A natural compound steps into the estrogen arena

Uterine fibroids and endometriosis are two of the most common gynecologic diseases, affecting 15% to 80% of women of childbearing age. Existing treatments—hormonal drugs and selective estrogen receptor modulators—have side effects and cannot reliably prevent recurrence. This is why less harmful non-hormonal therapies are urgently needed.

As we progressed in our research on these issues of women’s health, we came to realize that what we were dealing with was far from a purely scientific exercise—it was very much a human experience. Each piece of statistical information on fibroids and endometriosis meant there was a person behind that data—a woman experiencing pain, a woman tired of many years of hormone therapy, a woman torn between her work and her body. We were left with only one question: Should a woman bear pain or the consequences of treatment?

To be completely honest, this project has never been just “research” for us. We have lived long enough to see how many women have had to deal with excessive bleeding, continual pelvic pain or even extreme therapies. Some of those women have been our friends, while others we met only in the statistics. All of this made us want to find something better, something more natural to work with.

One moment that truly shifted the direction of our work was when we analyzed transcriptomic data. In endometriosis, ESR1 expression was dramatically reduced—with a p-value around 10⁻¹², leaving no room for doubt. In contrast, fibroid samples showed significantly increased ESR1 expression (p < 0.02). These opposing patterns felt like two clear signals: ESR1 is not only involved in these diseases, but its behavior might be the key to understanding their distinct pathophysiology. This insight convinced us to focus on ESR1 and search for a natural inhibitor that could modulate its activity.

To explore this idea, we turned to computational simulation—our virtual laboratory. For us, these tools are not just software; they are controlled environments where we can observe physics and chemistry at the atomic scale before stepping into a real lab.

We selected 40 plant-derived secondary metabolites—compounds long discussed for their estrogenic or anti-estrogenic effects. Choosing them wasn’t trivial. Each molecule needed scientific justification, validated structural data and evidence of potential ESR1 interaction. Once our library was ready, we docked all 40 compounds into the ESR1 binding pocket to see which ones could truly engage with the receptor.

Docking felt like watching a movie. Each molecule entered the ESR1 pocket, rotated, vibrated and tried to find its place. Using AutoDock Vina and PyRx and benchmarking everything against raloxifene, we compared natural candidates to a well-established SERM. The results surprised us because procyanidin had a better binding affinity than raloxifene (−11.1 kcal/mol), at −12.1 kcal/mol.

However, we have never used only one parameter. Docking analysis is only one still image out of many. Thus, we decided to use molecular dynamics (MD) simulations as well. The simulation lasted 10 nanoseconds, and to ensure reproducibility, we conducted three additional independent simulations that each lasted 10 nanoseconds. In total, we obtained 40 nanoseconds of dynamic data, which helped us see how procyanidin behaved over time under physiological conditions.

Everything seemed very promising. Stability was demonstrated by the RMSD, RMSF, SASA and Rg parameters. The ligand consistently stayed in the same binding pocket and formed interactions with key residues. Hydrogen bonds consistently formed with Glu353 and Arg394, along with hydrophobic interactions with Leu387 and Ala350. It seemed that procyanidin had chosen its comfortable niche in ESR1.

Of course, the journey wasn’t always smooth. Some days, docking results didn’t match our expectations. Other days, protein structures needed careful inspection. We spent hours troubleshooting unstable RMSD curves, trying to understand whether the issue came from simulation parameters or from the molecule’s natural behavior. But these challenges strengthened our connection to the project. Each problem overcome was another step toward unraveling the mysteries of these ailments.

Upon conducting MM-PBSA calculations, the binding energy of −22.66 kJ/mol supported our suspicions: The reaction is spontaneous, stable and thermodynamically feasible. This was a golden lead for us—a proof that an organic, hormone-free compound can someday be developed to help manage estrogen-related illnesses.

ESR1 has always fascinated us. It acts like a command center: When its expression rises, fibroids grow; when it falls, endometriosis disrupts normal signaling. These contrasting behaviors felt like two puzzles that might be solved with a single natural key. This duality made ESR1 an especially compelling therapeutic target.

In future work, we envision our study contributing toward more individualized care. Just think of being able to select a natural substance according to a patient’s genetic makeup—an option that matches their body chemistry, rather than forcing them to adjust to hormone manipulation. Procyanidin may be the first step, but it certainly won’t be the last. For us, this project marks the beginning of a long journey—one that we hope will lead to therapies that are both effective and gentle.

This study taught us something important: Nature still has many untold stories, and computational simulation may be the key to hearing them.

This story is part of Science X Dialog, where researchers can report findings from their published research articles. Visit this page for information about Science X Dialog and how to participate.

Who’s behind this story?

Lisa Lock

BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021.

Full profile →

Advertisements

Andrew Zinin

Master’s in physics with research experience. Long-time science news enthusiast. Plays key role in Science X’s editorial success.

Full profile →

Zahra Maravandi is an enthusiastic bioinformatician and researcher passionate about decoding biological complexities through data. With an M.Sc. in Bioinformatics and deep expertise in oncology and drug design, she navigates the intersection of biology and computational science. Maravandi specializes in in silico pharmacology, leveraging tools like Python, R, and molecular dynamics to uncover innovative, natural therapeutic candidates. An obsessive problem-solver with a detail-oriented mindset, Maravandi thrives on analyzing high-dimensional genomic data to solve complex medical puzzles. Driven by the belief that nature holds untapped therapeutic stories, she is dedicated to pushing the boundaries of biomedical innovation through data-driven research to improve patient outcomes.