Engineering and application of a recombinant Aspergillus niger CECT 2907 tannase to enhance the antioxidant activity of black tea

Abstract

BACKGROUND

Tannases (tannin acyl hydrolases, EC 3.1.1.20) catalyze the hydrolysis of galloylated polyphenols, releasing gallic acid and other bioactive compounds with antioxidant properties. These enzymes have a wide range of applications in the food and beverage industry, particularly for improving tea quality. In this study, a tannase gene from Aspergillus niger CECT 2907 was cloned and heterologously expressed in Komagataella phaffii. Two recombinant variants were generated: a double‐chain enzyme (AnTanDC) and a single‐chain variant (AnTanSC) obtained by mutating two Kex2 protease cleavage sites. Both enzymes were biochemically characterized, and the potential of AnTanSC to enhance the antioxidant properties of black tea was evaluated.

RESULTS

Both recombinant tannases were efficiently secreted and produced in 5 L bioreactors, reaching activities of 4238 ± 16 U L⁻¹ (AnTanDC) and 6505 ± 362 U L⁻¹ (AnTanSC). AnTanDC was proteolytically processed into two subunits, whereas AnTanSC remained as a single‐chain enzyme. Both variants showed similar pH and temperature activity profiles, with maximal activity around pH 6.0 and 40 °C. AnTanSC showed improved thermal stability compared with AnTanDC, although lower activity toward tannic acid was observed. Treatment of black tea infusions with AnTanSC increased soluble phenolic content 1.94‐fold and enhanced antioxidant capacity 5.76‐fold and 3.12‐fold according to TEAC and FRAP assays, respectively.

CONCLUSION

The engineered single‐chain tannase AnTanSC combined improved thermal stability with effective enhancement of the antioxidant properties of black tea, supporting its potential application as a biocatalyst for the processing of tea beverages and other polyphenol‐rich foods. © 2026 Society of Chemical Industry.

INTRODUCTION

Tannases (tannin acyl hydrolases, EC 3.1.1.20) hydrolyze the ester and depside bonds of gallotannins, ellagitannins, and gallic acid esters, releasing gallic acid (3,4,5‐trihydroxybenzoic acid). These enzymes are ubiquitous in nature, being identified in plants, animals, bacteria, yeasts, and filamentous fungi. Tannases have broad industrial applications, including the clarification of wine, beer, and fruit juices; the improvement of the nutritional value of animal feed; the production of gallic acid for the chemical and pharmaceutical industries; and the treatment of tannin‐rich industrial effluents.^{1, 2}

One of their most important applications is improving the organoleptic properties of tea, particularly in the production of instant tea powders and ready‐to‐drink beverages.^{1, 2} Tea is the second most consumed beverage worldwide after water and is widely recognized for its health‐promoting properties, particularly due to its high catechin content. ³ Catechins are a group of tannins with strong metal‐chelating and free radical‐scavenging activities, which help reduce oxidative stress.^{4, 5} Regular consumption of tea and other tannin‐rich foods has been associated with a reduced risk of chronic diseases, including cardiovascular disorders, cancer, diabetes, and osteoporosis.^{4, 6}

The composition and physicochemical properties of tea tannins depend strongly on the processing method. Green tea, produced from unfermented leaves, contains high levels of monomeric catechins such as (−)‐epigallocatechin gallate (EGCG), (−)‐epigallocatechin, (−)‐epicatechin gallate, and (−)‐epicatechin. EGCG is generally the predominant catechin in green tea, often representing more than 50% of the total catechin content.^{3, 7} In contrast, during black tea fermentation, about 75% of the catechins undergo oxidation and polymerization, resulting in complex polyphenolic compounds such as theaflavins and thearubigins. These compounds, which constitute up to 5% and 20% of the black tea solid extract, respectively, are responsible for the characteristic flavor, brightness, and reddish‐orange color of the infusion.^{3, 5}

Despite their beneficial bioactivities, tannins show limited bioavailability and may act as antinutritional factors by forming complexes with proteins and polysaccharides, thereby decreasing nutrient absorption and even affecting the pharmacokinetics of clinically used medications.^{8, 9} In addition, tannins contribute to excessive astringency and bitterness in tea infusions, which negatively influence consumer acceptance. Enzymatic treatment with tannases reduces bitterness and turbidity, prevents ‘cream’ formation during cooling, and improves the brightness and color stability of tea beverages. ² Moreover, tannase treatment can enhance the antioxidant capacity and functional properties of tea extracts. ⁷

Commercial tannases are mainly produced through fermentation with Aspergillus spp. and Penicillium spp. However, using native microorganisms often results in low yields and complex downstream processing. This drives up costs and hinders industrial‐scale use. ² To address this, culture optimization, bioprospecting novel microbial sources, and enzyme immobilization are being explored to boost efficiency and reusability.^{10, 11, 12, 13, 14}

Recombinant expression has emerged as a powerful alternative for tannase production, enabling expression in well‐characterized hosts under optimized conditions. Tannase genes from Aspergillus niger, A. oryzae, Penicillium longicatenatum, and the yeasts Rhodosporidium diobovatum and Debaryomyces hansenii have been cloned and successfully expressed in heterologous hosts, including Komagataella phaffii (syn. Pichia pastoris), Yarrowia lipolytica, and industrial Aspergillus spp. strains.^{13, 14, 15, 16, 17, 18, 19} Additionally, bacterial tannases have been expressed in Escherichia coli.^{20, 21, 22, 23}

In parallel, protein engineering strategies, including site‐directed mutagenesis and rational design, have been applied both to elucidate catalytic mechanisms and to improve enzyme properties such as thermostability and substrate specificity.^{14, 24, 25, 26} While recombinant production has significantly advanced tannase research, reports on practical applications in the tea industry remain scarce. Nonetheless, recent efforts include the use of recombinant A. niger tannase for enzymatic extraction and quality improvement of green tea beverages, ¹⁸ the application of P. longicatenatum tannase to enhance antioxidant properties in tea products, ¹⁴ and the immobilization of recombinant A. oryzae tannase expressed in K. phaffii for improving green tea antioxidant activity. ¹³

Despite the growing interest in tannases for improving tea quality, most studies have focused on native enzymes or recombinant tannases applied to green tea extracts; however, applications in black tea processing remain limited. Furthermore, strategies to improve enzyme stability and simplify purification are still needed to facilitate industrial implementation.

In this study, we cloned and engineered a tannase gene from A. niger CECT 2907 to express two enzyme variants in K. phaffii: a double‐chain tannase (AnTanDC) and an engineered single‐chain form (AnTanSC) generated by mutating Kex2 protease cleavage sites. The recombinant enzymes were produced in a 5 L bioreactor, purified by single‐step affinity chromatography, and biochemically characterized. Finally, the potential of AnTanSC to improve the antioxidant properties of black tea infusions was evaluated.

MATERIALS AND METHODS

Chemicals

Unless otherwise specified, all chemicals were of analytical grade and purchased from Sigma‐Aldrich (St Louis, MO, USA), Anedra (Research AG, Buenos Aires, Argentina), or Sintorgan SA (Buenos Aires, Argentina).

Microorganism

Aspergillus niger CECT 2907 (syn. A. awamori NRRL 3112 = ATCC 22342 = MUCL 28815) was obtained from the Colección Española de Cultivos Tipo (CECT), University of Valencia, Spain, and stock cultures were maintained on potato dextrose agar slants at 4 °C.

