Engineering a Protease K for Efficient Degradation of Wool Scale Layer Using a Substrate Pocket Modification

Abstract

The outermost surface of wool is covered by a scale layer, posing challenges to some steps of fabric processing. This layer, primarily composed of keratin, resists degradation by conventional proteases due to its high disulfide bond content. Protease K, an extracellular serine endo-proteinase derived from Tritirachium album Limber (tPRK), is known for its ability to digest native keratin. However, its limited activity against keratin has restricted its application in wool scale layer treatment. In this study, the substrate-binding pocket of tPRK was engineered, yielding the mutant N162A, which demonstrated an 84% increase in catalytic activity toward keratin. Additionally, the catalytic efficiency (k_cat/K_m) of N162A on keratin improved by 44.52%. Structural analysis indicated that modifications in the substrate-binding pocket reduced steric hindrance during substrate entry while enhancing substrate binding. Additionally, 3.3 mg/mL of amino acids were released within 6 h, which were catalyzed by N162A, with a 61% increase compared to the native tPRK. Moreover, the N162A variant effectively reduced the scale layer thickness without compromising the tensile strength of the wool, maintaining its mechanical properties. The findings provide a sustainable strategy for the wool industry while broadening the scope of biotechnological applications in the textile sector.

1. Introduction

Wool as a natural protein fiber has been one of the earliest materials used in the textile industry due to its outstanding properties, including warmth, moisture absorption, fire retardancy, high elasticity, and antistatic characteristics. The structure of wool fibers consists of a central cellular core, which is encased by an outer layer of cuticle cells, known as the scale layer [1]. When the wool fibers’ outer scale layer is subjected to constant mechanical forces in a humid environment, the scales interlock and insert into each other. This interlocking action makes the wool fibers more prone to felting and shrinking, particularly during washing [2]. As a result, wool fabrics in industrial processes are prone to issues such as size shrinkage, reduced softness, and distorted colors and patterns, leading to defective products. Consequently, developing effective treatments for the scale layer of wool remains a marked challenge in the textile industry.

So far, different types of chemicals, including dichloroisocyanuric acid, sodium hypochlorite, gaseous chlorine [1], thiols [3], deep eutectic solvents, and ionic liquids [4], have been used to impart anti-felting properties to wool textiles. Nevertheless, the applications of these chemicals in the textile industry have been limited by the damage to wool fibers, the remaining of toxic chemicals in products, the release of unpleasant odor, and their non-recyclability. Thus, developing an eco-friendly approach to treat the scale layer of wool without damaging its color, pattern, inherent structure, and hand feel has become an urgent problem that needs to be addressed.

Enzymatic treatment has garnered significant attention as an eco-friendly alternative to chemical methods, owing to its mild treatment conditions and pollution-free properties. Various types of proteases have been investigated for wool degradation, including bromelain [2], papain [5], microbial enzymes of Savinase 16 L [6], Esperase 8.0 L [7], PATO5T [8], keratinase [9], and Protease K [10]. However, the scale layer of wool predominantly consists of about 82% keratinous proteins with high cysteine content and highly crosslinked disulfide bonds, making it resistant to hydrolysis by common proteases [11]. Protease K is known for its ability to degrade keratin, while its application in wool degradation is limited compared to keratinase, likely due to inherent catalytic deficiencies such as low catalytic activity. Therefore, there is a pressing need to engineer a novel protease K with enhanced activity toward keratin to achieve effective anti-felting results.

Protein engineering is a versatile and powerful approach to creating a diverse range of biocatalysts by modifying protein structures to enhance their properties, such as stability, specificity, and catalytic activity [12]. In recent years, reforming the substrate-binding pocket of enzymes has gained significant attention and has led to notable successes in increasing catalytic activity [13], extending substrate specificity [14], and improving in enantioselectivity [15]. Particularly, pocket reshaping engineering based on steric hindrance modification can effectively enhance enzyme–substrate interactions and catalytic activity [13,16]. Using alanine or glycine scanning to reduce steric hindrance has emerged as a valid mutagenesis strategy for creating enhanced biocatalysts. This approach has successfully conferred improved substrate specificity and catalytic activity to several enzymes such as transaminase ATA-117 [17] and alcohol dehydrogenase [18,19].

To our knowledge, molecular modification of protease K, particularly via a steric hindrance reduction strategy, has not been systematically explored. In this study, the substrate-binding pocket of the Tritirachium album limber protease K (tPRK) was reshaped by reducing steric resistance to enhance its activity toward keratin. The mechanism behind the increased keratin activity of the mutant was further elucidated through structure-based computational analysis. Additionally, the degradation efficiency, morphology, structural characteristics, and tensile characteristics of wool treated using different enzymes were studied.

2. Materials and Methods

2.1. Materials

The DNA extraction and recovery kits were sourced from OMEGA (Norcross, GA, USA). DNA polymerase, all restriction enzymes, and T4 DNA ligase were provided by TAKARA (Kyoto, Japan). Unless otherwise specified, all chemicals were supplied by Sinopharm Chemical Reagent (Shanghai, China). Degreased Australian wool was supplied by Jiangnan University (Wuxi, China).

2.2. Gene Construction and Mutagenesis

The gene sequence of tPRK from T. album Limber (GenBank: P06873.2) was codon-optimized for expression in Pichia pastoris. The gene fragment was synthesized by GENEWIZ (Suzhou, China) and then cloned into the pPIC9K expression vector (Invitrogen, Carlsbad, CA, USA) between the EcoR I and Not I sites with a (His)₆ tag added at the C-terminus. Variants were created using polymerase chain reaction-based site-directed mutagenesis, with the pPIC9K-tPRK plasmid serving as the template. The sequences of the mutagenesis primers were provided in Table S1. All plasmids were confirmed by sequencing (GENEWIZ, China).

