Effects of The Constructed (+)-Limonene Synthesis Pathway in Sake Yeast on The Brewing Flavor of Sake

Article Information

Hao Xu, Nuo Xu, Yanhui Li, Jie Sun, Tingheng Zhu*, Shijin Wu*

College of Biotechnology and Bioengineering, Zhejiang University of Technology, Hangzhou 310014, Zhejiang Province, China

*Corresponding author: Tingheng Zhu, College of Biotechnology and Bioengineering, Zhejiang University of Technology, Hangzhou 310014, Zhejiang Province, China.

Shijin Wu, College of Biotechnology and Bioengineering, Zhejiang University of Technology, Hangzhou 310014, Zhejiang Province, China

Received: 17 May 2024; Accepted: 24 May 2024; Published: 21 June 2024

Citation: Hao Xu, Nuo Xu, Yanhui Li, Jie Sun, Tingheng Zhu, Shijin Wu. Effects of The Constructed (+)-Limonene Synthesis Pathway in Sake Yeast on The Brewing Flavor of Sake. Archives of Microbiology and Immunology. 8 (2024): 221-232.

View / Download Pdf Share at Facebook


Sake is a traditional alcoholic beverage in Japan, and the improvement of sake flavour and quality can be achieved by metabolic engineering modification of strains. Limonene, a terpenoid compound with an aromatic odour, has been detected in plant-based alcoholic products such as wine and brandy, and has flavour-modifying and bioactive effects. As limonene has not been investigated in sake, we used CRISPR/Cas9 technology to construct a limonene synthesis pathway in an industrial sake yeast strain and analysed the modification of sake flavour and composition. Eight strains with limonene yields ranging from 2 μg/L to 2.6 mg/L were obtained using the strategies of increasing the copy of the limonene expression cassette, weakening the competitive pathway, and enhancing the mevalonate (MVA) pathway. Comparative analyses of the sake brewing performance of the starting strain SY00 and the engineered bacteria with different limonene yields, as well as the composition of the sake samples, were carried out using gas chromatography-mass spectrometry (GC-MS) and other analytical methods. The result showed that the total amount of higher alcohols decreased, the phenylethanol content increased, and the flavour was significantly improved in the samples brewed with the limonene synthesising strains. Limonene was detected in the production of sake by two-phase fermentation of the engineered limonene producing strain SY08 (2.6 mg/L), but it was not detected in the final sake samples after fermentation, probably due to volatilisation. In conclusion, this study provides a new strategy for the construction of limonene producing strains by metabolic modification and its application to sake brewing.


sake; sake yeast; MVA pathway; limonene; bioactive molecules

sake articles; sake yeast articles; MVA pathway articles; limonene articles; bioactive molecules articles

Article Details

Graphical Abstartct


1. Introduction

Sake is a traditional alcoholic beverage in Japan with a history of more than 1,300 years [1, 2] and its production process is based on rice and water as the main raw materials, which are produced by the fermentation of Aspergillus oryzae and sake yeast [3, 4]. The brewing process can be divided into five steps, namely brewing, fermentation, filtration, bakelite and maturation, with fermentation being the key step. During this process, sake yeast converts glucose into ethanol under anaerobic conditions, producing a variety of flavour compounds such as higher alcohols, esters, acids and ketones, which give sake its complex flavour [4-7]. The flavour of sake is influenced by a variety of factors, including rice quality, degree of milling, water quality, brewing method and yeast strain used, with the degree of milling of the rice and the yeast strain selected having the most significant effect on the overall flavour of sake [1, 8, 9].

Yeast strains play a crucial role in flavour and other qualities in sake brewing, and improving flavour through strain modification is a common strategy. The improvement of sake yeast strains mainly focuses on increasing the efficiency of alcoholic fermentation, improving aroma and flavour, imparting new functional properties, reducing foam formation and adjusting the colour of sake [10, 11]. In the early days of sake brewing, new yeast strains were mainly obtained by tolerance screening, natural selection, mutagenesis and hybridisation [12-14], but with the development of gene editing technology, genetic engineering has also begun to be applied to sake yeast improvement [15]. Limonene is a plant terpene [16] with a citrus-like aroma and a variety of physiological activities, including anti-inflammatory, antibacterial, anticancer, antioxidant, anti-diabetic, hepatoprotective, and choleolytic [16-19], and is a GRAS food additive [20-22]. Due to its widespread occurrence in plants, limonene is frequently detected in some plant alcoholic products such as brandy and wine [23-26]. Therefore, limonene is a potential flavour modifier and bioactive molecule for alcohol. Studies have also been reported on the production of limone-containing alcoholic products by yeast modification, e.g. Hu et al [27] synthesised limonene by modifying Saccharomyces cerevisiae to synthesise limonene and brewed white wine based on health and flavour considerations and found that the construction of the limonene synthesis pathway had no significant effect on the fermentation characteristics of white wine. However, studies on limonene in alcoholic products are still relatively limited and have not been reported in sake.

