QSAR model for the evaluation of biological active molecules on estrogen positive breast cancer

Breast cancer remains one of the most persistent challenges in modern medicine, particularly when the disease is driven by the overexpression of specific hormone receptors. For decades, the standard approach has been to block these receptors with hormonal therapies, yet cancer cells often learn to resist these treatments, rendering them ineffective. In the search for new solutions, scientists have turned their attention to the microscopic machinery inside cells, specifically a protein called topoisomerase IIα. This protein acts as a molecular manager, untangling and managing the long strands of DNA that carry genetic instructions. When this protein is overactive, as it often is in cancer cells, it becomes a prime target for new drugs. Researchers also look at molecules that can slip between the layers of DNA, much like a book slipped between pages, to disrupt the cell's ability to copy itself. The goal is to find compounds that can lock onto this protein and the DNA it manages, stopping the cancer in its tracks without harming the patient.

To find these potential medicines, a team of researchers at the Universidad de Guanajuato in Mexico developed a sophisticated computer-based model. Instead of testing thousands of chemicals in a lab, which is slow and expensive, they used a method called quantitative structure-activity relationship, or QSAR. This approach treats the shape and electronic makeup of a molecule as a set of clues that predict how well it will fight cancer. The team focused on estrogen-positive breast cancer, using a specific cell line known as MCF-7, which mimics the behavior of the disease in patients. They began by creating a digital blueprint, or pharmacophore, based on known drugs that successfully interact with DNA and topoisomerase IIα. This blueprint identified the essential features a molecule must have to work, such as specific aromatic rings that allow it to stack against DNA and chemical groups that form hydrogen bonds to hold it in place.

Using this blueprint, the researchers screened a vast library of chemical structures from public databases. They selected a group of molecules that matched the required features and possessed the right level of potency against the cancer cells. To understand how these molecules behave, the team performed detailed computer simulations. They calculated the three-dimensional shape of each molecule and analyzed its electronic properties, looking at how electrons are distributed across the structure. A key discovery in their work was the importance of a specific electronic feature known as the trace of the quadrupole moment. In simple terms, this describes how the electrical charge is spread out within the molecule. The researchers found that molecules with a specific distribution of this charge were much better at binding to the cancer target. They combined this electronic insight with traditional measurements of molecular shape to build a predictive equation. This equation successfully linked the physical characteristics of the molecules to their ability to kill cancer cells, with statistical tests confirming that the model was robust and reliable.

The study did not stop at prediction; the team also looked at how these molecules would behave inside a human body. They ran computer simulations to estimate how well the drugs would be absorbed, how they would be broken down by the liver, and whether they might cause toxicity. Among the many candidates tested, four molecules stood out as particularly promising. These included compounds labeled 4a, 5a, 5b, and a specific chemical identified as CID2948. These candidates showed a high likelihood of being absorbed by the body if taken orally and demonstrated a lower risk of triggering dangerous side effects compared to existing drugs. The researchers also compared their best candidate, CID2948, against a well-known reference drug, doxorubicin, to see how they held up over time. They ran a massive simulation lasting 500 nanoseconds, which is an incredibly long time in the world of molecular motion, to watch how the drug, the protein, and the DNA interacted.

The results of these long simulations revealed a fascinating difference in how the drugs worked. The reference drug, doxorubicin, tended to bind tightly to the protein itself, forming several hydrogen bonds that kept it anchored. In contrast, the new candidate, CID2948, showed a different behavior. It interacted more strongly with the DNA strand, forming stable connections that held the complex together. Both molecules kept the protein stable throughout the simulation, suggesting they could effectively lock the machinery in place. The study suggests that the new molecule's ability to interact with the DNA, combined with its favorable electronic properties and lower toxicity profile, makes it a strong candidate for further development. By proving that the shape and electronic charge of a molecule can be tuned to improve its ability to fight cancer, this work offers a clear path forward for designing the next generation of treatments that could overcome the resistance seen in current therapies.