Amino transferase mutant and application thereof in leucine production

Abstract

The invention discloses an aminotransferase mutant from thermophilus thermophilus and application of the aminotransferase mutant in production of L-leucine, the activity of the aminotransferase of the mutant is high, the yield of the L-leucine is increased, and meanwhile, accumulation of a byproduct valine is effectively reduced.

Description

Technical Field

[0001] This invention belongs to the fields of genetic engineering and fermentation engineering technology, specifically relating to an aminotransferase mutant and its application in leucine production. Background Technology

[0002] L-Leucine, one of the eight essential amino acids, has wide applications and market demand in pharmaceuticals, food, feed, and cosmetics. The global L-leucine market continues to grow. Compared to traditional hair hydrolysis for L-leucine production, microbial fermentation offers advantages such as low cost, environmental friendliness, and high efficiency. However, the long biosynthetic pathway and strict feedback control mechanism of L-leucine in microorganisms mean that microbial fermentation for L-leucine production still has significant drawbacks, including low yield, low conversion rate, long fermentation cycle, and the accumulation of branched-chain amino acid byproducts such as L-valine, limiting its economic benefits.

[0003] Aminotransferases (ATFs) in bacteria constitute a large and diverse family of enzymes that catalyze the transfer of the amino group from one amino acid to another α-keto acid, generating new keto acids and amino acids. They play crucial roles in core life activities such as nitrogen metabolism, amino acid synthesis and degradation, and cell wall synthesis. In bacteria such as *E. coli*, the final step in L-leucine biosynthesis is a critical rate-limiting step: branched-chain amino acid aminotransferases encoded by ilvE catalyze the transamination of the precursor α-ketoisocaproic acid (α-KIC) to L-leucine. However, this enzyme has low catalytic efficiency and weak substrate specificity, and it also participates in the synthesis of other branched-chain amino acids, easily leading to the accumulation of L-valine and L-isoleucine (metabolic pathway see [link]). Figure 1 ).

[0004] To specifically enhance L-leucine synthesis, a common strategy is to replace endogenous branched-chain amino acid aminotransferases with enzymes that have a stronger affinity for the L-leucine precursor α-KIC and higher catalytic efficiency. CN116355818A employs a strategy of introducing leucine dehydrogenase (NCBI protein database ID NP_390288.1 and EC:1.4.1.9, encoding gene UniProt named leuDH and NCBI named bcd) from *E. coli*. Utilizing the asymmetric reductive amination reaction catalyzed by this enzyme, with the aid of the coenzyme NADH, an oxygen atom on the keto acid molecule is replaced with an amino group (-NH2), ultimately generating L-amino acids. The engineered strain constructed achieved an L-leucine yield of 85.6 g / L in a fermenter, with a conversion rate of 0.36 g / g, making it the strain with the highest reported production performance to date. However, Bacillus leuDH not only has a high affinity for the L-leucine precursor α-ketoisocaproic acid (α-KIC), but also for the L-valine precursor α-ketoisovalerate (α-KIV). High expression of α-KIV can easily lead to the accumulation of L-valine, thereby reducing leucine production performance and increasing separation and purification costs. CN110607268B utilizes this characteristic of LeuDH to introduce the Bacillus subtilis leucine dehydrogenase gene to replace the Escherichia coli branched-chain amino acid aminotransferase gene ilvE in constructing a genetically engineered bacterium that produces high levels of L-valine.

[0005] Furthermore, studies have found that aromatic amino acid aminotransferases (encoding gene tyrB) in *E. coli* and other bacteria primarily responsible for aromatic amino acid synthesis also possess the function of specifically catalyzing the synthesis of L-leucine from α-KIC. Moreover, compared to the leucine dehydrogenase encoded by leuDH, the aromatic amino acid aminotransferase TyrB encoded by tyrB has a significantly higher affinity for the leucine substrate α-KIC than for the valine substrate α-KIV. However, research on the direct catalysis of α-KIC to L-leucine using TyrB is limited, and there is little in-depth exploration and optimization of this function. Although some researchers have used tyrB to replace the ilvE gene to construct L-leucine-producing strains, the leucine yield is not ideal and fails to meet the requirements for industrial production. Therefore, discovering and constructing aminotransferases with high enzyme activity and strong substrate specificity is an important direction for improving the quality and efficiency of L-leucine synthesis.