In plants, limonene is produced from geranyl pyrophosphate (GPP) as a precursor in the mevalonate (MVA) pathway catalysed by limonene synthase [28], and the naturally occurring MVA pathway in S. cerevisiae facilitates the biosynthesis of limonene. However, the MVA pathway, as an important isoprenoid synthesis pathway, involves a variety of secondary metabolites, such as carotenoids, vitamin E, and steroids, which may potentially affect the aroma, flavour and colour of wine products. The aim of this study was to investigate the effect of the limonene synthesis pathway on the flavour of sake. We constructed the limonene synthesis pathway in the sake yeast strain SY00 to obtain strains with different limonene yields and analysed the effect of limonene synthesis on the improvement of sake flavour by comparing these constructed strains with the wild-type strains in terms of organoleptic evaluations, fermentation levels and changes in aroma compounds.

2. Materials and Methods

Strains, plasmids, media and chemicals

The plasmids and strains used in this study are listed in Table 1. Sake yeast (S. cerevisiae) SY00 was obtained from the strain conservation centre of Shaoxing Yuanzu Brewing Co. Yeast strains were cultured in YPD medium (yeast extract 10 g/L, peptone 20 g/L, glucose 20 g/L, pH natural) at 30°C and 180 rpm. Sake fermentation was carried out anaerobically at 30°C in rice semi-solid fermentation medium (rice (g): water (mL) = 1:2, saccharase: 2 g enzyme/1 kg rice). Escherichia coli DH5α was used for plasmid construction.

Table 1: Strains and plasmids

Strains and plasmids



Saccharomyces cerevisiae


For sake brewing

Our lab


SY00, Χ4:: (PPGK1-tLS- TCYC1)-(PTEF1-tCPT1- TCYC1)

This study


SY01, Χ3:: (PPGK1-tLS- TCYC1)-(PTEF1-tCPT1- TCYC1)

This study



This study




SY03, ΧI3:: PPGK1-tHMG1- TCYC1

This study



This study


SY05 ,Χ2:: (PTEF1-ERG20WW- TCYC1)- (PTEF1-ERG13- TCYC1)

This study


SY05, Χ2:: (PTEF1-ERG20WW- TCYC1)- (PTEF1-ERG13- TCYC1)-(PTEF1-ERG12- TCYC1)

This study


SY00, Χ2:: (PTEF1-ERG20WW- TCYC1)- (PTEF1-ERG13- TCYC1)-(PTEF1-ERG12- TCYC1)

This study

Escherichia coli


Φ80 lacZΔM15 ΔlacU169 recA1 endA1 hsdR17 supE44 thi-1 gyrA relA1

Our lab




This study


HygR;2μ ori with PTEF1,PT3,PAmpR;AmpR; PSNR52-sgRNA-TSUP4

Our lab

X4- gRNA

gRNA plasmid carrying X4 sgRNA

This study


gRNA plasmid carrying X3 sgRNA

This study


gRNA plasmid carrying XII2 sgRNA

This study


gRNA plasmid carrying XI3 sgRNA

This study

X2- gRNA

gRNA plasmid carrying X2 sgRNA

This study



Our lab



This study


pESC-HIS-(X2-ERG20WW), (PTEF1- ERG13 - TCYC1) , AmpR

This study


pESC-HIS-(X2-ERG20WW-ERG13), (PPGK1- ERG12 - TCYC1) , AmpR

This study

Method for preparing semi-solid fermentation medium for rice

Semi-solid fermentation medium for rice. a) Weigh the appropriate quantity of rice, wash with ultrapure water (UP water), wait until the amalgamated water is no longer cloudy, add the appropriate quantity of UP water and soak for 30 minutes, wrap in gauze to shake dry, then spread out and drain for 15 minutes; b) Place in a steamer for a total steaming time of 1 hour; c) Pour the stewed rice into triangular conical flasks containing 2 times the weight of the rice in sterile UP water. Depending on the quality of the rice before steaming, saccharase is added at a rate of 2 g of enzyme/1 kg of rice, bound with histopathic film and placed in a water bath at 60°C for 10 h for saccharification, shaking well every 2 h.When the saccharification was completed and cooled down to room temperature, it was obtained as a semi-solid fermentation medium of rice.

Plasmids construction

The primers used are listed in supplementary Table S1. The N-terminal signal peptides of the limonene synthase gene LS(GenBank ID: AF514287.1) from Citrus limon and the nerolidol pyrophosphate synthase gene CPT1(GenBank ID: NM001247704.1) from Solanum lycopersicum were codon optimised and synthesised (Tianlin Biotechnology,China) on pESC-HIS-(tLs-tCPT1) plasmid. Primers X2Z-F/-R, X2Y-F/R, TEF1-F/R, ERG20WW1-F/R, ERG20WW2-F/R, ERG20WW3-F/R were used to amplify the upstream and downstream homology arms of the X2 locus, the promoter TEF1 fragment and the farnesyl pyrophosphate synthase mutant (ERG20F96W-N127W,ERG20WW) fragment from the genome of strain SY00, respectively. The ERG20WW expression cassette with sequences upstream and downstream of the X2 locus was obtained by overlapping PCR and constructed as pESC-HIS-(X2-E20WW). The TEF1 gene promoter and the ERG13 gene fragment were obtained from the SY00 genome using primers TEF1-F/R and E13-F/R and constructed as plasmid pESC-HIS-(X2-E20WW-E13). The PGK1 gene promoter fragment was obtained from plasmid pESC-HIS-(tLs-tCPT1) using primer PGK1-F/R; the ERG12 gene fragment was obtained from the SY00 strain genome using primer E12-F/R and constructed as plasmid pESC-HIS-(X2-E20WW-E13-E12). gRNA plasmids were all constructed using reverse PCR, and the corresponding PAM sequences are shown in supplementary Table S2. The plasmids were constructed using a one-step cloning kit (Novozymes, China).