2.3. Recombinant Protein Expression and Purification

The plasmids were individually transformed into P. pastoris GS115 after linearization with Sal I and plated on YPD agar containing G418 with concentrations from 0.5 to 2 mg/mL. A single colony was inoculated into YPD media and grown at 30 °C overnight, with shaking at 220 rpm. The seed culture was then cultivated in 50 mL of BMGY for an additional 24 h, followed by induced in BMMY media at 30 °C with shaking at 220 rpm for several days. Methanol (0.5%) was added every 24 h. Samples were taken every 24 h, and cells were pelleted by centrifugation (8000× g for 10 min). The resultant supernatant was then used to analyze the activity of tPRK.

Cell-free extracts were harvested via 20 min of centrifugation at 20,000× g. The purification process was carried out following our previously established method [20], and the supernatant was precipitated by adding 50% saturation of (NH₄)₂SO₄. The precipitated protein was resuspended in 20 mM Tris-HCl (pH 10) and dialyzed overnight against the same buffer to remove excess (NH₄)₂SO₄. The target protein was then purified using gel filtration with a 5 mL HiTrap™ Q HP column (GE Healthcare, Fairfield, CT, USA). The fractions containing the target protein were concentrated appropriately for subsequent experiments. Protein concentration was estimated using the Bradford Protein Content Assay Kit (BOXBIO, Beijing, China) with bovine serine albumin as the standard. Protein purity (>95%) was confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

2.4. Enzyme Activity Measurement

The Folin-phenol reagent method was utilized to measure the proteinase K activity, with keratin (CAS: 69430-36-0) as the substrate. The reaction mixture was composed of 1 mL of 1.0% (w/v) keratin and 1 mL of appropriately diluted enzyme solution in Glycine-NaOH Buffer (100 mM, pH 10.0). The reaction was conducted for 10 min at 60 °C, after which the reaction was terminated by adding 2 mL of 0.4 M trichloroacetic acid. After centrifuging the mixture at 20,000× g for 2 min, the supernatant (500 μL) was mixed with 2.5 mL of Na₂CO₃ solution, followed by adding 500 μL of Folin-phenol reagent. To develop the color, the mixture was incubated for 20 min at 60 °C. Absorbance at 680 nm was monitored after cooling the mixture to room temperature. One unit (U) of protease K activity was designated as the quantity of enzyme required to yield 1 μmol of tyrosine per minute under standard conditions (pH 10.0, 60 °C, and 10 min).

2.5. Computational Design and Molecular Dynamics Simulations

The tertiary structure of tPRK was modeled with SWISS-MODEL (https://swissmodel.expasy.org/), employing the closest available structure, PRTK_PARAQ (PDB: P06873.1; sequence identity, 95.57%), as the template. The 3D structure of L-Cystine was downloaded from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/). The OPTIMIZE plugin of PyMOL was used to minimize the energy of the ligand. Molecular docking files for tPRK and L-Cystine were generated using AutoDockTools 1.5.6 software [21]. AutoDock Vina 1.2.2 was utilized to perform the molecular docking, with the active center of both the wild type (WT) and its mutants chosen as the docking region. The grid dimensions were set to 50 Å × 40 Å × 40 Å, and the grid center was defined at coordinates x = 5.225, y = −29.116, and z = −7.429. Further structural feature analysis and a molecular dynamics simulation were carried out using the lowest-energy complex model.

GROMACS 2018.8 software was utilized to perform all-atom molecular dynamics (MD) simulations of the dimeric complexes, with the general AMBER ff14SB force field [22] applied to all molecular dynamics. The protein complex was placed in truncated octahedron boxes, ensuring a 10 Å minimum distance between the box boundary and any protein atom. The TIP3P water molecules were used to solvate the system, and counterions (Na⁺ or Cl⁻) were introduced to neutralize any excess charges. The steepest descent method was utilized to carry out energy minimization, followed by the conjugate gradient method. The system was heated from 0 to 330 K with a constant number of particles (N) under constant temperature (T) and volume (V) conditions (NVT ensemble) for 50 ps. In the resulting complex conformation, amino acid residues that interacted with the substrate and formed the substrate access tunnel were selected for mutation. The final MD simulations were performed for 100 ns. Structural visualization and figure construction were conducted using PyMOL.

The Particle Mesh Ewald (PME) method was utilized to calculate the long-range electrostatic interactions [23]. The LINCS algorithm was employed to restrict all covalent bonds involving hydrogen atoms. The structural analysis included the calculation of the Radius of Gyration (Rg), Root Mean Square Fluctuations (RMSF), Solvent Accessible Surface Area (SASA), and Root Mean Square Deviation (RMSD), and was carried out using the CPPTRAJ module in AMBER 18.

The substrate-binding pocket was computed with the structure of tPRK and its mutant via the PockDrug-Server (http://pockdrug.rpbs.univ-paris-diderot.fr) [24].

A custom script was utilized to calculate the contacts between the protein and ligand. It was considered a contact if the distance between any two atoms on the protein and the ligand was less than 3.5 Å in a given simulation frame.

2.6. Biochemical Characterization

The optimum temperatures of tPRK and its best mutant variant, N162A, were measured within the range of 30 to 70 °C using a Gly-NaOH buffer (50 mM, pH 10.0). The purified enzymes were preincubated at 40, 50, and 60 °C for a specific duration to assess their thermostability. The activity of the enzymes without preincubation was considered 100%. Similarly, the optimum pH was estimated by testing enzyme activity at a pH ranging from 7.0 to 11.0 with 50 mM Gly-NaOH buffer.

