The achievement is notable because of what the protein lacks. Most well-studied sweet proteins carry stabilizing features that honey truffle sweetener simply does not have. Thaumatin, mabinlin, and brazzein contain multiple disulfide bonds and mixtures of alpha-helices and beta-sheets; neoculin is reinforced by extensive disulfide crosslinks; even monellin, which is largely beta-rich, retains a short stabilizing helix. Honey truffle sweetener, by contrast, is cysteine-free and helix-free—a compact bundle of beta-strands held together by sheet hydrogen bonding and loop geometry. Comparisons of residue-contact density showed the fold is not uniformly loose; its overall packing is comparable to thaumatin and denser than brazzein. But cavity analysis revealed a larger pocket volume per residue than reference proteins, and the absence of classic stabilizing elements left the protein vulnerable to heat.

The first step was to see the molecule in atomic detail. The team crystallized the protein and determined its structure at 1.58 angstrom resolution using synchrotron radiation at the Shanghai Synchrotron Radiation Facility, solving the structure by molecular replacement with an AlphaFold-predicted model as the search template. The structure revealed a characteristic arrangement of seven major beta-strands forming a slightly twisted antiparallel sheet, flanked by three shorter strands—a so-called 7-plus-3 topology—connected by solvent-exposed loops, including an extended and apparently flexible loop between strands three and four. This high-resolution map became the common reference for every subsequent design decision, from mutation evaluation to disulfide-bond planning.

With the structure in hand, the researchers screened every one of the 121 positions for potentially stabilizing substitutions using two complementary computational methods. Rosetta, a physics-based framework, estimated the energetic effect of each mutation after side-chain repacking and constrained relaxation, while ThermoMPNN, a deep-learning model, judged whether each amino acid substitution would be compatible with the local structural environment. From the overlapping and divergent predictions, the team selected 72 single-point variants for experimental testing. The results validated the strategy: 41 of the 72 variants showed a positive shift in melting temperature, and 27 exceeded a one-degree gain. The single best mutations, Q37A and S55G, raised the melting temperature by 3.91 and 4.23 degrees respectively.

But the raw computational scores turned out to be only modestly predictive. Correlations between predicted scores and measured stability changes across the 72 variants were weak, with Pearson coefficients of just 0.206 for Rosetta and 0.248 for ThermoMPNN. That gap is precisely where the study’s most innovative element comes in: a machine-learning framework the authors call SMRC-Net, which learns from the experimental data itself. The system combines a baseline melting-temperature estimate derived from a pretrained protein language model, ESM2, with a residual correction built from molecular dynamics descriptors—flexibility changes, solvent accessibility, residue contacts, and global perturbation measures calculated from simulations of the wild-type protein and 65 mutants.

Under repeated strict nested cross-validation, the sequence-only baseline explained about 44 percent of the variance in melting temperature, with a mean absolute error of 1.35 degrees. Adding the molecular dynamics residual correction lifted the explained variance to nearly 60 percent, cut the error to 1.15 degrees, and improved the rank correlation of variants from 0.540 to 0.698. When the frozen model was tested prospectively on ten Rosetta-designed mutants it had never seen, it correctly predicted the direction of the stability change for nine of the ten—a 90 percent directional accuracy. The single miss, T115I, was predicted to stabilize the protein but produced a small experimental decrease of 0.67 degrees.

Armed with validated stabilizing mutations, the team assembled them stepwise rather than all at once, a modular strategy that kept each combination experimentally testable. Ten substitutions were grouped into four modules of two or three mutations, each verified to raise the melting temperature before further assembly. Combining the three-residue modules produced a six-site variant at 67.77 degrees, and the gains proved nearly additive—the measured improvement differed from the sum of individual contributions by only 0.12 degrees. Two parallel assembly routes then produced eight-site intermediates, and their union yielded the ten-site variant Mut10-1 at 73.03 degrees, 18.43 degrees above the wild type.

The final flourish was structural rather than chemical: an engineered disulfide bond. Screening candidate cysteine pairs by geometric criteria with Rosetta’s DisulfidizeMover identified Q37C/D46C as the most favorable linkage. Introducing that single bond added another 6.95 degrees on its own, and when combined with the ten mutations it produced the final variant, FM, with a melting temperature of 82.07 degrees. Crucially, sensory evaluation by eight trained panelists showed that both Mut10-1 and FM retained sweetness thresholds similar to the wild-type protein—evidence that the stabilizing changes did not disrupt the receptor-binding surface responsible for the protein’s intense taste.