Strains construction

The limonene expression cassette (PPGK1-tLS- TCYC1)-(PTEF1-tCPT1- TCYC1) was amplified from the plasmid pESC-HIS-(tLs-tCPT1) with the primers TYLim-F/R. X4Lim, X3Lim and XII2Lim expression cassettes were constructed by overlapping PCR, and cassette homology arm sequences were obtained from the S cerevisiae genome using primers X4Z-F/-R, X4Y-F/R, X3Z-F/-R, X3Y-F/R, XII2Z-F/-R, XII2Y-F/R. Primers XI3Z-F/R, XI3Y-F/R, tHMG1-F/R, and CYC1-F/R were used to amplify the upstream and downstream homologous sequence fragments of the XI3 locus, the truncated HMG1 gene fragment, and the CYC1 terminator fragment, respectively, from the genome of strain SY00, and the overlapping PCR was used to obtain the XI3tHMG1 expression cassette with homologous sequences at the ΧI3 locus. The expression cassettes ERG20WW, ERG20WW-ERG13 and ERG20WW-ERG13-ERG12 with homologous sequence at the ΧI2 locus was obtained from plasmids pESC-HIS-(X2-ERG20WW), pESC-HIS-(X2-ERG20WW-ERG13) and pESC-HIS-(X2-ERG20WW-ERG13-ERG12) using primers X2Z-F and X2Y-R, respectively. The above gene expression cassettes were integrated into the corresponding loci on the chromosome to obtain a series of strains (Table 1) using the CRISPR/Cas9 system mediated by lithium acetate transformation. Transformants were initially screened using YPD medium containing G418(700 μg/mL) and hygromycin (650 μg/mL) and then confirmed by molecular identification.

Determination of limonene production in recombinant strains 

Shake flask fermentation was performed. Single colonies of yeast cells were collected in 5 mL YPD liquid test tubes for activation at 30°C, 180 rpm for 12 h. Yeast cells (1 × 107 cfu) of were transferred to 250 mL triangular conical flasks containing 50 mL of YPD liquid and 10% (v/v) dodecane was added as an organic extractant for a two-phase fermentation, at 30°C, 200 rpm, for 4 d.

Limonene Detection Method 

The fermentation broth was centrifuged at 8000 rpm to collect the upper extracted organic phase, and an appropriate amount of anhydrous sodium sulphate was added to remove the water, which was then filtered through a 0.22 μm membrane and analyzed by GC. GC conditions: Shimadzu GC-2010 PlusAF system equipped with an HP-5 column (30 m×0.25 mm×0.25 μm); the carrier gas was N2 at a flow rate of 3 mL/min, and the flow rate of hydrogen was 35 mL/min, the injection volume was 1 μL, the split ratio was 1:20, the gas chamber temperature was 250°C, and the tail-blow flow rate was 30 mL/min. The carrier gas was N2 at a flow rate of 3 mL/min, hydrogen at a flow rate of 35 mL/min, air at a flow rate of 350 mL/min, and a tail blow at a flow rate of 30 mL/min. The injection volume was 1 μL, the split ratio was 1:20, and the temperature of the gasification chamber was 250 °C, and the temperature of the detector was 250 °C. The initial temperature of the column was 100 °C, which was maintained for 1 min, and then it was increased to 110 °C at 10 °C/min, which was maintained for 3 min, then it was increased to 180 °C at 30 °C/min, which was maintained for 2 min. The column temperature was 100°C for 1 min, then increased to 110°C at 10°C/min for 3 min, and then increased to 180°C at 30°C/min for 2 min.

