An immutable target: the molecular model will help to eliminate cancer drugs from side effects

Russian scientists have created a computational model describing the interaction of antitumor drugs with the proteins of diseased and healthy cells. These proteins are similar to each other, so medications can affect not only the tumor, but also healthy tissues, causing unpleasant side effects such as nausea, hair loss and circulatory system disorders. Experts have already analyzed the effects of four drugs through development and proposed an approach that will make their effect more selective and reduce harm to the body. This can save patients from many unpleasant consequences of treatment. However, experts note that even a complex mathematical model is not able to take into account all the metabolites and cell types.

Drug interaction with cells

MIPT scientists have developed a computational model that, in a virtual environment, makes it possible to assess how selectively an antitumor drug will act on a target without affecting healthy cells. With its help, you can select the characteristics of the drug that will reduce its side effects. Technology has already helped to find a way to reduce the harmful effects of drugs used, in particular, in the treatment of breast and lung cancer. For patients, this may mean a reduction in the severity of nausea and other effects affecting the circulatory system, spinal cord, nails and hair.

Many antitumor inhibitors act not only on tumor cells, but also on healthy cells. One of the reasons is the high similarity of cyclin—dependent kinases (CDK), proteins that regulate cell division. CDK1 and CDK2 are almost 90% identical, so the drug molecules can bind to both proteins at once, leading to severe side effects.

To understand how to increase the selectivity of drug action, the experts built models of molecular dynamics and monitored the interaction of proteins with medicinal substances for 50 nanoseconds. Four CDK2 inhibitors developed by Pfizer, AstraZeneca, Incyclix Bio and Blueprint Medicines were selected for the study.

The scientists analyzed thousands of protein-ligand contacts and found that the ability of drugs to distinguish between CDK1 and CDK2 is particularly affected by two amino acid residues — alanine-31 (A31) and lysine-33 (K33).

— When the drug molecule interacts with the CDK2 protein of tumor cells, it forms a strong bond through the lysine-33 residue and is fixed in the protein "pocket". But such a connection is not formed with the protein of healthy CDK1 cells. Moreover, in CDK1, alanine-31 attracts the inhibitor to itself, knocks it off the correct trajectory and prevents it from gaining a foothold. These patterns allowed us to formulate recommendations that can help in the design of new selective inhibitors," said Philip Maksimov, head of the Laboratory for Structural and functional Research of innovative antitumor drugs at MIPT.

Like from a villi: Gut-mimicking membrane will speed up drug testing

The new development will make it possible to assess the degree of assimilation of drugs by the body

Instructions for reducing side effects

The results obtained were compared with published data on the relationship between the chemical structure of molecules and their biological activity. The scientists' conclusions are consistent with each other.

— In fact, the found features are ready—made instructions for creating safer CDK2 inhibitors. For example, calculations have shown that a volumetric hydrophobic group that provides strong contact with the isoleucine-10 residue can increase the selectivity of some molecules. Or, on the contrary, a polar atom directed at alanine-31 can weaken the unwanted bond," said Elena Peterson, head of the Laboratory of Cellular and Molecular Technologies at MIPT.

Photo: IZVESTIA/Eduard Kornienko

The new approach makes it possible to evaluate the prospects of their structure in a virtual environment even before the synthesis of compounds. This can accelerate the development of anticancer drugs and reduce research costs. The next stage of the work is an experimental verification of the proposed recommendations and the creation of new compounds with improved selectivity.

According to Maria Vedunova, director of the Institute of Biology and Biomedicine at UNN, although the metabolism of cancer and healthy cells differs markedly, there are many tissues in the body whose cells are constantly dividing. Therefore, antitumor drugs inevitably affect healthy cells.

Photo: IZVESTIA/Dmitry Korotaev

— Bioinformatics and chemoinformatics are actively developing now. This makes it possible to significantly reduce the cost of chemical synthesis and the choice of a chemical agent, but it does not provide an unambiguous answer. We still have to conduct experiments on cells and animals. It is not yet possible to create a medicine without clinical research. Even a very complex mathematical model cannot account for all metabolites and all cell types," said Maria Vedunova.

Stem tags: laser taught to "hit" the most tenacious tumor cells

Photodynamic therapy has shown high efficacy against sources of cancer recurrence

According to immunologist Mikhail Bolkov, reducing side effects is an extremely important area of research. In the future, this will reduce the negative effects of therapy on the body and make treatment more accessible to patients with disabilities.