Analysis of AdeSR positional entropy and morbidostat variant distributions. Credit: Antimicrobial Agents and Chemotherapy (2025). DOI: 10.1128/aac.00809-25

The bacteria Acinetobacter baumannii (A. baumannii) is a haunting presence in many hospitals in the United States, where more than one in 100 patients are treated for A. baumannii infections. This species of bacteria is known for its dynamic genome and ability to gain antibiotic resistance.

“This is a deadly pathogen that is notorious for its resistance to traditional drugs,” said Andrei Osterman, Ph.D., a professor in the Center for Data Science and Artificial Intelligence at Sanford Burnham Prebys Medical Discovery Institute and vice dean and associate dean of Curriculum in the Graduate School of Biomedical Sciences.

Prior research revealed that one-third of hospital-acquired A. baumannii infections in the U.S. were resistant to a common antibiotic called carbapenem. Patients suffering from these hard-to-treat infections were more likely to die in the hospital, be hospitalized for longer and more commonly transferred to other health care facilities rather than discharged to their homes.

Scientists at Sanford Burnham Prebys and their collaborators at Roche Pharmaceuticals published findings October 29, 2025, in Antimicrobial Agents and Chemotherapy demonstrating the use of an experimental evolution approach to map genetic mutations in A. baumannii treated with one of two uncommon antibiotics.

“Tigecycline and colistin are part of physicians’ last line of defense for A. baumannii infections,” said Osterman, the senior and corresponding author of the manuscript. “They are rarely used in the U.S. and existing resistance is comparatively low but rising, which prompted us to study how the bacteria acquire new antibiotic resistances.”

The research team conducted their experiments using a morbidostat, a device that enables bacteria to grow continuously over multiple generations while being placed under gradually increasing survival pressure by antibiotic treatment. The morbidostat is controlled by a computer that monitors the culture growth and introduces more and more of the antibiotic drug as long as the culture is growing rapidly.

“It works like an evolution machine that more closely mimics the conditions in the human body versus other methods,” said Osterman. “When combined with genomic sequencing, this approach allows us to achieve our goal of creating as comprehensive a map as possible of all theoretically possible mutations that offer resistance to the drugs.”

The scientists’ mapping efforts confirmed and extended the existing knowledge regarding the major mechanisms of resistance toward each of these classes of antibiotics. The first primary source of acquired resistance to tigecycline was driven by mutations affecting how bacteria isolate and remove drugs before they damage bacterial cells. These systems are called efflux pumps.

“This is a well-established mechanism of tigecycline resistance, and our findings expand the known spectrum of these types of resistance-driving mutations observed in A. baumannii,” said Osterman.

The second major area of resistance was linked to the evolution of colistin resistance. The team identified mutations affecting the activity of an enzyme that can prevent colistin from reaching its intended target, a component of the bacterial cell wall.

“Once we have a complete map of most possible mutations that provide resistance, we can compare this map with other sequenced genomes to make predictions, including sequences from patients suffering from A. baumannii infections,” said Osterman.

More than 10,000 genomes of various isolates of A. baumannii are publicly available to researchers, and all of them were included in this study’s comparative analysis.

The research team views this study and prior research on the evolution of resistance in A. baumannii and other bacterial pathogens as steps on the path to genomics-based predictions of drug resistance and susceptibility that could be used in hospitals and clinics.

“A serious problem occurs when patients are treated by trial and error and given an antibiotic the bacteria are already resistant or even partially resistant to,” said Osterman. “This further promotes bacterial resistance, and you lose time that patients don’t always have.

“The data we are accumulating would allow physicians to order a sequencing test and prescribe an antibiotic the bacteria are least likely to resist, which will be good for individual patients and, more globally, help us slow down the overall evolution of antibiotic resistance.”

More information: J. E. Kent et al, Mutations in two-component signaling systems drive experimental evolution of tigecycline and colistin resistance in Acinetobacter baumannii, Antimicrobial Agents and Chemotherapy (2025). DOI: 10.1128/aac.00809-25

Citation: Experimental evolution uncovers how bacteria develop drug resistance (2025, November 6) retrieved 6 November 2025 from https://phys.org/news/2025-11-experimental-evolution-uncovers-bacteria-drug.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.