Sake fermentation 

Sake fermentation methods include simultaneous saccharification fermentation (SSF) and separate hydrolysis and fermentation (SHF) (29, 30). Simultaneous saccharification fermentation involves the production of amylase by fungi such as A. oryzae to hydrolyze starch into fermentable sugars, while yeast utilizes the fermentable sugars for alcoholic fermentation in an anaerobic environment. Separate hydrolysis and fermentation requires thorough saccharification and hydrolysis of the starch in the fermented raw material before the yeast is accessed for alcoholic fermentation to ensure optimal conditions for each step, and the SHF method was used in this study. A single colony of sake yeast was inoculated into a test tube containing 5 mL of YPD liquid for activation at 30°C, 180 rpm for 12 h. After activation, the cells were transferred to a 500 mL triangular conical flask containing 250 mL of YPD liquid at 30°C, 180 rpm for 24 h. The yeast was inoculated into the semi-solid fermentation medium with rice at 1×107 cfu/g rice, and a gas check valve was installed to control the gas flow in and out and to record the initial weight. The fermentation was carried out at 30°C and 180 rpm for 24 h to allow the yeast to rapidly consume oxygen and adapt to the fermentation environment. The fermentation was then transferred to a 30°C incubator and weighed every 24h and the weight loss was recorded as CO2 emission. In other parallel experiments, the incubator placed at 15~20°C.The CO2 emission was used to monitor the degree of fermentation. When the total amount of CO2 emission reached 30 g / 100 g rice (DW), the fermentation was considered to be at the end of the fermentation period and the residual glucose was measured. When the residual glucose was less than 25 g/l, the fermentation was stopped and the alcohol content was measured.

Glucose Test Methods 

Reducing sugars of the fermentation broth were determined using Ferring's reagent according to GBT13662-2018 (31).

Alcohol Testing  

Alcoholic strength of the fermentation broth was determined by the alcoholic gravity method according to GB/T 13662-2018 (31).

Sensory Evaluation of Sake 

The sensory evaluation of sake wine is performed in four aspects: color, aroma, taste and style. The temperature of the wine sample is controlled between 15°C and 25°C, and the clarity and colour of the wine are observed in a bright place. The glass is gently shaken to smell the aroma. A small amount of the wine sample is introduced, and the taste is waited until it is clear, then swallowed and the texture and aftertaste are tasted.

Sake Ingredient Testing 

The sample was distilled in a distillation flask, and the distillate was collected to 50 mL, transferred to a separating funnel and extracted with 100 mL, 50 mL and 30 mL of dichloromethane. The organic phase was collected and water was removed by adding an appropriate amount of anhydrous sodium sulfate, and then concentrated by spinning at 22 °C and 80 rpm to 0.5 mL, and analyzed by GC-MS (Agilent 8890/7250 GC/Q-TOF system). GC conditions: capillary column DB-5MS (30 m*0.25 mm*0.25 μm); inlet temperature of 280 °C; programmed temperature increase of 40 °C for 2 min, and then increase the temperature to 250 °C at 20 °C/min for 5 min; injection volume of 0.1 μL; shunt ratio of 200:1; carrier gas He, flow rate of 1.0 mL/min; MS conditions: the water was removed from sodium sulfate, and then concentrated to 0.5 mL by spinning at 22 °C and 80 rpm, and then analyzed by GC-MS (Agilent 8890/7250 GC/Q-TOF system). /MS conditions: full scan acquisition mode, mass range m/z 25-600 Da; electron bombardment energy 70 eV; source temperature 200 °C; quadrupole temperature 150 °C.


Construction of a limonene synthesis pathway in sake yeast

We constructed the limonene synthesis pathway in the sake yeast strain SY00 using CRISPR/Cas9 technology. Strains with different limonene yields were constructed using strategies to increase the copy of the limonene expression cassette, weaken the downstream competing pathway of MVA, and strengthen the upstream pathway of MVA (Fig. 1). The limonene expression cassette X4Lim was integrated into the chromosome Χ4 locus, and an engineered strain SY01 showed a limonene yield of 2.3 μg/L, indicating that the limonene synthesis pathway was successfully constructed. The strategies and engineered strains with limonene titers (Fig. 2) are described below:

1) Increase the copy of limonene expression cassette

The MVA pathway involves a variety of flavour metabolites, and modification of this pathway affects the aroma and compositional changes of wine, with priority given to the strategy of increasing limonene production by increasing the copy of the limonene expression cassette. The recombinant strains SY02 and SY03 were obtained by integrating the limonene expression cassette into chromosome Χ3 and ΧII2 of strain SY00, respectively, and the limonene titer was 6.9 μg/L and 11 μg/L, respectively, with a limited increase in yield.


Figure 1: Schematic diagram of genetically modified sake yeast for the production of sake with the bioactive molecule limonene. Black arrows represent native pathways; green dashed lines represent exogenous pathways (or genes); overexpressed genes are highlighted in red. Metabolite abbreviations: Ac-CoA, Acetyl coenzyme A; AcAc-CoA, Acetoacetyl coenzyme A; HMG-CoA, Hydroxymethylglutaric acid monoacyl coenzyme A; MVA, mevalonate; MVA-P, Mevalonate phosphate; MVA-PP, Mevalonate diphosphate; IPP, isopentenyl pyrophosphate; DMAPP, dimethylallyl pyrophosphate; NPP, neryl diphosphate; GPP, geranyl diphosphate; FPP, isopentenyl pyrophosphate. Gene abbreviations: ERG10, Acetoacetyl coenzyme A sulfatase; ERG13, hydroxymethylglutaryl coenzyme A; tHMG1, truncated HMG-CoA reductase; HMGR, HMG-CoA reductase; ERG12, Mevalonate kinase; ERG8, Mevalonate kinase; MVD1, Mevalonate-5-bisphosphate decarboxylase; IDI, IPP isomerase; tCPT1, truncated neryl diphosphate synthase from Solanum lycopersicum; tCILS, truncated  limonene synthase from Citrus; ERG20WW, mutated farnesyl diphosphate synthase; ERG20, FPP synthase; ERG9, squalene synthase.