Cloning of AnTanCECT2907 coding sequence

Total gDNA was extracted from A. niger CECT 2907 mycelia using the Quick‐DNA Fungal/Bacterial kit (Zymo Research, Irvine, CA, USA) according to the manufacturer's instructions and used as the template for polymerase chain reaction (PCR) amplification of the complete coding sequence for tannase (AnTanCECT2907). PCR was performed using Pfu DNA polymerase (GDSBio, Guangzhou, China) and primers TanFw (5′‐TGAACACTCCTTGACCTCTGC‐3′) and TanRv (5′‐CTAGTATACAGGGACCTTGAAG‐3′). Primers were designed based on the A. oryzae tannase gene (GenBank accession D63338.1) and on genes encoding putative tannases identified in genomic sequences of other Aspergillus species available in the GenBank database. The amplified product of ~1.7 kbp was gel‐purified using Wizard SV Gel and PCR Clean‐Up System (Promega, Madison, WI, USA) and cloned into pGEM‐T Easy Vector (Promega) to obtain pGEMT‐AnTanCECT2907. The identity of the cloned DNA fragment was confirmed by sequencing (Macrogen Inc., Seoul, Korea) and comparing the encoded protein with fungal tannases previously reported in the GenBank database (Supporting Information, Fig. S1). Prediction of signal peptide and processing site in the translated protein was performed using SignalP 6.0 software (https://services.healthtech.dtu.dk/services/SignalP-6.0/), and prediction of N‐glycosylation sites with NetNGlyc 1.0 Server (http://www.cbs.dtu.dk/services/NetNGlyc).

Cloning of AnTanDC and AnTanSC variants in K. phaffii

The sequence encoding the mature double‐chain A. niger CECT 2907 tannase (AnTanDC) was amplified by PCR from pGEMT‐AnTanCECT2907 using Pfu DNA polymerase (GDSBio, Guangzhou, China) and primers 5awTANPIC (5′‐TATGAATTCGCTTCTTTTACCGATGTGTG‐3′) and 3awTANPIC (5′‐ATGCGGCCGCTTAGTATACAGGGACC‐3′), which contain EcoRI and NotI restriction sites, respectively (underlined). The amplified DNA fragment was purified, digested with EcoRI and NotI, and cloned into the corresponding sites of the pPIC9NHis vector ²⁷ (Supporting Information, Figs S2 and S3) to obtain plasmid pPICNHIS‐AnTanDC. The recombinant plasmid was verified by restriction enzyme mapping and sequencing (Macrogen Inc.).

To construct a single‐chain version of A. niger CECT 2907 tannase (AnTanSC), containing R311T R316G mutations at the two exposed KR sites recognized by Kexin protease KEX2 (Supporting Information, Fig. S3), PCR amplifications were performed using pPICNHis‐AnTanDC as the template. Primers 5′AOX1 (5′‐GACTGGTTCCAATTGACAAGC‐3′) and TanmutRv (5′‐TTACTAGTCTTGCCATTGCTGAAG3′, SpeI restriction site underlined) were used to amplify the 5′ DNA fragment of the coding sequence, and primers TanmutFw (5′‐AGACTAGTAATGTCAAGGGTCAGGC‐3′, SpeI restriction site underlined) and 3′AOX1 (5′‐GCAAATGGCATTCTGACATCC‐3′) to amplify the 3′ DNA fragment. The 5′ fragment was digested with EcoRI and SpeI, and the 3′ fragment with SpeI and NotI, then purified and ligated to EcoRI‐NotI restriction sites of pPIC9NHis to obtain pPICNHis‐AnTanSC. The correct assembly of the AnTanSC coding sequence in the plasmid was verified by restriction enzyme mapping and sequencing (Macrogen Inc.).

Vectors pPICNHis‐AnTanDC and pPICNHis‐AnTanSC were linearized with DraI restriction endonuclease to favor integration at the AOX1 (alcohol oxidase 1) locus in the GS115 strain of K. phaffii, which results in a His⁺ Mut^s phenotype. The linearized vectors were then used to transform electrocompetent K. phaffii GS115 cells, prepared according to the manufacturer's instructions (Thermo Fisher Scientific, Carlsbad, CA, USA), using a MicroPulser Electroporator (Bio‐Rad Laboratories Inc., Hercules, CA, USA). Recombinant clones reverting histidine auxotrophy were selected by growth on minimal dextrose medium plates (3.4 g L⁻¹ yeast nitrogen base without amino acids, 10 g L⁻¹ (NH₄)₂SO₄, 20 g L⁻¹ dextrose, 0.4 mg L⁻¹ biotin, and 20 g L⁻¹ agar). Clones expressing and secreting active tannase were identified by the presence of degradation halos on minimal methanol medium plates (3.4 g L⁻¹ yeast nitrogen base without amino acids, 10 g L⁻¹ (NH₄)₂SO₄, 0.4 mg L⁻¹ biotin, and 20 g L⁻¹ agar) supplemented with 10 g L⁻¹ tannic acid under induction of the AOX1 promoter achieved by adding 100 μL of pure methanol to the plate lid every 24 h.

Ten clones from each construct showing the largest tannic acid degradation halos on minimal methanol plates were selected to evaluate recombinant tannase expression under shake‐flask cultivation. Pre‐inoculum was prepared in 5 mL YPD medium (10 g L⁻¹ yeast extract, 20 g L⁻¹ peptone, 20 g L⁻¹ dextrose) and used to inoculate 10 mL of BMGY medium (10 g L⁻¹ yeast extract, 20 g L⁻¹ peptone, 3.4 g L⁻¹ yeast nitrogen base without amino acids, 10 g L⁻¹ (NH₄)₂SO₄, 0.4 mg L⁻¹ biotin, 10 g L⁻¹ glycerol, 100 mmol potassium phosphate buffer, pH 6.0) in 250 mL shake flasks and grown for 48 h at 30 °C and 220 rpm (3.7 s⁻¹). Cells were harvested by centrifugation at 1500 × g for 5 min and resuspended in 25 mL BMMY medium (10 g L⁻¹ yeast extract, 20 g L⁻¹ peptone, 3.4 g L⁻¹ yeast nitrogen base without amino acids, 10 g L⁻¹ (NH₄)₂SO₄, 0.4 mg L⁻¹ biotin, 100 mmol ⁻¹ potassium phosphate buffer, pH 6.0) to a final OD₆₀₀ of 10. Cultures were then incubated in 250 mL shake flasks at 30 °C and 220 rpm (3.7 s⁻¹). Sterile methanol was added every 24 h to a final concentration of 7.9 g L⁻¹ (1% v/v) to maintain induction conditions. Tannase activity was measured every 24 h as described in the 'Tannase activity assay' section, below. Clones showing the highest tannase activity after 4 days of cultivation were selected for bioreactor production.

Production of AnTanDC and AnTanSC

Production of the recombinant proteins was performed in a 5 L Biostat B5 Plus bioreactor (Sartorius, Göttingen, Germany) using a four‐stage strategy comprising batch (I), fed‐batch (II), transition (III), and induction (IV) phases. Seed cultures were obtained by inoculating YPD medium with previously prepared K. phaffii working cell bank for each recombinant clone and incubating for 24 h at 30 °C and 200 rpm (3.3 s⁻¹).

