Genome-Guided Computational Profiling of Selected Antibiotics for Repurposing against Multiple Resistance Proteins in Multidrug-Resistant Stenotrophomonas maltophilia

Bacteria have evolved a formidable defense system that allows them to survive the medicines designed to kill them. This phenomenon, known as antibiotic resistance, turns common infections into life-threatening crises. Some bacteria carry specific proteins that act like molecular tools, either chopping up antibiotics, changing the bacterial parts the drugs target, or chemically altering the medicine so it no longer works. Stenotrophomonas maltophilia is one such bacterium. It is a tough, naturally resistant germ that often infects people with weakened immune systems or those on ventilators. Because it resists many standard treatments, doctors have very few options left, and finding new ways to fight it is a matter of urgent global health.

To tackle this problem without waiting years for new drugs to be invented, scientists are exploring "drug repurposing." This approach involves taking existing, safe antibiotics and testing whether they might work against the specific resistance tools of a dangerous bug. Instead of building a new key from scratch, researchers ask if an old key might fit a new lock. However, bacteria are complex, and a drug that looks promising on paper might fail once it enters the chaotic environment of a living cell. To bridge this gap between theory and reality, researchers use powerful computer simulations. These digital experiments allow scientists to watch how drug molecules interact with bacterial proteins in slow motion, revealing whether they stick together tightly or slide apart, long before any physical test tube is used.

In a recent study, a team of researchers from Malaysia and Bangladesh focused on three specific resistance proteins found in a strain of Stenotrophomonas maltophilia. These proteins represent three different ways the bacterium defends itself: one modifies aminoglycoside antibiotics, another changes the bacterial machinery to ignore a drug called trimethoprim, and the third is an enzyme that breaks down a broad class of drugs known as beta-lactams. The researchers wanted to see if a small selection of existing antibiotics could bind to these proteins and potentially stop them from working. They chose a focused group of drugs, including kanamycin C, trimethoprim, and two carbapenems, along with cefiderocol, a newer drug used for difficult infections.

The scientists began by building detailed 3D computer models of the three bacterial proteins. Since the actual physical structures of these specific proteins had not been captured in a laboratory, the team used high-precision software to construct them based on very similar, known structures. They rigorously checked these models to ensure they were physically realistic, confirming that the atoms were arranged in stable, natural positions. Once the digital proteins were ready, the team ran a series of docking simulations. Imagine placing a puzzle piece into a slot; the computer tested how well each antibiotic fit into the most likely binding pocket of each protein. They looked for the tightest fits and the most stable connections.

The initial results were striking. One drug, kanamycin C, consistently ranked as the best fit across all three different resistance proteins. It showed the strongest predicted binding scores in every case, suggesting it had a natural affinity for these bacterial defenses. However, a good fit in a static snapshot does not guarantee a stable bond in a moving system. To understand what would happen over time, the researchers subjected the most promising combinations to molecular dynamics simulations. This process is akin to running a high-speed movie of the drug and protein interacting, allowing them to see if the drug stays locked in place or if it wobbles and drifts away.

The movie revealed a more complex story. When kanamycin C was paired with the first protein, the aminoglycoside-modifying enzyme, it held on remarkably well. The drug settled into the protein's pocket and stayed there, maintaining a stable position and a strong network of chemical connections throughout the entire simulation. This suggested that kanamycin C could effectively target this specific resistance mechanism. However, the story changed with the other proteins. When the same drug was tested against the second protein, the dihydrofolate reductase, it became restless. Instead of staying in one spot, it shifted positions multiple times, exploring different areas of the binding site. In the third case, involving the enzyme that breaks down beta-lactams, the drug eventually drifted far away from its starting position, indicating that the binding site chosen for the test was not a stable home for the molecule.

Interestingly, another drug, cefiderocol, behaved differently. While it did not top the initial ranking list, it showed surprising stability when paired with the second protein, maintaining a compact and steady presence that kanamycin C lacked. This highlighted a crucial lesson: a drug that looks best in a quick computer scan is not always the most stable in a dynamic environment. The researchers also calculated the energy of these interactions to understand the forces at play. They found that the strong bond between kanamycin C and the first protein was driven by powerful electrical attractions, while the stability of cefiderocol with the second protein relied more on the physical closeness of the molecules.

The study concluded that while the computer models pointed to kanamycin C as a highly consistent candidate for further investigation, the results were not a final proof of success. The simulations showed that the drug's behavior depended entirely on which bacterial protein it encountered. The team emphasized that these findings are hypotheses generated by computers, not confirmed facts of biology. The strong binding seen in the simulation of kanamycin C with the first protein is a promising lead, but it requires real-world laboratory experiments to verify if the drug can actually stop the bacteria from resisting treatment. The researchers did not claim to have found a cure, but rather a set of specific, testable ideas about how these drugs might interact with the enemy's defenses.

Ultimately, this work provides a clear roadmap for future experiments. It identifies kanamycin C as a primary candidate to test against the specific resistance protein that modifies aminoglycosides, while suggesting that other drugs might be better suited for the other targets. By combining static docking with dynamic motion analysis, the researchers were able to separate the drugs that simply looked good from those that appeared to hold on. This integrated approach offers a more reliable way to prioritize which old drugs deserve the time and resources of a physical trial, bringing science one step closer to outsmarting the evolving defenses of multidrug-resistant bacteria.