Figure 2: Production of limonene by recombinant strains

2) Weak promoter PCYC1 weakens MVA downstream competition pathway

The recombinant strain SY04 was obtained using the weak promoter PCYC1 to replace the endogenous promoter of the ERG20 gene of strain SY03, with a limonene yield of 49 μg/L, indicating that the attenuation of the downstream pathway of MVA improved limonene production, but a severe growth slowdown of strain SY04 was observed, and it was speculated that the attenuation of the downstream pathway affected the synthesis of squalene, a key substance for cell growth, and thus cell biomass. Due to the low cell biomass and the specificity of the application in this study, the inducible promoters such as the glucose-sensitive promoter PHXT1 and the Fe2+-deterrent promoter PFET3 could not be used to dynamically regulate the downstream pathway during sake brewing, so the regulation of the competitive downstream pathway was no longer considered.

3) Overexpression of truncated trihydroxy-3-methylglutarate (HMG-CoA) reductase (tHMG1)

The truncated HMG1 gene is the key to terpene modification. Integration of the tHMG1 expression cassette into the ΧI3 locus of strain SY03 resulted in strain SY05 with a limonene yield of 59 μg/L, a slight improvement over strain SY03.

4) Overexpression of mutant farnesyl pyrophosphate synthase (ERG20F96W-N127W)

The integration of an ERG20F96W-N127W mutant at the Χ2 locus of strain SY05 was used to reduce the use of farnesyl diphosphate (FPP) for sterol biosynthesis, and strain SY06 was obtained with a limonene yield of 0.18 mg/L, which was further improved compared to the limonene yield of strain SY05.

5) Overexpression of hydroxymethylglutaryl coenzyme A synthase (ERG13)

ERG13 is a key enzyme in the MVA pathway of S. cerevisiae that catalyses the synthesis of hydroxymethylglutaryl coenzyme A (HMG-CoA) from acetoacetyl coenzyme A. By regulating the expression level or enzyme activity of ERG13, the production of HMG-CoA in the MVA pathway can be regulated, thereby increasing the carbon flux of the MVA pathway. The ERG13 gene expression cassette was increased at the Χ2 locus of strain SY06 to obtain strain SY07, which had a limonene yield of 0.4 mg/L, and the limonene yield was greatly increased compared to strain SY06.

6) Overexpression of mevalonate kinase (ERG12)

Mukherjee et al (33) identified ERG12 as a key enzyme gene affecting the flux of the MVA pathway and its overexpression affects the amount of pentenyl pyrophosphate (IPP) production.IPP is an intermediate of the MVA pathway used for the synthesis of terpenoids such as terpenic alcohols, aldehydes, ketones and acids. These terpenoids play an important role in the metabolism of S. cerevisiae, for example, by influencing factors such as the aroma and flavour of wine. Finally, we increased the ERG12 gene expression cassette at the Χ2 locus of strain SY07 to obtain strain SY08 (Supplementary file Fig. S1), which had a limonene yield of 2.6 mg/L, a significant increase in limonene production compared to strain SY07. It was hypothesised that the previous overexpression of tHMG1, ERG20F96W-N127W, and ERG13 had caused the strain to accumulate a higher amount of mevalonate, and that the overexpression of ERG12 relieved this part of the pressure.

Effect of different limonene titres on sake flavor

We selected the original strain SY00 and limonene-producing strains SY06, SY07 and SY08 for sake brewing. Sensory evaluation of the samples showed that the colour, aroma and style of the samples brewed with strain SY00 were within the characteristics of conventional rice wine. Samples brewed with strain SY06 showed a slight improvement in fruity flavour. Strain SY07 showed a further improvement in the fruity flavour, and the rice wine style was in harmony with the fruity components. On the other hand, the samples brewed with strain SY08 further improved the fruity flavour and the coordination between rice wine and fruity flavours, receiving the highest score of 92 points (Table 2). These results suggest that sake flavour was further improved with the increased limonene synthesis capacity. However, these strains differed in the intensity of MVA pathway modification in addition to the limonene synthesis capacity. To further determine whether the limonene synthesis pathway was the main influence on flavour rather than the enhancement of the MVA pathway, we constructed a strain SY09 with the same intensity of MVA pathway modification as strain SY08 and without the limonene synthesis pathway, and then performed sensory evaluations on the wine samples brewed by this strain and found that the wine samples brewed by this strain had some fruity aroma but were slightly astringent and bitter in taste.

Table 2: Sensory evaluation of wine samples brewed with sake yeast strains


Liquor characteristics



Slightly yellowish and transparent, with the colour and lustre that rice wine should have; pure rice wine aroma, fruity aroma is not obvious; full body, acidity and astringency are appropriate; rice wine style is obvious, and the body is more harmonious.



