Biosynthesis of (R)-(-)-1-Octen-3-ol in Recombinant Saccharomyces cerevisiae with Lipoxygenase-1 and Hydroperoxide Lyase Genes from Tricholoma matsutake

Tricholoma matsutake is an ectomycorrhizal fungus, related with the host of Pinus densiflora. Most of studies on T. matsutake have focused on mycelial growth, genes and genomics, phylogenetics, symbiosis, and immune activity of this strain. T. matsutake is known for its unique fragrance in Eastern Asia. The most major component of its scent is (R)-(-)-1-octen-3-ol and is biosynthesized from the substrate linoleic acid by the sequential reaction of lipoxygenase and peroxide lyase. Here, we report for the first time the biosynthesis of (R)-(-)- 1-octen-3-ol of T. matsutake using the yeast Saccharomyces cerevisiae as a host. In this study, cDNA genes correlated with these reactions were cloned from T. matsutake, and expression studies of theses genes were carried out in the yeast Saccharomyces cerevisiae. The product of these genes expression study was carried out with Western blotting. The biosynthesis of (R)-(-)- 1-octen-3-ol of T. matsutake in recombinant Saccharomyces cerevisiae was subsequently identified with GC-MS chromatography analysis. The biosynthesis of (R)-(-)-1-octen-3-ol with S. cerevisiae represents a significant step forward.


Introduction
Tricholoma matsutake, found in the Korean peninsula, Japan and China, have been regarded as precious food for a long time. Nowadays, it is famous symbol viand in Korea and Japan. T. matsutake is high-value food because not only its taste and smell but also its notoriously difficulty of artificial cultivation. The fruiting body of T. matsutake is formed by the complex interaction of several ecological factors in specific forest of Pinus densiflora [11,21]. Unfortunately, the occurrence of T. matsutake fruiting body in nature is steadily decreasing and the demand for the artificial cultivation of T. matsutake is rising. In 1983, Hiroshima Forestry Examination Center had succeeded in the artificial cultivation but failed to preserving this cultivating artifact. Since 1983, lots of studies have been reported and recently the National Forest Research Institute, Korea and Taki Chemical Industry, Japan had succeeded the artificial cultivation of T. matsutake in 2010 and 2018, respectively [17,20]. But most of studies had failed to maintain the cultivation and little is known about its artificial cultivation [23,31]. Moreover, transplantation of P. densiflora trees infected with T. matsutake attempted by Japan and Korea had failed to yield the mushroom with a meaningful number [12,16]. The genomic study of T. matsutake was carried out by Dr Min's research group in January 2020 [28]. Thus, in this study, not artificial cultivation but rather production of the T. matsutake flavor is the focus of this study. We expect the outcome of our work to help producing the (R)-(-)-1-octen-3-ol at a low price and at a significant yield by cloning the genes encoding the lipoxygenase and hydroperoxide lyase of T. matsutake into industrial microorganisms such as Escherichia coli and Saccharomyces cerevisiae.
E. coli is a gram negative, facultative anaerobic, coliform bacterium that commonly found in the lower intestine of warm-blooded organisms. As harmless yeast, its benefits hosts by producing nutrient and preventing invasion of pathogenic bacteria. Since 1885, E.coli has been used for experiments because it could be grown and cultured easily and inexpensively in a laboratory setting [13,36]. S. cerevisiae has been known as desirable microorganism closely related with humans because of participation in producing the fermented food like beer, wine and bread. The genetic information of S. cerevisiae is well known and the yeast can easily be engineered, which has been an attractive microorganism in industrial biotechnology [6,35]. In addition with transgenic S. cerevisiae, successful researches have been reported for the production of high value materials such as fuel and chemical [18,24,29]. Therefore, the biosynthesis of (R)-(-)-1-octen-3-ol using engineered S. cerevisiae to manufacturing the various T. matsutake flavor products has a considerable commercial value. In this study, gene sets introduced into S. cerevisiae for synthesizing the highest amount of (R)-(-)-1-octen-3-ol were identified among lipoxygenase-1, lipoxygenase-2, lipoxygenase-3 and hydroperoxide lyase genes of T. matsutake. Furthermore, the optimal biosynthesis conditions of (R)-(-)-1-octen-3-ol in transgenic yeast were studied to effectively generate and extract (R)-(-)-1-octen-3-ol.

