Thaumatin-like protein

Thaumatin-like proteins (TLPs) constitute the pathogenesis-related protein family 5 (PR-5), a class of plant defense proteins with molecular masses of 20–26 kDa, structurally defined by a conserved β-sheet-rich fold stabilized by 8–10 disulfide bridges. Unlike the intensely sweet homolog thaumatin, which activates the human T1R2/T1R3 sweet taste receptor at a threshold concentration of 50 nM, the vast majority of TLPs are non-sweet and instead exhibit antifungal, β-1,3-glucanase, xylanase-inhibitory, or allergenicity properties.

Thaumatin-Like Protein (TLP) Procurement Guide: Sourcing a Multifunctional PR-5 Protein for Non-Sweet Industrial Applications

Thaumatin-like proteins (TLPs) constitute the pathogenesis-related protein family 5 (PR-5), a class of plant defense proteins with molecular masses of 20–26 kDa, structurally defined by a conserved β-sheet-rich fold stabilized by 8–10 disulfide bridges [1]. Unlike the intensely sweet homolog thaumatin, which activates the human T1R2/T1R3 sweet taste receptor at a threshold concentration of 50 nM, the vast majority of TLPs are non-sweet and instead exhibit antifungal, β-1,3-glucanase, xylanase-inhibitory, or allergenicity properties [2]. This functional divergence, rooted in discrete sequence and surface-charge differences within a conserved three-dimensional scaffold, makes TLP selection a matter of precise functional specification rather than generic substitution.

• [1] Liu JJ, Sturrock R, Ekramoddoullah AK. The superfamily of thaumatin-like proteins: its origin, evolution, and expression towards biological function. Plant Cell Rep. 2010;29(5):419-36. doi:10.1007/s00299-010-0826-8. View Source
• [2] Kaneko R, Kitabatake N. Structure-sweetness relationship in thaumatin: importance of lysine residues. Chem Senses. 2001;26(2):167-177. View Source

Why Generic Thaumatin Substitution Fails: Functional Evidence Against Interchanging Thaumatin and Thaumatin-Like Proteins

Although thaumatin and thaumatin-like proteins share 30–80% sequence identity and a common three-domain architecture, they cannot be interchanged for practical applications because key surface residues governing receptor activation and enzymatic activity differ systematically. Thaumatin possesses five lysine residues (K78, K97, K106, K137, K187) essential for sweet taste receptor binding that are absent in non-sweet TLPs [1]. Conversely, many TLPs harbor an acidic interdomain cleft with a Glu-Asp catalytic pair that confers endo-β-1,3-glucanase activity and antifungal potency, features entirely absent from thaumatin [2]. Even among TLPs themselves, quantitative differences in antifungal IC₅₀, glucanase specific activity, pH/thermal stability, and IgE epitope content preclude simple within-class interchange [3]. These molecular differences translate directly into divergent industrial utility, making evidence-based selection mandatory.

• [1] Kaneko R, Kitabatake N. Structure-sweetness relationship in thaumatin: importance of lysine residues. Chem Senses. 2001;26(2):167-177. View Source
• [2] Menu-Bouaouiche L, Vriet C, Peumans WJ, et al. A molecular basis for the endo-β1,3-glucanase activity of the thaumatin-like proteins from edible fruits. Biochimie. 2003;85(1-2):123-131. View Source
• [3] Grenier J, Potvin C, Trudel J, Asselin A. Some thaumatin-like proteins hydrolyse polymeric beta-1,3-glucans. Plant J. 1999;19(4):473-80. doi:10.1046/j.1365-313x.1999.00551.x. View Source

Thaumatin-Like Protein Quantitative Differentiation Evidence: Head-to-Head Data vs. Thaumatin, Brazzein, Osmotin, and Zeamatin

Sweetness Absence in TLPs vs. Thaumatin: A Structural Determinant for Non-Sweet Functional Selection

Thaumatin elicits a sweet taste in humans at a threshold concentration of 50 nM (approximately 100,000 times sweeter than sucrose on a molar basis), whereas thaumatin-like proteins are non-sweet [1]. Chemical modification studies identified five lysine residues—K78, K97, K106, K137, and K187—as essential for sweetness; phosphopyridoxylation of any of these residues abolished sweetness, which was restored upon dephosphorylation (except K106) [2]. Sequence alignment reveals that these five lysine residues are present in sweet thaumatin but absent in non-sweet thaumatin-like proteins [2]. The D21N mutant of thaumatin further lowered the sweetness threshold to 31 nM, confirming the electrostatic basis of receptor activation that TLPs lack [1].