Slightly yellowish and transparent, with the colour and lustre that rice wine should have; pure rice wine aroma, with a more pronounced fruity aroma; fuller and richer body; obvious rice wine style, with a harmonious body.



Slightly yellowish and transparent, with the colour and lustre that rice wine should have; its aroma is pure, fruity and harmonious; its body is full, smooth and harmonious; its acidity and astringency are appropriate; its style is clear and its fruity components are harmonious.



Light yellow and transparent, with the colour and lustre that rice wine should have; its aroma is pure, fruity and harmonious; its body is full, smooth and harmonious; its acidity and astringency are appropriate; its style is clear and harmonious with its fruity components.



Slightly yellowish and transparent, with the colour and lustre that rice wine should have; pure rice wine aroma, with a more pronounced fruity aroma; full-bodied, slightly astringent and slightly bitter; obvious rice wine style, with a more harmonious body.


Fermentation performance of recombinant strains

To study the effect of genetically modified strains on fermentation performance, three aspects were analysed: CO2 emission, alcoholic strength determination and residual glucose. In terms of fermentation speed, as the number of overexpressed genes increased, the fermentation speed of sake gradually decreased, with the original strain SY00 completing brewing in 9 days, the highest limonene yielding strain SY08 taking 11 days, and the control genetically modified strain SY09 taking 13 days (Table 3), suggesting that there is a large accumulation of a certain intermediate metabolite in the MVA pathway, enhanced by the overexpressed genes, which has a toxic or inhibitory effect on cells, causing the slow growth of the strain.

Table 3: Fermentation performance of sake yeast strains with different limonene yields


CO2 cumulative weight loss (g)

Alcoholic content (20°C) (%/vol)

Residual glucose (g/L)

Days of fermentation (d)


























Sake Sample Composition Identification

To analyse the causes of changes in the flavour of sake brewed with recombinant strains, samples of sake brewed with the original strain SY00, the limonene strain SY08 and the genetically modified control strain SY09 were analysed by GC-MS, and 16 common flavour compounds were selected for comparative analysis (Table 4). Among the four lipids, compared to the original strain SY00, the relative levels of ethyl acetate and phenylethyl acetate were slightly increased in samples brewed by strain SY08, and the relative levels of ethyl caproate and ethyl benzoyl formate were slightly decreased, with ethyl benzoyl formate below the detection threshold. In strain SY09, the relative levels of these four lipids decreased, except for phenylethyl acetate, for which the relative levels increased slightly.

Table 4: Relative contents of major aroma components in samples of sake brewed with different yeast strains


Molecular formula




Ethyl acetate




















Isoamyl alcohol










Isobutyric acid















L-(+)-lactic acid





2-Methylbutyric acid





Ethyl benzoylformate





Ethyl caproate





2-Methylhexanoic acid





Phenethyl alcohol





Phenethyl acetate





Comparative analysis of brewing samples of strain SY08 with the original strain SY00 for seven alcohols revealed the following characteristics: the relative contents of isobutanol, 3-methyl-3-buten-1-ol and isopentyl alcohol decreased, with the relative content of isopentyl alcohol decreasing by up to 14%; the relative contents of 2-methylbutanol, isopentenyl alcohol, 3-pentanol and phenylethanol increased, with the relative content of 2-methylbutanol increasing by about 7% and the relative content of phenylethanol increasing by about 4%. The relative content of 2-methylbutanol, isopentenol, 3-pentanol and phenethyl alcohol increased, with the relative content of 2-methylbutanol increasing by about 7% and that of phenethyl alcohol by about 4%. In general, the total amount of higher alcohols decreased, and the decrease in the total amount of higher alcohols helped to reduce the "headiness", and the increase in the relative content of phenylethanol, which is one of the main aromas of sake flavour, helped to improve the quality of sake. In the samples brewed with strain SY09, the total relative content of higher alcohols increased and the relative content of phenylethanol, the main aroma of sake, decreased.

Among the remaining five substances, the relative contents of 3-hydroxy-2-butanone, L-(+)-lactic acid, 2-methylbutyric acid and 2-methylhexanoic acid increased in the brewing sample of strain SY08, while the relative content of isobutyric acid decreased, and the most significant change was in 2-methylhexanoic acid, which, as a substance with a creamy fatty taste, has a positive effect on the improvement of sake quality with an increase in the content of 2-methylhexanoic acid. In the brewing samples of strain SY09, the relative contents of these five substances were slightly increased or decreased, with no major changes. It is noteworthy that limonene was not detected in the final brewing sample of strain SY08, but the yield of limonene in the organic phase was found to be 15.8 mg/L after dodecane extraction during sake brewing, and it was hypothesised that limonene might be volatilised during sake fermentation along with carbon dioxide emission.