Kinetic parameters for tPRK and the N162A mutant were determined using a modified assay of protease K reaction rates. Briefly, reaction rates (V) were measured at various keratin concentrations (0, 2.5, 5, 7.5, 10, 12.5, 15, 17.5, and 20 mM) for both the mutant and the wild type. The maximum reaction rate (V_max) and Michaelis constant (K_m) were obtained by nonlinear fitting to the Michaelis–Menten equation. From these values, the catalytic rate constant (k_cat) and catalytic efficiency (k_cat/K_m) were calculated to evaluate enzyme performance.

2.7. Effect of Protease Treatment on Wool

Enzymatic treatment of wool. Enzymatic shrink-resistant finishing of wool was performed by treating wool samples with tPRK or the N162A mutant. A reaction mixture consisting of 1 mL of protease K and 50 mM Gly-NaOH buffer (pH 10) was prepared. Wool samples (5.0 cm, ~0.5 g) were placed into a 20 mL solution for 6 h of incubation at 40 °C. Two control groups were also set up for comparison: the water control group, where wool was treated with buffer without enzymes, and the enzyme control group, where the enzyme culture mixture was used in the absence of wool. After the incubation, the enzyme was inactivated by heating the wool for 5 min at 90 °C. The wool was then thoroughly washed using distilled water and dried at 40 °C. This treatment aimed to evaluate the shrink-resistance properties imparted by the enzymatic process.

Amino acid analysis. The total content of polypeptides and amino acids in the supernatants was evaluated using a ninhydrin colorimetric assay following a described method [25]. Leucine was utilized as the standard to quantify the total amino acid content. The absorbance of the reaction mixtures was measured, and concentrations were calculated based on the standard curve obtained from known leucine concentrations.

Weight loss. The treated wool samples were dried in a 70 °C oven for 1 h to ensure all moisture was removed. After drying, the samples were weighed to determine their weight loss or changes in mass, which can indicate the degree of enzymatic treatment, such as the removal or modification of the wool’s scale layer.

Microstructure and spectroscopic characterization of wool samples. After gold sputter coating, the fiber morphology of the treated wool samples was identified utilizing scanning electron microscopy (SEM) (JSM-6010 LA, Jeol Ltd., Tokyo, Japan) at an acceleration voltage of 10.0 kV. An FT-IR spectrometer (Tensor 27, Bruker, Billerica, MA, USA) was utilized to monitor the Fourier-transform infrared (FT-IR) spectra of wool samples at room temperature.

Tensile test. The tensile strength was detected with the YG006 single-fiber strength tester (Ningbo Textile Instrument Factory, Ningbo, China) to investigate the influence of enzymatic treatment on the mechanical characteristics of wool samples. Before testing, each wool sample was clamped at a gauge length of 10 mm, followed by conditioning at 23 °C and 50 ± 5% relative humidity (R.H.) for 24 h to stabilize the fiber. The tensile measurements were carried out at a tensile speed of 30 mm/min in the warp direction. The maximum extension at breaking was recorded for each sample. Three specimens were assayed for each measurement to acquire reliable measurements and ensure the reproducibility of the data.

3. Results and Discussion

3.1. Structural Analysis and Prediction of Key Mutagenesis Sites with High Activity

To enhance the catalytic activity of tPRK toward keratin substrates, we identified the need to modify the binding pocket to facilitate better substrate binding. Although tPRK possesses the catalytic triad (D39, H69, S224), which is crucial for hydrolyzing keratin, the binding pocket appears too narrow for effective substrate binding, particularly for larger keratin molecules such as L-cystine. Structural alignment of tPRK with its homologous complex (PDB: P06873.1) revealed that the catalytic triad is located within a pocket formed by four α-helix and seven β-sheets (Figure 1a). The molecular docking of L-cystine showed that the substrate binds within a crevice near the catalytic triad (Figure 1b), but the entrance to this pocket is somewhat restricted, hindering the proper entry of the substrate.

Additionally, three flexible loops adjacent to the active site form two tunnel-like structures that further affect substrate accessibility (Figure 1c). This suggests that the tight fit and restricted access to the substrate binding site may limit the enzyme’s efficiency in hydrolyzing keratin. To address this issue, we employed rational design strategies such as alanine scanning to reduce steric hindrance within the binding pocket. By replacing bulky amino acid residues with smaller alanine residues, the goal was to open the binding site and allow for better substrate entry, ultimately enhancing the enzyme’s activity [26,27].

To enhance the activity of tPRK toward keratin substrates, we focused on modifying the residues in the substrate-binding pocket that directly interact with the substrate and contribute to forming the substrate access channels. In light of the structural analysis of the tPRK-L-cystine complex model, we identified 27 residues within the substrate-binding pocket, excluding the active site residues H69 and S224, that could influence substrate binding and enzyme activity. These residues were selected for potential modification, as they are involved in either direct interaction with the substrate or shaping the access channels.

Given that residues closer to the substrate are more likely to cause steric hindrance, we further refined our selection by focusing on the residues that directly engage with the substrate and contribute to the formation of substrate access tunnels. This led to the simplification of the 27 residues into a smaller set of 11 residues, which were retained for further analysis and mutation (Figure 1d).

3.2. Confirmation of the Role of Alanine Mutation in Keratinase Activity

Using computer-aided design, we identified eleven specific residues within the substrate-binding pocket of wild-type tPRK (WT), which were then subjected to alanine scanning mutagenesis. The goal of this approach was to micro-adjust the steric resistance within the substrate pocket, making it easier for the substrate to bind and enter the active site. Site-directed mutagenesis was utilized to mutate the selected residues to alanine, and the resultant mutants were then expressed in P. pastoris GS115. After a 5 d induction period, the crude enzyme solutions were harvested and purified, and the specific activities of the variants were determined using keratin as a substrate.

