ITMO Researchers Find New Molecular Target for Popular Cancer Treatment

In our body, proteins play a major role: they are responsible for the creation, growth, and recovery of cells. At the same time, some proteins participate in the development and progression of diseases, including cancer. When researchers work on new cancer treatments, they look for compounds that deactivate the “guilty” protein by connecting to it and suppressing its activity, effectively suppressing tumor growth. Lately, one popular subject of study has been the protein hnRNPA2B1 or A2B1, originally discovered in 2011; it participates in the development of colorectal and lung cancer, as well as carcinoma. Research shows that cancer cells have a higher amount of this protein and that it is responsible for their stability. Thus, A2B1 can become a new biological target and expand cancer therapy methods. However, until recently,  the ChEMBL dabatase of bioactive molecules with drug-like properties didn’t contain any records on compounds that could bind to the protein.

Using computer modeling methods, ITMO scientists have found a drug potentially capable of stopping the growth of cancer cells when it binds to A2B1; the drug turned out to be the popular antitumor treatment irinotecan.

To select the medicinal compound, chemists used data on the binding of an irinotecan derivative – a campothecin molecule – to the A1 protein, which is related to A2B1. It is known that campothecin binds well to A1 and is considered a biologically active compound that has shown activity against breast, ovarian, colorectal, lung, and stomach cancers. Scientists used this compound as a reference against which they compared irinotecan.

To verify the binding of proteins and drug molecules, scientists used two computational modeling methods: molecular docking and molecular dynamics. In the first case, researchers sought the optimal position of the protein and the molecule’s 3D structures wherein they fit together and form a stable complex; in the second, they checked the stability of this complex over time.

Simulated hnRNPA2B1 complex with the anticancer treatment irinotecan made using molecular dynamics. Video courtesy of the researchers

Computer simulation has shown that the medicinal formula connects firmly to the protein, even better than the reference molecule. For example, in one of the simulations, the binding constant to the protein (lower value indicates better binding) was 25 nanomolar for the new compound with irinotecan and 26 nanomolar for the known compound with campothecin. The team also proved these results experimentally using the earlier developed protein immobilization method; with it, it’s possible to preserve protein properties and structure, keeping them fit for experiments.

“We compared the way two molecules with anticancer properties bind to related proteins that participate in cancer development. This helped us demonstrate that the anticancer treatment irinotecan has an additional target, the A2B1 protein. In theory, by deactivating this protein with the drug we can selectively inhibit the development of various cancer types and use a gentler approach to cancer therapy, such as lower radiation dozes,” explains Olga Volkova, the paper’s first author and a PhD student at ITMO’s Infochemistry Scientific Center.

In their study, the team uses international databases of treatments and bioactive compounds, including those issued by the U.S. Food and Drug Administration. In the future, the team is planning to find out how other drugs in this database bind to the A2B1 protein.

“We are currently screening a database of thousands of medicinal compounds. Then, using the molecular docking software GNINA and the molecular dynamics library Gromacs, we will model each compound to produce a list of the most promising ones – which we will later test experimentally. We hope to find other promising molecules with a toxicity lower than that which is usually found in cancer treatments,” shares Anastasia Serova, one of the paper’s authors and a student at ITMO’s Infochemistry Scientific Center.

This study was supported by the Russian Science Foundation's grant No. 22-65-00022.