In the field of sake brewing, improving flavour and quality has always been the goal of brewers. Bioactive molecule research has been a hot research topic in recent years, and the brewing field has begun to focus on how to maintain health and prevent disease through the intake of bioactive molecules from alcoholic beverages. This study is the first to successfully construct the limonene synthesis pathway in sake yeast, which provides a new strategy for metabolically modifying yeast strains and applying them to sake brewing and demonstrates the potential of limonene as a flavour modifier and bioactive molecule. Limonene is a monoterpenoid widely found in plants and has various physiological functions such as antioxidant, anti-inflammatory, antibacterial and anticancer, etc. Limonene is also an important flavouring agent, imparting a fresh lemony flavour to foods and beverages. Sake is a traditional Japanese rice wine and its flavour is mainly composed of compounds such as alcohols, esters, higher alcohols and phenols. The results of this study showed that the construction of a limonene synthesis pathway in sake yeast was able to reduce the content of higher alcohols in sake, increase the fruity aroma of sake, and improve the flavour and mouthfeel of sake. This may be because limonene synthesis in yeast consumes precursors such as pyrophosphate and acetyl-coenzyme A, thereby reducing the synthesis of higher alcohols.

The strategy used in this study has some generality and can be used to construct different terpene synthesis pathways in other yeasts to produce alcoholic products with specific flavours and functions. Terpenoids are a class of natural products with diverse structures and functions that are widely distributed in plants, animals, microorganisms and other organisms, and the extension of terpene synthesis pathways in yeast for the production of different types of terpenoids can be achieved by introducing terpene synthases from other sources into the yeast MVA pathway.

There are still some limitations to this study, such as there is still room for improvement in the production of limonene, and the effect of the limonene synthesis pathway on other aroma compounds in sake has not been studied in depth. The results in this paper show that limonene production is related to the number of integrated limonene expression cassettes in yeast, the strengthening of the MVA pathway and the weakening of the competitive pathway, and that limonene produced by the modified strains may volatilise during sake brewing. Therefore, the retention of appropriate amounts of limonene in sake is an important direction for future research, and strategies such as the optimisation of limonene metabolic pathways and the development of new sake brewing processes could be adopted. In addition, the physiological activity and safety of limonene in sake need to be further verified. The content and bioavailability of limonene in sake, as well as its interaction with other components in sake, may affect the physiological effects and side effects of limonene. Therefore, more in-depth and detailed studies on the bioactivity and safety of limonene in sake are needed in the future to ensure the efficacy and safety of limonene application in sake. In conclusion, the study in this paper provides a new approach and evidence for improving the flavour of sake by constructing a limonene synthesis pathway in sake yeast, and also opens up a new field and direction for the application of terpenoids in sake.


This work was supported by the National Natural Science Foundation of China (32272538).


