Scientists in China have built a bionic electrochemical sensor that can follow the changing umami-related flavor profile of salmon, crucian carp, and cod as the fish spoils, and it keeps working even when its surface is bombarded by the proteins, fats, and small molecules that normally wreck such measurements. The device, described in Food Chemistry: X, pairs a zwitterionic antifouling hydrogel with a conductive gold nanoparticle–MXene framework and an immobilized umami taste receptor, reaching detection limits as low as 0.33 picomolar for representative umami substances while retaining 81 percent of its signal after twenty days of refrigerated storage.

The central problem the team set out to solve is biofouling. Real foods are messy matrices: they contain proteins, carbohydrates, and fats that readily adsorb onto electrode surfaces, blocking the specific recognition between a sensing element and its target analyte, raising background noise, and degrading the signal-to-noise ratio. According to the authors, this fouling has become a major bottleneck preventing bionic taste sensors from moving out of laboratory platforms and into practical food analysis. Existing approaches—sensory panels, instrumental analysis, and electronic tongues—are constrained by subjectivity, time-consuming procedures, or insufficient specificity, while earlier bionic electrochemical sensors built on nanomaterials and taste receptors lose receptor activity over long-term storage.

The sensing interface combines three functional components. Sulfobetaine methacrylate, or SBMA, is a strongly hydrophilic zwitterionic compound that forms a tightly bound hydration layer on surfaces, preventing impurities from directly contacting the electrode; its internal balance of positive and negative charges also minimizes electrostatic attraction to charged contaminants. The drawback is that this high hydrophilicity impedes electron transfer and weakens electrochemical signals. To compensate, the researchers embedded a framework of gold nanoparticles grown on Ti3C2 MXene, a two-dimensional transition metal carbide whose layered structure and surface functional groups confer exceptional conductivity. The gold nanoparticles provide favorable interfacial sites for immobilizing T1R1-VFT, the Venus flytrap domain of the human umami taste receptor, while the MXene offers abundant anchoring points for the receptor–nanoparticle complex.

Fabrication was deliberately simple. SBMA monomer, a crosslinker, and an initiator were dissolved in water, mixed with the AuNPs/MXene solution and the receptor solution, and shaken at 4 degrees Celsius for eight hours before being gelled in one step. A five-microliter drop of the pre-hydrogel was dispensed onto a polished glassy carbon electrode, incubated for twenty minutes to form a film, and rinsed. Structural characterization by scanning electron microscopy showed that adding AuNPs/MXene transformed the hydrogel’s irregular porous architecture into a finer, more interconnected network. X-ray diffraction confirmed the gold and MXene crystal phases, infrared spectroscopy and X-ray photoelectron spectroscopy verified receptor incorporation, and rheological testing showed gel-like viscoelastic behavior with the storage modulus consistently exceeding the loss modulus.

Electrochemical measurements confirmed each construction step. Coating the bare electrode with pure SBMA hydrogel sharply reduced current and increased charge-transfer resistance because the water-rich layer blocks electron flow. Adding AuNPs/MXene restored conductivity, and immobilizing the receptor protein modestly increased impedance again. When the umami ligand inosine 5′-monophosphate (IMP) was added, the peak current fell further, consistent with a binding-induced interfacial blocking mechanism: as more ligand molecules occupy the immobilized receptors, the redox probe increasingly loses access to the electrode surface. Crucially, control experiments showed the signal was not an artifact. With 10^-8 molar IMP, the native receptor electrode produced a current change of 16.27 microamperes, versus only 2.48 microamperes for a receptor-free hydrogel, 0.93 microamperes for a receptor denatured by heating, and 0.37 microamperes for a receptor pre-saturated with excess IMP.

The antifouling performance was the study’s most striking result. Water contact angle measurements showed the SBMA hydrogel was the primary source of surface hydrophilicity, dropping the angle from 87.77 degrees on a bare electrode to 19.78 degrees, while zeta potential analysis confirmed the composite remained nearly electrically neutral at −0.30 millivolts. Exposed to representative foulants—bovine serum albumin, hemoglobin, lysozyme, crude salmon protein extract, the lipids lecithin and phosphatidylcholine, and the fish metabolites lactic acid and hypoxanthine—the finished sensor’s signal changed by only 0.18 to 4.42 percent even after forty minutes of incubation. By comparison, signal change rates for the conductive framework without SBMA ranged from 40.74 to 68.59 percent, and for bare electrodes from 45.14 to 84.14 percent. Fluorescence microscopy with labeled albumin corroborated the result: bare electrodes glowed with an average intensity of 6.395, while the hydrogel-coated electrodes registered just 0.264.

