Mushroom Peptides That Make Less Salt Taste Like More, Found by Virtual Screening

Salt is the ingredient many of us love and loathe in equal measure. Public health agencies around the world have spent decades urging people to cut back on sodium, because chronic high intake is firmly linked to hypertension, cardiovascular disease and stroke. Yet salt is not just a flavoring; it is a fundamental amplifier of nearly every other taste in a dish. Remove it and food turns flat, metallic and unappealing, which is precisely why so many low-sodium products languish on supermarket shelves. A new study published in npj Science of Food offers an appetizing way out of this dilemma: researchers have identified short protein fragments from an edible mushroom that make salt taste saltier, potentially allowing food manufacturers to achieve the same flavor impact with less sodium.

The mushroom in question is Hypsizygus marmoreus, often sold under the names beech mushroom or bunapi-shimeji. It is prized in East Asian cuisine for its firm texture and mild, nutty taste, and like many edible fungi it is a rich source of small, bioactive peptides. Previous work on other mushrooms, including king oyster mushroom (Pleurotus eryngii), had hinted that peptides liberated from fungal proteins could boost saltiness perception. The team behind the new study, led by Wenping Kong of Gannan Medical University together with colleagues from Shanxi Medical University, set out to apply a modern, multi-tool discovery pipeline to find the best candidates in H. marmoreus and to explain, at the molecular level, how they work.

Their strategy is a textbook example of how computational screening can accelerate food science. Rather than laboriously testing every possible peptide fragment in the mushroom’s protein repertoire, the researchers combined sensory-guided screening, machine learning and bioinformatics to triage candidates in silico. Peptide sequences were evaluated for the physicochemical features that typically correlate with taste activity, such as molecular size, charge distribution and the abundance of acidic and hydrophilic residues. Machine learning models trained on known taste-active peptides ranked the candidates, and only the most promising survivors were advanced to laboratory and sensory testing. This virtual-first approach dramatically reduces the number of expensive, slow human taste experiments needed, and it mirrors workflows already transforming drug discovery and flavor chemistry.

The screening funnel converged on three short peptides with striking names rendered in single-letter amino acid code: ASHGEGF, a seven-residue fragment; GDDWT, a five-residue fragment; and LDDGF, another five-residue peptide. A quick glance at their sequences reveals a common theme: two of the three begin with consecutive acidic residues, aspartic acid and aspartic acid, and all three are rich in polar, hydrogen-bonding side chains. Acidic, negatively charged motifs have long been associated with salt-taste modulation, and the computational filters clearly homed in on that chemistry. What matters most, however, is what happened when the peptides met human tasters and electronic instruments.

To confirm that the peptides were genuinely active, the researchers deployed a two-pronged evaluation. First, trained sensory panels tasted the peptides in sodium chloride solutions and rated the perceived saltiness intensity and how long the sensation lingered. Second, an electronic tongue, an array of cross-sensitive potentiometric sensors that produce a digital fingerprint of a solution’s overall taste profile, provided an objective, instrument-based readout that complements and cross-checks human perception. Both methods pointed in the same direction: the peptides did more than simply add their own faint taste. They acted synergistically or additively with sodium chloride, increasing both the intensity of the saltiness perceived by panelists and the duration of the sensation. In practical terms, a solution containing the peptides tasted saltier than its sodium content alone would predict.

Understanding why required a journey into structural biology. Salt taste transduction in mammals is a famously murky field, but one emerging player is TMC4, a member of the transmembrane channel-like protein family that has been proposed as a sodium-sensitive channel in taste receptor cells. The researchers built a model of TMC4 and used molecular docking, an algorithm that predicts how small molecules nestle into protein binding sites, to explore whether their peptides could physically interact with the channel. The docking results showed that all three peptides could bind the TMC4 model stably, and the interactions were dominated by hydrogen bonds rather than strong electrostatic locks or hydrophobic packing. Four residues emerged as recurring contact points: Lys486, Gln377, Glu235 and Gln482. The presence of lysine, a positively charged residue, alongside glutamine and glutamate suggests a network of complementary polar contacts that could anchor a peptide near the channel’s sodium-permeating machinery.

Docking gives a static snapshot, but proteins are restless machines that flex and breathe in solution. To test whether the peptide-TMC4 complexes would hold together under realistic, warm, watery conditions, the team ran molecular dynamics simulations, which track every atom of the system over time. The simulations supported the picture painted by docking: the peptides remained closely associated with the channel model, and the complexes were stable over the simulated trajectories. Stable, persistent association does not prove enhancement of salt taste on its own, but it is exactly what you would expect if a peptide is genuinely modulating the channel rather than bumping into it fleetingly and drifting away.

The study added one more layer of quantum chemistry for good measure. Using density functional theory, or DFT, the researchers calculated the frontier molecular orbitals and electrostatic potential maps of the three peptides. These quantities describe where a molecule’s electrons are most easily donated or accepted and where its positive and negative electrostatic regions lie, which in turn govern how it approaches and engages a binding partner. The analysis revealed distinct electrostatic patterns among the three peptides, offering a computational framework for rationalizing why some saltiness-enhancing peptides outperform others and how future candidates might be designed rather than merely discovered.

Why does this matter beyond the laboratory bench? Excess dietary sodium is a genuine global health problem, and most of it hides in processed and restaurant foods rather than the salt shaker on the dinner table. The food industry’s standard workarounds, potassium chloride substitutes and flavor masking, often introduce bitterness or metallic off-notes that consumers notice immediately. Taste-enhancing peptides from edible mushrooms sidestep that problem: the enhancers themselves are food-derived, generally recognized as safe by their dietary origin, and they work by amplifying the salt signal rather than replacing it. If ASHGEGF, GDDWF-style fragments and their relatives can be produced economically, for example through controlled enzymatic hydrolysis of mushroom protein, they could find their way into soups, sauces, snacks and ready meals, letting formulators trim sodium without sacrificing the savory satisfaction that keeps consumers coming back.

The authors are careful about scope, and rightly so. The work identifies candidate peptides and supports a plausible mechanism through modeling of a proposed salt-taste channel, but translating those findings into a commercial low-sodium ingredient will require further validation, including studies in physiological taste tissue, dose-response work in real food matrices and regulatory assessment. Still, the study’s integrated pipeline, sensory-guided screening fused with machine learning, bioinformatics, electronic tongue analysis, docking, molecular dynamics and quantum chemical calculation, is arguably as important as the three peptides it delivered. It provides a reusable blueprint that other labs can point at any food protein source to hunt for taste modulators quickly and cheaply. As the global push to reduce sodium intensifies, the humble beech mushroom has just handed flavor scientists a promising new toolkit, one tiny peptide at a time.

Subject of Research: Saltiness-enhancing peptides from Hypsizygus marmoreus identified through virtual screening, sensory evaluation and molecular simulation

Article Title: Discovery of saltiness-enhancing peptides from Hypsizygus marmoreus via virtual screening, multisensory evaluation, electronic tongue, and molecular simulation

Article References: Kong, W., Lin, L., Lai, L., Ding, M., & Wang, M. (2026). Discovery of saltiness-enhancing peptides from Hypsizygus marmoreus via virtual screening, multisensory evaluation, electronic tongue, and molecular simulation. npj Science of Food. https://doi.org/10.1038/s41538-026-01088-9

Image Credits: AI Generated

DOI: 10.1038/s41538-026-01088-9

Keywords: Hypsizygus marmoreus, saltiness-enhancing peptides, sodium reduction, TMC4, molecular docking, molecular dynamics, electronic tongue, sensory evaluation, machine learning, food science, bioactive peptides, density functional theory