Evidence Dimension	Human sweetness detection threshold and structural determinants
Target Compound Data	Thaumatin-like proteins: non-sweet; critical lysine residues K78, K97, K106, K137, K187 absent
Comparator Or Baseline	Thaumatin: sweet, threshold 50 nM (D21N mutant: 31 nM); five critical lysine residues present
Quantified Difference	Qualitative binary: sweet (thaumatin) vs. non-sweet (TLPs); threshold difference >100,000-fold sucrose equivalence for thaumatin vs. no detectable sweetness for TLPs
Conditions	Human sensory panel evaluation; chemical modification (phosphopyridoxylation) and site-directed mutagenesis; far-UV circular dichroism confirmed no secondary structure disruption

Why This Matters

For procurement in sweetener applications, TLPs are categorically unsuitable; conversely, for antifungal or enzymatic applications where sweetness is an undesirable off-target property, TLPs are the only viable choice within the PR-5 family.

• [1] Masuda T, Ohta K, Mikami B, Kitabatake N. A hypersweet protein: removal of the specific negative charge at Asp21 enhances thaumatin sweetness. Sci Rep. 2016;6:20255. doi:10.1038/srep20255. View Source
• [2] Kaneko R, Kitabatake N. Structure-sweetness relationship in thaumatin: importance of lysine residues. Chem Senses. 2001;26(2):167-177. View Source

Antifungal Activity: Castanopsis TLP IC₅₀ 0.5 µM vs. French Bean and Kiwi TLPs

A thaumatin-like protein purified from seeds of Castanopsis chinensis (Kweilin chestnut) exerted antifungal activity against Fusarium oxysporum with an IC₅₀ of 0.5 µM, and was also active against Botrytis cinerea, Mycosphaerella arachidicola, and Physalospora piricola [1]. In direct comparative assays, Castanopsis TLP was more potent than both French bean TLP and kiwi fruit TLP against F. oxysporum and M. arachidicola [1]. By contrast, thaumatin itself exhibits no antifungal activity, despite its structural homology to TLPs [2]. Castanopsis TLP also inhibited HIV-1 reverse transcriptase with an IC₅₀ of 1.6 µM, a property markedly superior to kiwi fruit TLP (IC₅₀ ≥ 27 µM) [1].

Evidence Dimension	Antifungal IC₅₀ against F. oxysporum
Target Compound Data	Castanopsis chinensis TLP: IC₅₀ = 0.5 µM against F. oxysporum; also active against B. cinerea, M. arachidicola, P. piricola
Comparator Or Baseline	French bean TLP and kiwi fruit TLP: higher IC₅₀ values (exact values not reported but stated as less potent); thaumatin: no antifungal activity; kiwi fruit TLP HIV-1 RT IC₅₀ ≥ 27 µM
Quantified Difference	Castanopsis TLP IC₅₀ = 0.5 µM vs. F. oxysporum; qualitatively more potent than French bean and kiwi TLPs; HIV-1 RT inhibition: 1.6 µM vs. ≥27 µM (≥16.9-fold difference)
Conditions	In vitro antifungal assay; HIV-1 reverse transcriptase inhibition assay; protein purified from seeds (yield 20 mg/kg seeds)

Why This Matters

When selecting a TLP for antifungal or antiviral applications, Castanopsis TLP offers quantitatively superior potency compared to fruit-derived TLPs, reducing the required effective concentration and formulation cost.

• [1] Wang HX, Ng TB. Isolation of a large thaumatin-like antifungal protein from seeds of the Kweilin chestnut Castanopsis chinensis. Biochem Biophys Res Commun. 2003;301(2):364-370. doi:10.1016/S0006-291X(02)03009-7. View Source
• [2] Gowda CT, Purama SNS, Kammara R. TLPdb: a resource for thaumatin-like proteins. Int J Pept Res Ther. 2020;26:301-307. doi:10.1007/s10989-019-09840-2. View Source

β-1,3-Glucanase Activity: TLP Enzymatic Activity (0.07–89 nkat/mg) vs. Zero Activity in Thaumatin and Zeamatin

A systematic survey of thaumatin and 12 purified TLP proteins using an in-gel glucanase assay revealed that six TLPs—including barley leaf permatin, two tomato fruit osmotins, a cherry fruit TLP, and two tobacco stigma TLPs—hydrolyzed carboxymethyl(CM)-pachyman with specific activities ranging from 0.07 to 89 nkat mg⁻¹ [1]. By contrast, thaumatin, zeamatin, tobacco leaf PR-R protein, and four stress-related TLPs from barley and pea exhibited no detectable β-1,3-glucanase activity [1]. Thin-layer chromatography confirmed that active TLPs function as endo-β-1,3-glucanases (EC 3.2.1.x), distinct from classical β-1,3-glucanases (EC 3.2.1.39) [1]. In a separate study, cherry and apple TLPs showed moderate endo-β-1,3-glucanase activity but no antifungal activity, whereas banana TLP inhibited Verticillium albo-atrum hyphal growth but lacked glucanase activity, demonstrating functional uncoupling of these two properties [2].