Practical stress tests underscored the difference. After one hour at 80 degrees Celsius, the wild-type protein had largely precipitated out of solution, while FM remained predominantly soluble; even after four hours, FM retained a substantial soluble fraction. Circular dichroism spectroscopy showed that FM’s secondary structure was largely preserved through the prolonged heating, and the engineered protein also stayed soluble and structurally intact across acidic, neutral, and alkaline conditions at pH 3, 6, and 9. Molecular dynamics simulations suggested why: the mutations did not rigidify the protein wholesale but redistributed local flexibility, modestly remodeled the residue-contact network, and stabilized a single dominant folded ensemble. An experimental test supported one long-range interaction in particular—disrupting the H16-F99 contact with a double mutant lowered the melting temperature by 5.2 degrees, confirming its role in tertiary stability.

The authors are careful about the limits. Sweetness was tested only on untreated samples, so whether the engineered protein survives heating and pH shifts while still tasting sweet in real foods remains unverified, and sensory testing in complex food matrices is the necessary next step. The SMRC-Net predictor is also calibrated specifically to this scaffold, though the authors argue the residual-learning architecture could be retrained for other compact proteins where molecular dynamics remains computationally feasible. Still, the demonstration is striking: a cysteine-free, helix-free beta protein with no natural stabilizing armor was transformed into a heat-tolerant, sweetness-preserving molecule through a coordinated workflow of crystallography, complementary computation, machine learning, and disciplined stepwise assembly. For a protein once undone by a warm afternoon, that is a remarkable turnaround—and a template for engineering the next generation of sugar substitutes.

Subject of Research: Computation-guided protein engineering to improve the thermal stability of the fungal sweet protein honey truffle sweetener

Article Title: Integrated Computation-Guided Thermostabilization of the Fungal Sweet Protein Honey Truffle Sweetener

Article References: Wang, Z., Wang, W., Zhu, F., Zhang, Y., Li, Y., Zhu, Z., Lu, Z., Huang, A., Yu, M., & Liu, S. (2026). Integrated computation-guided thermostabilization of the fungal sweet protein Honey Truffle Sweetener. Current Research in Food Science, 13, Article 101584. https://doi.org/10.1016/j.crfs.2026.101584

Image Credits: AI Generated

DOI: 10.1016/j.crfs.2026.101584

Keywords: sweet proteins, honey truffle sweetener, protein engineering, thermostability, crystal structure, Rosetta, ThermoMPNN, molecular dynamics, machine learning, disulfide bond, sugar substitute, food science

Phytocytokines, the plant counterparts of animal cytokines, are small secreted peptides that coordinate defense responses when pathogens or herbivores wound plant tissue. They act like molecular alarms: once released, they bind to cell-surface receptor kinases and trigger a cascade of intracellular events, including calcium influx, the production of reactive oxygen species, and the activation of defense genes. CTNIP peptides belong to a family of these signals that has intrigued structural biologists because their sequences do not resemble the classical cysteine-rich peptide motifs that dominate the phytocytokine world. Without an obvious templating pattern, the question of how CTNIP4 achieves a stable, recognizable structure has lingered since the family was first connected to immune signaling.

The new work answers that question with a combination of structural biology, biochemistry, and genetics. Using nuclear magnetic resonance spectroscopy, the team determined the solution structure of mature CTNIP4 and found that it adopts a compact, well-defined fold anchored by an intramolecular disulfide bond, a covalent bridge formed between two cysteine residues within the same peptide chain. Rather than serving as an interchangeable decoration, this bond acts as the load-bearing element of the structure, pinning the peptide into a looped architecture that presents its receptor-binding face in an exact orientation. When either of the two participating cysteines was mutated to a non-reactive residue, the peptide lost its ordered conformation and, critically, its ability to be recognized.

