Results

Cloning and bioinformatics analysis of EGT synthase gene

Cloning and homology analysis of EGT synthase gene

Based on the amino acid sequences of NcEgt2 and GfEgt2from N. crassa and G. frondosa , three hypothetical proteins with certain similarity were obtained by Protein Blast in the protein database of P. eryngii . The specific protein information is as follows: PLP-dependent transferase [P. eryngii ](KAF9497900.1), PLP-dependent transferase [P. eryngii ](KAF9497887.1), PLP-dependent transferase [P. eryngii ](KAF9493115.1). The corresponding genes were named asPeEgt2a , PeEgt2b and PeEgt2c , which were 1290 bp, 1290 bp and 1245 bp, respectively. The coverage/similarity ofNcEgt2 amino acid sequence alignment with N. crassa was 94%/36.95%, 93%/37.78% and 96%/36.64%, respectively. The coverage/similarity of GfEgt2 amino acid sequence alignment withG. frondosa was 94%/53.65%, 94%/50.99% and 93%/50.56%, respectively. Our previous studies have mined and clonedPeEgt1 [30] (MAZY01000061.1) in the EGT synthesis pathway of P. eryngii .
Because P. eryngii is a eukaryote, most of the DNA sequence corresponding to its expressed proteins have introns. When heterologously expressed in other model species, whole exon genes should be used. Therefore, it is necessary to extract the total RNA of P. eryngii (Supplementary data: Fig. S1). DNase was not used to process the sample during RNA extraction, so there will be a shallow band above the gel electrophoresis picture for genomic DNA. In the subsequent reverse transcription process, gDNA Remover will be added to remove genomic DNA. Then, the cDNA was obtained by reverse transcription with total RNA as template, and then the potential EGT synthase genes were amplified with cDNA as a template. The results are shown in Fig. S1, and the bands ofPeEgt2a , PeEgt2b and PeEgt2c were about 1300 bp, which were consistent with the expected band size. The target genes were recovered by cutting gel.
After adding A to the end of the recovered products, they were connected to the T vector using the pMD19-T cloning vector kit, and then transformed into E. coli DH5α. The transformants were selected and identified by colony PCR using the universal primers M13-47/RV-M. The correct transformants were cultured and the plasmids were extracted for sequencing. The sequencing results showed that the lengths ofPeEgt2a and PeEgt2b are 1290 bp, and the length ofPeEgt2c is 1245 bp, both of which are complete CDS.
In P. eryngii , the nucleotide similarity of PeEgt2a ,PeEgt2b and PeEgt2c was 72.57 %, and the amino acid similarity was 69.86 %, as shown in Fig. S2 and Fig. S3. The nucleic acid sequence and amino acid sequence of Egt1 gene of P. eryngii and Pleurotus ostreatus and Pleurotus nebrodensiswere highly homologous. The nucleic acid sequence homology was 97.03 %, and the amino acid sequence homology was as high as 97.93 %[30].

Physicochemical properties analysis and structure prediction of EGT synthase

The physicochemical properties were analyzed by ExPASy ProtParam as follows: PeEgt2a , PeEgt2b and PeEgt2c were composed of 429, 429 and 414 amino acids, respectively, and the molecular weights were 47.99, 47.55 and 46.12 kDa, respectively. The isoelectric points were predicted to be 6.6, 6.53 and 6.06, respectively. The total negatively charged residues (Asp + Glu) were 44, 41 and 46, respectively, which were higher than the total positively charged residues (41, 38 and 38, respectively). The predicted instability indexes were 33.08, 39.48 and 40.25, respectively, indicating thatPeEgt2a was relatively stable. In yeast, the estimated half-life was greater than 20 h. The total fat indexes were 89.77, 90.44 and 85.05, respectively. The average values of hydrophilicity were -0.179, -0.09 and -0.138, respectively. The hydrophilicity of the protein was analyzed by ExPASy ProtScale. PeEgt2a , PeEgt2b , andPeEgt2c were the most hydrophobic between the 150-200 amino acids, located at 181 and 182, 181, 161, respectively. As shown in Fig. S4, the greater the ordinate positive value, the stronger the hydrophobicity; the smaller the negative value, the stronger the hydrophilicity, and the hydrophilic region in the figure is larger than the hydrophobic region. According to the predicted average hydrophilicity, it can be seen that PeEgt2 belongs to hydrophilic protein.
PredictProtein was used to predict the secondary structure of EGT synthases, as shown in Fig. S5. PeEgt2a , PeEgt2b andPeEgt2c were mainly random coil, accounting for 48.72%, 53.15% and 48.79% respectively. Followed by α-helix, accounting for 37.76%, 34.27% and 37.44% respectively. The β-sheet was the least, accounting for 13.52%, 12.59%, and 13.77%, respectively.
Through CD-SEARCH prediction analysis of domains and superfamilies,PeEgt2a , PeEgt2b , and PeEgt2c all contain CsdA domains and belong to CsdA superfamily members. They are described as selenocysteine lyases or cysteine desulfurases that mainly play a role in amino acid transport and metabolism in cells. Using SWISS-MODEL, the protein tertiary structure was constructed by searching homologous templates, as shown in Fig. S6.