The batch phase (I) was initiated by inoculating seed cultures to an OD₆₀₀ of 0.6–0.8 in 1.0 L of LSM medium (15.6 g L⁻¹ H₃PO₄, 0.23 g L⁻¹ CaSO₄, 4.55 g L⁻¹ K₂SO₄, 3.73 g L⁻¹ MgSO₄.7H₂O, 1.03 g L⁻¹ KOH, 40 g L⁻¹ glycerol, 4.9 mg L⁻¹ biotin, and 4.35 mL L⁻¹ PTM1 trace salts: 6 g L⁻¹ CuSO₄·5H₂O, 0.08 g L⁻¹ KI, 3 g L⁻¹ MnSO₄.H₂O, 0.2 g L⁻¹ Na₂MoO₄.2H₂O, 0.02 g L⁻¹ H₃BO₃, 0.5 g L⁻¹ CoCl₂.6H₂O, 20 g L⁻¹ ZnCl₂, 65 g L⁻¹ FeSO₄.7H₂O, and 9 g L⁻¹ H₂SO₄).

The fed‐batch phase (II) was started after dissolved oxygen (DO) increased over 65% by applying a DO‐stat control strategy to prevent substrate accumulation. The feeding solution (756 g L⁻¹ glycerol, supplemented with 12.25 mL L⁻¹ PTM1 trace salts and 4.9 mg L⁻¹ biotin) was supplied using a feeding pump automatically activated when the DO exceeded 45% and stopped when it dropped below 35%. Process control was performed using MFCS/Win 3.1 software (Sartorius).

The transition phase (III) was initiated at OD₆₀₀ ~ 150 by adding a pulse of pure methanol to a final concentration of 1.98 g L⁻¹ (0.25% v/v) supplemented with 12.25 mL L⁻¹ PTM1 trace salts and 4.9 mg L⁻¹ biotin, to adapt K. phaffii metabolism to methanol utilization.

The induction phase (IV) started once the DO increased above 65%, applying a DO‐stat control strategy as described for the fed‐batch phase, but using methanol as both carbon source and inductor of AOX1 promoter.

The batch and fed‐batch phases were conducted at 30 °C, while the transition and induction phases were performed at 25 °C.

The total cultivation time was 96 h, including 72 h of methanol induction. Agitation was maintained between 300 and 1000 rpm (5–16.7 s⁻¹) with an airflow rate of 0.5–2 vvm (8.3 × 10⁻³–3.3 × 10⁻² s⁻¹). Dissolved oxygen was monitored using a polarographic probe (Mettler Toledo, Giessen, Germany) and maintained at 20% saturation with filter‐sterilized air. Foam formation was automatically controlled using an integrated level probe and the addition of antifoam (100 mg mL⁻¹ stock solution of simethicone). The pH was set at 5.0 by automatic addition of 14% (v/v) NH₄OH (~35 g L⁻¹ as NH₃) or 15% (v/v) H₃PO₄ (~215 g L⁻¹) using a pH probe (Mettler Toledo, Giessen, Germany).

After cultivation, the culture was centrifuged at 17 000 × g for 15 min at 4 °C, and the supernatant was filtered through a hydrophilic cellulose acetate capsule filter (0.2–0.45 μm). The cell‐free supernatant was concentrated and diafiltered using a tangential flow filtration system (Centramate 500 S, Paul Corp., Port Washington, NY, USA) equipped with a 10 kDa cutoff membrane against 50 mmol L⁻¹ sodium phosphate buffer, pH 8.0, and stored at −20 °C for further analysis. Tannase activity was measured at each processing step.

Purification of AnTanDC and AnTanSC

Recombinant tannases were purified from concentrated culture supernatants by gravity‐flow nickel–nitrilotriacetic acid (Ni‐NTA) affinity chromatography using His‐Select Nickel Affinity Gel (Sigma‐Aldrich, St Louis, MO, USA), following a protocol previously described for His‐tagged laccases. ²⁸ Briefly, binding was performed in equilibration buffer (17.5 g L⁻¹ NaCl, 50 mmol L⁻¹ sodium phosphate buffer, pH 8.0), followed by two washes with washing buffer (17.5 g L⁻¹ NaCl, 50 mmol L⁻¹ sodium phosphate buffer, pH 6.5), and elution with washing buffer supplemented with 17 g L⁻¹ imidazole (250 mmol L⁻¹).

Tannase activity assay

Tannase activity was determined by quantifying the release of gallic acid after incubation of the enzyme with 10 mmol L⁻¹ propyl gallate or 0.5 mmol L⁻¹ tannic acid at 40 °C for 10 min in a 50 mmol L⁻¹ sodium phosphate buffer, pH 6.0. Gallic acid was detected spectrophotometrically through the formation of a chromogenic complex with rhodanine (2‐thio‐4‐ketothiazolidine), following the method described by Sharma et al. ²⁹ Briefly, 120 μL of the enzyme reaction was mixed with an equal volume of 6.67 g L⁻¹ methanolic rhodanine solution and incubated for 5 min. Then, 120 μL of 0.5 mol L⁻¹ KOH was added to enhance color development. After a further 5 min incubation, absorbance was measured at 520 nm. Gallic acid concentration was calculated using a standard curve. Control reactions containing substrate and enzyme extract previously inactivated at 100 °C for 10 min were performed under identical conditions, and their values were subtracted from the experimental results. Total protein in purified fractions was determined using the Bradford method with bovine serum albumin as the standard, ³⁰ and specific activity was determined for purified enzyme fractions as enzyme activity units per milligram of total protein. All assays were performed in triplicate. Statistical analysis was performed using an unpaired two‐tailed t‐test, with P < 0.05 considered significant. Analyses were conducted using GraphPad Prism version 5.00 (GraphPad Software, San Diego, CA, USA).

One unit (U) of enzyme activity was defined as the amount of enzyme required to release 1 μmoL of gallic acid per min under the specified assay conditions.

Biochemical characterization

The effect of pH on the activity of AnTanDC and AnTanSC was evaluated at 30 °C using 10 mmol L⁻¹ propyl gallate as the substrate in 50 mmol L⁻¹ sodium citrate buffer at pH 4.5, 5.0, and 5.5 or in 50 mmol L⁻¹ sodium phosphate buffer at pH 6.0, 6.5, 7.0, 7.5, and 8.0. The effect of temperature on enzymatic activity was assessed by incubating the reactions at temperatures ranging from 30 to 70 °C in a 50 mmol L⁻¹ sodium phosphate buffer, pH 6.0. The release of gallic acid was quantified using the rhodanine‐based assay described in the 'Tannase activity assay' section, above. All assays were performed in triplicate. Relative activity values shown in the figures were normalized to the maximum activity of each enzyme solely for graphical representation and comparison of activity profiles.

Thermostability was assessed by pre‐incubating AnTanDC and AnTanSC at 30, 40, and 50 for up to 2 h. Samples were collected at 5, 10, 15, 30, 60, and 120 min, and stored at 4 °C until further analysis. Residual tannase activity was measured at 40 °C in a 50 mmol L⁻¹ sodium phosphate buffer, pH 6.0, as described in the 'Tannase activity assay' section, above. Residual activity was expressed as a percentage relative to the activity at t = 0. All experiments were conducted in triplicate.

Statistical analyses were performed on absolute activity values (U mL⁻¹) using one‐way ANOVA, followed by Tukey's multiple comparison. Differences were considered significant at P < 0.05. Analyses were conducted using GraphPad Prism version 5.00 (GraphPad Software).

Deglycosylation assay

Recombinant AnTanDC and AnTanSC were denatured in 5 g L⁻¹ sodium dodecyl sulfate (SDS) and 40 mmol L⁻¹ dithiothreitol at 100 °C for 5 min, and subsequently incubated with Endo Hf endoglycosidase (New England BioLabs, Ipswich, MA, USA) following the manufacturer's instructions. Deglycosylated proteins were analyzed by western blotting.

