Beta-galactosidase mutant with high enzyme activity and thermal stability and application of beta-galactosidase mutant

Abstract

The invention provides a beta-galactosidase mutant with high enzyme activity and thermal stability and application of the beta-galactosidase mutant. According to the invention, mutation site prediction is carried out on wild type beta-galactosidase by using a machine learning algorithm, and a beta-galactosidase mutant with high enzyme activity and thermal stability is screened out; the amino acid sequence of the beta-galactosidase mutant is as shown in SEQ ID NO: 4, and the nucleotide sequence for coding the beta-galactosidase mutant is as shown in SEQ ID NO: 3. Compared with a wild type, the beta-galactosidase mutant has higher enzyme activity, the application range of beta-galactosidase can be widened, a more efficient biocatalyst is provided for industrial production, and the beta-galactosidase mutant is expected to be widely applied to the fields of food processing, agriculture and feed and has good practicability.

Description

A β-galactosidase mutant with high enzyme activity and thermostability and its applications Technical Field

[0001] This invention belongs to the field of bioengineering technology, specifically relating to a β-galactosidase mutant with high enzyme activity and thermal stability and its applications. Background Technology

[0002] β-Galactosidase (EC 3.2.1.23) is a class of hydrolases that specifically catalyze the hydrolysis of lactose into galactose and glucose, and it has wide applications in the food, pharmaceutical, and biotechnology fields. For example, in the food industry, this enzyme can be used to produce low-lactose dairy products, improving milk consumption for lactose-intolerant individuals; in the functional food field, it can catalyze the synthesis of prebiotic galactooligosaccharides; and in biomanufacturing, it can be used for the high-value utilization of whey resources and the synthesis of certain carbohydrate-based drugs.

[0003] However, natural β-galactosidases generally suffer from limited catalytic activity and insufficient stability, severely restricting their efficiency and economic viability in industrial applications. For example, in the production of low-lactose milk, low enzyme activity leads to a slow hydrolysis process, requiring higher enzyme dosages or longer reaction times, increasing production costs. Their poor heat resistance and pH stability also make the enzymes prone to inactivation during processing or storage, affecting product consistency and process adaptability. Therefore, it is necessary to develop β-galactosidases with high enzyme activity and high heat stability. Summary of the Invention

[0004] To address some shortcomings in existing technologies, this invention provides a β-galactosidase mutant with high enzyme activity and thermostability, and its applications. This invention utilizes machine learning algorithms to predict mutation sites in wild-type β-galactosidase, screening for β-galactosidase mutants with high enzyme activity and thermostability. The amino acid sequence of the β-galactosidase mutant is shown in SEQ ID NO:4, and the nucleotide sequence encoding the β-galactosidase mutant is shown in SEQ ID NO:3. Compared to the wild type, the β-galactosidase mutant exhibits higher enzyme activity, which can broaden the application range of β-galactosidase, providing a more efficient biocatalyst for industrial production. It is expected to achieve wider applications in food processing, agriculture, and feed, demonstrating excellent practicality.

[0005] To achieve the above-mentioned technical objectives, the present invention employs the following technical means:

[0006] This invention first provides a β-galactosidase mutant with high enzyme activity and thermostability, wherein the β-galactosidase mutant is based on the wild-type β-galactosidase with the amino acid sequence shown in SEQ ID NO:2, and undergoes one or more of the following mutations:

[0007] (1) Mutate the 254th position from threonine (Thr, T) to proline (Pro, P);

[0008] (2) Mutate the 256th position from threonine (Thr, T) to leucine (Leu, L);

[0009] (3) Mutate position 296 from leucine (Leu, L) to tyrosine (Tyr, T);

[0010] (4) Mutate the 339th position from threonine (Thr, T) to methionine (Met, M);

[0011] (5) Mutate the 648th position from glutamine (Gln, Q) to methionine (Met, M).

[0012] Preferably, the β-galactosidase mutant has undergone five mutations simultaneously, and the amino acid sequence of the β-galactosidase mutant is shown in SEQ ID NO:4.

[0013] The present invention also provides a biomaterial, said biomaterial comprising polynucleotides, recombinant vectors, or recombinant engineered bacteria;

[0014] The polynucleotide encodes the above-mentioned β-galactosidase mutant;

[0015] The recombinant vector contains the above-mentioned polynucleotides or expresses the above-mentioned β-galactosidase mutant;

[0016] The recombinant engineered bacteria contain the above-mentioned polynucleotides, or contain the above-mentioned recombinant vector, or express the above-mentioned β-galactosidase mutant.

[0017] Preferably, the nucleotide sequence of the polynucleotide includes SEQ ID NO:3 or its degenerate sequence.