Among the 11 mutants, four variants (G100A, S101A, Q103A, and N162A) showed increased specific activity compared to WT, with the N162A mutant exhibiting a significant improvement (425.62 U/mg), which was 1.84 times higher than the WT (231.32 U/mg) (Figure 2a). We hypothesized that the enhanced activity of the N162A mutant was due to a reduction in steric hindrance within the substrate-binding pocket, allowing better access to the keratin substrate. Structural analysis of the N162A mutation revealed that the distance between site 162 and Ser101, which was located at the opposite end of the pocket, increased from 9.95 angstroms to 16.1 angstroms after mutation (Figure 2b). This structural change led to the opening of the substrate entrance, particularly between Ala162 and Gly134, where the distance increased from 7.88 Å to 10.71 Å (Figure 2c,d). This larger gap facilitated easier substrate entry, thus improving the enzyme’s catalytic efficiency.

The substrate-binding pockets of both the N162A mutant and the WT were compared, and a significant difference in pocket volume was observed. The pocket volume of the N162A mutant increased to 1582.26 ų compared to the WT’s 1440.72 ų. This increase in pocket size likely played a key role in enhancing enzyme activity, as a more spacious binding pocket can accommodate the substrate more effectively, reducing steric hindrance and promoting better binding. Additionally, the substitution of asparagine (hydropathy index −3.5) at position 162 with alanine (hydropathy index 1.8) likely contributed to the observed increase in enzyme activity. Asparagine, with its polar amide structure, may hinder interactions with hydrophobic acyl chains, exacerbating steric hindrance. Replacing it with alanine, which is smaller and non-polar, reduces steric bulk and allows for easier substrate binding, thereby enhancing the catalytic efficiency of the enzyme. The mutation of glycine to alanine also explains part of the enhanced activity. Glycine, with its achiral α-carbon, imparts greater flexibility to the protein structure, which could interfere with optimal enzyme–substrate interactions. Substituting glycine with alanine, which has a chiral side chain, helps stabilize the enzyme’s structure and improves the overall efficiency of the enzyme in binding and catalyzing the reaction [28]. Thus, these structural and amino acid changes lead to significant improvements in enzyme activity.

The observation that the S170A, P171A, and G160A mutants exhibited significant reductions in activity (below 50%) suggests that these residues play an essential role in substrate binding and catalysis. These positions, along with the catalytic center, contribute to the formation of a substrate-binding tunnel, which is essential for the enzyme’s interaction with the keratin substrate (Figure S1). When these residues were mutated to alanine, their interactions with the substrate were likely weakened due to the loss of the side chain functional groups involved in stabilizing the substrate-binding pocket or contributing to the formation of the access tunnel. These changes likely impaired the enzyme’s ability to bind the substrate efficiently, leading to a significant drop in catalytic activity.

However, further iterations of the mutants with enhanced activity did not yield additional improvements (Figure S2), potentially due to the enlarged substrate pocket resulting from the mutation. Consequently, N162A, which showed the most significant activity enhancement, was selected for further investigation. SDS-PAGE analysis of the purified WT and N162A revealed that both were purified to homogeneity, displaying an apparent molecular mass of 28 kDa, in line with their theoretical molecular mass (Figure S3a). As shown in Figure S3b,c, N162A displayed the same optimum temperature and pH as WT, and it retained high relative activity (>80%) across a wide pH range (pH 7.0–11.0). The relationship between pH and structure changes remains unclear; however, we hypothesize that the increased hydrophobicity near the catalytic site may enhance alkali adaptability (Figure S4). It is well established that high-performing proteases require both exceptional catalytic activity and thermostability, which often exhibit a trade-off when mutations are introduced into the enzyme [29]. The similar thermostability profile suggests that the activity enhancement resulting from the Asn162Ala mutation does not compromise the enzyme’s stability (Figure S3d).

The kinetic parameters of purified mutants for keratin were analyzed to explore the biochemical basis for the observed increase in activity (Figure S3). As shown in Table 1, the catalytic efficiency (k_cat/K_m) of the WT enzyme toward L-cystine was relatively low, while the mutant N162A demonstrated a notable improvement in catalytic efficiency. The K_m value of N162A (4.492 μM·mL⁻¹) was moderately greater than that of the WT enzyme (3.964 μM·mL⁻¹), indicating that the mutation had minimal impact on substrate binding affinity. In contrast, the k_cat values showed significant differences, with N162A achieving a k_cat value of 49.067 min⁻¹, 1.64 times higher than that of WT. These findings suggest that the rational design of tPRK was effective and that engineering specific sites can substantially enhance the catalytic efficiency of tPRK for L-cystine.

3.3. Mechanism Elucidation for the Enhanced Catalytic Efficiency of N162A Toward Keratin

The alanine mutation at site 162 effectively enlarged the substrate-binding pocket, enhancing N162A’s activity toward keratin (Figure 2). To further investigate the catalytic behavior of WT and N162A, MD simulations were conducted. A 100 ns dynamic simulation was performed on protein-L-cystine complexes for both variants to identify the structures with the lowest energy and highest stability, which were subsequently used for studying the catalytic mechanisms. As shown in Figure S5a, the RMSD values of both models increased rapidly before stabilizing. Notably, model N162A exhibited greater RMSD fluctuations, potentially increasing the likelihood of disulfide bond cleavage in the substrate and moderately enhancing enzymatic activity. Additionally, the solvent-accessible surface area (SASA) was calculated over the 100 ns simulations, as SASA quantifies the protein’s surface area that is exposed to the solvent [30]. The radius of gyration (Rg) provides insight into the compactness of a protein molecule, with higher SASA and Rg values generally indicating increased flexibility. The SASA values for WT and N162A were 116.6 nm² and 120.8 nm², respectively (Figure S5b), with the higher SASA of N162A suggesting a more flexible enzyme structure. Similarly, the higher Rg values of N162A (1.8 Å) compared to WT (1.5 Å) (Figure S5c) further support its increased flexibility. The enhanced flexibility likely facilitates substrate binding, contributing to the improved catalytic activity observed in N162A.