Evidence Dimension	Endo-β-1,3-glucanase specific activity on CM-pachyman
Target Compound Data	Six TLPs: specific activity range 0.07–89 nkat mg⁻¹ (barley permatin, tomato osmotins, cherry TLP, tobacco stigma TLPs)
Comparator Or Baseline	Thaumatin: 0 nkat mg⁻¹; zeamatin: 0 nkat mg⁻¹; tobacco PR-R: 0 nkat mg⁻¹; four barley/pea stress TLPs: 0 nkat mg⁻¹
Quantified Difference	Activity range 0.07–89 nkat mg⁻¹ for active TLPs vs. 0 nkat mg⁻¹ for thaumatin, zeamatin, and PR-R; some active TLPs also lysed fungal spores
Conditions	In-gel glucanase assay with CM-pachyman substrate; thin-layer chromatography for product analysis; activity confirmed as endo-β-1,3-glucanase

Why This Matters

Only a subset of TLPs possess β-1,3-glucanase activity; procurement must specify the exact TLP isoform, as thaumatin and zeamatin are enzymatically inert and unsuitable for applications requiring fungal cell wall hydrolysis.

• [1] Grenier J, Potvin C, Trudel J, Asselin A. Some thaumatin-like proteins hydrolyse polymeric beta-1,3-glucans. Plant J. 1999;19(4):473-80. doi:10.1046/j.1365-313x.1999.00551.x. View Source
• [2] Menu-Bouaouiche L, Vriet C, Peumans WJ, et al. A molecular basis for the endo-β1,3-glucanase activity of the thaumatin-like proteins from edible fruits. Biochimie. 2003;85(1-2):123-131. View Source

pH and Thermal Stability: TLXI Stable at pH 1–12 and 100°C vs. Thaumatin Limited to pH 2–8 and Heat-Stable Only Below pH 5.5

The wheat thaumatin-like xylanase inhibitor TLXI, a short-chain TLP with five disulfide bridges, is stable for at least 120 minutes across pH 1–12 and withstands 100°C for at least 2 hours without loss of structural integrity or inhibitory activity [1]. This stability far exceeds that of the other two wheat xylanase inhibitors, TAXI and XIP [1]. In contrast, commercial thaumatin is stable in aqueous solution only between pH 2–8 and retains heat stability solely below pH 5.5; above pH 7.0, heating causes irreversible aggregation and loss of sweetness [2]. Atomic structure analysis of thaumatin I at pH 8.0 revealed that the large disulfide-rich region in domain II undergoes conformational changes that destabilize the protein at alkaline pH [3]. Comparative spectroscopic studies of sweet-thaumatin, zeamatin, and a Cassia didymobotrya PR-5 confirmed a hierarchic scale of structural stability, with sweet-thaumatin exhibiting the greatest tendency to denature [4].

Evidence Dimension	pH stability range and thermal stability at 100°C
Target Compound Data	TLXI (thaumatin-like xylanase inhibitor): stable ≥120 min at pH 1–12; stable ≥2 h at 100°C
Comparator Or Baseline	Thaumatin: stable aqueous pH 2–8; heat-stable only below pH 5.5; loses sweetness and aggregates above pH 7.0 upon heating; zeamatin and Cassia PR-5: intermediate stability (hierarchic scale, exact values from spectroscopic unfolding)
Quantified Difference	TLXI: pH 1–12 stability range and 100°C/2 h thermal tolerance; thaumatin: pH 2–8 range and heat-labile above pH 5.5; TLXI stable at pH levels 3 units more acidic and 4 units more alkaline than thaumatin
Conditions	Biochemical characterization: inhibition activity assay after pH/temperature treatment; circular dichroism spectroscopy; X-ray crystallography at pH 8.0 for thaumatin

Why This Matters

For industrial processes involving extreme pH or high-temperature conditions (e.g., alkaline cleaning, high-temperature biocatalysis, or baked product formulations), TLXI-class TLPs offer operability windows that thaumatin cannot tolerate, directly impacting process design and product stability.

