivery, not only for conditions involving melanin but also for targeting other specialized structures within the body.
Moving forward, researchers intend to explore avenues for extending the duration of drug action, assess the success rate of the AI model’s drug delivery predictions with different medications, and ascertain the safety of these advancements in human subjects.
Conlcusion:
The groundbreaking research conducted by Wilmer Eye Institute, Johns Hopkins Medicine, utilizing artificial intelligence models and machine-learning algorithms to predict safe drug delivery to eye cells signifies a significant advancement in the market. This development holds immense potential for revolutionizing the treatment landscape for chronic blinding eye diseases such as glaucoma and macular degeneration.
By enabling more tolerable and sustained drug therapies, this breakthrough technology has the capacity to enhance patient outcomes and significantly improve the quality of life for millions of individuals affected by these conditions. Moreover, the successful application of AI in predicting effective peptide sequences for drug delivery opens doors for similar innovative approaches in other areas of healthcare, promising far-reaching implications for the market’s future.