[0018] Preferably, the recombinant vector includes a recombinant prokaryotic vector, wherein the prokaryotic vector includes a pET(+) plasmid or a pGEX vector.

[0019] Preferably, the host bacteria of the recombinant engineered bacteria include Escherichia coli, and more preferably Escherichia coli BL21(DE3).

[0020] The present invention also provides a method for preparing a β-galactosidase mutant, the method comprising: inoculating the above-mentioned recombinant engineered bacteria into a fermentation medium, fermenting and culturing, and then centrifuging to collect the supernatant to obtain a crude product of the β-galactosidase mutant.

[0021] The present invention also provides the application of the above-mentioned β-galactosidase mutant or the above-mentioned biological material in food processing and feed.

[0022] Preferably, in the application, food processing includes any one or more of the following:

[0023] (1) Production of low-lactose / lactose-free dairy products;

[0024] (2) Hydrolyze lactose into glucose and galactose;

[0025] (3) Production of functional galactooligosaccharides;

[0026] (4) Comprehensive utilization of whey.

[0027] Preferably, the application includes adding the above-mentioned β-galactosidase mutant to the feed of young animals (such as piglets and calves) or to feed containing whey powder to aid in the digestion of lactose, reduce diarrhea in young animals, and promote growth.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0029] This invention combines machine learning and genetic engineering techniques, leveraging the screening advantages of directed evolution, to systematically modify β-galactosidase. First, molecular dynamics simulations and sequence co-evolutionary analysis are used to predict key sites affecting enzyme activity and stability. Second, machine learning models are employed to pre-evaluate mutation effects, narrowing the size of the mutant library. Based on this, a high-quality, small mutant library is constructed. This strategy is expected to yield β-galactosidase mutants with significantly improved catalytic efficiency, enhanced thermal stability, and improved acid-base tolerance, thereby promoting its more efficient and economical application in food, pharmaceutical, and other industrial fields.

[0030] This invention utilizes machine learning technology to rapidly and accurately predict mutation sites that enhance enzyme activity. Based on wild-type β-galactosidase, five site-directed mutations of a single amino acid were performed to construct a β-galactosidase single-amino acid mutant (Gal-5M) with five mutation sites. The β-galactosidase mutant (Gal-5M) exhibits higher enzyme activity compared to the wild type, which is significant for improving production efficiency and reducing costs. The β-galactosidase mutant described in this invention not only broadens the application range of β-galactosidase but also provides a more efficient biocatalyst for industrial production, thus holding promise for wider applications in food processing, agriculture, and animal feed.

[0031] This invention uses an Escherichia coli expression system to express a β-galactosidase mutant. The E. coli cells have strong metabolic capabilities and rich cellular mechanisms, enabling them to efficiently synthesize and fold proteins. Compared with other expression systems, the E. coli expression system is simple to operate and low in cost. The recombinant engineered bacteria culture and protein expression cycle of this invention is relatively short, making it suitable for rapid large-scale production. Attached Figure Description

[0032] Figure 1 shows the overall folding conformation diagram in PyMOL, which is a cartoon / surface view of the three-dimensional structure predicted by AlphaFold2 based on the amino acid sequence of wild-type β-galactosidase. In the figure, the red highlighted residues are the screened and identified mutation sites T254, T256, L296, T339, and Q648.

[0033] Figure 2 shows the protein expression of wild-type β-galactosidase and β-galactosidase mutant; M is the protein marker; L is the whole-cell lysate containing wild-type (Gal) and its mutant (Gal-5M); S is the protein in the supernatant after centrifugation at 4℃; P is the precipitate or insoluble protein fraction after centrifugation at 4℃.

[0034] Figure 3 shows the electrophoresis diagrams of the purified wild-type (A) and β-galactosidase mutant (B) proteins; M is the protein marker; L is the whole-cell lysate containing wild-type (Gal) and β-galactosidase mutant (Gal-5M); S is the protein in the supernatant after centrifugation at 4 °C; P is the precipitate or insoluble protein fraction after centrifugation at 4 °C; 150, 250, and 450 are different concentrations of imidazole elution buffer used in Ni column purification.

[0035] Figure 4 shows the specific activity of β-galactosidase wild-type (Gal) and β-galactosidase mutant (Gal-5M).

[0036] Figure 5 shows the thermal stability of wild-type β-galactosidase (Gal) and its mutant (Gal-5M) after heat treatment for 30 minutes.

[0037] Figure 6 shows the preservation stability of wild-type β-galactosidase (Gal) and its mutant (Gal-5M).

[0038] Figure 7 shows a comparison of the local conformations of the wild-type (left) and mutant (right) T254P sites (modeled using PyMOL). The main chain is shown in cartoon form, key residues are shown in bar form, and the interaction distances are marked to characterize the local microenvironment changes caused by Thr254→Pro254.

