AI and supercomputer simulations reveal how a bacterial energy-converting enzyme pumps sodium ions, paving the way for new antibiotics

The Na⁺-NQR enzyme is vital for energy production in pathogenic bacteria like the one that causes cholera, making it a highly promising target for new antibiotics. Researchers combined modified artificial intelligence techniques with extensive supercomputer simulations to visualize the hidden, dynamic movements of this enzyme during sodium transport. The study revealed that sodium binding and electron transfer act as a precise dual trigger, driving specific subunits of the enzyme through a cycle of shape changes to pump sodium ions across the cell membrane. Unlocking this pumping mechanism provides a powerful new computational framework for understanding cellular energy-converting molecular machines and could accelerate the design of targeted antibacterial drugs.

Pathogenic bacteria, including the species responsible for cholera, rely on a specialized enzyme called Na⁺-pumping NADH-quinone oxidoreductase (Na⁺-NQR) to generate energy. This enzyme works like a molecular machine, moving sodium ions out of the cell to create an energy gradient that powers essential microbial functions like swimming and nutrient uptake. Because the overall structure of Na+-NQR is completely different from the equivalent energy-converting enzymes found in human cells, it is widely considered an excellent target for developing highly selective antibiotics. However, scientists have struggled to fully understand the exact moving parts of this enzyme, specifically how it changes shape to pump sodium ions while transferring electrons. Mapping these rapid, complex shape changes is crucial for grasping the fundamental energy requirements of the enzyme and for eventually designing drugs that can effectively jam its machinery.

To overcome the challenge of observing fleeting molecular movements, the research team employed a novel computational approach combining artificial intelligence with supercomputer simulations. Because the default settings of the AI structure-prediction tool AlphaFold3 generated only one static shape, the team used modified techniques, such as shallow sequence alignments and specific structural templates, to coax the AI into predicting the rare, hidden transition states of the enzyme. Using these diverse AI-generated shapes as starting points, the researchers ran extensive molecular dynamics simulations and applied a mathematical technique called Markov state modeling to chart the precise pathways and energy landscapes of the enzyme's motion over time.

The simulations successfully revealed the precise energy requirements and step-by-step pathways of the enzyme's specific sodium-transporting subunits, NqrD and NqrE. The researchers discovered that these subunits operate through an alternating-access mechanism, acting like a gated channel that opens to the inside of the cell, closes to temporarily trap the sodium, and then opens to the outside. This shape-shifting cycle is directly driven by the binding of a sodium ion near a specific iron-sulfur cluster within the enzyme, but only after that cluster gains an electron. The mathematical models identified that transitioning from a closed, trapped state to an outward-opening state is the slowest, rate-limiting step, while the entire cycle from an inward-facing to an outward-facing state takes approximately 1.5 milliseconds. When the iron-sulfur cluster loses its electron, the enzyme resets its shape without needing sodium, readying itself for the next pumping cycle.

By clarifying the intricate conformational cycle of Na⁺-NQR, this study provides a detailed blueprint of how cellular pumps powered by electron transfer function at the atomic level. This deep mechanistic understanding offers vital clues for drug developers seeking to create novel antibiotics that can specifically block the sodium-pumping action in harmful bacteria. Furthermore, the research demonstrates that integrating modified AI-based structure prediction with advanced molecular simulations is a highly effective strategy for capturing the hidden dynamics of complex proteins. This innovative computational framework can now be applied to investigate a wide variety of other challenging molecular machines and transporter proteins in biological systems.