[0006] In recent years, artificial intelligence (AI) technology has demonstrated tremendous potential in enzyme discovery and protein design and modification. The intervention of deep learning technology has shifted protein engineering from the traditional "random mutation + high-throughput screening" model to a new paradigm based on "model navigation - data-driven." By using multi-dimensional collaborative searches based on sequence, structure, and language models to obtain efficient candidate enzymes, the blind spots and waste of wet experiments have been significantly reduced. With the release of AlphaFold, through a structure-aware protein language model, full-site saturation mutation prediction of target proteins is performed, enabling intelligent initial screening of massive mutants and providing a revolutionary tool for enzyme modification and evolution.

[0007] To address the problems of low yield and numerous byproducts in existing L-leucine production, this invention breaks through the limitations of traditional protein modification methods. By combining systems metabolic engineering with artificial intelligence-assisted protein design, it identifies and modifies key rate-limiting enzymes, providing technical support for the industrial production of L-leucine. Summary of the Invention

[0008] To improve the yield and conversion rate of L-leucine, this invention provides a novel aminotransferase mutant derived from the thermophilic bacterium *Tepidiphilus thermophilus* and its application in L-leucine production. This mutant not only increases aminotransferase activity and L-leucine yield but also reduces the accumulation of the byproduct valine. To achieve the objectives of this invention, in a first aspect, this invention provides an amino acid aminotransferase mutant derived from the thermophilic bacterium, said mutant comprising or consisting of an amino acid sequence selected from or composed of the following: 1) A mutation from H to R in amino acid position 233 of the aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). H233R ); 2) A mutation from P to D at amino acid position 75 of an aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). P75D ); 3) A mutation from P to L in amino acid position 75 of an aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). P75L ); 4) A mutation from T to A at amino acid position 55 of the aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). T55A ); 5) A mutation from L to A in amino acid position 236 of the aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). L236A ); 6) A mutation from R to P at amino acid position 152 in an aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). R152P ); 7) A mutation from A to D at amino acid position 169 of the aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). A169D ); 8) A mutation from H to A in amino acid position 234 of the aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). H234A ); 9) A mutation from P to S at amino acid position 75 of an aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). P75S ); 10) A mutation from K to R at amino acid position 54 of an aminotransferase from thermophilic bacteria, as shown in SEQ ID NO:1 (TyrB). K54R ) In a second aspect, the present invention provides a nucleic acid molecule encoding the amino acid aminotransferase mutant.

[0009] Thirdly, the present invention provides biological materials containing the nucleic acid molecules, including but not limited to recombinant DNA, expression cassettes, transposons, plasmid vectors, viral vectors, or engineered bacteria.

[0010] Fourthly, the present invention provides a recombinant microorganism, which is constructed by integrating a nucleic acid molecule encoding the aminotransferase mutant into the chromosome of Escherichia coli through genetic engineering.

[0011] Fifthly, the present invention provides the application of the aminotransferase mutant, the biological material, or the recombinant microorganism in the production of L-leucine.

[0012] By employing the above technical solution, the present invention has at least the following advantages and beneficial effects: This invention mutates the amino acid aminotransferase TyrB derived from thermophilic bacteria, screening for mutants with strong substrate specificity and higher aminotransferase activity. This not only increases the yield of L-leucine but also eliminates the accumulation of the byproduct L-valine, which is of great significance for the large-scale production of L-leucine.

[0013] This invention discovers that utilizing a two-stage dissolved oxygen control mode for L-leucine fermentation is beneficial for improving sugar-acid conversion. Aerobic fermentation is carried out in the early stages of fermentation to primarily supply the normal growth of the cells. When the biomass accumulates to a certain level, microaerobic fermentation is initiated. At this point, the tricarboxylic acid cycle is weakened, significantly reducing precursor consumption and promoting L-leucine accumulation, thereby increasing the sugar-acid conversion rate. Furthermore, the use of microaerobic fermentation in the middle and later stages of fermentation allows for a reduction in stirring speed, significantly lowering fermentation energy consumption.