Polyacrylamide gel electrophoresis and immunoblotting

Fermentation samples and purified enzymes were resolved by SDS–polyacrylamide electrophoresis (PAGE) under reducing conditions using 12% (w/v) (120 g L⁻¹) and 10% (w/v) (100 g L⁻¹) polyacrylamide gels, respectively, and visualized by Coomassie Brilliant Blue staining. For western blot analysis, purified AnTanDC and AnTanSC were resolved by SDS‐PAGE under reducing conditions on 10% (w/v) (100 g L⁻¹) polyacrylamide gels and subsequently transferred onto 0.45 μm nitrocellulose membranes (Bio‐Rad Laboratories) for western blotting. Membranes were probed with 0.1 μg mL⁻¹ monoclonal mouse anti‐His6 antibody (Roche, Basel, Switzerland), followed by a 1:15 000 dilution of alkaline phosphatase‐conjugated goat anti‐mouse antibody (Sigma‐Aldrich, St Louis, MO, USA). Detection was performed using 5‐bromo‐4‐chloro‐3‐indolyl phosphate and nitroblue tetrazolium as chromogenic substrates.

Zymography

Native PAGE was performed on 10% (w/v) polyacrylamide gels (100 g L⁻¹) at pH 8.8 under non‐denaturing conditions. Duplicate samples of purified AnTanDC and AnTanSC were loaded onto the same gel and run under identical conditions. Following electrophoresis, the gel was divided into two halves for protein visualization by Coomassie Brilliant Blue staining and for activity detection. For activity staining, the gel was equilibrated twice for 20 min in 50 mmol L⁻¹ sodium acetate buffer, pH 5.5, then incubated for 30 min in 25 mmol L⁻¹ propyl gallate prepared in the same buffer. After removing the substrate solution, the released gallic acid was visualized by incubation for 5 min in 6.67 g L⁻¹ rhodanine solution, followed by development in 0.5 mol L⁻¹ KOH. The resulting purple rhodanine–gallic acid complex was used as an indicator of tannase activity.

Tannase activity on black tea

A black tea infusion was prepared by steeping 0.5 g of commercially dried tea leaves in 50 mL of boiling distilled water for 5 min, followed by filtration and cooling to 24 °C. The pH was adjusted to 6.0 with NaOH. To assess tannase activity, 0.9 mL of tea infusion was incubated with 0.1 mL of AnTanSC (85 U mL⁻¹ of the purified enzyme solution in 50 mmol L⁻¹ sodium phosphate buffer, pH 6.0) at 30 °C for 25 min. Control reactions were prepared under identical conditions by replacing the enzyme solution with an equal volume of 50 mmol L⁻¹ sodium phosphate buffer, pH 6.0. Non‐hydrolyzed polyphenols were precipitated by adding 0.5 mL of cold food‐grade gelatin solution (2.5 g L⁻¹ in an acid‐supersaturated NaCl solution at 80 °C) and 3.5 mL of supersaturated NaCl solution, followed by centrifugation at 1177 × g and 4 °C for 10 min. The resulting supernatant was used for subsequent analyses.

Total phenol content was determined by the Folin–Ciocalteu method as described by Singleton et al., ³¹ with minor modifications. Briefly, 0.2 mL of sample, control, or gallic acid standard (0–0.45 mmol L⁻¹) was mixed with 0.2 mL Folin–Ciocalteu reagent and 1.8 mL distilled water. After pH neutralization with 0.2 mL of 200 g L⁻¹ sodium carbonate saturated solution, samples were incubated at room temperature in the dark for 1 h. Absorbance was measured at 765 nm. Results were expressed as milligrams of gallic acid equivalents per milliliter.

Antioxidant capacity was evaluated using the Trolox equivalent antioxidant capacity (TEAC) and ferric reducing antioxidant power (FRAP) assays.

For TEAC, the 2,2′‐azino‐bis(3‐ethylbenzothiazoline‐6‐sulfonic acid) radical cation (ABTS^⦁+), scavenging activity was determined following the procedure described by Re et al., ³² with minor modifications. An ABTS^⦁+ stock solution was prepared by reacting 7 mmol L⁻¹ ABTS with 2.45 mmol L⁻¹ potassium persulfate in the dark for 16 h, then diluted to an absorbance of 0.75 ± 0.1 at 735 nm. A 0.1 mL aliquot of sample, control, or Trolox (6‐hydroxy‐2,5,7,8‐tetramethylchroman‐2‐carboxylic acid) standard (7–30 μmol L⁻¹) was mixed with 1.9 mL of the diluted ABTS^⦁+ solution, incubated in the dark for 15 min, and the absorbance was read at 735 nm. The percentage of ABTS^⦁+ absorbance inhibition was calculated using Eqn (1):

$$\%\text{Inhibition}=\frac{\left(A_0-A_{\mathrm{f}}\right)}{A_0}\times 100$$ (1)

where A ₀ is the initial absorbance at t = 0 and A _f is the absorbance after 15 min of incubation. TEAC results were expressed as milligrams of Trolox equivalents per milliliter.

For FRAP, the reduction of ferric ions (Fe³⁺) to ferrous ions (Fe²⁺) was determined according to Benzie and Strain, ³³ with minor modifications. Briefly, 0.1 mL of sample, control, or ascorbic acid standard (0–150 mg L⁻¹) was added to 3 mL of freshly prepared FRAP reagent (300 mmol L⁻¹ acetate buffer, pH 3.6; 10 mmol L⁻¹ TPTZ (2,4,6‐Tris(2‐pyridyl)‐s‐triazine) in 40 mmol L⁻¹ HCl; 20 mmol L⁻¹ FeCl₃, 10:1:1 v/v/v), incubated in the dark for 10 min, and absorbance read at 593 nm. Results were expressed as milligrams of ascorbic acid equivalents per milliliter.

Ferrous ion chelating ability was determined following Hossain et al., ³⁴ with minor modifications. A 50 μL aliquot of sample, control, or ethylenediaminetetraacetic acid (EDTA) standard (30–50 mg L⁻¹) was mixed with 50 μL of 2 mmol L⁻¹ FeCl₂ and 2.35 mL water, incubated for 5 min, and then 100 μL of 5 mmol L⁻¹ ferrozine was added. After 10 min in the dark, absorbance was measured at 562 nm. Chelating ability was expressed as a percentage of inhibition of ferrous–ferrozine complex according to Eqn (2):

$$\text{Ferrous}\mathrm{ion}\text{chelating ability}\left(\%\right)=\left[1-\frac{A_1}{A_0}\right]\times 100$$ (2)

where A ₀ is the blank absorbance and A ₁ is the sample absorbance. Results were expressed as milligrams of EDTA equivalent per milliliter using a standard curve.

All determinations were performed in triplicate. Statistical analysis was performed using an unpaired two‐tailed t‐test, with P < 0.05 considered significant. Analyses were conducted using GraphPad Prism version 5.00 (GraphPad Software).

Protein structural modeling and visualization

The structure of AnTanSC was predicted using the AlphaFold Server ³⁵ and the amino acid sequence of the mature enzyme. Model confidence was evaluated using the predicted Local Distance Difference Test (pLDDT) scores provided by the server.

The predicted structure was compared with the crystallographic structure of the tannase An‐Tan from Aspergillus niger SH‐2 (PDB ID: 7K4O). Structural visualization, superposition, and figure preparation were performed using Mol* 3D Viewer. ³⁶ Structural models were used exclusively for qualitative comparative analysis.