The degrees of freedom for each residue in WT and N162A were further examined using the RMSD values from the final 10 ns of the MD simulations (Figure 3a). Most residues in N162A, particularly those in Loop 1 (D184-D200), exhibited higher RMSF values, indicating increased mobility in this region compared to WT. Notably, Loop 1 is adjacent to the substrate pocket and plays an essential role in shaping the larger binding pocket (Figure 1c). The Asn162Ala mutation led to an increase in the distance between site 162 and Loop 1, from 3.76 Å in WT to 5.7 Å in the N162A, accompanied by a reduction in the number of contacts and hydrogen bonds (Figure 3b–f). These changes decreased the restrictive influence of Loop 1 on the substrate pocket, enhancing its flexibility and facilitating more effective substrate binding [31].

The distance between the substrate and the enzyme’s active cavity is critical for catalytic reactions. After performing MD simulations, the distances were analyzed using PyMol ver. 2.5.9 software (Figure 4a). The average distance in the WT enzyme was 1.84 Å, whereas it decreased to 1.2 Å in N162A. Both distances were within the reactive range (≤3.5 Å), meeting the requirements for effective catalysis. A shorter distance in N162A facilitated increased contact between the substrate and the binding pocket (Figure 4b,c). The interaction network between the substrate and the surrounding binding site residues plays a pivotal role in stabilizing the substrate and enabling proton transfer. Hydrogen bonds, crucial for enzyme−substrate binding, were more abundant in N162A, with six bonds compared to five in the WT (Figure 4). Additionally, a new salt bridge was formed between the catalytic residue Asp39 and the substrate, further enhancing substrate binding. These structural adaptations in N162A significantly improved substrate binding, a key factor for effective keratin degradation.

3.4. Mechanistic Insight into the Action of WT and N162A on Wool

The wool samples were significantly degraded by both WT and N162A after 6 h of treatment at 40 °C, demonstrated by a greater loss of wool weight (Figure 5a). The WT enzyme caused an 8.13% reduction in wool weight, while N162A resulted in a more pronounced 13.41% weight loss. In comparison, wool samples treated without protease K retained their initial weight after 6 h, demonstrating no degradation.

To intuitively observe the degradation of wool by WT and mutant N162A, the residual hydrolysates from various wool treatments were harvested and analyzed for their composition. During the enzymatic treatment, the scale layer of the wool released amino acids or oligopeptides into the system, which could be detected to confirm the treatment’s effectiveness (Figure 5b). In the sample treated in the absence of enzyme, the hydrolysate showed low amino acid content, as the wool scale layer was not degraded. In contrast, the amino acid concentration significantly increased after treatment with WT or N162A, with values of 2.05 mg/mL and 3.30 mg/mL, respectively. The higher amino acid content following N162A treatment indicated its greater hydrolysis efficiency. Figure 5c presents the FTIR spectra of enzyme-treated and untreated wool samples over the range of 4000–700 cm⁻¹, excluding the untreated fabric. The characteristic peaks of wool were observed at 1629, 1514, and 1228 cm⁻¹, corresponding to Amide I, Amide II, and Amide III, respectively. Amide I was associated with the stretching vibration of the carbonyl group (C=ONH–R), Amide II was linked to N–H bending deformation, and Amide III was associated with N–H in-plane bending and C–N stretching [32]. The S–O–S derivatives of wool suggest a peak around 1044 cm⁻¹ [33]. The intensity of the peak at 1044 cm⁻¹ weakened following enzyme treatment, suggesting modifications to the disulfide bonds, particularly the partial breakage of the sulfur bonds and their conversion into salt derivatives. Additionally, the intensity of the Amide III peak at 1228 cm⁻¹ changed as a result of the enzymatic treatments. Notably, for N162A treatment, the peak became broader, indicating a reduction in C–N stretching. The characteristic Amide I peak at 1629 cm⁻¹ remained unchanged after enzyme treatment, confirming that no major chemical modifications occurred in the wool. Similarly, the –CH₃ stretching vibrational absorption peaks 2920 and 2850 cm⁻¹, which are related to the keratin backbone structure of the wool, also remained unaffected [34]. This result is consistent with our findings shown in Figure 6, where both WT and N162A effectively degraded the wool scale layer with minimal impact on the wool’s mechanical properties.

Additionally, the degradation of the wool scale layer was observed through Scanning Electron Microscopy (SEM). As demonstrated in Figure 5d, the untreated wool displayed a visible, overlapping scale layer with angular, thick edges. After treatment with WT, the scale layer softened and the roughness of the wool decreased, likely due to the opening of the disulfide bond, resulting in slight exfoliation of the scale layer (Figure 5e). In Figure 5f, the N162A treatment led to the complete removal of the rough scale layer, leaving the wool surface smoother and the scale layer thinner. This effect is attributed to the enhanced keratinase activity of the protease K mutant, which promoted the peeling of the scale layer, leading to its successful degradation and further confirming the effectiveness of N162A. Additionally, the scale structures treated with protease K exhibited a stepped appearance, providing clear evidence that the scales were gradually thinned from the outer to the inner layers. This observation suggests that the action of both WT and N162A was more selectively confined to the outer scale layer rather than penetrating deeper into the inner layers.

Figure 6 exhibits the tensile properties of WT-treated, untreated, and N162A-treated wool fabric samples. The enzyme-treated fabrics exhibited lower tensile strength as compared with the untreated samples, likely caused by the area density and reduced thickness of the wool after treatment. As a result, the load needed to break the treated samples was lower, leading to decreased tensile strength. However, the enzymatic cleavage of the surface scales reduced the probability of felting, which significantly improved the extension of the wool samples, resulting in higher elongation at break. Additionally, the N162A-treated sample showed higher extension and only a slight decrease in breaking strength (4.5%), suggesting that the Asn-to-Ala mutation enhanced the activity of protease K toward keratin. This indicates that N162A has potential applications in the anti-felting treatment of wool.