[0014] Compared with existing L-leucine engineered strains, the engineered strain provided by this invention has the following advantages: high yield, high conversion rate, reduced production cost, and good industrialization prospects. Attached Figure Description

[0015] Figure 1 Metabolic modification diagram of the L-leucine synthesis pathway in Escherichia coli Figure 2 Results of relative L-leucine content in shake flasks of heterologous tyrB Figure 3 Results of relative L-valine content in shake flasks of heterologous tyrB Figure 4 Shake-flask fermentation results for the tyrB1 mutant Figure 5 Results of L-leucine production in fermenters Detailed Implementation

[0016] The strains used in the embodiments of this invention are shown in the table below: Table 1. Strains used in this invention The plasmids used in the embodiments of this invention are shown in the table below: Table 2. Plasmids used in this invention The primer sequences used in the embodiments of this invention are shown in the table below: Table 3 Primers used for constructing tyrB overexpression strains Example 1: Construction of starting strain L1 According to the construction method of engineered bacteria (LEU30) described in reference 1 (Hao Y, Pan X, Li G, et al. Construction of a plasmid-free L-leucine overproducing Escherichia coli strain through reprogramming of the metabolic flux. Biotechnol Biofuels Bioprod. 2023;16(1):145. doi:10.1186 / s13068-023-02397-x), Escherichia coli MG1655 was used instead of E. coli W3110ΔlacI as the starting strain. LeuA, derived from E. coli and resistant to feedback inhibition, was overexpressed at the ycjV, yciQ, and yghE sites. fbr Overexpressing leuBCD at the yjgX site and replacing the leuA promoter with promoter P trc Overexpression of ilvIH, derived from E. coli and resisting feedback inhibition, at the yjiT and ylbE sites fbr The feedback inhibition and transcriptional attenuation of key enzymes in the leucine synthesis pathway were relieved; the ldhA, pflB, and poxB genes were knocked out to block the production of byproducts lactic acid, formic acid, and acetic acid; pntAB was overexpressed at the yjiP site to increase the level of intracellular cofactor NADPH; brnQ was knocked out to block intracellular leucine uptake; and brnFE from Corynebacterium glutamicum was overexpressed to enhance leucine efflux, resulting in strain L0 (Table 1).

[0017] Following the construction method of the engineered strain (LXH-21) described in Reference 2 (Ding X, Yang W, Du X, et al. High-level and 1-yield production of L-leucine in engineered Escherichia coli by multistep metabolic engineering. Metab Eng. 2023;78:128-136. doi:10.1016 / j.ymben.2023.06.003), strain L0 was used instead of E. coli W3110ΔlacI in the reference as the starting strain, and ilvBN, an anti-feedback inhibition antagonist derived from Corynebacterium glutamicum, was overexpressed. CP This involves relieving feedback inhibition of key enzymes in the leucine synthesis pathway; knocking out the ackA gene to block the production of the byproduct acetic acid, increasing the supply of pyruvate and acetyl-CoA, and reducing carbon loss; and knocking out ilvC and integrating ilvC. EMBy integrating rocG from Bacillus subtilis at the yghX site, the intracellular NADPH / NADH levels were balanced, and the redox flux was optimized. Knocking out livK blocked the uptake of intracellular leucine, and overexpressing leuE enhanced the efflux of leucine, resulting in strain L0' (Table 1).

[0018] According to the construction method of engineered strain (Val) described in reference 3 (Park JH, Kim TY, Lee KH, et al. Fed-batch culture of Escherichia coli for L-valine production based on in silico flux response analysis. Biotechnol Bioeng. 2011;108(4):934-46. doi: 10.1002 / bit.22995.), strain L0' was used instead of E. coli W3110ΔlacI in the reference as the starting strain. ilvA was knocked out to block isoleucine synthesis, and finally the basic strain L1 was obtained (Table 1).

[0019] Example 2: Fermentation Condition Test of Basic Strains 1. Culture medium preparation LB medium (L): 5 g yeast extract, 10 g peptone, 10 g NaCl, and 20 g agar powder added to the solid medium.

[0020] Shake flask seed culture medium (L): yeast extract 5 g, peptone 10 g, NaCl 10 g.

[0021] Shake flask fermentation medium (L): glucose 30 g, yeast extract 2.5 g, (NH4)2SO4 5 g, citric acid 1 g, KH2PO4 1 g, MgSO4·7H2O 1 g, MnSO4·H2O 10 mg, FeSO4·7H2O 10 mg, biotin 0.2 mg, vitamin B1 2 mg, MOPS 42 g, pH 7.2.