[0039] Figure 8 shows a comparison of the local conformations of the wild-type (left) and mutant (right) T256L site (modeled by PyMOL), which shows the changes in local side chain occupancy and close-contact network after Thr256→Leu256, and can be used to explain the potential impact of this site on structural stability and catalytic efficiency.

[0040] Figure 9 shows a comparison of the local conformations of the wild-type (left) and mutant (right) L296Y site (modeled by PyMOL), demonstrating the local stacking and directional interaction trends and spatial matching relationships caused by the introduction of aromatic side chains after Leu296→Tyr296.

[0041] Figure 10 shows a comparison of the local conformations of the wild-type (left) and mutant (right) T339M site (modeled by PyMOL), showing the local hydrophobic occupancy and contact network adjustment after Thr339→Met339, which supports the inference that this site plays a role in structural compaction and the inhibition of conformational fluctuations.

[0042] Figure 11 shows a comparison of the local conformations of the wild-type (left) and mutant (right) sites at the Q648M site (PyMOL modeling), demonstrating the changes in solvent accessibility and interaction networks caused by the local conversion from polar side chains to hydrophobic occupants after Gln648→Met648, which is used to explain its contribution to the inhibition of inactivation pathways and the improvement of overall performance. Detailed Implementation

[0043] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited thereto. Unless otherwise specified in the following embodiments, all conditions are performed according to conventional and known conditions or conditions recommended by the manufacturer. Reagents or instruments used, unless otherwise specified, are all commercially available products. Unless otherwise specified, the present invention employs existing technology in this field.

[0044] Example 1: Prediction and Acquisition of β-galactosidase Mutants

[0045] Machine learning-assisted computation methods utilize bioinformatics data from multiple protein sequences to predict one or more amino acid sequences. These sequences can be substituted based on ancestral or consensus homology to generate stable mutants, making them experimentally attractive in enzyme engineering research.

[0046] This embodiment utilizes machine learning-assisted computation to predict and design β-galactosidase mutants of wild-type β-galactosidase (Lactobacillus sp. B164, accession number JF345715.1), aiming to obtain β-galactosidase mutants with higher enzyme activity. The amino acid sequence of the wild-type β-galactosidase is shown in SEQ ID NO:2, and the nucleotide sequence encoding the β-galactosidase is shown in SEQ ID NO:1.

[0047] SEQ ID NO:1

[0048]

[0049] SEQ ID NO:2

[0050] .

[0051] Enzyme activity enhancement mutation criteria:

[0052] 1) Correlation score ≥ 0.5 (high correlation with activity);

[0053] 2) Conserved sites (Conservation ≤ 6) are allowed to be mutated in terms of structure / function;

[0054] 3) ΔΔG (FoldX) < –0.5 (significantly improves stability (the larger the negative value, the more stable);

[0055] 4) The higher the SCMTTP, the better, as it comprehensively represents stability and activity potential;

[0056] 5) isLowCombined = True FireProt is assessed as a low-risk setting.

[0057] As shown in Figure 1, the three-dimensional structure of the target β-galactosidase was predicted based on AlphaFold2, and its overall folding conformation was shown in the form of a cartoon and surface diagram. The residue sites marked in red are the mutation sites (T254, T256, L296, T339, Q648) screened and determined in this invention, which are used for subsequent mutant construction and enzymatic performance optimization analysis.

[0058] Using the change in folding free energy ΔΔG predicted by FoldX as the preferred stability indicator, ΔΔG < −0.5 kcal / mol indicates that the mutation can improve protein stability, and the larger the negative value, the more significant the stability improvement. The five sites screened under this standard (T254P, T256L, L296Y, T339M, and Q648M) all showed significant negative ΔΔG (−1.03 to −3.52) (Table 1). The conservations were 4, 5, 2, 5, and 1, respectively, indicating that the mutations did not interfere with the conserved functional regions of β-galactosidase. This suggests that these substitutions are expected to reduce the folding free energy by enhancing local packaging, reducing adverse solvent exposure, or reducing conformational fluctuations. This “computational stabilization prediction” is highly consistent with subsequent experimental phenotypes: the mutants not only exhibited better heat treatment / storage stability, but also showed significantly higher enzyme activity / specific enzyme activity in the ONPG assay system than the wild type (Gal-5M approximately 2.37×). The rationale is that, for many enzymes, improved stability is not only reflected in "greater heat resistance," but also in reducing local loose conformations and decreasing the probability of inactivation caused by partial unfolding and aggregation, allowing more molecules to remain in the "set of correct catalytic conformations," thus exhibiting higher apparent enzyme activity and specific enzyme activity under the same assay conditions. Therefore, the high degree of overlap between FoldX's ΔΔG screening and experimental results is consistent with the coupling law of protein thermodynamics and enzyme conformation kinetics.