4. Conclusions

In summary, the catalytic efficiency of tPRK was successfully enhanced by reshaping its substrate-binding pocket. In this study, a more open binding pocket was engineered to facilitate the efficient degradation of the wool scale layer. The most beneficial variant, N162A, demonstrated an 84% increase in enzyme activity. Structural analysis revealed that the N162A mutant had a broader entrance for substrate entry and a larger substrate-binding pocket compared to the WT. Additionally, MD simulation results suggested that the improved catalytic efficiency towards keratin was due to increased flexibility in the substrate pocket and improved interaction with the substrate. Furthermore, wool treatment tests showed that N162A effectively descaled the wool with minimal impact on its tensile properties. Our study elucidated the mechanism of the pocket steric hindrance of protease K on the adaptability of keratin substrates based on the implementation of pocket reshaping, with potential applications in the keratin-containing substrate acceptance evolution of other proteases, and provided a highly efficient and sustainable biological catalyst and method for wool degradation.

Author Contributions

Conceptualization, L.Z. and Y.L.; methodology, L.Z., Y.D. and K.Z.; software, X.M.; validation, Y.D. and K.Z.; formal analysis, K.W., L.Z., X.M., Y.D. and K.Z.; investigation, L.Z. and X.M.; resources, F.L. and Y.L.; data curation, Y.D.; writing—original draft preparation, L.Z. and X.M.; writing—review and editing, L.Z.; supervision, Y.L.; project administration, Y.L.; funding acquisition, F.L. and Y.L. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Conflicts of Interest

Author Kefen Wang was employed by the company Shandong Lonct Enzymes Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as potential conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Table

Figure 1. Structure analysis and site selection. (a) Structural overview of wild-type tPRK complexed with L-cystine. The red sticks are catalytic triad residues. (b) Binding mode analysis. L-cystine substrate and residues that interact with substrates are indicated using red sticks and green balls, respectively. (c) Close-up view of the substrate access tunnel of tPRK. Two potential substrate access tunnels are shown as a mesh model. (d) Screening process of mutation sites.

Figure 2. (a) Relative activities of WT and eleven mutants toward keratin. (b) Distance between site 162 and Ser101 in WT and N162A, respectively. (c,d) Analysis of substrate entrance diameter (between Ala162 and Gly134) and substrate pocket volume of WT and N162A. Substrate keratin is a red stick, while the substrate pocket is represented by green foam. All experiments were conducted in triplicate, with the error bars suggesting standard deviations.

Figure 3. The effects of N162A mutation on the local structure of WT. (a) RMSF values; (b) the distance between Loop 1 and Site 162; (c,d) relative positions of Loop1 and Site 162 in WT and N162A structures; the number of contacts (e) and hydrogen bonds (f) between Loop 1 and Site 162. Loop 1 is highlighted in red, and the substrate pocket is illustrated using blue foam with the replaced residue 162 highlighted by a yellow stick.

Figure 4. Analysis of the source of improved kcat value of N162A by MD simulation. (a) The distance between the WT or N162A and the L-cystine; the number of contacts (b) and hydrogen bonds (c) between WT or N162A and L-cystine. (d,e) Interactions formed in the complexes of WT and mutant N162A with L-cystine. Substrate L-cystine is shown with an orange stick, hydrogen-bonding interactions are displayed as green dotted lines, hydrophobic interaction is exhibited by gray dotted lines, and salt bridge distance is labeled with red dotted lines.

Figure 5. Wool degradation by different enzymes. Weight loss (a) and content of amino acids (b) of the hydrolysates from various treatments of wool. FT-IR analysis (c) of wool samples. SEM image of (d) untreated, (e) WT-, and (f) N162A-treated wool at nearly 2500× magnification. All experiments were carried out in triplicate, with the error bars indicating standard deviations.

Figure 6. Tensile properties of wool sample with different treatments. All experiments were performed in triplicate with the error bars indicating standard deviations.

Table 1

Kinetic parameters of WT and N162A using keratin as a substrate.

Enzyme	Vmax(μM·mL−1·min−1)	Km(μM·mL−1)	kcat(min−1)	kcat/Km(μM−1·min−1·mL)
WT	455.286 ± 10.27 1	3.964 ± 0.48	29.958 ± 1.06	7.558
N162A	745.678 ± 12.11	4.492 ± 0.65	49.067 ± 3.65	10.923

¹ Activity was detected using 0–20 mM substrates in Gly-NaOH buffer (50 mM, pH 10) at 60 °C. All determinations were carried out in triplicate, and the values are exhibited as the mean ± standard deviation.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/fermentation11020051/s1, Figure S1: The relative position of three residues Gly160, Ser170, and Pro171 in the substrate pocket of tPRK; Figure S2: Assessments of catalytic activity of combinatorial mutants; Figure S3: Enzymatic properties of WT and N162A; Figure S4: Hydrophobicity analysis of tPRK and mutant N162A; Figure S5: Analysis of MD simulation for WT and N162A; Table S1: Primer for the mutations.