[0022] 2. Shake-flask fermentation test Shake-flask seed culture: Cryopreserved glycerol-based bacteria were streaked onto LB solid medium for activation and incubated overnight at 37°C. Fresh single colonies were inoculated into 5 mL shake-flask seed cultures and incubated overnight at 37°C and 220 rpm to obtain seed cultures.

[0023] Shake-flask fermentation culture: Seed culture was inoculated into shake-flask fermentation medium and fermented under the following three conditions.

[0024] Shake-flask aerobic fermentation: controlling the initial inoculation OD 600 =1 was inoculated into 20 mL of shake flask fermentation medium and cultured at 37℃ and 220 rpm for 24 h.

[0025] Shake-flask microaerobic fermentation: controlling the initial inoculation OD 600 =1 was inoculated into 20 mL of shake flask fermentation medium and cultured at 37℃ and 160 rpm for 24 h.

[0026] Shake-flask aerobic-microaerobic two-stage fermentation: controlling the initial inoculation OD 600 =1 was inoculated into 20 mL of shake flask fermentation medium and cultured at 37℃ and 220 rpm; after 8 h, it was cultured at 37℃ and 160 rpm for 16 h; the total fermentation cycle was 24 h.

[0027] Fermentation broth processing and detection: 1 mL of the fermentation broth sample was taken into a 1.5 mL centrifuge tube, centrifuged at 12000 rpm for 10 min, and the supernatant was collected and diluted 10 times with water. The diluted solution was filtered through a 0.22 μm PES filter membrane, and the amount of L-leucine was detected by HPLC.

[0028] Table 4 Comparison of OD, acid production, and conversion rate in shake-flask fermentation under different oxygen conditions Table 4 shows that the OD of the basic strain L1 under aerobic-microaerobic dual-stage fermentation conditions... 600 The yields of L-leucine and the sugar-acid conversion rate reached 8.45 g / L, 5.7 g / L, and 19%, respectively, significantly higher than those in aerobic and microaerobic modes. Therefore, aerobic-microaerobic dual-stage culture conditions were selected for subsequent shake-flask fermentation experiments and tests to screen and validate highly efficient enzymes.

[0029] Example 3: Selection and identification of key rate-limiting enzymes In *E. coli*, the reaction from α-ketoisocaproic acid (α-KIC) to L-leucine is catalyzed by a branched-chain aminotransferase encoded by the endogenous gene *ilvE*. However, due to its low activity and poor specificity, this gene is unsuitable for large-scale leucine production. Therefore, when constructing leucine-producing strains, the endogenous branched-chain aminotransferase gene *ilvE* was replaced with either the highly active aromatic aminotransferase (TyrB) gene *tyrB* or the leucine dehydrogenase (LeuDH) gene *leuDH*. The effects of *tyrB* and *leuDH* on L-leucine yield and byproducts were then tested.

[0030] 1. Construction of strains overexpressing tyrB and leuDH Starting with strain L1, the endogenous gene ilvE was replaced with either the tyrB gene from wild-type Escherichia coli MG1655 (Uniport ID: P04693) or the leuDH gene from Bacillus subtilis 168 (Uniport ID: P54531). Specifically: (1) Preparation of recombinant DNA fragments Using the genome sequence of Escherichia coli MG1655 as a template, the upstream and downstream homologous arms of the ilvE gene were amplified using primers ilvE-U-1 / ilvE-U-2 and ilvE-D-1 / ilvE-D-2. The Ptrc-tyrB fragment containing the Ptrc promoter and the tyrB gene was amplified using primers Ptrc / tyrB-1 / tyrB-2. Then, the upstream and downstream homologous arms and the tyrB gene were fused by overlap PCR using primers ilvE-U-1 / ilvE-D-2 to obtain the integrated fragment Ptrc-tyrB.

[0031] Using the Bacillus subtilis 168 genome sequence as a template, the upstream and downstream homologous arms of the ilvE gene were amplified using primers ilvE-U-1 / ilvE-U-2 and ilvE-D-1 / ilvE-D-2. The Ptrc-leuDH fragment containing the Ptrc promoter and the leuDH gene was amplified using primers Ptrc / leuDH-1 / leuDH-2. Then, the upstream and downstream homologous arms and the leuDH gene were fused by overlap PCR using primers ilvE-U-1 / ilvE-D-2 to obtain the integrated fragment Ptrc-leuDH.