[0059] To improve the stability of β-galactosidase, machine-aided computational methods using a web server (http: / / pmlabstack.pythonanywhere.com / SCMTPP) were used to predict and design β-galactosidase mutants with potential for improved stability. The results showed that the SCMTTP values ​​of the T254P, T256L, L296Y, T339M, and Q648M mutants were 389.82, 388.91, 389.15, 389.14, and 388.99, respectively. Different amino acid mutations led to different SCMTTP values, with higher values ​​indicating better thermostability (Table 1).

[0060] Table 1. Predictive analysis results of β-galactosidase mutants

[0061] Mutation Site Location ΔG (FoldX) (kcal / mol) Conservatism Low Risk Design SCMTPPT254P254–1.244 389.82T256L256–1.374 388.91L296Y296–1.032 389.15T339M339–2.185 389.14Q648M648–3.521 388.99 surface

[0062] Based on site analysis using the AlphaFold2 structural model, combined with FoldX fold free energy prediction (ΔΔG) and SCMTPP machine learning-assisted evaluation, five key mutation sites (T254P, T256L, L296Y, T339M, and Q648M) were successfully screened. Mutations were made at these five key mutation sites on the wild-type β-galactosidase to obtain the β-galactosidase mutant with the amino acid sequence shown in SEQ ID NO:4. The nucleotide sequence encoding the β-galactosidase mutant is shown in SEQ ID NO:3.

[0063] All mutation sites in the aforementioned β-galactosidase mutants were predicted to have the potential to enhance stability (ΔΔG < -0.5 kcal / mol) and exhibited low conservation (Conservation 1–5), indicating that they are located far from functionally conserved regions and are structurally feasible. SCMTPP values ​​further support the positive effect of these mutations on improving heat resistance. Theoretical analysis suggests that these mutations may enhance structural stability by improving local stacking, reducing exposure to adverse solvents, or decreasing conformational fluctuations, thereby maintaining more enzyme molecules in the catalytically active conformation.

[0064] SEQ ID NO:3

[0065]

[0066] SEQ ID NO:4

[0067] .

[0068] In summary, this embodiment successfully obtained a β-galactosidase mutant with high enzyme activity and good stability.

[0069] Example 2. Construction of recombinant vectors (Gal, Gal-5M)

[0070] The wild-type β-galactosidase (Gal) and the β-galactosidase mutant (Gal-5M) obtained in Example 1 were synthesized by Nanjing GenScript Biotech Co., Ltd. BamHI and EcoRI restriction sites were added to the 5' and 3' ends of these two gene sequences, respectively, to obtain the nucleotide sequences of the two genes with BamHI and EcoRI restriction sites. These sequences were then constructed into the pET28a(+) plasmid, denoted as pET28a(+)-Gal and pET28a(+)-Gal-5M. E. coli were transformed using pET28a(+)-Gal and pET28a(+)-Gal-5M, respectively. The specific steps are as follows:

[0071] Remove competent E. coli BL21(DE3) cells from the -80 ℃ freezer and thaw them on ice. Add 5 μL of each of the two plasmids mentioned above, and gently tap the bottom of the EP tube to mix them. Place the tubes on ice for 30 minutes, incubate at 42 ℃ for 45 seconds, and then immediately place them on ice for 2 minutes. After incubation, add 700 μL of preheated LB medium in a clean bench, mix well, and shake at 37 ℃ and 200 rpm for 2 hours to revive the BL21(DE3) cells. Then, centrifuge at 10,000 rpm for one minute at room temperature to collect the cells. Discard 600 μL of supernatant. Mix the remaining supernatant with the cells and spread the mixture on LB solid medium containing kanamycin. Incubate overnight at 37 ℃ to obtain recombinant E. coli transformed with the Gal and mutant (Gal-5M) plasmids, respectively, i.e., recombinant E. coli expressing Gal and Gal-5M.

[0072] Example 3. Large-scale culture of Escherichia coli expressing Gal and Gal-5M

[0073] (1) The recombinant Escherichia coli expressing Gal and Gal-5M obtained in Example 2 were transferred to solid LB medium containing 50 μg / mL kanamycin and cultured overnight at 37 °C. After the culture, single colonies in the medium were transferred to 5 mL of liquid LB medium containing 50 μg / mL kanamycin and cultured overnight at 37 °C and 200 rpm. After the culture, 3 mL of recombinant Escherichia coli expressing Gal and Gal-5M were inoculated into 300 mL of liquid LB medium containing 50 μg / mL kanamycin and grown at 37 °C and 200 rpm for about 3 h until the bacterial OD 600 When the pH reaches between 0.4 and 0.8, liquid LB medium containing recombinant Escherichia coli seed culture expressing Gal and Gal-5M is obtained.