Data availability

AnTanCECT2907 nucleotide sequence was deposited at DDBJ/EMBL/GenBank under the accession number PX520001.

RESULTS

Cloning and heterologous expression of the tannase from A. niger CECT 2907 in K. phaffii

Aspergillus niger CECT 2907 was previously identified as a tannase producer during a screening of bacterial and fungal strains performed in our laboratory (manuscript in preparation). PCR amplification using genomic DNA from A. niger CECT 2907 as the template yielded a 1767 bp DNA fragment containing an intronless open reading frame encoding a 588‐amino‐acid protein, including a predicted 18‐residue N‐terminal signal peptide. The deduced protein (designated AnTanCECT2907) shares high sequence identity with characterized tannases within the Aspergillus genus: 99.49% with AoTanA from A. oryzae RIB40 (GenBank accession BAA09656.1), ³⁷ 77.15% with KM71ANT from A. niger GH1 (GenBank accession AKM52312.1), ³⁸ 49.75% with AoTanB from A. oryzae RIB40 (GenBank accession XP_001820636.1), ¹⁶ 49.66% with a tannase from A. niger N5‐5 (GenBank accession AKC01963.1), ³⁹ and 48.44% with An‐Tan (Tan7) from A. niger SH‐2 (GenBank accession XP_001401809.1). ⁴⁰

Protein sequence alignment of AnTanCECT2907 with these characterized tannases (Supporting Information, Fig. S1) showed the catalytic triad composed of Ser195, Asp455, and His501 within the CS‐D‐HC signature, which is a characteristic feature of fungal tannases and feruloyl esterases. ⁴¹ In addition, the catalytic serine is located within the conserved GXSXG motif typical of serine hydrolases. ⁴² The two cysteine residues from the CS‐D‐HC signature (Cys194–Cys502) form a disulfide bridge connecting the catalytic serine and histidine. AnTanCECT2907 has eight other conserved cysteine residues, which are likely to form four additional disulfide bridges (Cys25–Cys71, Cys248–Cys294, Cys261–Cys278, and Cys550–Cys566), according to crystallographic studies on An‐Tan from A. niger SH‐2. ⁴³ Residues contributing to the galloyl substrate‐binding site are also conserved in AnTanCECT2907, including Gln227, Gln228, His231, Ser456, and Ile457. Additionally, AnTanCECT2907 is likely to be highly glycosylated, showing multiple predicted N‐linked glycosylation sites (N‐X‐S/T).

In microbial tannases, a highly variable region composed of a two‐stranded antiparallel β‐sheet and a surface‐exposed flexible loop was predicted to form a lid that regulates substrate accessibility, as revealed by structure studies.^{43, 44} In AnTanCECT2907, this region contains two Kex2‐like protease recognition sites (KR motifs) at positions K310–R311 and K315–R316, which may be susceptible to post‐translational proteolytic processing (Supporting Information, Fig. S1). To obtain a recombinant tannase for industrial use, AnTanCECT2907 was expressed in K. phaffii, where the surface‐exposed KR motifs are expected to be cleaved by the endogenous Kex2 protease, yielding a double‐chain protein (AnTanDC) (Supporting Information, Fig. S3). Additionally, a single‐chain version of AnTanCECT2907 (AnTanSC; Supporting Information, Fig. S3), was designed by mutation of the arginine residues in the surface‐exposed KR motifs to threonine (R311T) and glycine (R316G). These substitutions were selected because both residues are among the most abundant in the flexible loop of AnTanCECT2907 and other Aspergillus tannases (Supporting Information, Fig. S1). This mutant was expected to remain as a single‐chain protein when expressed in K. phaffii, as it has been previously demonstrated that any amino acid substitution of the arginine residue within the KR motif prevents recognition by Kex2 proteases.^{45, 46} The sequences encoding both mature AnTanDC and AnTanSC were cloned in‐frame with the Saccharomyces cerevisiae α‐factor secretion signal and an N‐terminal 6 × His tag into the pPIC9NHIS vector, under the control of the methanol‐inducible AOX1 promoter (Supporting Information, Fig. S2), and used to transform the GS115 strain of K. phaffii. Both constructs generated K. phaffii transformants that secreted active tannase, as evidenced by clear degradation halos on agar plates containing tannic acid (Supporting Information, Fig. S4).

Recombinant tannase production and purification

For the recombinant production of AnTanDC and AnTanSC, an initial screening step was performed to select K. phaffii clones showing the highest tannase activity under shake‐flask cultivation. To this end, ten different clones obtained from agar plate screening on tannic acid‐containing medium were evaluated. Tannase activity in the culture supernatants ranged from 16 ± 2 U L⁻¹ in low‐producing clones, to 43 ± 2 U L⁻¹ and 77 ± 6 U L⁻¹ for the highest‐producing clones of AnTanDC and AnTanSC, respectively, which were selected for scale‐up production. After 72 h of methanol induction in 5 L bioreactors, tannase activity in the culture supernatants reached 4238 ± 16 and 6505 ± 362 U L⁻¹ for AnTanDC and AnTanSC, respectively, measured at 40 °C and pH 6.0 using propyl gallate as the substrate. This represents a 98‐fold and 84‐fold increase in AnTanDC and AnTanSC production, respectively, compared to shake‐flask fermentation.

SDS‐PAGE protein profiles of fermentation samples collected after 24, 48, and 72 h of methanol induction showed an enrichment of bands with apparent molecular weights ranging from 45 to 60 kDa for AnTanDC and approximately 120 kDa for AnTanSC as determined by comparison with protein molecular weight markers (Fig. 1(A),(B)). Similar bands were observed after Ni‐NTA affinity chromatography purification, suggesting that they correspond to glycosylated forms of the recombinant enzymes (Fig. 2(A)). Following treatment with EndoHf endoglycosidase, a narrowing of the smeared bands was observed (Fig. 2(B)). In AnTanDC samples, two major bands migrating between the 26 and 37 kDa molecular weight markers were detected, consistent with the predicted molecular masses of the deglycosylated N‐terminal and C‐terminal subunits (33 and 31 kDa, respectively). In AnTanSC samples, a major band corresponding to the molecular mass of the deglycosylated single‐chain enzyme (64 kDa) was observed, although its visualization was partially affected by overlap with the EndoHf enzyme (~70 kDa). Western blot analysis using an anti‐His antibody revealed the glycosylated forms of the N‐terminal subunit of AnTanDC and the full‐length AnTanSC, as well as the corresponding deglycosylated forms after EndoHf treatment (Fig. 2(C)). Additionally, faint lower‐molecular‐weight bands were also detected.

Figure 1.

Electrophoretic characterization of fermentation samples obtained at different induction times. Reducing sodium dodecyl sulfate–polyacrylamide gel electrophoresis (12% w/v polyacrylamide) stained with Coomassie Brilliant Blue showing AnTanDC (A) and AnTanSC (B). M1: Precision Plus Protein All Blue Prestained Protein Standards (Bio‐Rad Laboratories). Lanes 0, 24, 48, and 72 correspond to the induction times (h), respectively.

Figure 2.