References

1. Hassan, M.M.; Carr, C.M. A review of the sustainable methods in imparting shrink resistance to wool fabrics. J. Adv. Res.; 2019; 18, pp. 39-60. [DOI: https://dx.doi.org/10.1016/j.jare.2019.01.014]

2. Kaur, A.; Chakraborty, J.N. Controlled eco-friendly shrink-resist finishing of wool using bromelain. J. Clean. Prod.; 2015; 108, pp. 503-513. [DOI: https://dx.doi.org/10.1016/j.jclepro.2015.07.147]

3. Wan, A.; Yu, W.D. Effect of wool fiber modified by ecologically acceptable ozone-assisted treatment on the pilling of knit fabrics. Text. Res. J.; 2012; 82, pp. 27-36. [DOI: https://dx.doi.org/10.1177/0040517511414973]

4. Shavandi, A.; Jafari, H.; Zago, E.; Hobbi, P.; Nie, L.; De Laet, N. A sustainable solvent based on lactic acid and l-cysteine for the regeneration of keratin from waste wool. Green Chem.; 2021; 23, pp. 1171-1174. [DOI: https://dx.doi.org/10.1039/D0GC04314A]

5. Shi, F.R.; Liu, Q.B.; Zhao, H.Z.; Fang, K.J.; Xie, R.Y.; Song, L.; Wang, M.Y.; Chen, W.C. Eco-friendly pretreatment to the coloration enhancement of reactive dye digital inkjet printing on wool fabrics. ACS Sustain. Chem. Eng.; 2021; 9, pp. 10361-10369. [DOI: https://dx.doi.org/10.1021/acssuschemeng.1c03486]

6. Mei, J.; Zhang, N.; Yu, Y.; Wang, Q.; Yuan, J.; Wang, P.; Cui, L.; Fan, X. A novel “trifunctional protease” with reducibility, hydrolysis, and localization used for wool anti-felting treatment. Appl. Microbiol. Biotechnol.; 2018; 102, pp. 9159-9170. [DOI: https://dx.doi.org/10.1007/s00253-018-9276-y]

7. Prajapati, C.D.; Smith, E.; Kane, F.; Shen, J.S. Selective enzymatic modification of wool/polyester blended fabrics for surface patterning. J. Clean. Prod.; 2019; 211, pp. 909-921. [DOI: https://dx.doi.org/10.1016/j.jclepro.2018.11.079]

8. Iglesias, M.S.; Sequeiros, C.; García, S.; Olivera, N.L. Eco-friendly anti-felting treatment of wool top based on biosurfactant and enzymes. J. Clean. Prod.; 2019; 220, pp. 846-852. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.02.165]

9. Wang, P.; Wang, Q.; Cui, L.; Gao, M.R.; Fan, X.R. The combined use of cutinase, keratinase and protease treatments for wool bio-antifelting. Fibers Polym.; 2011; 12, pp. 760-764. [DOI: https://dx.doi.org/10.1007/s12221-011-0760-6]

10. Li, W.J.; Zhang, N.; Wang, Q.; Wang, P.; Yu, Y.Y.; Zhou, M. A sustainable and effective bioprocessing approach for improving anti-felting, anti-pilling and dyeing properties of wool fabric. Fibers Polym.; 2021; 22, pp. 3045-3054. [DOI: https://dx.doi.org/10.1007/s12221-021-1148-x]

11. Brebu, M.; Spiridon, I. Thermal degradation of keratin waste. J. Anal. Appl. Pyrolysis; 2011; 91, pp. 288-295. [DOI: https://dx.doi.org/10.1016/j.jaap.2011.03.003]

12. Bornscheuer, U.T.; Huisman, G.W.; Kazlauskas, R.J.; Lutz, S.; Moore, J.C.; Robins, K. Engineering the third wave of biocatalysis. Nature; 2012; 485, pp. 185-194. [DOI: https://dx.doi.org/10.1038/nature11117]

13. Wu, T.; Mu, X.Q.; Xue, Y.Y.; Xu, Y.; Nie, Y. Structure-guided steric hindrance engineering of Bacillus badius phenylalanine dehydrogenase for efficient l-homophenylalanine synthesis. Biotechnol. Biofuels; 2021; 14, 207. [DOI: https://dx.doi.org/10.1186/s13068-021-02055-0] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34689801]

14. Xu, J.; Lou, Y.J.; Wang, L.L.; Wang, Z.G.; Xu, W.H.; Ma, W.Q.; Chen, Z.C.; Chen, X.Y.; Wu, Q. Rational design of biocatalytic deuteration platform of aldehydes. ACS Catal.; 2021; 11, pp. 13348-13354. [DOI: https://dx.doi.org/10.1021/acscatal.1c03659]

15. Wu, K.; Yang, Z.J.; Meng, X.G.; Chen, R.; Huang, J.K.; Shao, L. Engineering an alcohol dehydrogenase with enhanced activity and stereoselectivity toward diaryl ketones: Reduction of steric hindrance and change of the stereocontrol element. Catal. Sci. Technol.; 2020; 10, pp. 1650-1660. [DOI: https://dx.doi.org/10.1039/C9CY02444A]

16. Lewis, J.C.; Mantovani, S.M.; Fu, Y.; Snow, C.D.; Komor, R.S.; Wong, C.H.; Arnold, F.H. Combinatorial alanine substitution enables rapid optimization of cytochrome P450BM3 for selective hydroxylation of large substrates. ChemBioChem; 2010; 11, pp. 2502-2505. [DOI: https://dx.doi.org/10.1002/cbic.201000565] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21108271]

17. Savile, C.K.; Janey, J.M.; Mundorff, E.C.; Moore, J.C.; Tam, S.; Jarvis, W.R.; Colbeck, J.C.; Krebber, A.; Fleitz, F.J.; Brands, J. et al. Biocatalytic asymmetric synthesis of chiral amines from ketones applied to sitagliptin manufacture. Science; 2010; 329, pp. 305-309. [DOI: https://dx.doi.org/10.1126/science.1188934]

18. Zheng, Y.C.; Li, F.L.; Lin, Z.; Lin, G.Q.; Hong, R.; Yu, H.L.; Xu, J.H. Structure-guided tuning of a hydroxynitrile lyase to accept rigid pharmaco aldehydes. ACS Catal.; 2020; 10, pp. 5757-5763. [DOI: https://dx.doi.org/10.1021/acscatal.0c01103]