[0074] (2) Place the liquid LB medium containing the recombinant Escherichia coli seed liquid expressing Gal and Gal-5M obtained in step (1) on ice for 20 minutes, add isopropyl β-D-1-thiogalactopyranoside (IPTG) to it to a final concentration of 0.4 mM, and culture at 25 ℃ with shaking at 180 rpm for 16-20 h (overnight) to induce the expression of Gal and Gal-5M. After the induction is completed, centrifuge the liquid medium at 3000 rpm and 4 ℃ for 20 minutes and discard the supernatant to obtain the cell precipitate cultured at 25 ℃. Store it at -80 ℃ for later use.

[0075] (3) Thaw the cell precipitate obtained in step (2) and resuspend it in 10 mL Tris-HCl (50 mM, pH 8.0). Then, centrifuge at 3000 rpm and 4 ℃ for 20 minutes, discard the supernatant, collect the cells, and wash the cells twice with 50 mM Tris-HCl with a pH of 8.0.

[0076] The bacterial cells were resuspended in 20 mL of Tris-HCl buffer containing 1 mM PMSF (200 μL) and lysed on ice using an ultrasonic cell disruptor for 30 min, followed by alternating sonication for 6 s and intermittent cooling for 6 s to obtain a solution containing cell pellet (total protein). The total protein was centrifuged at 13000 rpm for 30 min at 4 ℃ to obtain the supernatant and pellet. The supernatant was transferred to a new EP tube, and the pellet was resuspended in 50 mM pH 8.0 Tris-HCl. SDS-PAGE was performed on the total protein, supernatant, and pellet to analyze the expression and solubility of the recombinant proteins. The results are shown in Figure 2. The results showed that under 25 ℃ induction conditions, significant bands appeared in the Gal and Gal-5M cell lysates (total protein), supernatant, and pellet relative to the pellet at 75 kDa, indicating that both recombinant proteins were successfully expressed; and the significant band in the supernatant indicated that Gal and Gal-5M are soluble.

[0077] Example 4. Purification of Gal and Gal-5M proteins using Ni-NTA nickel column chromatography

[0078] The supernatants containing Gal and Gal-5M obtained in Example 3 were purified using Ni-NTA columns, and the specific steps are as follows:

[0079] Add 10 mL of nickel removal buffer to the column and wash twice until the column turns from blue to white. Wash the column twice with 10 mL of deionized water. Then, slowly add 10 mL of NiSO4 buffer to the column, cover with both caps, and incubate in a chromatography cabinet at 4 °C for 30 min by rotation. Open the bottom cap to allow excess NiSO4 to flow out, and then wash once with 10 mL of deionized water and once with 10 mL of PBS.

[0080] Add 4 mL of the supernatant solution of Gal or Gal-5M after cell lysis and centrifugation to the column, and incubate at 4 °C in a chromatography cabinet for 2 h. Open the bottom cap to allow unbound protein samples to flow out, and wash again with 15 mL of PBS buffer. After washing, elute with 2 mL of 150 mM, 250 mM, and 450 mM imidazole buffer, respectively. Prepare protein samples for loading, and determine the purification results by SDS-PAGE. The results are shown in Figure 3. As can be seen from Figure 3, Gal and Gal-5M show significant bands at 75 kDa when eluted with 250 mM imidazole buffer, and there are relatively few impurities, indicating that soluble Gal and Gal-5M have been successfully purified.

[0081] Example 5: Detection of Gal and Gal-5M enzyme activities

[0082] The activity of β-galactosidase is determined by measuring the amount of 2-nitrophenyl-β-D-galactopyranoside (ONPG) substrate catalyzed by the enzyme per minute.

[0083] The specific activity of the enzyme (U / mg) was determined by utilizing the molar absorptivity of β-galactosidase at 4500 mol / L / cm. One unit of enzyme activity (U) was defined as the amount of enzyme required to catalyze the hydrolysis of ONPG and release 1 nmol of o-nitrophenol per minute under the stated assay conditions.

[0084] 90 μL of ONPG was added to 400 μL of NaAc-HAc buffer (50 mM, pH 7.5), and the mixture was incubated at 40 °C for 10 minutes. Then, 10 μL of purified β-Gal solution and Gal-5M solution were added, and the reaction was stopped after 10 minutes by adding 500 μL of 1 M Na₂CO₃ solution. The absorbance and enzyme activity of the solutions were then measured. β-Gal solution was used as a control.