The researchers then turned to the other half of the pairing: HSL3, a receptor-like kinase embedded in the plasma membrane. Receptor-like kinases form the frontline of plant perception, typically comprising an extracellular domain that captures ligands, a single membrane-spanning helix, and an intracellular kinase domain that relays the message. By mapping the interaction surface, the study showed that HSL3 recognizes not merely the chemical identity of CTNIP4 but its geometry. Binding assays demonstrated that the oxidized, disulfide-bonded form of the peptide engages the extracellular domain with high affinity, whereas the reduced or mutant forms bind poorly or not at all. In other words, the disulfide bond functions as a quality-control stamp, ensuring that only properly folded peptide molecules can sound the alarm.

This mechanism has implications that reach beyond a single ligand-receptor pair. Small peptides are intrinsically floppy; without stabilizing features, they sample many conformations in solution, and a receptor must either tolerate that flexibility or demand a specific shape. The CTNIP4-HSL3 system illustrates the second strategy, sometimes described as conformational selection with a structural gate: the receptor waits for, and exclusively accepts, the correctly folded state. The disulfide bond effectively reduces the search space, pre-organizing the peptide so that entropic costs of binding are minimized. For the plant, this design offers both sensitivity and specificity, allowing the immune system to respond rapidly to genuine danger signals while ignoring misfolded or degraded fragments that could otherwise trigger wasteful false alarms.

Genetic evidence reinforced the structural picture. Plants in which CTNIP4 or HSL3 function was disrupted showed blunted defense responses, and plants expressing mutant versions of CTNIP4 lacking the disulfide-forming cysteines failed to complement those defects, despite accumulating the peptide to normal levels. The deficiency was not in production but in presentation: the peptide was made, secreted, and present, yet structurally illegible to its receptor. This separation between abundance and functionality echoes a growing theme in peptide biology, where post-translational modifications and folding states, not raw expression levels, determine biological activity. It also suggests that breeding or engineering efforts aimed at boosting immune peptides must account for the machinery, such as the secretory pathway’s oxidative folding environment, that allows them to fold correctly in the first place.

The study also clarifies the evolutionary logic of the CTNIP family. Sequence comparisons across species reveal that the cysteine residues are among the most conserved positions, even as the surrounding amino acids vary considerably. That conservation pattern makes sense now: the cysteines are under selection because the fold they create is the substrate of recognition, while other positions can drift as long as the overall architecture and receptor-contacting residues are preserved. Such structural constraints explain how a family of peptides can diversify into multiple members with distinct receptor partners, expanding the vocabulary of plant immune communication without breaking the existing grammar. Each new pair of cysteines is a scaffold; the surface painted on that scaffold determines who listens.

From a biotechnological standpoint, the results arrive at a moment of intense interest in peptide-mediated immunity. Researchers are exploring synthetic phytocytokines as disease-resistance tools, either by applying peptides directly to crops or by engineering plants to produce enhanced versions. The CTNIP4-HSL3 work supplies a design principle for such efforts: any engineered variant must respect the disulfide architecture that defines the fold. It also suggests opportunities, since the disulfide-bonded loop is a discrete, portable module that could in principle be grafted onto other peptide scaffolds to create novel ligands, or targeted by small molecules that lock the peptide into active or inactive conformations. Structural knowledge of this kind turns peptide immunity from a black box into an engineerable system.

There are also lessons for understanding receptor kinase signaling more broadly. HSL3 belongs to a large superfamily of receptors, many of which bind small, post-translationally modified peptides through extracellular domains whose ligand-binding mechanisms remain unresolved. Demonstrating that a single covalent bond can be the decisive determinant of recognition provides a template hypothesis for other orphan receptors: when a peptide ligand seems too short or too featureless to encode specificity, look for disulfides, hydroxyprolines, tyrosine sulfations, or other modifications that impose order. The new study adds weight to the idea that the extracellular space of the plant cell is a chemically decorated landscape where modifications are not embellishments but the very language of communication.

As with any strong structural study, questions remain. How exactly does the HSL3 kinase domain translate ligand binding at the cell surface into intracellular phosphorylation events, and does CTNIP4 require a co-receptor, as many immune peptides do? What proteases liberate CTNIP4 from its precursor in vivo, and do those processing steps couple directly to the oxidizing environment that permits disulfide formation? Answering these will require further structural work on full-length receptor complexes and careful cell biology of the apoplast. But the core achievement stands: the study identifies the disulfide bond of CTNIP4 as the sculptor of its fold and, through that fold, the gatekeeper of its recognition by HSL3. In doing so, it reveals how plants squeeze remarkable information density into molecules barely a dozen amino acids long, and it hands researchers a blueprint for reading, and perhaps rewriting, the molecular conversations that keep crops alive.