• [1] Fierens E, Gebruers K, Brijs K, et al. Biochemical and structural characterization of TLXI, the Triticum aestivum L. thaumatin-like xylanase inhibitor. J Enzyme Inhib Med Chem. 2009;24(3):646-54. doi:10.1080/14756360802321831. View Source
• [2] ChemicalBook. Thaumatin (CAS 53850-34-3): solubility and stability. Accessed 2026. View Source
• [3] Masuda T, Ohta K, Mikami B, Kitabatake N. Atomic structure of the sweet-tasting protein thaumatin I at pH 8.0 reveals the large disulfide-rich region in domain II to be sensitive to a pH change. Biochem Biophys Res Commun. 2012;419(1):72-76. View Source
• [4] Perri F, Romitelli F, Rufini F, Secundo F, Di Stasio E, Giardina B, Vitali A. Different structural behaviors evidenced in thaumatin-like proteins: a spectroscopic study. Protein J. 2008;27(1):13-20. doi:10.1007/s10930-007-9103-2. View Source

Allergenicity Risk Profile: TLP IgE Sensitization Prevalence 1.9% with Low Clinical Reactivity vs. Thaumatin's Weak Sensitizer Status

In a cross-sectional study of 7,176 patients from Italian tertiary allergy centers, TLP sensitization (IgE reactivity to Act d 2 and/or Mal d 2) occurred in only 1.9% of individuals (n = 137), with isolated TLP sensitization being rare (n = 20) and associated with low IgE levels (mean 0.61 ± 0.98 kUA/L) and minimal clinical symptoms [1]. Co-sensitized patients (n = 117) showed higher IgE (mean 1.41 ± 3.49 kUA/L) but ROC analysis demonstrated limited discriminatory power of TLP-specific IgE for predicting reaction severity (AUC 0.51–0.61) [1]. In silico analysis of sapodilla TLP 1 identified two IgE epitopes (residues 36–48 and 51–63), with comparative analysis across allergenic plant TLPs revealing 2–3 IgE epitopes per protein, a lower epitope density than many classical food allergens [2]. By comparison, thaumatin was historically identified as a weak sensitizer in rats upon systemic administration, and allergy testing of exposed laboratory workers showed intermittent reactivity [2]. Among the seven known sweet proteins, only thaumatin, miraculin, monellin, and brazzein have published in silico and in vitro allergenicity data, while pentadin, mabinlin, and curculin remain largely uncharacterized [3].

Evidence Dimension	Clinical IgE sensitization prevalence and quantitative IgE levels
Target Compound Data	TLP (Act d 2, Mal d 2): sensitization prevalence 1.9% (137/7176); isolated TLP IgE mean 0.61 ± 0.98 kUA/L; co-sensitized IgE mean 1.41 ± 3.49 kUA/L; AUC for severity prediction 0.51–0.61
Comparator Or Baseline	Thaumatin: weak sensitizer in rat models; intermittent reactivity in occupationally exposed workers; no large-cohort IgE prevalence data available; brazzein and monellin: recombinant forms assessed for acute/subchronic toxicity and allergenicity in animal models
Quantified Difference	TLP clinical sensitization rate of 1.9% provides a quantitative benchmark; low IgE levels (mean 0.61 kUA/L) and poor severity prediction (AUC ~0.56) indicate low clinical allergenicity risk
Conditions	ALEX2 multiplex IgE detection; 7176-patient Italian cohort; in silico epitope prediction using FARRP, Allermatch, Evaller, BCpreds, BepiPred; homology modeling with banana TLP template (1Z3Q)

Why This Matters

When selecting a TLP for food, pharmaceutical, or cosmetic applications, the low clinical sensitization rate and minimal symptom profile of isolated TLP allergy provide a quantitative risk-assessment benchmark that is absent or less favorable for several alternative sweet proteins, supporting regulatory dossier preparation.

• [1] Scala E, Till SJ, Asero R, et al. Thaumatin-like proteins in molecular allergy diagnostics: uncommon, co-sensitized, and clinically inconspicuous? Insights from an Italian cohort. Allergy. 2025. doi:10.1111/all.70114. View Source
• [2] Ashok Kumar HG, Venkatesh YP. In silico analyses of structural and allergenicity features of sapodilla (Manilkara zapota) acidic thaumatin-like protein in comparison with allergenic plant TLPs. Mol Immunol. 2014;57(2):119-128. View Source
• [3] Palupi PD, Andarwulan N, Lioe HN, et al. Allergenicity assessment of plant-derived sweet proteins—in silico, in vitro, in vivo, and clinical approach: a systematic review. Molecules. 2026;31(9):1424. doi:10.3390/molecules31091424. View Source

Evidence-Based Application Scenarios for Thaumatin-Like Proteins in Research and Industry