[0032] (2) Construction of gRNA plasmids Using pGRB plasmid as a template, PCR amplification was performed using primers pGRB-ilvE-1 and pGRB-ilvE-2. The PCR product was purified and recovered, transferred into DH5α, plated on LB plates containing ampicillin (50 μg / mL), and incubated at 37℃ for 12 h. Positive clones were then identified to obtain the recombinant plasmid pGRB-ΔilvE.

[0033] (3) Gene editing The pGRB-ΔilvE plasmid obtained in step (2) and the integration fragment obtained in step (1) were co-transformed into competent cells of the chassis strain containing the pCas9 plasmid. The cells were plated on LB agar plates containing ampicillin (50 μg / mL) and spectinomycin (100 μg / mL) for screening. Sequencing was performed using primers ilvE-U-1 / ilvE-D-2 to verify the selection of positive recombinants. The positive recombinants were then transferred to LB medium containing 0.2% arabinose and cultured overnight at 37°C to induce the loss of the pGRB-ΔilvE plasmid. They were then transferred to antibiotic-free LB medium and cultured overnight at 42°C to induce the loss of the pCas9 plasmid, ultimately yielding the tyrB overexpressing strain L1-TyrB and the leuDH overexpressing strain L1-LeuDH.

[0034] 2. Shake-flask fermentation test of tyrB and leuDH overexpression strains The preparation, specific operation, sample processing and detection methods of the shake-flask fermentation medium are described in Example 2. The fermentation conditions selected are aerobic-microaerobic two-stage fermentation.

[0035] After fermentation, the fermentation broth samples were processed, and the L-leucine and L-valine yields of L1-TyrB and L1-LeuDH strains were detected by HPLC.

[0036] Table 5 Comparison of fermentation strains Table 5 shows that the starting strain L1 uses the branched-chain amino acid aminotransferase encoded by the endogenous gene ilvE to catalyze the production of L-leucine from α-KIC. The yield of L-valine is basically equivalent to that of L-leucine, reaching about 95% of the L-leucine yield. Although the yield of L-leucine increased by about 40% and the accumulation of L-valine decreased after replacing ilvE with leuDH, the amount of L-valine in the final product still reached more than 20% of L-leucine. The yield of L-leucine in strain L1-TyrB, which replaced ilvE with tyrB, was 50.7% higher than that of L1, and there was almost no accumulation of L-valine. This result indicates that compared with the leucine dehydrogenase LeuDH, the aromatic amino acid aminotransferase TyrB has higher specificity for the leucine production substrate α-KIC and has greater potential for leucine fermentation production. Next, based on the Escherichia coli aromatic amino acid aminotransferase TyrB, we will further explore heterologous aminotransferases with strong substrate specificity and high enzyme activity as efficient leucine production enzymes.

[0037] Example 4: Discovery and initial screening of aminotransferases suitable for leucine production 1. Discovery of heterologous aminotransferases (1) Enzyme mining based on sequence similarity Using the amino acid sequence of the aromatic amino acid aminotransferase (Uniport ID: P04693) encoded by *Escherichia coli* strain K12 tyrB as the base sequence, a comparison search was performed in the UniRef90 database using Diamond software to screen sequences with 30%–90% identity to the base sequence. The screened sequences were then clustered using USEARCH software at a 90% identity threshold, and the centroids of each cluster were extracted using the "-centroids" parameter as representative sequences, ultimately obtaining a TyrB candidate sequence set containing 98 sequences from different sources.

[0038] (2) Enzyme mining based on structural information and functional annotation Using the AA+3Di vocabulary of the A chain in the crystal structure of TyrB from *E. coli* K12 (PDB ID: 3FSL) as input, the Swiss-Prot sequence database was searched using the protein language model search tool ProTrek, and simultaneously, the PDB and AFDB structure databases were searched using the Foldseek tool based on structural information. Sequences containing the term "aminotransferase" were selected based on functional annotation information. The obtained sequences were compared with candidate set 1 using Diamond software, and redundant sequences with a sequence identity higher than 90% were removed. Clustering was performed again using USEARCH (90% identity threshold), and the centroids of each cluster were extracted using the "-centroids" parameter as representative sequences, resulting in a TyrB candidate sequence set 2 containing 14 sequences.