19. Tian, K.Y.; Li, Z. A simple biosystem for the high-yielding cascade conversion of racemic alcohols to enantiopure amines. Angew. Chem. Int. Ed.; 2020; 59, pp. 21745-21751. [DOI: https://dx.doi.org/10.1002/anie.202009733] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32776678]

20. Ma, X.Y.; Wang, L.Y.; Chen, J.Y.; Guo, E.P.; Zheng, H.C.; Zhao, L.; Lu, F.P.; Liu, Y.H. Surfactant-tolerance evolution of Bacillus clausii protease for enhancing activity and stability by reshaping the substrate access tunnel. Colloids Surf. B Biointerfaces; 2025; 245, 114289. [DOI: https://dx.doi.org/10.1016/j.colsurfb.2024.114289] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39366107]

21. Sanner, M.F. Python: A programming language for software integration and development. J. Mol. Graph. Model.; 1999; 17, pp. 57-61. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/10660911]

22. Maier, J.A.; Martinez, C.; Kasavajhala, K.; Wickstrom, L.; Hauser, K.E.; Simmerling, C. ff14SB: Improving the accuracy of protein side chain and backbone parameters from ff99SB. J. Chem. Theory Comput.; 2015; 11, pp. 3696-3713. [DOI: https://dx.doi.org/10.1021/acs.jctc.5b00255] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26574453]

23. Sagui, C.; Darden, T.A. Molecular dynamics simulations of biomolecules: Long-range electrostatic effects. Annu. Rev. Biophys.; 1999; 28, pp. 155-179. [DOI: https://dx.doi.org/10.1146/annurev.biophys.28.1.155]

24. Hussein, H.A.; Borrel, A.; Geneix, C.; Petitjean, M.; Regad, L.; Camproux, A.C. PockDrug-Server: A new web server for predicting pocket druggability on holo and apo proteins. Nucleic Acids Res.; 2015; 43, pp. 436-442. [DOI: https://dx.doi.org/10.1093/nar/gkv462] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25956651]

25. Okamoto, M.; Yonejima, Y.; Tsujimoto, Y.; Suzuki, Y.; Watanabe, K. A thermostable collagenolytic protease with a very large molecular mass produced by thermophilic Bacillus sp. strain MO-1. Appl. Microbiol. Biotechnol.; 2001; 57, pp. 103-108. [DOI: https://dx.doi.org/10.1007/s002530100731] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11693905]

26. Liu, G.F.; Tong, X.; Wang, J.L.; Wu, B.; Chu, J.L.; Jian, Y.; He, B.F. Reshaping the binding pocket of purine nucleoside phosphorylase for improved production of 2-halogenated-2′-deoxyadenosines. Catal. Sci. Technol.; 2021; 11, pp. 4439-4446. [DOI: https://dx.doi.org/10.1039/D0CY02424D]

27. Shao, Z.H.; Su, B.M.; Yang, S.L.; Ye, L.D.; Yu, H.W. Rational design of the carbonyl reductase EbSDR8 for efficient biosynthesis of enantiopure (R)-3-chloro-1-phenyl-1-propanol. Appl. Microbiol. Biotechnol.; 2020; 104, pp. 9219-9228. [DOI: https://dx.doi.org/10.1007/s00253-020-10904-5]

28. Wang, H.Y.; Wang, S.Y.; Wang, J.; Shen, X.L.; Feng, X.D.; Yuan, S.G.; Sun, X.X.; Yuan, Q.P. Engineering a prokaryotic non-P450 hydroxylase for 3′-Hydroxylation of flavonoids. ACS Synth. Biol.; 2022; 11, pp. 3865-3873. [DOI: https://dx.doi.org/10.1021/acssynbio.2c00430] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36321874]

29. Daniel, R.M.; Danson, M.J. A new understanding of how temperature affects the catalytic activity of enzymes. Trends Biochem. Sci.; 2010; 35, pp. 584-591. [DOI: https://dx.doi.org/10.1016/j.tibs.2010.05.001] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20554446]

30. Durham, E.; Dorr, B.; Woetzel, N.; Staritzbichler, R.; Meiler, J. Solvent accessible surface area approximations for rapid and accurate protein structure prediction. J. Mol. Model.; 2009; 15, pp. 1093-1108. [DOI: https://dx.doi.org/10.1007/s00894-009-0454-9]

31. Bunzel, H.A.; Anderson, J.L.R.; Hilvert, D.; Arcus, V.L.; van der Kamp, M.W.; Mulholland, A.J. Evolution of dynamical networks enhances catalysis in a designer enzyme. Nat. Chem.; 2021; 13, pp. 1017-1022. [DOI: https://dx.doi.org/10.1038/s41557-021-00763-6] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34413499]

32. Gogoi, P.; Kakoti, P.; Saikia, J.; Sarma, R.K.; Yadav, A.; Singh, B.P.; Saikia, R. Plant growth-promoting Rhizobacteria in management of soil-borne fungal pathogens. Management of Fungal Pathogens in Pulses: Current Status and Future Challenges; Springer: Cham, Switzerland, 2020; pp. 1-13.

33. Fernández-d’Arlas, B. Tough and functional cross-linked bioplastics from sheep wool keratin. Sci. Rep.; 2019; 9, 14810. [DOI: https://dx.doi.org/10.1038/s41598-019-51393-5] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31616010]

34. Zhang, P.; Wang, Q.; Shen, J.S.; Wang, P.; Yuan, J.G.; Fan, X.R. Enzymatic thiol-ene click reaction: An eco-friendly approach for MPEGMA-grafted modification of wool fibers. ACS Sustain. Chem. Eng.; 2019; 7, pp. 13446-13455. [DOI: https://dx.doi.org/10.1021/acssuschemeng.9b02902]