  1. Zhang K, Wu W and Yan Q. Research advances on sake rice, koji, and sake yeast: A review. Food Sci Nutr (2020): 2995-3003.
  2. Ohya Y and Kashima M. History, lineage and phenotypic differentiation of sake yeast. Biosci Biotechnol Biochem 83 (2019): 1442-1448.
  3. Katou T, Namise M, Kitagaki H, Akao T and Shimoi H. QTL mapping of sake brewing characteristics of yeast. J Biosci Bioeng 107 (2009): 383-393.
  4. Kanauchi M. Sake Alcoholic Beverage Production in Japanese Food Industry. Food Industry (2013).
  5. Yoshizawa K.: Sake: production and flavor. Food reviews international 15 (1999): 83-107.
  6. Tatsukami Y, Morisaka H, Aburaya S, Aoki W, Kohsaka C, Tani M, et al. Metabolite profiling of the fermentation process of "yamahai-ginjo-shikomi" Japanese sake. PLoS One 13 (2018): 0190040.
  7. Akaike M, Miyagawa H, Kimura Y, Terasaki M, Kusaba Y, Kitagaki H et al. Chemical and Bacterial Components in Sake and Sake Production Process. Curr Microbiol 77 (2020): 632-637.
  8. Okuda M. Rice used for Japanese sake making. Biosci Biotechnol Biochem 83 (2019): 1428-1441.
  9. Dahabieh MS, Husnik JI and Van Vuuren HJ. Functional enhancement of Sake yeast strains to minimize the production of ethyl carbamate in Sake wine. J Appl Microbiol 109 (2010): 963-973.
  10. Kitagaki H and Kitamoto K. Breeding research on sake yeasts in Japan: history, recent technological advances, and future perspectives. Annu Rev Food Sci Technol 4 (2013): 215-235.
  11. Negoro H and Ishida H. Development of sake yeast breeding and analysis of genes related to its various phenotypes. FEMS Yeast Res 22 (2022): 057.
  12. Asano T, Inoue T, Kurose N, Hiraoka N and Kawakita S. Improvement of isoamyl acetate productivity in sake yeast by isolating mutants resistant to econazole. J Biosci Bioeng 87 (1999): 697-699.
  13. Kotaka A, Sahara H and Hata Y. The construction and application of diploid sake yeast with a homozygous mutation in the FAS2 gene. J Biosci Bioeng 110 (2010): 675-678.
  14. Takahashi T, Ohara Y and Sueno K. Breeding of a sake yeast mutant with enhanced ethyl caproate productivity in sake brewing using rice milled at a high polishing ratio. J Biosci Bioeng 123 (2017): 707-713.
  15. Mizoguchi H, Yamauchi T, Watanabe M, Yamanaka H, Nishimura A and Hanamoto H. Different missense mutations in PDR1 and PDR3 genes from clotrimazole-resistant sake yeast are responsible for pleiotropic drug resistance and improved fermentative activity. J Biosci Bioeng 93 (2002): 221-227.
  16. Anandakumar P., Kamaraj S and Vanitha MK. D-limonene: A multifunctional compound with potent therapeutic effects. J Food Biochem 45 (2021): 13566.
  17. Eddin LB, Jha NK, Meeran MFN, Kesari KK, Beiram R and Ojha S. Neuroprotective Potential of Limonene and Limonene Containing Natural Products. Molecules 26 (2021): 4535.
  18. Da Silva CEH. Gosmann G and De Andrade SF. Limonene and Perillyl Alcohol Derivatives: Synthesis and Anticancer Activity. Mini Rev Med Chem 21 (2021): 1813-1829.
  19. Denkova-Kostova R, Teneva D, Tomova T, Goranov B, Denkova Z, Shopska V, et al. Chemical composition, antioxidant and antimicrobial activity of essential oils from tangerine (Citrus reticulata L.), grapefruit (Citrus paradisi L.), lemon (Citrus lemon L.) and cinnamon (Cinnamomum zeylanicum Blume). Z Naturforsch C J Biosci 76 (2020): 175-185.
  20. Ciriminna R, Lomeli-Rodriguez M, Demma Carà P, Lopez-Sanchez JA and Pagliaro M. Limonene: a versatile chemical of the bioeconomy. Chem Commun (Camb) 50 (2014): 15288-15296.
  21. Kim YW, Kim MJ, Chung BY, Bang duY, Lim SK, Choi SM, et al. Safety evaluation and risk assessment of d-Limonene. J Toxicol Environ Health B Crit Rev 16 (2013): 17-38.
  22. Ravichandran C, Badgujar PC, Gundev P and Upadhyay A. Review of toxicological assessment of d-limonene, a food and cosmetics additive. Food Chem Toxicol 120 (2018): 668-680.
  23. Slaghenaufi D and Ugliano M. Norisoprenoids, Sesquiterpenes and Terpenoids Content of Valpolicella Wines During Aging: Investigating Aroma Potential in Relationship to Evolution of Tobacco and Balsamic Aroma in Aged Wine. Front Chem 6 (2018): 66.
  24. Antonella DS, Federico D, Grazia SM and Gabriela M. Antimutagenic and antioxidant activities of some bioflavours from wine. Food Chem Toxicol 60 (2013): 141-146.
  25. Chigo-Hernandez MM and Tomasino E. Aroma Perception of Limonene, Linalool and α-Terpineol Combinations in Pinot Gris Wine. Foods 12 (2023): 2389.
  26. Januszek M, Satora P and Tarko T. Oenological Characteristics of Fermented Apple Musts and Volatile Profile of Brandies Obtained from Different Apple Cultivars. Biomolecules 10 (2020): 853.
  27. Hu Z, Lin L, Li H, Li P, Weng Y, Zhang C, et al. Engineering Saccharomyces cerevisiae for production of the valuable monoterpene d-limonene during Chinese Baijiu fermentation. J Ind Microbiol Biotechnol 47 (2020): 511-523.
  28. Dudareva N, Klempien A, Muhlemann JK and Kaplan I. Biosynthesis, function and metabolic engineering of plant volatile organic compounds. New Phytol 198 (2013): 16-32.
  29. Wu Q, Chen B and Xu Y. Regulating yeast flavor metabolism by controlling saccharification reaction rate in simultaneous saccharification and fermentation of Chinese Maotai-flavor liquor. Int J Food Microbiol 200 (2015): 39-46.
  30. Chen B, Wu Q and Xu Y. Filamentous fungal diversity and community structure associated with the solid state fermentation of Chinese Maotai-flavor liquor. Int J Food Microbiol 179 (2014): 80-84.
  31. GB/T 13662-2018. Yellow Wine. Beijing: State Administration for Market Supervision and Administration of the People's Republic of China, National Standardization Administration of the People's Republic of China (2018).
  32. Mukherjee M, Blair RH and Wang ZQ. Machine-learning guided elucidation of contribution of individual steps in the mevalonate pathway and construction of a yeast platform strain for terpenoid production. Metab Eng 74 (2022): 139-149.

Figure S1: Gas chromatography analysis profiles of the fermentation broths of the wild strain SY00 and the engineered strain SY08. In strain SY08, limonene peaked at 4.5 min.

Table S1: Primers used in this study


Sequences  (5'→3')

Lengths (bp)





































































































Table S2: Integration sites used in this study

S. cerevisiae loci

Cas9 target sites

















© 2016-2024, Copyrights Fortune Journals. All Rights Reserved