In a laboratory in Ulsan, South Korea, scientists have quietly pulled off what might become one of the most important advances in the race against infectious disease.
They’ve built a test that can identify deadly bacteria—quickly, accurately, and without culturing. In a clinical world where hours can mean the difference between life and death, especially in conditions like sepsis, the implications are very serious.
The new technology, developed by a team at the Ulsan National Institute of Science and Technology (UNIST), identifies bacterial pathogens in under three hours and does so with near-perfect accuracy—between 96% and 99.9%, even in complex samples where different species coexist.
“This method will aid in the diagnosis of infections requiring immediate antibiotic treatment, such as sepsis, urinary tract infections, and pneumonia, while also helping to reduce unnecessary antibiotic usage,” said Hajun Kim, one of the lead researchers on the project.
Time Is of the Essence
The most common way to detect a bacterial infection today involves growing cultures in the lab. That process can take 24 to 72 hours—an eternity in medical emergencies. Faster techniques like PCR help, but they still struggle with speed, accuracy, and the ability to work well in mixed bacterial samples.
In contrast, the new test from UNIST requires no culturing or sequencing. It relies on a technique called fluorescence in situ hybridization, or FISH, but with a tweak.
The researchers took a fresh look at a type of molecule called peptide nucleic acid, or PNA. These synthetic molecules mimic DNA but are sturdier and better suited to infiltrating bacterial cells. More importantly, they are extremely sensitive to mismatches, making them powerful tools for telling one species from another—even if they look almost identical under a microscope.
What sets this approach apart is how the PNA probes are used: in pairs. One PNA molecule transfers energy to the other in a process called Förster Resonance Energy Transfer (FRET). This only happens when both molecules successfully attach to the target bacteria, which acts like a two-factor authentication. If one key is wrong, no light is emitted. That greatly reduces false positives, especially in complex infections.
A New Class of Probes
The team led by Sungho Kim and colleagues at UNIST focused on seven species frequently implicated in bloodstream infections: Escherichia coli, Staphylococcus aureus, Klebsiella pneumoniae, Proteus mirabilis, Enterococcus faecalis, Pseudomonas aeruginosa, and Bacillus subtilis.
Each bacterial species carries unique patterns in their 16S ribosomal RNA—a genetic workhorse critical to protein synthesis. The researchers exploited these subtle differences to design species-specific 15-base PNA probes that bind only to their intended target. That precision matters. In blood infections like sepsis, misidentifying a pathogen can mean administering the wrong treatment—sometimes with fatal consequences.
In real-world samples, infections often involve multiple pathogens. Detecting them one at a time is slow. So, the team developed a clever workaround: they introduced chemically cleavable fluorescent tags on the probes. Once bound to bacterial RNA inside a sample, they emit light, which is captured using confocal microscopy. Distinct signals that act as glowing fingerprints of individual pathogens. After imaging one species, a mild chemical treatment breaks the link between the dye and the probe, erasing the signal and making room for the next round.
This could reach hospitals soon
Right now, the test is in the lab. But the team is preparing for clinical validation using blood samples from real patients. If successful, it could become part of the diagnostic toolkit in intensive care units, emergency departments, and remote clinics. There’s still a bit of confirming to do, but this is something that could make a difference in the not-too-distant-future.
The study, published in Biosensors and Bioelectronics, involved a collaborative effort between faculty members, postdoctoral researchers, and graduate students, with support from the National Research Foundation of Korea, the Institute for Basic Science, and the Korea National Institute of Health.