Materials and Methods
Cloning the Genes of Lipoxygenase and Hydroperoxide Lyase from Tricholoma matsutake For producing a yeast expression vectors, the recombinant plasmid was prepared and whole RNA extract from T. matsutake fruiting bodies was isolated using liquid nitrogen and TRIzo1 reagent. Then cDNA was synthesized from whole RNA by Accuscript High Fidelity 1  USA). The lipoxygenase and hydroperoxide lyase genes were amplified from the synthesized cDNA by RT-PCR using PrimeSTAR HS Polymerase (TaKaRa, Japan). The primer sequences and restriction enzymes used for each gene are described in Table 1. PCR reaction conditions for each gene were set as follows: pre-denaturation at 98 Because the PrimeSTAR HS Polymerase (TaKaRa) has a 3' to 5' exonuclease activity, A-tailing of the amplified PCR product was conducted with the TA-cloning Reagent Set for PrimeSTAR. The A-tailed PCR products were cloned into pGEM Easy T-vectors (Promega, USA) and ligated vectors of each gene were transformed into E. coli DH5α competent cells (TaKaRa). After the transformation reaction, the E. coli were incubated overnight in a Luria Broth (LB) agar containing ampicillin, isopropyl β-D-1-thiogalactopyranoside (IPTG) and X-gal at 37 o C. From the incubated E. coli, plasmids were extracted by Higene Plasmid Mini Prep Kit (BioFact, Korea). Then agarose gel electrophoresis was conducted to confirm the insertion of the target genes. By using ABI BigDye Terminator v3.1 Cycle Sequencing Kits (Applied Biosystems, USA) and ABI 3730xl DNA Analyzer (Applied Biosystems), DNA sequencing of cloned vectors were conducted.

Yeast Transformation
With C-terminal peptide that encoding a poly-histidine tag and V5 epitope to detect the recombinant protein, pYES3/CT and pYES2/CT vectors were used for expressing target genes in yeast.
The T-vectors inserted with lipoxygenase-1, lipoxygenase-2, lipoxygenase-3 and hydroperoxide lyase genes, which have 100 percentage of homology by comparison with cDNA sequences of T. matsutake, were selected by DNA sequencing analysis of the cloned vectors. Using specific restriction enzymes, each gene of the T-vectors was cut for ligation to yeast expression vectors. Using a S.c EasyComp Transformation Kit (Invitrogen, USA), each cloned vector was transformed into S. cerevisiae. The transformed cells were incubated at 30 o C for 2~3 days in SC minimal medium (0.67% yeast nitrogen base, 0.192% yeast synthetic drop-out medium supplement, 2% glucose and 2% agar) in separate situation because of the different auxotrophic markers in each expression vector for the selection of yeast transformants: without tryptophan for pYES3/CT transformants and without uracil for pYES2/CT transformants. Each cloned gene was detected by a colony PCR method.

