Researchers have discovered a new class of antibiotic that shows promise against multidrug-resistant bacteria, marking the first new class found in 60 years and offering hope in the fight against deadly “superbugs”.
AI breakthrough leads to antibiotic capable of killing resistant pathogens
Using artificial intelligence, scientists at MIT and a startup called Generate Biomedicines have identified a compound named zosurabalpin that acts against Acinetobacter baumannii, a dangerous pathogen with increasing antibiotic resistance (1). Zosurabalpin is the first in a new class of antibiotics in 60 years and works differently than all other known antibiotics – by neutralizing bacteria’s ability to build an outer membrane (2).
In lab tests, it eliminated A. baumannii within 3 hours – including strains resistant to all other antibiotic options. It was also effective against other Gram-negative pathogens like E. coli and Klebsiella pneumoniae (3).
This discovery demonstrates AI’s potential to accelerate antibiotic development. The researchers used machine learning to screen over 100 million chemical compounds to identify promising candidates. They plan to screen billions more as they work to combat resistance (4).
|Blood, wound, urinary tract, pneumonia
|Urinary tract, blood
|Lung, wound, blood stream
Table 1. Common antibiotic resistant pathogens that zosurabalpin acted against in initial testing
New antibiotic polymer family kills drug-resistant superbugs
Scientists at the University of Sheffield have developed antibacterial polymers that can kill superbugs without resistance developing (5). The polymers penetrate the outer membrane of Gram-negative bacteria like E. coli and stimulate lethal oxidative stress.
After 30 exposures over 10 days, no resistance developed in the bacteria. Traditional antibiotics usually develop resistance within a day (6). Additionally, the polymers retained effectiveness when applied directly to chronic diabetic wounds in mice (7).
The researchers believe these flexible polymers can avoid resistance by attacking bacteria through multiple mechanisms simultaneously, making adaptation extremely difficult. They hope to progress the polymers toward human trials within 2 years (8).
Calls to incentivize antibiotic development after AI breakthrough
Zosurabalpin’s discovery has renewed calls for governments to incentivize antibiotic research, which has lagged over the past few decades (9). Only 1-3 new antibiotic classes have been found per decade recently, while resistance continues rising.
Many major pharma companies abandoned the field due to low return on investment compared to drugs for chronic illnesses. But health leaders argue lifesaving antibiotics are invaluable from a societal perspective and keeping pace with resistance is crucial (10).
Proposed incentives include increased public funding, prizes for key discoveries, licensing agreements to ensure stewardship, and reimbursement models delinked from antibiotic volumes sold (11). The UK recently implemented subscription style payments for antibiotics, in which companies are paid upfront based on the value of their product rather than volume sold.
Outlook: Cautious optimism as antibiotic pipeline builds
The latest advances offer hope of renewed productivity after decades of minimal antibiotic innovation. But history warrants caution against premature declarations of victory (12).
Bacteria have shown the ability to develop resistance against every new antibiotic discovered in time. While zosurabalpin and the new polymers use innovative approaches to attacking bacteria, resistance remains an ever-present threat (13).
However, the pipelines look more promising than in recent memory. Alongside the latest breakthroughs, scientists are exploring unconventional sources for new antibiotic leads like fungi, plants, and soil bacteria. And computational drug discovery unlocks the ability to screen billions of chemicals for candidates (14).
While cautious, health leaders express renewed optimism at the momentum building in antibiotic innovation. Continued investment and stewardship of new discoveries will be key to making recent hopes a long-term reality (15). The pipeline bears watching in the coming years.
To err is human, but AI does it too. Whilst factual data is used in the production of these articles, the content is written entirely by AI. Double check any facts you intend to rely on with another source.