[0085] As shown in Figure 4, the enzyme activity of the Gal-5M mutant was 2.37 times that of the wild-type Gal. This indicates that the β-galactosidase mutant (Gal-5M) constructed by mutating threonine (Thr, T) at position 254 to proline (Pro, P); threonine (Thr, T) at position 256 to leucine (Leu, L); leucine (Leu, L) at position 296 to tyrosine (Tyr, T); threonine (Thr, T) at position 339 to methionine (Met, M); and glutamine (Gln, Q) at position 648 to methionine (Met, M) significantly increased the enzyme's catalytic activity.

[0086] Example 6: Thermal stability and storage stability of Gal and Gal-5M

[0087] This embodiment examines the thermal stability of Gal and Gal-5M to evaluate the stability of the β-galactosidase mutant, as detailed below.

[0088] The purified Gal enzyme and Gal-5M were preheated with 50 mM NaAc-HAc (pH 7.5) buffer at 20, 30, 40, 50, 60, and 70 °C for 30 min, respectively. A certain volume of substrate ONPG (final concentration 10 mM) was added, and the reactions were carried out at their respective temperatures for 10 min. The enzyme activities of Gal and Gal-5M were detected using the method described in Example 5. The maximum enzyme activity during the reaction was taken as 100%, and other enzyme activities were plotted as percentages against this value. The results are shown in Figure 5. The figure shows that from 40 °C, the activity of Gal enzyme decreased significantly faster than that of Gal-5M; at 50 °C, Gal-5M maintained 78.10% of its activity, while Gal only maintained 19.29%; at 60 °C, the activity of Gal enzyme was almost zero, while Gal-5M maintained 12.30% of its activity. This indicates that the Gal-5M mutant obtained by introducing five key amino acid mutations has significantly improved thermostability compared to the wild-type Gal enzyme. At higher temperatures, Gal-5M can maintain higher catalytic activity, especially in the range of 40–60 °C, where its enzyme activity retention rate is significantly higher than that of the wild type, indicating that this mutant is more suitable for maintaining functional stability in high-temperature or large-temperature-fluctuation environments.

[0089] This embodiment also investigated the storage stability of Gal and Gal-5M. Two samples of each Gal and Gal-5M enzyme were prepared, one stored at 4 °C and the other at 25 °C. Enzyme activity was then measured at 0, 4, 8, 12, 16, 20, 24, and 28 days of storage, following the method described in Example 5. The highest enzyme activity was taken as 100%, and other enzyme activities were compared with the highest activity. A time-vs-relative enzyme activity curve was plotted, and the results are shown in Figure 6. The figure shows that at both 4 °C and 25 °C, the activities of both Gal and Gal-5M decreased with prolonged storage time, but the activity of Gal-5M remained consistently higher than that of Gal. This indicates that the amino acid mutation improved the storage stability of Gal-5M.

[0090] From the perspective of amino acid chemical properties and local structural microenvironment, the five mutation sites have a clear molecular basis for affecting enzyme activity and stability.

[0091] (1) T254P: Conformation entropy locking suppresses local fluctuations and increases the proportion of effective catalytic conformations. In wild-type Thr254, the side chain hydroxyl groups form multiple close-range interactions (approximately 3.2–3.5 Å) locally, suggesting their participation in switchable hydrogen bond and water bridge networks. Although such networks can provide some support, they are prone to introducing microscopic conformational polymorphisms during temperature changes or conformational respiration, leading to increased local conformational entropy and ineffective wobbling. After mutation to Pro254, the local close-range contact is still maintained (approximately 3.3–3.5 Å), but the ring structure of Pro has a strong constraint effect on the φ angle of the main chain, making this region more inclined to the fixed turn ring conformation, i.e., achieving "conformation entropy locking". This locking effect reduces the probability of local unfolding at the initiation of flexible rings and reduces the early inactivation pathway under thermal storage conditions. On the other hand, it makes the geometry near the backbone or channel entrance more consistent, reducing non-productive conformations for substrate entry and localization, thereby improving apparent enzyme activity and specific enzyme activity (Figure 7).

[0092] (2) T256L: Enhanced hydrophobic packing and drainage reduce "breathing looseness" and improve apparent catalytic efficiency. Wild-type Thr256 has multiple short contacts (approximately 2.7–2.9 Å) around it, indicating that the site is in a relatively compact interaction environment. Thr is a weakly polar residue, which may introduce water-mediated interactions in the semi-buried region. Its occupation and release can cause small geometrical drifts, thus becoming a source of local fluctuations. After mutation to Leu256, the local area still maintains a tight contact (approximately 2.8–2.9 Å), and no obvious cavity or looseness is observed. At the same time, Leu is larger and strongly hydrophobic, which is conducive to filling microcavities, enhancing van der Waals packing, and improving drainage effect, thereby reducing solvent penetration and water bridge fluctuations and inhibiting local "breathing looseness". This compact microenvironment can reduce skeletal noise and energy barrier deviation during substrate entry / localization, thereby improving apparent catalytic efficiency and specific enzyme activity (Figure 8).

