June 18, 2024

AI-Designed Drug Candidate Shows Promise for Gastric Acid Inhibition

A team of researchers at Nagoya University in Japan has made significant progress in the development of a new gastric acid inhibitor, using artificial intelligence (AI) to design and synthesize a candidate compound with better binding affinity than existing drugs. The findings, published in Communications Biology, demonstrate the potential of AI in pharmaceutical development and highlight the collaborative approach between humans and AI in drug discovery.

The balance of gastric mucosal secretion plays a crucial role in food digestion. However, when this balance is disrupted, excess stomach acid can lead to discomfort and various conditions, including gastric ulcers and reflux esophagitis. To alleviate these symptoms, many individuals turn to gastric acid suppressants, which target the gastric proton pump responsible for acid secretion, neutralizing stomach acid and providing relief from heartburn and related conditions.

The research team, led by Associate Professor Kazuhiro Abe and Professor Satoshi Yokoshima of the Graduate School of Pharmaceutical Sciences at Nagoya University, took a unique approach to drug development. They focused on the steric structure of the gastric proton pump, a complex protein structure in the stomach lining that transports protons responsible for gastric acid secretion. Using an AI-driven drug discovery platform called Deep Quartet, the team designed candidate compounds with unique chemical structures that effectively targeted the gastric proton pump.

The goal was to identify compounds that could bind to multiple sites on the proton pump simultaneously, enhancing the overall effectiveness of the drug. The candidate compounds were synthesized and analyzed using cryo-electron microscopy to examine their binding structures with proteins. The compounds were then further modified to improve their binding ability.

Using AI, the researchers generated over 100 candidate compounds with unique chemical structures. Expert chemists and structural biologists selected the most promising candidates for synthesis and tested their binding and inhibitory effects on the gastric proton pump. The sixth compound synthesized, called DQ-06, exhibited stronger binding than existing reference compounds.

At first, there were reservations about the strange chemical structures suggested by AI, including DQ-02. However, the researchers realized that AI was designing compounds tailored to the flexible binding site of the proton pump. Further analysis using cryo-electron microscopy revealed that the binding strength could be improved. Based on this knowledge, a new compound called DQ-18 was synthesized, resulting in even stronger binding by introducing a chlorine atom into DQ-06.

Although the results confirmed the expected binding of the compound, there was still room for improvement in terms of bonding. The researchers believe that filling these gaps will result in a tighter fit and stronger bonding. DQ-18 showed a binding affinity nearly 10 times higher than that of SCH28080, a prototype compound for gastric acid inhibitors.

The success of this innovative approach highlights the synergy between humans and AI in drug discovery. Associate Professor Abe emphasizes that while AI is essential for structure-based drug design, the final decision-making still requires human knowledge. He believes that this collaborative approach can lead to more efficient and reliable treatments for gastric acid-related conditions.

The research conducted by Nagoya University represents a significant step forward in the field of pharmaceutical development. The use of AI in drug discovery has the potential to revolutionize medicine, providing new insights and approaches to treatment. This collaborative effort between researchers and AI showcases the power of combining human expertise with AI capabilities to improve human health and advance medical science.


1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it