Expression of EGT synthase in S. cerevisiae

Construction of Egt1 single-gene engineered strains

The integration site of PeEgt1 was selected as XI-5, and the corresponding knockout plasmid was pQc009. After extracting IMX581 genome and plasmid pQc009, the product was tested by 1% agarose gel electrophoresis, as shown in Figure 1. There is a band at the top of the lane 1 and 2, which is the genome of S. cerevisiae , and the bottom diffuse band is degraded RNA. Using S. cerevisiae genomes as template, the amplification results showed that upstream homologous arm Ⅺ-5-us was 663 bp, promoter TEF1p was 412 bp and downstream homologous arm Ⅺ-5-ds was 468 bp; using pYES2 as template, the amplification result showed that terminator CYC1t was 248 bp; the length of PeEgt1 was about 2600 bp, which was amplified by primers plus their homologous arms. As shown in Figure 1, the PCR products corresponded with the expected band sizes. After the above bands were cut and recovered, the five fragments were assembled into the homologous repair fragment XI-5-us-TEF1p-PeEgt1 -CYC1t-XI-5-ds of the PeEgt1 expression frame by overlap extension PCR, as shown in Figure 1, the length is about 4500 bp. There are many miscellaneous bands below the target band, because the annealing temperature is uncertain during the overlap extension, so the method of adding 0.5°C per cycle is adopted. There may be some base pairing between the five short segments. The high temperature leads to duplex unwinding, and the low temperatures, these complementary fragments may be mismatched; in addition, the primers may also have individual bases combined with the template, resulting in base mismatch, so it is necessary to cut the gel to recover the purified product.
The knockout plasmid pQC009 and homologous repair fragment XI-5-us-TEF1p-PeEgt1 -CYC1t-XI-5-ds were co-transformed into S.cerevisiae competent cells. After resuscitation, they were evenly coated on Sc-Ura auxotrophic medium for screening. The results are shown in Figure 2. Multiple single colonies grow on the transformation plate, and the negative control did not grow. Because the transformed strain contained the plasmid pQC009 and the expression frame of Ura3, it could grow on the Sc-Ura plate, while the control strain did not contain the knockout plasmid, it was still a Ura auxotrophic strain and could not grow on Sc-Ura auxotrophic plates. Using homologous arm primers on the genome, single colonies were selected for PCR identification to ensure successful transformation of the engineered strain.
The protein of the engineered strain IMX581-PeEgt1 and the original strain IMX581 were extracted respectively. The results of SDS-PAGE were shown in Figure 2. The protein size of PeEgt1 was 96.8 kDa [30]. There was no significant difference in the expression protein between the engineered strain and the original strain, which may be due to the low expression of the protein. It is necessary to verify the activity of PeEgt1 by detecting EGT by HPLC.
By HPLC detection, as shown in Figure 3, the peak of EGT standard substance appeared at 21.077 min. Control strain IMX581 did not produce EGT without the integration of EGT synthetase gene. The peak of IMX581-PeEgt1 indicated thatS. cerevisiae could synthesize EGT only by integratingPeEgt1 , which proved that PeEgt1 was active. After 7 days of fermentation, the EGT yield of IMX581-PeEgt1 was 2.63±0.05 mg/L.

Construction of double-gene engineered strains

The three potential PeEgt2were integrated into the IMX581-PeEgt1 which had lost the knockout plasmid to construct the double-gene engineered strains. X-2 of IMX581 was selected as the integration site of PeEgt2 gene expressing strain, and the corresponding knockout plasmid was pQc029. The homologous recombinant fragments were amplified, and the five fragments were assembled by overlap extension PCR. Three X-2-us-TDH3p-PeEgt2 -TEF1t-X-2-ds five-fragment homologous repair templates were constructed, respectively, as shown in Fig. S7.
The above homologous repair templates and the knockout plasmid pQc029 were co-transformed into the IMX581-PeEgt1 competent cells by electroporation. After resuscitation, they were coated on the Sc-Ura auxotrophic medium. The transformants were picked up and the integration of the two genes was verified by colony PCR with homologous arm primers. The successfully transformed double-gene engineered strains were IMX581-PeEgt1 -2a, IMX581-PeEgt1 -2b , IMX581-PeEgt1 -2c . After 7 days of fermentation, as shown in Fig. S8, the yield of EGT was increased to 4.04±0.23 mg/L and 4.30±0.23 mg/L by integratingPeEgt2a and PeEgt2b , respectively, which increased by 53.61% and 63.50%. After integrating PeEgt2c , the yield of EGT was 2.63±0.09 mg/L, which was not significantly different from that of IMX581-PeEgt1 . The results showed that PeEgt2c had no enzyme activity, while PeEgt2a and PeEgt2b had enzyme activity.
PeEgt2a , PeEgt2b andPeEgt2c were separately integrated into the S.cerevisiaeIMX581 to construct the IMX581-PeEgt2 single-gene engineered strains. EGT was not detected after fermentation, indicating that only the single-gene engineered strains containing PeEgt2 could not synthesize EGT.

Construction of multi-gene engineered strains

In fungi, the synthesis pathway of EGT only requires the participation of two genes Egt1 andEgt2 , but three potential PeEgt2 genes are obtained by bioinformatics mining in P. eryngii . Based on the above construction of double-gene engineered strains, we speculate thatPeEgt2a and PeEgt2b have enzyme activity, andPeEgt2 contains at least two isozymes. Therefore, we further constructed a triple-gene engineered strain to explore the significance of the presence of multiple PeEgt2 genes. The XI-3 was selected as the knockout site, and the corresponding knockout plasmid was pQc006.PeEgt2b was integrated into IMX581-PeEgt1 -2a, which had lost the knockout plasmid, to obtain triple-gene engineered strain IMX581-PeEgt1 -2a -2b . After 7 days of fermentation, the EGT yield was 4.45±0.15 mg/L, which was not statistically significant compared with the yield of double-gene engineered strains. A total of eight engineered strains were constructed, and their EGT production was summarized in Table 1.