[0093] (3) L296Y: The introduction of aromatic ring stacking and directional hydrogen bonding "anchoring" converges the local conformation and enhances catalytic efficiency. Wild-type Leu296 has contacts on the order of 2.9–3.3 Å, suggesting that it mainly participates in local packaging through hydrophobicity and van der Waals interactions. However, the Leu interaction is relatively weak in directionality and has limited ability to "directionally lock" the conformation. After mutation to Tyr296, the local contact is still maintained. At the same time, the Tyr aromatic ring can enhance hydrophobic adhesion and may form a more stable stacking and clamping effect. Its phenolic hydroxyl group further provides hydrogen bond sites with stronger directionality, forming a "conformation anchoring" effect. Thus, without destroying the original hydrophobic core, the local conformational diversity can be reduced and the geometric distribution related to catalysis can be more concentrated, thereby improving the average turnover efficiency and specific enzyme activity (Figure 9).

[0094] (4) T339M: A key node for stabilizing enzyme activity and stability through hydrophobic "gap-filling" site, reducing long-range fluctuations and promoting simultaneous improvement in enzyme activity and stability. The comparison of T339M shows that there are multiple dense contacts of 2.9–3.4 Å around wild-type Thr339, indicating that this site is in a relatively critical stacking environment. The variable polarity interaction and water bridge rearrangement involved by Thr hydroxyl groups may make this node more sensitive to dynamic changes. After mutation to Met339, the side chain occupancy is significantly increased and multiple close contacts (about 2.9–3.4 Å) are still maintained, which is consistent with the "soft but occupancy" gap-filling characteristics of Met: the local core is improved through more sufficient hydrophobic occupancy and van der Waals adhesion, the interface tightness is improved, and the variable polarity network initiated by hydroxyl groups is removed, reducing the source of fluctuations. If this position participates in the coupling of structural elements or affects the overall conformational respiration, its stabilization can reduce ineffective conformations and non-productive deformation of pockets / channels, thus contributing more significantly to the improvement of enzyme activity and specific enzyme activity, and showing a synchronous trend with the improvement of stability (Figure 9).

[0095] (5) Q648M: Inhibits dynamic hydrogen bond rearrangement and solvent penetration, reduces irreversible inactivation and increases the proportion of effective active proteins. Wild-type Gln648 has a polar contact of about 3.0–3.4 Å with its surroundings, but the Gln amide side chain is highly polar and has a high degree of conformational freedom. It is easy to form a dynamic hydrogen bond network in semi-buried or interfacial regions and allow water molecules to intermittently penetrate, thus triggering the "loosening-rearrangement-irreversible inactivation" pathway during thermal stress or long-term storage. After mutation to Met648, the side chain becomes more hydrophobic and has a clear occupancy. It still maintains a close contact of 3.0–3.5 Å locally, and no cavity formation is observed. Instead, it becomes more compact. At the same time, hydrophobicity significantly reduces solvent accessibility, inhibits the formation and rearrangement of water-mediated networks, thereby reducing the probability of inactivation and increasing the proportion of effective active proteins. This mechanism can improve average turnover efficiency by reducing conformational noise and inactivation pathways without directly altering catalytic chemical steps, ultimately forming a closed loop consistent with the result of a significant increase in enzyme activity / specific enzyme activity (Figure 10).

[0096] In summary, none of the five mutation sites selected in this invention introduced obvious cavities or structural damage after mutation. On the contrary, they maintained a tight contact network in their respective local areas within the range of 2.7–3.5 Å, indicating that these mutations are more of a "reinforcement and convergence" rather than a "conflict and loosening" in terms of space and energy. Based on this, the effects of the five mutations are clearly complementary and hierarchical: T254P mainly achieves local "conformational entropy locking" through main chain conformational constraints, suppressing ineffective oscillations and early local unfolding of flexible ring regions (Figure 7); T256L reduces local breathing looseness and water-mediated network fluctuations by enhancing hydrophobic volume filling and drainage (Figure 8); L296Y introduces aromatic ring stacking and potential directional hydrogen bond "anchoring" while maintaining the original hydrophobic core compactness, further converging the conformational distribution of the local microenvironment (Figure 9); T339M strengthens the coupling of key nodes / structural elements with "hydrophobic gap-filling occupancy", reducing long-range fluctuations and unproductive conformations (Figure 10); Q648M significantly inhibits solvent penetration and dynamic hydrogen bond rearrangement by replacing the multi-configuration polar amide side chains with hydrophobic occupancy, reducing the irreversible inactivation pathways during thermal stress and storage from the source (Figure 11).