Expression and Detection of the Recombinant Protein
The pYES3/CT or pYES2/CT vectors have only a GAL1 promoter. Thus, the transcription of each gene could be induced by adding galactose as the carbon source. The transformants were pre-cultivated overnight at 180 rpm in the appropriate SC selectable medium with 2% raffinose as a carbon source at 30 o C. Pre-cultivated cells were inoculated into 100 ml of fresh SC induction medium with 2% galactose for carbon source and incubated at 30 o C for 8 h at 180 rpm. The induced cells were harvested and disrupted with bead beater that consist of sodium phosphate lysis buffer (50 mM sodium phosphate, 1 mM Phenylmethanesulfonyl fluoride (PMSF), 5% glycerol and 2% Triton X-100; pH 6.5) and acid-washed glass beads (0.4-0.6 mm size). Then, the crude protein supernatant was obtained by centrifugatinn at 4 o C for 20 min and the proteins were quantified with a Pierce BCA Protein Assay Kit (Invitrogen).
Using western blot analysis, the recombinant proteins were detected and the protein samples were separated by SDS-PAGE with a 10% polyacrylamide gel at 100 V and transferred to nitrocellulose membranes for 2 h at 50 V. The membranes were blocked with 5% skim milk in 1× TBS buffer containing 0.05% Tween-20 for 90 min at room temperature. To detect the recombinant proteins in membranes, reaction with a 1:5,000 dilution of Anti-V5 Mouse monoclonal antibody (Invitrogen) and a 1:1,000 dilution of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (Invitrogen) as the loading control antibody was incubated at 4 o C overnight. After the membranes were reacted with a 1:50,000 dilution of Rabbit Anti-Mouse polyclonal secondary antibody (Abcam Inc., USA) at room temperature for 90 min. In this process, signals of the target proteins were detected using the Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare Life Sciences, USA).

Analysis of (R)-(-)-1-Octen-3-ol Production based on Lipoxygenase and Hydroperoxide Lyase Combination by Gas Chromatography-Mass Spectrometry
To determine which gene set is capable for synthesizing the highest amount of (R)-(-)-1-octen-3-ol, various combinations of gene encoding proteins were reacted with the substrate, linoleic acid. Before inoculation into 100 ml of fresh SC induction medium and incubation at 30

Co-Transformation and Protein Expression
In order to biosynthesize the (R)-(-)-1-octen-3-ol in yeast, the lipoxygenase-1 and hydroperoxide lyase genes were introduced into S. cerevisiae. The expression vectors lipoxygenase-1/pYES3 and hydroperoxide lyase/pYES2 were mixed in a 1:1 ratio and transformed into S. cerevisiae by S.c EasyComp Transformation Kit (Invitrogen). The co-transformants were spread on SC minimal medium without tryptophan and uracil, and incubated at 30 o C for 2~3 days. After incubation, the presence of the two genes from the co-transformants was determined by colony PCR.
In order to determine the expression level of the lipoxygenase-1 and hydroperoxide lyase proteins based on incubation time, cotransformants were pre-cultivated overnight at 30 o C in SC selectable medium without tryptophan and uracil. The pre-cultivated cotransformants were incubated in fresh SC induction medium without tryptophan and uracil at 30 o C. Then induced cell isolates were collected separately at 0, 8, 16, 28 and 32 h in incubation. After crude proteins were extracted by bead beater with sodium phosphate lysis buffer from the induced cells, they were quantified with a Pierce BCA Protein Assay Kit (Invitrogen). Quantified samples were separated by SDS-PAGE in a 4~20% gradient polyacrylamide gel at 100 V and transferred to nitrocellulose membranes for 2 h at 50 V. The membranes were blocked with 5% skim milk in 1× TBS buffer containing 0.05% Tween-20 for 90 min at room temperature. To detect the recombinant proteins on the membranes, reaction with a 1:5,000 dilution of Anti-V5 Mouse monoclonal antibody (Invitrogen) and a 1:1,000 dilution of GAPDH antibody (Invitrogen) as the loading control antibody was incubated at 4 o C overnight. Next, the membranes were reacted with a 1:50,000 dilution of Rabbit Anti-Mouse polyclonal secondary antibody (Abcam Inc., USA) at room temperature for 90 min. In this process, the signals of the target proteins were detected by using the Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare Life Sciences).
In order to identify the (R)-(-)-1-octen-3-ol biosynthesis from the yeast with lipoxygense-1 and hydroperoxide lyase genes, the co-transformants were incubated in the presence of the linoleic acid, as a substrate. The co-transformants used for identifying the (R)-(-)-1-octen-3-ol biosynthesis were pre-cultivated overnight in SC selectable medium without tryptophan and uracil at 30 o C. After overnight cultivation, the cells were inoculated into 100 ml of SC induction medium containing 1.5 mM linoleic acid and 0.2% Tween-20, followed by an incubation at 30 o C for 20 h. Next, the co-transformants in the presence of the linoleic acid, cells and medium were separated by centrifugation at 1,500 rpm for 10 min. The harvested cells were disrupted by the bead beater with sodium phosphate lysis buffer (50 mM sodium phosphate, 1 mM PMSF, 5% glycerol and 2% Triton X-100; pH 6.