Subject of Research: Structural basis of CTNIP4 phytocytokine folding via a disulfide bond for recognition by the receptor kinase HSL3

Article Title: A disulfide bond sculpts the CTNIP4 phytocytokine fold for recognition by the receptor kinase HSL3

Article References: Jiménez-Sandoval, P., Johanndrees, O., Snoeck, S., Harshith, C. Y., Omary, M., Broyart, C., Rhodes, J., Bender, K. W., Zipfel, C., & Santiago, J. (2026). A disulfide bond sculpts the CTNIP4 phytocytokine fold for recognition by the receptor kinase HSL3. Nature Plants, 12(9), 1688-1697. https://doi.org/10.1038/s41477-026-02380-y

Image Credits: AI Generated

DOI: 10.1038/s41477-026-02380-y

Keywords: CTNIP4, HSL3, phytocytokine, disulfide bond, receptor-like kinase, plant immunity, peptide folding, structural biology, ligand recognition, defense signaling, apoplast, secreted peptides

The research, led by Zhiyun Wang, Lihao Wan, Siqi Tang, Xiao Yu and Shutong Xu of Huazhong Agricultural University in Wuhan, China, focused on a signalling peptide called SCREW—short for SMALL PHYTOCYTOKINES REGULATING DEFENSE AND WATER LOSS. SCREW belongs to a recently identified family of phytocytokines that helps plants balance two competing priorities: sealing stomata and mounting chemical defences during infection, while still regulating water loss and growth. Previous work had shown that SCREW is perceived at the cell surface by a receptor kinase known as NUT, or PLANT SCREW UNRESPONSIVE RECEPTOR, which partners with a shared co-receptor called BAK1, a central hub of plant innate immunity. But the molecular choreography by which SCREW brings these two receptor proteins together had never been visualized.

To capture the interaction in atomic detail, the team used X-ray crystallography to solve the three-dimensional structure of a ternary immune complex from the model plant Arabidopsis thaliana, comprising the extracellular domains of NUT and BAK1 bound to the SCREW peptide. The resulting crystal structure, deposited in the Protein Data Bank, exposes for the first time the precise geometry of the ligand-receptor-coreceptor assembly—and it delivered a surprise. Unlike most well-characterized cysteine-rich peptides, which contain multiple disulfide bonds that fold them into compact, rigid structures, SCREW carries only two conserved cysteine residues forming a single intramolecular disulfide bond. That lone bond, together with a strategically placed proline residue, constrains the peptide’s flexible loop into what the researchers describe as a neck-ring-like or noose-like conformation.

This ring architecture turns out to be far more than a structural curiosity. Within the complex, the carboxy-terminal cyclic region of SCREW wedges itself into the junction between NUT and BAK1, wedging the two receptor ectodomains together while burying an unusually large contact surface on BAK1. In effect, the peptide acts as a molecular clasp: its ringed C-terminus is gripped by the receptor, while the same region simultaneously makes extensive interactions with the co-receptor, nucleating the assembly of the active signalling complex. Mutating the disulfide-forming cysteines, disrupting the critical proline, or altering any of the key interface residues abolished both complex formation in vitro and downstream immune signalling in planta, including MAP kinase activation—demonstrating that the neck-ring conformation is not merely permissive but strictly required for receptor activation.

The finding challenges a long-standing assumption in the field. Because CRPs with multiple disulfide bonds—such as defensins, EPF stomatal peptides, RALF peptides and pollen-attracting LURE peptides—fold into well-defined compact folds that are recognized as intact structural units, researchers have largely assumed that all CRP ligands operate this way. The new structure shows that a two-cysteine CRP can instead rely on a single disulfide to create a constrained loop that functions as a recognition epitope in its own right. The team’s biochemical analyses, including thiol-labeling assays of both recombinant and synthetic SCREW peptides, confirmed that the disulfide bond is formed in the extracellular environment, and surface plasmon resonance experiments quantified how each mutation erodes binding affinity, often eliminating detectable interaction altogether.