[0097] Since the aforementioned effects target different limiting factors such as "local conformational entropy," "solvent accessibility drainage," "hydrophobic packing cavity filling," "structural element coupling stability," and "remote inactivation pathway inhibition," they do not compete for the same set of interaction resources in the same microregion, and therefore are unlikely to cancel each other out. On the contrary, when multiple types of conformational noise are suppressed simultaneously, the conformational energy landscape of the protein will shift from "polymorphic distribution" to "a distribution more concentrated in the active conformation," which manifests in the following ways at the kinetic level: ① The proportion of effective active molecules increases, reducing the "false concentration" caused by partial folding, micro-unfolding, and aggregation; ② The substrate entry and localization pathways are more stable, non-productive binding is reduced, and the apparent energy barrier and geometric deviation are lowered; ③ The geometric fluctuations of the active center converge, and the average turnover efficiency is improved. The aforementioned synergistic effect at the structural level is consistent with the significant increase in enzyme activity and specific enzyme activity of the mutant shown in the experimental results. In particular, it explains why the 5-site superposition Gal-5M can achieve significantly higher enzyme activity and specific enzyme activity than the wild type under the same assay system, and this increase occurs simultaneously with enhanced stability. The essence is that multiple sites reduce inactivation pathways and conformational noise in a complementary manner, allowing the enzyme molecule to be in the catalytically effective conformation set for a larger proportion and for a longer period of time under the assay conditions, thereby achieving a "significant increase after superposition" rather than "mutual cancellation".

[0098] In summary, this invention utilizes machine learning algorithms to train the amino acid sequence and three-dimensional structure data of β-galactosidase, and selects specific amino acid sites for site-directed mutagenesis based on the prediction results of the machine learning model to obtain β-galactosidase mutants. The amino acid sequence of the β-galactosidase mutant is shown in SEQ ID NO:4, and the nucleotide sequence encoding the β-galactosidase mutant is shown in SEQ ID NO:3. Compared with the wild type, the β-galactosidase mutant has higher degradative activity and thermal stability, which is of great significance for improving production efficiency and reducing costs, and has good practicality.

[0099] The embodiments described above are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A β-galactosidase mutant with high enzyme activity and thermostability, characterized in that, The β-galactosidase mutant is based on the wild-type β-galactosidase with the amino acid sequence shown in SEQ ID NO:2, and undergoes one or more of the following mutations: (1) mutating the 254th position from threonine to proline; (2) mutating the 256th position from threonine to leucine; (3) mutating the 296th position from leucine to tyrosine; (4) mutating the 339th position from threonine to methionine; (5) mutating the 648th position from glutamine to methionine.

2. The β-galactosidase mutant according to claim 1, characterized in that, The β-galactosidase mutant underwent five mutations simultaneously, and the amino acid sequence of the β-galactosidase mutant is shown in SEQ ID NO:

4.

3. A biomaterial, characterized in that, The biological material includes a polynucleotide, a recombinant vector, or a recombinant engineered bacterium; the polynucleotide encodes the above-mentioned β-galactosidase mutant; the recombinant vector contains the above-mentioned polynucleotide or expresses the above-mentioned β-galactosidase mutant; the recombinant engineered bacterium contains the above-mentioned polynucleotide, or contains the above-mentioned recombinant vector, or expresses the above-mentioned β-galactosidase mutant.

4. The biomaterial according to claim 3, characterized in that, The nucleotide sequence of the polynucleotide includes SEQ ID NO:3 or its degenerate sequence.

5. The biomaterial according to claim 3, characterized in that, The recombinant vector includes a recombinant prokaryotic vector, wherein the prokaryotic vector includes a pET(+) plasmid or a pGEX vector.

6. The biomaterial according to claim 3, characterized in that, The host bacteria of the recombinant engineered bacteria include Escherichia coli.

7. A method for preparing a β-galactosidase mutant, characterized in that, The method includes: inoculating recombinant engineered bacteria expressing the β-galactosidase mutant of claim 1 into a fermentation medium, fermenting and culturing, then centrifuging and collecting the supernatant to obtain the crude product of the β-galactosidase mutant.

8. The application of the β-galactosidase mutant of claim 1 or the biomaterial of claim 3 in food processing and feed.

9. The application according to claim 8, characterized in that, In the application, food processing includes any one or more of the following: (1) production of low-lactose / lactose-free dairy products; (2) hydrolysis of lactose into glucose and galactose; (3) production of functional galactooligosaccharides; and (4) comprehensive utilization of whey.

10. The application according to claim 8, characterized in that, The application includes adding the β-galactosidase mutant of claim 1 to young animal feed or feed containing whey powder.

Images