Researchers use AI to predict compounds that could counteract baldness

Hair loss is undesirable for many men -; and women -; because her hairstyle is often intimately linked to her self-confidence. And while some people embrace it, others wish they could regrow their lost locks. Now the researchers who report in ACS’ Nano-letters used artificial intelligence (AI) to predict compounds that could neutralize reactive oxygen species causing baldness in the scalp. Using the best candidate, they constructed a proof-of-concept microneedle patch and effectively regenerated mouse hair.

Most people with severe hair loss suffer from androgenic alopecia, also known as male or female pattern baldness. In this condition, hair follicles can be damaged by androgens, inflammation, or an overabundance of reactive oxygen species, such as free oxygen radicals. When levels of oxygen free radicals are too high, they can overwhelm the body’s antioxidant enzymes that usually control them. Superoxide dismutase (SOD) is one such enzyme, and researchers have recently created SOD mimics called “nanozymes.” But so far, the ones that have been reported are not very effective in removing oxygen free radicals. So Lina Wang, Zhiling Zhu and their colleagues wanted to see if machine learning, a form of AI, could help them design a better nanozyme to treat hair loss.

The researchers chose transition metal thiophosphate compounds as potential candidates for the nanozymes. They tested machine learning models with 91 different combinations of transition metals, phosphates and sulfates, and the techniques predicted that MnPS3 would have the strongest SOD type ability. Next, MnPS3 nanosheets were synthesized by chemical vapor transport of manganese, red phosphorus and sulfur powders. In early tests with human skin fibroblast cells, the nanosheets significantly reduced levels of reactive oxygen species without causing harm.

Based on these results, the team prepared MnPS3 microneedle patches and treated with them mouse models affected by androgenic alopecia. Over 13 days, the animals regenerated thicker strands of hair that more densely covered their previously bald bottoms than mice treated with testosterone or minoxidil. The researchers say their study has both produced a nanozyme treatment for hair regeneration and indicated the potential for computational methods to be used in the design of future nanozyme therapies.


American chemical society

Journal reference:

Zhang, C. et al. (2022) Machine learning guided discovery of superoxide dismutase nanozymes for androgenetic alopecia. Nano-letters.

Leave a Comment