Analysis of Optimal Reaction Condition for (R)-(-)-1-octen-3-ol Biosynthesis
In order to optimize the (R)-(-)-1-octen-3-ol biosynthesis in yeast with the lipoxygense-1 and hydroperoxide lyase genes, the co-transformants were incubated under different conditions. The co-transformants for incubation were pre-cultivated overnight in SC selectable medium without tryptophan and uracil at 30 o C. Then, pre-cultured cells were collected by centrifugation at 1,500 rpm for 10 min and suspended to an optical density at 600 nm (OD

Results
Cloning the Genes Encoding Lipoxygenase-1, Lipoxygenase-2, Lipoxygenase-3 and Hydroperoxide Lyase pGEM Easy T-vectors (Promega), pYES3/CT and pYES2/ CT vectors (Invitrogen) were used for cloning each gene. DNA sequencing of each of the genes inserted into the vectors confirmed a perfect 100% homology compared with the cDNA sequences of T. matsutake analyzed. The lengths of cDNA of lipoxygenase-1, -2, -3 and hydroperoxide lyase were 3,159 bases, 3,330 bases, 3,855 bases and 1,560 bases, respectively. These translates to protein sequences of 1,052 amino acids, 1,110 amino acids, 1,284 amino acids and 519 amino acids, respectively. Furthermore, molecular weights of recombinant proteins are expected to be approximately 122, 130, 150, and 63 kDa, respectively.
For expression of each gene in yeast, the lipoxygenase-1, -2, -3 genes were inserted in the pYES3/CT vector and the hydroperoxide lyase gene was inserted in the pYES2/CT, followed by transformation of recombinant DNAs into S. cerevisiae. The bands of each amplified gene by colony PCR were confirmed to match to the lenght of each gene resolved by electrophoresis on an agarose gel.

Expression of the Recombinant Protein from Transformants with Each Gene
To analyze the expression of recombinant proteins, transformants that include the genes of lipoxygenase-1, lipoxygenase-2, lipoxygenase-3 and hydroperoxide lyase were incubated respectively in the appropriate SC induction medium with 2% galactose at 30 o C for 8 h. After incubation, the induced proteins were analyzed by western blot with an Anti-V5 Mouse monoclonal antibody. The signal of the recombinant proteins encoded by each gene were confirmed to match to the expected molecular weight and lipoxygenase-2 showed the highest level while lipoxygenase-3 showed the lowest level ( Fig. 2A).

Determination of the Gene Set Capable of Synthesizing the Highest amount of (R)-(-)-1-Octen-3-ol
To analyze the production of (R)-(-)-1-octen-3-ol based on protein combinations that react with linoleic acid, various combinations of protein extractions were mixed overnight with 1.5 mM linoleic acid at 4 o C in a total reaction volume of 3 ml. Then the reactants were analyzed using gas chromatography-mass spectrometry. The results showed that lipoxygenase-1, -2, -3 and hydroperoxide lyase expressed in transformants had specific functions and activity. Furthermore, a distinct difference was observed in the amount of (R)-(-)-1-octen-3-ol produced from all protein combinations that reacted with linoleic acid ( Table 2). Among the results of (R)-(-)-1-octen-3-ol concentration, the set of lipoxygenase-1 and hydroperoxide lyase showed the highest efficiency in generating (R)-(-)-1-octen-3-ol.