Antifungal Biopesticide Development Using Castanopsis TLP (IC₅₀ 0.5 µM)

For agricultural biocontrol formulations targeting Fusarium oxysporum, Castanopsis chinensis TLP provides an IC₅₀ of 0.5 µM, quantitatively superior to French bean and kiwi fruit TLPs [1]. Its additional activity against Botrytis cinerea, Mycosphaerella arachidicola, and Physalospora piricola supports broad-spectrum crop protection. The yield of 20 mg purified protein per kg of chestnut seeds provides a benchmark for sourcing feasibility [1]. Unlike thaumatin, which lacks antifungal activity, Castanopsis TLP is the only PR-5 family member with demonstrated dual antifungal and anti-HIV-1 reverse transcriptase activity (IC₅₀ 1.6 µM), enabling potential dual-use applications [1].

High-Temperature Industrial Biocatalysis with TLXI-Class TLPs (Stable pH 1–12, 100°C for 2 h)

Processes requiring xylanase inhibition under extreme conditions—such as high-temperature cereal processing, alkaline cleaning-in-place (CIP) systems, or lignocellulosic biomass pretreatment—can employ TLXI or TLXI-class thaumatin-like proteins. TLXI retains full inhibitory activity after 120 minutes across pH 1–12 and after 2 hours at 100°C, performance unattainable with thaumatin (limited to pH 2–8, heat-labile above pH 5.5) [2]. The five-disulfide-bridge architecture of TLXI provides the structural basis for this stability [2].

Fungal Cell Wall Hydrolysis Using β-Glucanase-Active TLPs (Specific Activity up to 89 nkat/mg)

For enzymatic degradation of fungal cell walls in research or industrial contexts, only TLPs with demonstrated endo-β-1,3-glucanase activity should be sourced—specifically barley permatin, tomato osmotins, cherry TLP, or tobacco stigma TLPs, which exhibit specific activities of 0.07–89 nkat mg⁻¹ on CM-pachyman [3]. Thaumatin, zeamatin, and tobacco PR-R are completely inactive and must be excluded from procurement specifications. Crucially, glucanase activity and antifungal activity are not correlated, as demonstrated by banana TLP (antifungal but enzymatically inactive) and cherry/apple TLPs (enzymatically active but non-antifungal), requiring functional validation for each intended application [4].

Low-Allergenicity Food Ingredient Sourcing with Quantified TLP IgE Risk Data

For food, beverage, or pharmaceutical formulations where allergen labeling and safety are paramount, TLPs characterized by low clinical sensitization rates (1.9% prevalence in a 7,176-patient cohort, with isolated sensitization associated with mean IgE of 0.61 ± 0.98 kUA/L and minimal symptoms) offer a quantitatively defined risk profile [5]. This contrasts with sweet proteins such as thaumatin and brazzein, which have documented IgE cross-reactivity with pollen allergens (e.g., Mus a 4, Ole e 13 for thaumatin; Art v 1 for brazzein) but lack large-cohort clinical prevalence data [6]. Sapodilla TLP, with only two predicted IgE epitopes, represents a candidate with intrinsically lower allergenic potential [6].

• [1] Wang HX, Ng TB. Isolation of a large thaumatin-like antifungal protein from seeds of the Kweilin chestnut Castanopsis chinensis. Biochem Biophys Res Commun. 2003;301(2):364-370. View Source
• [2] Fierens E, Gebruers K, Brijs K, et al. Biochemical and structural characterization of TLXI, the Triticum aestivum L. thaumatin-like xylanase inhibitor. J Enzyme Inhib Med Chem. 2009;24(3):646-54. View Source
• [3] Grenier J, Potvin C, Trudel J, Asselin A. Some thaumatin-like proteins hydrolyse polymeric beta-1,3-glucans. Plant J. 1999;19(4):473-80. View Source
• [4] Menu-Bouaouiche L, Vriet C, Peumans WJ, et al. A molecular basis for the endo-β1,3-glucanase activity of the thaumatin-like proteins from edible fruits. Biochimie. 2003;85(1-2):123-131. View Source
• [5] Scala E, Till SJ, Asero R, et al. Thaumatin-like proteins in molecular allergy diagnostics: uncommon, co-sensitized, and clinically inconspicuous? Allergy. 2025. doi:10.1111/all.70114. View Source
• [6] Ashok Kumar HG, Venkatesh YP. In silico analyses of structural and allergenicity features of sapodilla (Manilkara zapota) acidic thaumatin-like protein in comparison with allergenic plant TLPs. Mol Immunol. 2014;57(2):119-128. View Source