[0039] (3) Screening based on enzyme kinetics prediction of target substrate Candidate sequence set 1 and candidate sequence set 2 were merged to construct a total candidate sequence set containing 112 sequences. Using the ion activation Smiles formula (CC(C)CC(=O)C(=O)[O-]) of the leucine precursor 2-KIC as the input substrate, the enzyme kinetic parameters (Kcat / Km) of the total candidate sequence set and the base sequence (UniportID: P04693) were predicted and ranked using UniKP, an enzyme kinetic parameter prediction tool based on language and machine learning models. Finally, 17 sequences with higher predicted values ​​than the base sequence and greater potential were selected as test sequences for experimental verification.

[0040] Table 6. TyrB candidate sequences and their Kcat / Km 2. Construction of strains overexpressing heterologous aminotransferases The TyrB1-TyrB17 sequences from 17 different sources listed in Table 6 were sent to Beijing Qingke Biotechnology Co., Ltd. for codon optimization and gene synthesis.

[0041] Starting with the basic strain L1, its branched-chain amino acid aminotransferase gene ilvE was replaced with the tyrB genes from the 17 different sources mentioned above. The specific methods and steps were the same as those described in Example 3, resulting in strains L1-TyrB1 to L1-TyrB17 (see Table 1).

[0042] 3. Screening test for heterologous aminotransferases L1-TyrB and L1 strains were selected as control strains, while L1-TyrB1 to L1-TyrB17 were used as experimental strains. Shake-flask fermentation was conducted according to the aerobic-microaerobic two-stage fermentation method described in Example 2. After fermentation, the fermentation broth samples were processed, and the yields of L-leucine and L-valine were detected by HPLC.

[0043] The results are as follows Figure 2 As shown, among the 17 strains that replaced the endogenous ilvE gene with heterologous aminotransferase genes, strains overexpressing tyrB1, tyrB3, tyrB4, tyrB6, tyrB7, tyrB8, tyrB9, and tyrB10 all showed increased L-leucine production compared to the control strain L1-TyrB. Except for the three strains overexpressing tyrB3, tyrB7, and tyrB11, the L-valine production of the remaining 14 strains was very low. Overexpression of the tyrB1 aminotransferase gene (UniProt ID: A0A0K6IX30, amino acid sequence as shown in SEQ ID NO:1) was the most effective, with the recombinant strain L1-TyrB1 exhibiting the highest leucine level, reaching twice that of the basal strain L1, and also increasing by 31% compared to the control strain L1-TyrB, with negligible accumulation of the byproduct L-valine. Therefore, the coding gene of TyrB1, derived from thermophilic bacteria, was chosen as the template gene for further evolution.

[0044] Example 5: Screening of TyrB1 mutants based on deep learning model 1. Construction of TyrB1 mutant library The three-dimensional structure of the target protein TyrB1 was predicted using AlphaFold 2. The model quality assessment parameter pTM value was 0.95, indicating that the predicted structure had extremely high confidence. Based on the predicted TyrB1 protein structure, the structural information was converted into an AA+3Di vocabulary using the utils.foldseek_util module. The structure-aware protein language model SaProt (pre-trained model version: SaProt_1.3B_AFDB_OMG_NCBI) was used to perform full-site saturation mutation simulation on the TyrB1 protein, covering 7638 mutants. The zero-shot prediction capability of this model was used to evaluate the mutational effect of each mutation site. The top 10 mutants were selected as candidate mutation sites based on their mutational effect prediction scores for subsequent experimental validation.

[0045] Table 7 TyrB candidate mutants 2. Construction of TyrB1 mutant strain Starting with L1-TyrB1, amino acid mutations were performed at positions 54, 55, 75, 152, 169, 233, 234, and 236 of TyrB1 using CRISPR / Cas9 technology, as described in Example 3. This resulted in 10 mutant strains, L1-TyrB1*, containing the TyrB mutation sites listed in Table 7 (see Table 1 for details).

[0046] 3. Screening test for tyrB1 mutant L1-TyrB1 strain was selected as the control strain, and 10 mutant strains were selected as experimental strains. Shake flask fermentation tests were conducted according to the aerobic-microaerobic two-stage fermentation method described in Example 2.