Expression of the Recombinant Proteins from the Transformants with Lipoxygenase-1 and Hydroperoxide Lyase Genes
In order to biosynthesize the (R)-(-)-1-octen-3-ol in yeast, lipoxygenase-1 and hydroperoxide lyase genes were co-transformed into S. cerevisiae competent cells and the expression of recombinant proteins were analyzed by western blot. With the colony PCR, target genes from the co-transformant were amplified and analyzed by electrophoresis with 0.7% agarose gel. As a result, the lipoxygenase-1 and hydroperoxide lyase genes from cotransformants were confirmed to correspond to the size of each gene (Fig. 2B). The co-transformants were incubated in SC induction medium without tryptophan and uracil containing 2% galactose at 30 o C for 0, 8, 16, 28, and 32 h. Then the induced proteins were analyzed by western blot and the signal of recombinant proteins encoded by each gene was confirmed to correspond to the expected molecular weight. Lipoxygenase-1 and hydroperoxide lyase from the co-transformants were expressed at higher levels in 16 h, were expressed at higher levels than the other incubation times and the signals of lipoxygenase-1 and hydroperoxide lyase decreased after 28 h of incubation (Fig. 2C).
1-octen-3-ol was detected at retention time of 38.27 min in cell lysates and medium incubated with linoleic acid and the detection showed the peak consistent with the mass
spectrum of standard 1-octen-3-ol. The peak comprised 16% of the total area in cell lysates incubated with linoleic acid and 0.8% of the total area in the supernatant incubated with linoleic acid. Therefore, it was identified that transformants with the lipoxygense-1 and hydroperoxide lyase genes are able to convert from linoleic acid to (R)-(-)-1-octen-3-ol. Biosynthesis of (R)-(-)-1-octen-3-ol from cells showed more efficiency than from the supernatant (Fig. 3).
The optimal condition for (R)-(-)-1-octen-3-ol biosynthesis in transgenic yeast was determined to be at a substrate concentration of 3 mM linoleic acid, incubation temperature of 30 o C and an incubation time of 24 h.

