AsianScientist (May. 27, 2026)– Bacteria can be reprogrammed to manufacture medicines, break down plastic waste or capture carbon. Doing so involves pushing microbes through cycles of genetic trial-and-error until useful traits emerge. It’s a tedious process, and it becomes far more difficult when entire clusters of interacting genes must be tuned rather than one gene at a time.
The standard approach is directed evolution, which accelerates natural selection in the lab by compressing millennia of evolution into days. However, even the fastest versions can only mutate short DNA stretches of around 8,000 base pairs, limiting their ability to optimise the large multi-gene pathways needed for industrial applications.
They also suffer from a reliability problem, where under selection pressure, bacteria can evolve unrelated mutations elsewhere in their genome that help them survive without improving the target genes. These so-called cheater mutations can obscure which genetic changes are actually responsible for better performance.
A new virus-based platform developed by researchers at the National University of Singapore (NUS) addresses both limitations. The platform, called Lytic Selection and Evolution (LySE), is described in Nature Microbiology by a team led by Assistant Professor Julius Fredens at the NUS Department of Biochemistry.
LySE harnesses bacteriophage T7—a virus that infects bacteria and ultimately bursts open, or lyses, bacterial cells—to mutate a target gene cluster and then carry the mutated genes into fresh bacterial hosts. That transfer step, known as transduction, is the key to eliminating cheater mutations. Since the bacterial host is replaced entirely at the end of each cycle, any off-target mutations the bacterium accumulated during the experiment are discarded.
The controllability of the system comes from an engineered version of T7’s DNA-copying enzyme, which is deliberately redesigned to be error-prone. Compared with the bacterium’s normal DNA-copying machinery, the engineered enzyme introduces mutations about 160,000 times faster while largely restricting them to the selected genes carried by the virus.
That same high error rate also weakens the virus itself, preventing it from spreading uncontrollably. Researchers can control when mutation cycles begin and end by adjusting the ratio of virus particles to bacterial cells, giving them tighter control than conventional continuous evolution systems.
“Traditionally, scientists had to choose between slow but highly controlled evolution methods or super-fast but uncontrollable continuous methods,” said Fredens. “Our goal was to create a best-of-both-worlds system: a tool that rapidly evolves large biological pathways while still letting us hit the pause button to control the process and prevent unwanted genetic error.”
To demonstrate LySE, the team evolved a five-gene metabolic pathway, spanning 9,715 base pairs, that enabled Escherichia coli to consume ethylene glycol, a building block of PET plastic, as its sole carbon source. After five rounds of evolution, the best-performing bacteria grew 50.9% better on ethylene glycol than the starting strain.
A parallel experiment using standard adaptive laboratory evolution showed bacterial growth gains but without any mutations in the target pathway itself. Instead, the bacteria acquired unrelated changes elsewhere in their genome, illustrating the cheater problem that LySE is designed to avoid.
With the capacity to handle gene clusters up to five times larger than the current phage-based evolution systems, the researchers plan to apply LySE to synthetic biological systems that do not yet exist in nature, including AI-designed enzymes for capturing carbon dioxide. Future studies will need to determine how stable these evolved pathways remain during long industrial processes and whether the platform functions efficiently in microbes beyond E. coli.
—
Source: National University of SingaporeImage: rawpixel.com/magnific
This article can be found at: Bridging continuous and discrete evolution through a controllable, hypermutagenic phage-bacteria system
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.