Structural comparisons with previously solved receptor complexes sharpened the picture. When the researchers superposed their NUT-SCREW-BAK1 structure onto related complexes such as HAESA-IDA-SERK1, HSL1-IDL1-SERK1, MIK2-SCOOP12-BAK1, PXY-CLE41-SERK2 and FLS2-flg22-BAK1, the overall architecture resembled the canonical mode in which a linear peptide bridges a receptor and a SERK-family co-receptor. But SCREW departs sharply from the mode seen in other receptor-CRP complexes, such as ERL1-EPF1-TMM or PRK6-LURE1.2, where multi-disulfide peptides bind through their folded cores. SCREW thus occupies an intriguing middle ground—structurally a CRP, but functionally behaving much like a linear peptide whose conformational constraint is imposed by disulfide chemistry rather than encoded in a compact fold.

Conservation across species adds an evolutionary dimension to the work. The team showed that the same assembly mechanism operates in rapeseed (Brassica napus), an economically vital crop, where BnSCREW1 induces heterodimerization of BnNUT and BnBAK1 in the same ring-mediated manner. Sequence alignments spanning the Brassicaceae, Solanaceae, Fabaceae and Poaceae families revealed that the two disulfide-forming cysteines and the structural proline are strictly conserved, and AlphaFold3-predicted models of NUT-SCREW-BAK1 complexes from pepper, potato, barrel medic and common bean reproduced the same ternary architecture with high confidence. The authors conclude that this disulfide-dependent recognition mechanism is probably widespread among dicot plants, suggesting it represents a general design principle for two-cysteine CRP signalling.

The implications extend well beyond basic structural biology. BAK1 is a co-receptor shared by dozens of immune and developmental receptor kinases, and understanding how different ligands engage it illuminates how plants achieve specificity with a limited molecular toolkit. Because SCREW signalling regulates both stomatal immunity and water loss, dissecting its activation mechanism at the atomic level could inform strategies for engineering disease-resistant crops that maintain yield under pathogen pressure and drought. The detailed map of the NUT-SCREW-BAK1 interfaces—now verifiable down to individual residues—provides a rational template for designing peptide analogues or screening for small molecules that tune this signalling axis in agriculturally important species.

The study also offers methodological lessons for the broader receptor-kinase community. The researchers had to engineer a functional quadruple-mutant version of the NUT ectodomain to obtain diffraction-quality crystals, carefully verifying by SPR that the engineered protein retained wild-type binding behaviour. Their structural comparisons drew on a decade of crystallographic work on plant receptor complexes, and their hybrid approach—combining crystallography, SPR kinetics, pull-down assays, size-exclusion chromatography, MAP kinase readouts and AlphaFold3 modelling—illustrates how modern structural biology integrates complementary techniques to resolve transient extracellular assemblies. What emerges is a vivid picture of plant immune signalling at the atomic scale: a tiny peptide, shaped like a molecular ring by one covalent bond and one proline, snapping two receptor proteins together to sound the alarm. It is a reminder that in biology, the smallest molecules often carry the heaviest structural burdens—and that a single disulfide bond can be the difference between silence and a full-blown immune response.

Subject of Research: Structural mechanism of SCREW peptide recognition by the NUT-BAK1 receptor complex in plant immunity

Article Title: Disulfide-bond-driven neck-ring-like conformation mediates SCREW recognition in plant immunity

Article References: Wang, Z., Wan, L., Tang, S., Wang, X., Yang, Y., Wu, H., Zhang, S., Yu, X., & Xu, S. (2026). Disulfide-bond-driven neck-ring-like conformation mediates SCREW recognition in plant immunity. Nature Plants, 12(9), 1756-1768. https://doi.org/10.1038/s41477-026-02377-7

Image Credits: AI Generated

DOI: 10.1038/s41477-026-02377-7

Keywords: plant immunity, SCREW peptide, cysteine-rich peptides, disulfide bond, receptor kinase, BAK1 co-receptor, crystal structure, phytocytokines, Arabidopsis thaliana, X-ray crystallography, Brassica napus, peptide signalling