Forget the nasal swabs. A team of German researchers has developed a molecular sensor that releases the taste of thyme when it encounters influenza virus in your saliva, potentially turning chewing gum into a diagnostic tool for catching infections before symptoms appear.
The innovation arrives as public health officials grapple with a persistent challenge: people spread influenza for roughly 48 hours before they feel sick. Current testing methods either catch infections too late or cost too much for mass screening. Lorenz Meinel and his colleagues at the University of Wurzburg decided the solution required abandoning sophisticated laboratory equipment entirely.
“We met these challenges by switching away from complex detectors and machinery and toward a detector that is available for anyone, everywhere and anytime: the tongue.”
The sensor exploits a vulnerability in the influenza virus itself. To break free from infected cells and spread through the body, flu viruses deploy an enzyme called neuraminidase (the “N” in designations like H1N1). The researchers synthesized a molecule that mimics the virus’s natural target but with a twist: they attached thymol, the compound that gives thyme its distinctive herbal punch. When viral neuraminidase cleaves the bond, it releases the flavor molecule directly onto the tongue.
Engineering Selectivity Into a Sugar Molecule
The elegance lies in the chemistry. The team started with N-acetylneuraminic acid, a sugar molecule the virus naturally recognizes, then modified it at two specific positions. These alterations proved crucial. Bacterial neuraminidases, which exist naturally in saliva and would trigger false positives, cannot process the modified molecule. Viral neuraminidase, however, cleaves it readily.
In laboratory tests using saliva from hospitalized flu patients, the sensor released detectable thymol within 30 minutes. The researchers calculated that viral concentrations present even in late-stage patients would release enough thymol to cross the human taste threshold, which sits around 1,100 to 1,700 parts per billion. Early infections, when viral loads peak, should generate even stronger signals.
The science required some molecular trial and error. Initial versions using unmodified N-acetylneuraminic acid responded to everything, rendering them useless for flu detection. Methylation at positions O4 and O7 solved the problem, creating a lock that only the viral key could open. Crystallographic analysis and molecular docking studies confirmed why: bacterial neuraminidase active sites cannot accommodate the bulkier modified molecule, while viral neuraminidase possesses a roomier binding pocket.
From Concept to Chewing Gum
The practical applications extend beyond mere novelty. Meinel envisions incorporating the sensor into chewing gum or lozenges, creating what he describes as a rapid first-line screening tool for high-risk environments. The approach could theoretically enable mass testing without laboratories, trained personnel, or expensive equipment. A positive result (that unexpected thyme taste) would prompt isolation and confirmatory testing.
The sensors showed no cytotoxicity in human kidney cells or mouse fibroblasts at concentrations up to 1.0 millimolar. They remained stable for four weeks under various storage conditions, maintaining at least 94 percent integrity even under stressed conditions of 50 degrees Celsius at 75 percent humidity. The researchers calculated that each test would require between 2.1 and 11.5 milligrams of sensor, depending on viral load.
Thymol might not be the final answer. The team notes that switching to denatonium benzoate, one of the most bitter substances approved for food use, could lower detection thresholds by 10 to 100-fold. This could prove valuable given that some flu patients experience temporary loss of taste. Alternatively, the sensor could release a visible dye instead of a flavor molecule.
The testing paradigm differs fundamentally from existing diagnostics. PCR tests offer high accuracy but require laboratory processing. Rapid antigen tests provide convenience but miss presymptomatic infections when viral loads remain low. The taste sensor occupies a middle ground: less definitive than PCR, but potentially capable of catching infections earlier than current at-home tests.
Clinical trials will determine whether the chemistry translates to real-world utility. The researchers plan to begin human studies within two years, focusing on whether people can reliably detect the taste difference between infected and uninfected states. They will also need to establish performance in presymptomatic versus symptomatic stages.
The broader question concerns scalability. Influenza killed an estimated 500,000 people globally in 2023, with pandemic strains historically proving far deadlier. The 1918 Spanish flu infected more than one-fourth of the world’s population. A 2009 H1N1 pandemic killed disproportionately young people, inverting typical mortality patterns. Avian flu continues jumping from birds to humans, with H5N1 cases reported from dairy cattle as recently as 2024.
“This sensor could be a rapid and accessible first-line screening tool to help protect people in high-risk environments.”
The approach sidesteps infrastructure requirements that limit pandemic responses in lower-income regions. Manufacturing these sensors requires conventional chemical synthesis rather than specialized biotech facilities. Distribution could follow existing supply chains for consumer products. The detectors themselves (human tongues) require no calibration or maintenance.
Whether chewing gum becomes a legitimate diagnostic platform remains an open question. The research establishes proof of principle: viral enzymes can be co-opted to generate sensory signals, and chemical modifications can tune selectivity. The path from laboratory bench to pharmacy shelf involves regulatory hurdles, manufacturing scale-up, and clinical validation. But the core innovation stands. Sometimes the most sophisticated solution involves the simplest tools.
ACS Central Science: 10.1021/acscentsci.5c01179
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.