Electrophoretic characterization of purified recombinant AnTanDC and AnTanSC. (A) Reducing sodium dodecyl sulfate–polyacrylamide gel electrophoresis (10% w/v polyacrylamide) stained with Coomassie Brilliant Blue showing purified AnTanDC and AnTanSC. M1: Precision Plus Protein All Blue Prestained Protein Standards (Bio‐Rad Laboratories). The molecular weights of the protein standards are shown on the right‐hand side of the gel. (B) Reducing sodium dodecyl sulfate–polyacrylamide gel electrophoresis (10% w/v polyacrylamide) stained with Coomassie Brilliant Blue showing purified AnTanDC and AnTanSC after EndoHf treatment. Arrows indicate the putative bands corresponding to the deglycosylated N‐terminal and C‐terminal chains of AnTanDC and to the deglycosylated single‐chain AnTanSC. EndoHf is indicated by an asterisk. M2: BenchMark Pre‐Stained Protein Ladder (Life Technologies, Carlsbad, CA, USA). (C) Western blot analysis of purified enzymes before (−) and after (+) EndoHf treatment, revealed using an anti‐His tag antibody. M2: BenchMark Pre‐Stained Protein Ladder (Life Technologies). The molecular weights of protein standards are shown on the right‐hand side of the blot. (D) Native polyacrylamide gel electrophoresis (10% w/v polyacrylamide). Left: activity staining using propyl gallate as the substrate. Right: purified AnTanSC loaded onto the same gel and stained with Coomassie Brilliant Blue to indicate the migration position of the single‐chain enzyme.

Native PAGE zymography using propyl gallate as the substrate revealed enzymatic activity only for the high‐molecular‐weight species of both enzymes, as determined by comparison with the migration of AnTanSC run under identical native conditions and revealed by Coomassie Brilliant Blue staining (Fig. 2(D)). These results indicate that both enzymes are secreted as highly glycosylated, active monomeric forms, with AnTanDC undergoing processing into two subunits linked by a disulfide bond. At the same time, AnTanSC remains a single‐chain enzyme due to mutations at the KR recognition sites.

Purified AnTanDC and AnTanSC did not differ significantly (P > 0.05) in the specific activity against propyl gallate (2.03 ± 0.01 U mg⁻¹ and 2.92 ± 0.15, respectively). However, AnTanDC showed a significantly higher specific activity than AnTanSC on tannic acid (P < 0.05), with values of 10.66 ± 0.01 and 8.38 ± 0.44 U mg⁻¹, respectively.

Biochemical characterization of recombinant tannases

AnTanDC and AnTanSCs showed similar pH and temperature activity profiles (Fig. 3(A),(B)). Statistical analysis of absolute activity values (U mL⁻¹), performed independently for each enzyme, revealed a broad pH activity range, with no significant differences between pH 5.0 and 7.0 for AnTanDC and between pH 5.5 and 6.5 for AnTanSC (P > 0.05), while both enzymes showed maximal activity around pH 6.0. Additionally, both enzymes exhibited significantly lower activity at pH values below or above the optimal pH range (P < 0.05–0.001). Regarding temperature, no significant differences in absolute activity were observed between 30 , 40, and 50 °C for either AnTanDC or AnTanSC (P > 0.05). However, for AnTanDC activity at 60 and 70 °C was significantly lower than that observed between 30 and 50 °C (P < 0.001), and activity at 70 °C was significantly lower than at 60 °C (P < 0.001). Similarly, for AnTanSC, activity at 60 °C was slightly reduced, although significant differences were only observed relative to 40 °C (P < 0.05). In contrast, activity at 70 °C was significantly lower than that measured at all other temperatures evaluated (P < 0.001).

Figure 3.

Activity profiles of AnTanDC and AnTanSC. Tannase activity was determined using propyl gallate as substrate at variable pH at 30 °C (A) or at variable reaction temperature at pH 6.0 (B). Activity is represented as a percentage relative to the maximum for each curve, with 100% assigned to the maximum. Data are presented as the mean ± standard deviation of triplicates. Statistical analyses were performed independently for each enzyme using absolute activity values (U mL⁻¹). Significant differences among conditions within each enzyme are described in the text.

Additionally, AnTanSC demonstrated greater thermal stability than AnTanDC, with no statistically significant loss of absolute activity after 2 h of pre‐incubation at 30 °C (P > 0.05). AnTanSC also showed half‐lives of 45 min at 40 °C, and 13 min at 50 °C. Conversely, AnTanDC showed half‐lives of 50 and 20 min at 30 and 40 °C, respectively, and less than 5 min at 50 °C (Fig. 4).

Figure 4.

Thermostability of recombinant tannases. Tannase activity was determined after pre‐incubation of the purified AnTanDC (A) and AnTanSC (B) at 30 °C (squares), 40 °C (circles), and 50 °C (triangles) for the indicated times. Activity is expressed as a percentage of the initial time (T _o), set to 100%. Data are presented as the mean ± standard deviation of triplicates.

Enhancement of antioxidant properties in black tea by AnTanSC

Since AnTanSC showed greater stability than AnTanDC, it was selected to assess the improvement of the antioxidant potential of black tea. Treatment of tea infusions with AnTanSC resulted in a 1.94‐fold increase in soluble phenolic content (Fig. 5(A)). In addition, the antioxidant capacity increased 5.76‐fold and 3.12‐fold, as determined by ABTS radical scavenging activity (TEAC assay, Fig. 5(B)) and ferric reducing antioxidant power (FRAP assay, Fig. 5(C)), respectively. In contrast, the ferrous ion chelating capacity did not differ significantly between treated and control samples (Fig. 5(D)). This suggests that the increase in antioxidant capacity mainly resulted from the greater availability of free phenolic compounds derived from the enzymatic hydrolysis of galloylated catechins, theaflavins, and thearubigins, rather than from enhanced metal‐chelating properties.

Figure 5.

Activity of AnTanSC on black tea. Total soluble phenols (A), ABTS radical scavenging activity expressed as TEAC (B), ferric reducing antioxidant power (FRAP) (C), and metal ion chelation capacity (D) were determined in black tea samples treated with tannase and controls after 25 min of incubation at 30 °C. Results are expressed as milligrams of quantified equivalents per milliliter of tea supernatant after polyphenol precipitation. Data are presented as the mean ± standard deviation of triplicates.

DISCUSSION

The present study describes the cloning, engineering, recombinant production in K. phaffii, and functional application of the tannase AnTanCECT2907 from A. niger CECT 2907 to improve the antioxidant properties of black tea beverages. To obtain a robust recombinant enzyme suitable for bioreactor production and industrial applications, two protein variants were generated: AnTanDC and AnTanSC. For AnTanDC, post‐translational processing by K. phaffii Kex2 protease was expected at two intramolecular surface‐exposed KR motifs. These KR sites are not fully conserved among Aspergillus tannases, as enzymes containing two, one, or none of these sites have been described.^{16, 38, 39, 40, 47} Proteolytic cleavage at surface‐exposed KR motifs has been demonstrated in the native tannase AoTanA from A. oryzae. ³⁷ Although the protease responsible was not directly identified in that study, the processing pattern is consistent with the activity of KexB, a homologue of yeast Kex2 protease involved in the maturation of secreted proteins in Aspergillus.^{48, 49, 50, 51} In contrast, the tannase KM71ANT of A. niger GH1, which contains one surface‐exposed KR motif, is secreted as a single‐chain protein in its native host, ⁵² but as a double‐chain recombinant enzyme in K. phaffii. ³⁸ This suggests that such post‐translational processing may not be essential for enzymatic function across all Aspergillus species. However, in previous studies, single‐chain recombinant variants of KM71ANT and AoTanA generated by mutation (KM71TanMut) or deletion (AoTanA ΔKR) of KR motifs exhibited increased specific activity and catalytic efficiency compared with their double‐chain counterparts.^{26, 53} On the basis of these observations, in the present work a single‐chain tannase (AnTanSC) was designed by mutation of the two surface‐exposed KR motifs present in AnTanCECT2907.