[0047] The results are as follows Figure 3 As shown, the mutant strain L1-TyrB1 A169D L1-TyrB1 H233R L1-TyrB1 H234A L1-TyrB1 L236A The L-leucine production of all strains was increased compared to the control strain, with the mutant strain L1-TyrB1 showing the highest yield. H234A The L-leucine production was the highest, reaching 1.55 times that of the unmutated control strain L1-TyrB1. The mutant strain L1-TyrB1... H234A It was named L2 and subsequently tested for fermentation.

[0048] Example 6: Fermenter Test of Engineered Strains To further investigate the fermentation performance of the engineered strain, strain L2 was tested in a 7 L fermenter.

[0049] 1. Culture medium preparation: Seed culture medium (L): glucose 30 g, corn steep liquor 8 mL, yeast powder 5 g, peptone 2 g, (NH4)2SO4 2 g, KH2PO4 1.5 g, K2HPO4 4 g, sodium citrate 10 g, MgSO4·7H2O 2 g, MnSO4·H2O 10 mg, FeSO4·7H2O 10 mg, biotin 0.2 mg, vitamin B1 2 mg, pH 7.2.

[0050] Fermentation medium (L): glucose 30 g, yeast extract 2.5 g, L-glutamic acid 0.5 g, L-valine 0.5 g, L-isoleucine 0.5 g, (NH4)2SO4 2 g, K2HPO4 7 g, sodium citrate 2 g, MgSO4·7H2O 2 g, MnSO4·H2O 20 mg, FeSO4·7H2O 20 mg, ZnSO4·7H2O 1 mg, CuSO4 0.2 mg, NiCl2·6H2O 0.02 mg, biotin 1 mg, vitamin B1 1 mg, vitamin B3 1 mg, vitamin B5 1 mg, vitamin B6 1 mg, pH 7.2.

[0051] 2. Fermentation culture in a fermenter: (1) Streak L2 onto LB solid medium and incubate overnight at 37°C.

[0052] (2) Inoculate a single colony with good growth into 20 mL of LB liquid medium for activation and incubate at 37℃ for 6-8 h.

[0053] (3) Transfer the activated bacterial solution to a seed tank containing 2 L of seed culture medium at an inoculation rate of 5-10‰. Control the temperature at 37℃, automatically add ammonia water to stabilize the pH at 7.0-7.2, control the dissolved oxygen at 25-45%, control the residual sugar at 1-3%, and culture for 12-20 h.

[0054] (4) Transfer all the seed culture medium to a 7 L fermenter containing 2.5 L of fermentation medium, control the temperature at 37℃, automatically add ammonia water to stabilize the pH at 7.0-7.2, maintain dissolved oxygen at 30%, and culture for 16-20 h.

[0055] (5) OD 600 Once the dissolved oxygen level reaches 20, it is reduced to 0.5% to allow for microaerobic fermentation, with a fermentation cycle of 40 hours.

[0056] The results from the fermentation tank showed that ( Figure 4 The strain was cultured in a 7 L fermenter in an aerobic-microaerobic dual-stage fermentation for 40 h, and the L-leucine yield reached 90.03 g / L, the sugar-acid conversion rate was 39.1%, and L-valine was not detected.

[0057] Table 8 Fermentation performance of L2 Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. An aminotransferase mutant, characterized in that, It is obtained by mutation of the amino acid sequence shown in SEQ ID NO:1, selected from the following sites or combinations thereof: 1) The 169th amino acid is mutated from A to D; 2) The 233rd amino acid is mutated from H to R; 3) The 234th amino acid is mutated from H to A; 4) The 236th amino acid is mutated from L to A.

2. A nucleic acid molecule encoding the mutant of claim 1.

3. A biomaterial containing the nucleic acid molecule of claim 2, characterized in that, The biological material is recombinant DNA, expression cassette, transposon, plasmid vector, or viral vector.

4. Recombinant microorganisms, characterized in that, The recombinant microorganism comprises the aminotransferase mutant of claim 1 or the nucleic acid molecule of claim 2.

5. Use of the aminotransferase mutant of claim 1, the nucleic acid molecule of claim 2, the biomaterial of claim 3, or the recombinant microorganism of claim 4 in the production of L-leucine.

6. A method for producing L-leucine, characterized in that, The aminotransferase mutant as described in claim 1, the nucleic acid molecule as described in claim 2, the biomaterial as described in claim 3, or the recombinant microorganism as described in claim 4 may be used.

Images