Analytical performance was competitive with previously reported umami assessment methods. The sensor detected monosodium glutamate from 10^-12 to 10^-4 molar with a limit of detection of 0.33 picomolar, IMP from 10^-11 to 10^-5 molar with a limit of 3.3 picomolar, the umami organic acid WSA over the same range with a 3.3 picomolar limit, and the umami peptide BMP from 10^-12 to 10^-6 molar with a 0.33 picomolar limit. Signal-to-noise ratios near the detection limits ranged from 4.5 to 5.5, with relative standard deviations of 10 to 18.18 percent. In selectivity tests at equimolar 10^-7 molar concentration, responses to IMP and MSG reached 43.18 and 38.14 microamperes, while sour, sweet, bitter, and salty tastants and structurally related analogues such as glutamine, asparagine, xanthosine monophosphate, and uridine monophosphate elicited only 2.88 to 5.74 microamperes. Across eighteen electrodes from three independently prepared batches, the overall relative standard deviation was 4.83 percent.

The team then applied the sensor to real fish. Minced samples of salmon, crucian carp, and cod were extracted, diluted, and measured during storage at room temperature for up to twenty-four hours and at 4 degrees Celsius for up to six days. The electrochemical response followed a consistent arc: during early storage, endogenous enzymes converted ATP into IMP and protein hydrolysis released flavor-enhancing amino acids such as glutamic and aspartic acid, strengthening the receptor-mediated signal. Later, IMP degraded into hypoxanthine while lipid oxidation and microbial proliferation altered the ligand profile and the signal declined. Refrigeration slowed but did not change this pattern, with the response peaking around three to four days before falling under the combined influence of cold-tolerant bacteria, lipid oxidation, and endogenous enzymes.

An auxiliary readout based on tryptophan fluorescence quenching of the receptor showed the same rise-and-fall trend in fish extracts, with correlation coefficients between the two methods ranging from 0.80 to 0.95 across the six storage conditions. The authors are careful about what the sensor actually measures. Because the T1R1-VFT interface responds broadly to multiple umami-related ligands with different sensitivities, the integrated signal reflects changes in both ligand composition and abundance but cannot resolve individual contributions or directly quantify total umami content. Fluorescence measurements in fish extracts can also be distorted by autofluorescent matrix components such as proteins, lipids, and pigments. The hydrogel interface reduces nonspecific matrix interference during the electrochemical readout, improving signal stability in real samples.

The researchers frame this as a platform for monitoring storage-dependent changes in overall umami-related response rather than a finished freshness meter, and they note that future work must correlate the receptor-mediated signal with targeted molecular markers and sensory panel results, expand the range of interfering substances tested, and adapt the coating so the sensor can handle real food samples with little or no pretreatment. Still, the combination of picomolar sensitivity, near-electroneutral antifouling chemistry, batch-to-batch reproducibility, and twenty-day storage stability addresses several of the most persistent obstacles keeping biomimetic taste sensors confined to the laboratory, and it offers food producers a rapid electrochemical route to tracking flavor evolution in one of the world’s most commercially important food categories.

Subject of Research: An antifouling conductive hydrogel electrochemical biosensor with an immobilized umami taste receptor for monitoring umami-related responses in fish during storage

Article Title: An electrochemical biosensor based on antifouling conductive hydrogel for umami detection in complex fish matrices

Article References: Qin, M., Luo, Y., Hong, F., Feng, X., Zhu, Y., Luo, X., Wang, G., Liu, J., & Liu, Y. (2026). An electrochemical biosensor based on antifouling conductive hydrogel for umami detection in complex fish matrices. Food Chemistry: X, 40, Article 104677. https://doi.org/10.1016/j.fochx.2026.104677

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104677

Keywords: electrochemical biosensor, antifouling hydrogel, umami, T1R1 receptor, MXene, gold nanoparticles, sulfobetaine methacrylate, fish freshness, food flavor analysis, differential pulse voltammetry, biofouling, food chemistry