SDS‐PAGE and Western blot analyses confirmed that recombinant AnTanDC underwent proteolytic processing into two subunits when expressed in K. phaffii, whereas recombinant AnTanSC remained as a single‐chain enzyme. Both enzymes showed extensive N‐glycosylation, consistent with previous reports of recombinant Aspergillus tannases expressed in K. phaffii.^{16, 38, 47} Biochemical characterization revealed similar activity profiles and optimal pH and temperature ranges for both enzymes. However, AnTanSC showed improved thermal stability at 30, 40, and 50 °C compared with AnTanDC. This is consistent with previous reports for the tannase from A. niger GH1 expressed in K. phaffii, in which the double‐chain variant KM71ANT retained 44% of the activity after 2 h of pre‐incubation at 30 °C, ³⁸ while a half‐life of 4 days at 30 °C was reported for KM71TanMut. ²⁶ Although thermal stability at higher temperatures was not evaluated for KM71ANT or KM71TanMut, the results obtained in the present study suggest that fusion of both subunits into a single polypeptide may exert a stabilizing effect.

In contrast, no significant differences in specific activity toward propyl gallate were observed between AnTanDC and AnTanSC (2.03 and 2.92 U mg⁻¹, respectively), and the single‐chain variant even showed lower specific activity toward tannic acid (8.38 vs. 10.66 U mg⁻¹). These activities were similar to those reported for AoTanA toward propyl gallate (4.6 U mg⁻¹), ⁴⁷ but lower than those reported for An‐Tan from A. niger SH‐2 toward propyl gallate and tannic acid (40.7 U mg⁻¹ and 59.4 U mg⁻¹, respectively).^{40, 43} Nevertheless, comparisons of specific activity among tannases should be interpreted cautiously due to differences in assay conditions and enzyme preparations. Although differences in purification degree cannot be completely excluded, SDS‐PAGE analysis after Ni‐NTA affinity chromatography revealed no major contaminating proteins, suggesting that the comparatively low specific activities are more likely associated with intrinsic structural features affecting substrate accessibility. Previous studies on microbial tannases suggest that differences in catalytic performance may be influenced by structural elements controlling access to the active site. Crystallographic analysis of the bacterial tannase LpTan from Lactobacillus plantarum and An‐Tan revealed a high structural conservation with the catalytic triad located within a conserved surface groove formed at the interface between an α/β‐hydrolase fold domain and a cap domain.^{43, 44} In LpTan, a flap or lid structure composed of a two‐stranded antiparallel β‐sheet and a flexible loop has been proposed to guide substrates toward the active site. ⁴⁴ This lid is conserved among Aspergillus tannases, although it shows substantial variability in length, mainly due to differences in the flexible loop region, which in some enzymes also contains the surface‐exposed KR motifs susceptible to proteolytic processing. The available evidence suggests that the impact of chain fusion on tannase activity may depend on the interplay between lid architecture, substrate size, and enzyme–substrate interactions. In An‐Tan, a relatively short flexible loop of ten residues leaves an exposed catalytic groove, suggesting a less regulated substrate accessibility. ⁴³ This is consistent with the similar specific activities reported for this enzyme toward methyl gallate, propyl gallate, and tannic acid.^{40, 43} In contrast, larger flexible loops have been predicted for KM71TanMut and AoTanA ΔKR (28 and 31 residues, respectively) and proposed to interact with the substrate, facilitating its accessibility to the active site and leading to an increase in specific activity toward small gallate esters.^{26, 53} Interestingly, no clear improvement in the catalytic performance of AoTanA ΔKR was observed for larger substrates such as epicatechin gallate and epigallocatechin gallate. ⁵³ These observations suggest that the effects of chain fusion may be substrate‐dependent and that for bulky substrates steric hindrance due to an enlarged lid may outweigh favorable enzyme–substrate interactions, resulting in a reduced catalytic efficiency. For AnTanSC, sequence analysis predicts a flexible loop of 36 amino acids, which would result in an even larger lid structure than those described for KM71TanMut and AoTanA ΔKR. Additionally, AlphaFold structural modelling predicts a highly flexible loop partially covering the catalytic groove (Supporting Information, Fig. S5). In this context, one possible explanation for the observed behavior of AnTanSC is that this enlarged lid may increase steric hindrance and alter substrate accessibility, particularly for bulky substrates such as tannic acid. For AnTanDC, proteolytic cleavage within this region could partially relieve this steric constraint. Importantly, the proposed effects of the lid region on substrate accessibility should be interpreted with caution, given the intrinsic limitations of structure prediction in highly flexible regions and the absence of experimental structural data for native or recombinant tannases from A. niger CECT2907. Additionally, a potential contribution of the amino acid substitutions introduced at the KR motifs in AnTanSC cannot be excluded. Further mutagenesis and structural studies will be required to fully elucidate the mechanisms underlying substrate recognition and catalysis in these enzymes. However, such analyses are beyond the scope of the present study, as the primary objective was not to establish detailed structure–function relationships but rather to evaluate the recombinant enzyme as a potential biocatalyst for enhancing the antioxidant properties of black tea infusions.

Due to its improved thermal stability, AnTanSC was selected for application assays in black tea infusions. Recent studies have explored the use of microbial tannases to enhance the antioxidant properties of black tea beverages. Aharwar and Parihar ¹¹ reported increases of 1.85‐ to 2.25‐fold in antioxidant capacity after incubation for 20 min at 60 °C with a tannase from Talaromyces verruculosus immobilized in alginate beads. Similarly, Govindarajan et al. ⁵⁴ observed increases of 1.55‐ to 1.62‐fold in radical scavenging activity in CTC (crush, tear, curl) and Kangra orthodox black teas treated with a tannase from Enterobacter cloacae for 1 h at 30 °C. Compared with these studies, treatment with AnTanSC produced substantial improvements in the antioxidant capacity of black tea infusions under relatively mild conditions (8.5 U mL⁻¹, 25 min, and 30 °C). The enzymatic treatment significantly increased the concentration of total soluble phenolic compounds, yielding 5.76‐fold and 3.12‐fold increases in radical scavenging activity according to the TEAC and FRAP assays, respectively. These results are consistent with the proposed mechanism of tannase action on tea polyphenols. Tannases catalyze the hydrolysis of ester bonds in galloylated catechins such as epigallocatechin gallate (EGCG) and epicatechin gallate, as well as other galloylated polyphenolic compounds present in black tea, including theaflavins and thearubigins. This enzymatic transformation generates smaller phenolic molecules, such as epigallocatechin, epicatechin, and gallic acid. These compounds show higher radical‐scavenging activity and greater solubility than their galloylated counterparts, thereby contributing to the observed increase in antioxidant capacity.^{2, 54, 55, 56, 57} On the other hand, gallic acid and epicatechin may exhibit lower metal‐chelating capacity than EGCG,^{5, 58, 59} which may explain the unchanged Fe²⁺‐chelating capacity observed after AnTanSC treatment.