Discussion
Tricholoma matsutake has a unique flavor and excellent taste and is treated as a high-value food. However, its artificial cultivation is unfortunately hardly possible. Since Hiroshima Forestry Examination Center had artificially cultivated T. matsutake by infection in 1983, over 101 studies had been published but none of them have been proven successful in its artificial cultivation. Recently, the National Forest Research Institute, Korea and Taki Chemical Industry, Japan have somewhat succeeded in the artificial cultivation of T. matsutake in 2010 and 2018, respectively [17,20]. Although considerable progressed have been achieved, there are still lots of problems to practically and efficiently cultivate artificially T. matsutake [23].
Rather than cultivating the whole mycelium of T. matsutake, synthesis of specific extract specially the flavor has been another approach to the problem. Many studies on the flavor of T. matsutake have been reported and in 2019, study about flavor fingerprint of T. matsutake has been published by National Engineering Research Center of Seafood, China [22]. But in their investigation, components of 3-octanone, 3-octano, 1-octen-3-one, 1-octanol, methanol and 1-pentanol became subject without analysis of gene sequencing information.
Here, we use another approach that centers on the genes responsible for the biosynthesis of 1-octen-3-ol. 1-octen-3-ol is a major flavor component among the none flavor components of T. matsutake [25]. Although 1-octen-3-ol is synthesized in other mushrooms or plants, synthesis of 1octen-3-ol in T. matsutake showed a better efficiency [26,32]. As lipoxygenase and hydroperoxide lyase participate in synthesizing 1-octen-3-ol, those two genes were selected for recombinant. Research of lipoxygenase and hydroperoxide lyase had been published. Dr Ding's research group analyzed the protein sequence of hydroperoxide lyase from tobacco with 497 amino acids in 2019 [7]. Also Dr Tiwari's research group analyzed the protein sequence of lipoxygenase from Eleusine coracana with 887 amino acids in 2016 [37]. In our study, the cDNA of lipoxygenase-1, -2, -3 and hydroperoxide lyase were translated to protein sequences of 1,052 amino acids, 1,110 amino acids, 1,284 amino acids and 519 amino acids, respectively. Also E. coli and S. cerevisiae were used for stable and efficient expression of recombinant protein from transformants with lipoxygenase and hydroperoxide lyase. With gene sequencing information about lipoxygenase and hydroperoxide lyase, lipoxygenase-1, lipoxygenase-2, lipoxygenase-3 and hydroperoxide lyase genes were synthesized from cDNA of T. matsutake and these genes were cloned into yeast expression vectors. The vectors from E. coli were then expressed successfully by the S. cerevisiae.
To investigate the efficiency of recombinant protein expression, various combinations of proteins encoded by each gene was mixed overnight with 1.5 mM linoleic acid at 4 o C in a total reaction volume of 3 ml. The reactants were then analyzed using gas chromatography-mass spectrometry. Lipoxygenase-1, -2, -3 and hydroperoxide lyase gene expressed in the transformants showed specific functions and activity. In analysis of the protein activity and stability of genes in yeast, the combination of lipoxygenase-1 and hydroperoxide lyase showed the highest efficiency in generating (R)-(-)-1-octen-3-ol.
Taken all together, introducing the set of lipoxygenase-1 and hydroperoxide genes into S. cerevisiae showed so far the best path to biosynthesize the flavor of T. matsutake in yeast. In order to achieve the highest efficiency of synthesis of lipoxygenase-1 and hydroperoxide lyase, co-transformants were incubated in SC induction medium without tryptophan and uracil containing 2% galactose at 30 o C for several hours. As results, expression of lipoxygenase-1 and hydroperoxide lyase was higher at 16 h than any other incubation time. The signals of lipoxygenase-1 and hydroperoxide lyase markedly decreased after 16 h. Having identified the best settings for recombinant expression of lipoxygenase-1 and hydroperoxide allow peak biosynthesis production of (R)-(-)-1-octen-3-ol. The analyzing (R)-(-)-1-octen-3-ol biosynthesis was proceeded according to whether incubation of cell lysates of supernatant and linoleic acid. By the peak of 1octen-3-ol, the result was analyzed using gas chromatographymass spectrometry. Without linoleic acid, (R)-(-)-1-octen-3ol was hardly detected in both incubation of cell lysates and supernatant. In addition of linoleic acid, (R)-(-)-1octen-3-ol was detected at a retention time of 38.27 min but biosynthesizing (R)-(-)-1-octen-3-ol in cell lysates was better than in the supernatant. It is obvious that incubation in cell lysates with linoleic acid is the most efficient way for (R)-(-)-1-octen-3-ol biosynthesis. Also, to determine optimum condition for (R)-(-)-1-octen-3-ol biosynthesis, sets of linoleic acid concentration, incubation time and incubation temperature were analyzed separately. The yield of (R)-(-)-1-octen-3-ol biosynthesis was optimized with 3 mM of linoleic acid, 24 h of incubation time and 30 o C of incubation temperature.
In summary, in this manuscript we investigated the optimal conditions for (R)-(-)-1-octen-3-ol biosynthesis. We report that the synthesis with the highest yield was achieved with the set of lipoxygenase-1 and hydroperoxide lyase contained in the cell lysate, 3 mM of linoleic acid, 24 h of incubation time and 30 o C of incubation temperature. Large scale production of (R)-(-)-1-octen-3-ol is needed but require additional studies in order to efficiently scale up the process.