A polarized charge model for RNA: enhanced stability of G- quadruplexes and improved folding free energy landscapes

Life on Earth relies on a complex molecular language written in ribonucleic acid, or RNA. While DNA is often celebrated as the master blueprint of heredity, RNA acts as the active worker, translating genetic instructions into proteins and regulating countless biological processes. To understand how RNA performs these tasks, scientists must see how it folds into specific three-dimensional shapes. A molecule's shape determines its function, much like a key must have the right contours to fit a lock. For decades, researchers have used powerful computer simulations to watch these molecules move and fold, hoping to predict their behavior without needing to build them in a lab. However, these digital models have struggled to capture the true nature of RNA, often producing shapes that are too loose or unstable, failing to match what is seen in real experiments.

The core of this problem lies in how the simulations handle electricity. RNA molecules are highly charged, and they exist in a watery environment where water molecules constantly shield and reshape these electrical forces. Standard computer models have traditionally treated the electrical charge of every atom as a fixed, unchanging value, ignoring how the surrounding water and other atoms might slightly shift that charge. This simplification causes the simulations to miscalculate how strongly different parts of the RNA attract or repel each other. As a result, the digital molecules often fall apart or fail to form the tight, intricate structures that nature creates.

In a recent study, a team of researchers from Jining University and Shenzhen University of Advanced Technology proposed a new way to fix this. They developed a method to calculate the electrical charges of RNA atoms not as fixed numbers, but as values that change depending on the molecule's specific shape and environment. They call this approach a polarized charge model. By applying this new method to two distinct types of RNA structures, the team demonstrated that their simulations could hold these molecules together much more effectively than previous models. The results suggest that accounting for these shifting electrical forces is essential for creating accurate digital twins of RNA.

The researchers first tested their model on a structure known as a G-quadruplex. These are complex, four-stranded shapes formed by RNA sequences rich in a building block called guanine. G-quadruplexes are stabilized by a stack of flat rings of guanine, and crucially, they require metal ions, such as potassium or sodium, to sit in the center channel to hold the structure together. In standard simulations using older models, these central ions often drift away into the surrounding water, causing the entire structure to unravel. When the researchers applied their new polarized charge model, the ions stayed firmly in place. The model successfully recreated the strong electrical attraction between the central ions and the surrounding RNA rings, keeping the structure stable for the duration of the simulation. This stability was observed across different types of ions and various standard simulation settings, indicating that the new method corrects a fundamental flaw in how these interactions were previously calculated.

The team then turned their attention to a simpler but equally important structure: the RNA hairpin. This is a common fold where a single strand of RNA loops back on itself to form a stem and a loop, resembling a hairpin. To test their model, the researchers simulated the process of a loose RNA strand folding into this hairpin shape. They used a specialized technique that allows the computer to explore many different shapes and calculate the energy required to reach each one. In the simulations using standard models, the energy landscape was messy and confusing, often failing to show that the folded hairpin was the most stable state. In contrast, when the researchers used their polarized charge model, the simulations consistently showed that the folded hairpin was the most energetically favorable shape. The model successfully guided the virtual RNA strand into its correct, compact form, matching what is observed in real-world experiments.

A key insight from this work is that the electrical charge of an atom is not a static property but depends on its surroundings. In the folded hairpin, atoms involved in hydrogen bonds become slightly more charged than they are in a loose, unfolded strand. The new model captures this change, whereas older models treat every atom as having the same charge regardless of whether it is buried inside a tight knot or exposed to water. By incorporating this sensitivity, the researchers found that their simulations could reproduce the correct stability of the molecule without needing to change the underlying mathematical rules of the simulation software. This means the method can be easily adopted by other scientists using standard tools, offering a practical upgrade to the field.

The study also highlighted a limitation: the new charge values depend on the specific shape of the molecule used to calculate them. If the researchers calculated the charges based on a loose, unfolded strand and then tried to simulate the folding process, the results were less accurate. The model works best when the charges are derived from a shape that closely resembles the final, stable structure. This suggests that for the most accurate results, scientists should use the known, stable shape of a molecule to generate the charge parameters before running simulations of how that molecule behaves.

Ultimately, this research provides a clearer window into the microscopic world of RNA. By refining how electrical forces are calculated, the team has shown that it is possible to create simulations that are far more faithful to reality. The ability to keep G-quadruplexes stable and to correctly predict the folding of hairpins brings scientists closer to understanding how RNA functions in health and disease. While the method is not a perfect solution for every scenario, it offers a significant step forward in the quest to model the dynamic and complex machinery of life at the atomic level.