Despite these promising results, several limitations should be considered. In this study, an 84‐ to 98‐fold increase in recombinant tannase production was reached after scaling up K. phaffii fermentation to a 5 L bioreactor compared to shake‐flask cultures. Although this increase is higher than that reported for the bioreactor‐scale production of the recombinant tannases from A. niger GH1 expressed in K. phaffii (35‐fold), ²⁶ the maximum volumetric activity in the cell‐free culture supernatant was considerably lower (~6500 U L⁻¹ vs. ~312 000 U L⁻¹), even when considering the differences in the substrates used for activity assays (propyl gallate and methyl gallate, respectively). However, the production levels obtained in this study are comparable to those reported for the tannase from A. oryzae expressed in K. phaffii in a 7 L bioreactor (~7000 U L⁻¹ using tannic acid as substrate). ⁶⁰ Importantly, both the tannases produced in this study and the A. oryzae enzyme were expressed using their native coding sequences, without codon optimization for the recombinant host. In contrast, the recombinant tannases from A. niger GH1 were expressed using a gene sequence optimized for K. phaffii codon usage.^{26, 38} This strategy is widely used to enhance recombinant protein expression in K. phaffii and has been reported to significantly increase protein yield and activity in recombinant enzymes from filamentous fungi.^{61, 62} This suggests that codon optimization may have contributed, at least in part, to the higher expression levels reported for the recombinant A. niger GH1 tannases. Beyond codon usage, further optimization of the production process will be required to increase enzyme yield – for example, by employing protease‐deficient strains such as SMD1168. In addition, enzyme activity was primarily evaluated using model substrates such as propyl gallate and tannic acid; therefore, a more comprehensive kinetic characterization using a wider range of natural tea polyphenols would provide deeper insight into substrate specificity and catalytic efficiency. Future studies should explore AnTanSC performance across different tea varieties (e.g., green, oolong, and specialty black teas) and evaluate the effects of temperature, enzyme dosage, and incubation time on both biochemical and sensory properties. Finally, sensory evaluation and stability studies during beverage storage will be necessary to fully assess the technological feasibility of incorporating this enzyme into industrial tea processing, as well as its potential impact on consumer acceptance.

CONCLUSIONS

A tannase gene from A. niger CECT 2907 was successfully cloned, and two enzyme variants were heterologously expressed in K. phaffii: a double‐chain tannase (AnTanDC) and a single‐chain variant (AnTanSC) generated by mutating Kex2 protease cleavage sites. Both recombinant enzymes were efficiently secreted and produced as active enzymes in bioreactor cultures. AnTanSC exhibited improved thermal stability compared with AnTanDC, while both enzymes showed comparable activity toward propyl gallate. In contrast, the single‐chain variant displayed lower activity toward tannic acid, indicating that the effects of chain fusion on catalytic performance may depend on the substrate evaluated. Importantly, AnTanSC significantly increased the soluble phenolic content and antioxidant capacity of black tea infusions under mild processing conditions. Although further optimization of enzyme production and performance is required, these results identify AnTanSC as a promising biocatalyst for improving the nutritional and functional quality of tea beverages and other polyphenol‐rich food products, supporting its potential for food processing applications.

FUNDING INFORMATION

This study was supported by the National Institute of Industrial Technology (INTI), Argentina and the National Agency for Science and Technology Promotion (ANPCyT), Argentina (Grant PICT 2020–0022).

Supporting information

Figure S1. Alignment of the predicted AnTanCECT2907 protein sequence with characterized tannases from Aspergillus. The sequences correspond to the enzymes AnTanCECT2907 from A. niger CECT 2907 (GenBank accession YBV31606.1), KM71ANT from A. niger GH1 (GenBank accession AKM52312.1), AoTanA and AoTanB from A. oryzae RIB40 (GenBank accessions BAA09656.1 and XP_001820636.1, respectively), An‐Tan from A. niger SH‐2 (GenBank accession XP_001401809.1), and the tannase from A. niger N5‐5 (GenBank accession AKC01963.1). The consensus sequence GXSXG found in serine hydrolases is indicated with asterisks. The CS‐D‐HC motif, including the catalytic triad (Ser195, Asp455, and His501) and two cysteine residues (Cys194 and Cys502) involved in the formation of a disulfide bond linking the catalytic serine and histidine, is highlighted in green. Additional conserved cysteine residues forming four other disulfide bridges (Cys25‐Cys71, Cys248‐Cys294, Cys261‐Cys278, and Cys550‐Cys566) are highlighted in magenta. Highly conserved residues involved in binding the hydroxyl groups of gallic acid (Gln227, Gln228, His231, Ser456, and Ile457) are highlighted in light blue, while the conserved Gly (Gly109) and the amino acid at the carboxyl side of the catalytic Ser (Asp196), which form the oxyanion hole, are highlighted in orange.

Figure S2. Vector pPICNHIS. (A) Map of the pPICNHIS vector showing genetic elements derived from the pPIC9 backbone and the position of the 6xHis sequence used to generate N‐terminal His‐tagged proteins. The positions of the DraI restriction sites used for plasmid linearization are indicated.

Figure S3. Sequence features of recombinant AnTanDC and AnTanSC expressed in K. phaffii. (A) Sequence encoding the double‐chain AnTanDC tannase, which contains two internal KR motifs that are post‐translationally processed by the endogenous Kex2 protease (black arrowheads), yielding two polypeptides (~33 and ~ 31 kDa). (B) Sequence encoding the single‐chain AnTanSC tannase variant, in which the KR motifs were mutated (R311T and R316G) to prevent Kex2‐mediated cleavage, resulting in a single ~64 kDa polypeptide. In both panels, only partial sequences are shown, including the N‐terminal fusion (MHHHHHHEF) generated after cloning into the EcoRI site of pPICNHIS, the region containing the KR motifs, and the C‐terminal region surrounding the stop codon, followed by the NotI cloning site. Omitted regions are indicated by dashed lines. Protein sizes are not drawn to scale. Both panels also indicate the Kex2 cleavage site within the α‐factor signal sequence (grey arrowhead).

Figure S4. Screening of recombinant K. phaffii clones. Clones selected by reversion of histidine auxotrophy were replica‐plated onto minimal medium supplemented with 1% tannic acid and induced with methanol. Clear halos indicate tannic acid hydrolysis by the secreted tannase. (A) Clones transformed with pPICNHIS‐AnTanDC. (B) Clones transformed with pPICNHIS‐AnTanSC. (−): P. pastoris transformed with an empty pPICNHIS vector. One representative plate per construct is shown.

Figure S5. Structural comparison of the lid region in fungal tannases. (A) Crystal structure of An‐Tan from A. niger SH‐2 (PDB ID: 7K4O), with the lid region highlighted in pink. Gallic acid located within the catalytic groove is shown in red. (B) AlphaFold predicted structure of AnTanSC generated using the amino acid sequence of the mature enzyme and colored according to Predicted Local Distance Difference Test (pLDDT) confidence scores. The enlarged region predicted within the lid is indicated by a dashed circle. Lower confidence scores are observed within the flexible loop region. (C) Structural superposition of An‐Tan and AnTanSC highlighting differences in the lid regions, shown in pink and yellow, respectively. (D) Surface representations and close‐up views of the catalytic groove of An‐Tan (upper panel) and AnTanSC (lower panel), with the lid regions shown in pink and yellow, respectively. An‐Tan shows a more exposed catalytic groove, whereas the enlarged loop predicted in AnTanSC partially occludes the substrate‐binding region. Gallic acid from the crystallographic structure of An‐Tan is indicated in red as a reference marker for the substrate‐binding site.

Predicted structures were visualized and structurally aligned using Mol^* 3D Viewer and used exclusively for qualitative comparative analysis.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.