Abstract
Lavandulol, the primary chemical constituent of lavender essential oil, is an irregular monoterpene present in Lavandula angustifolia. It has been employed in the spice and cosmetic industries owing to its pleasing aromatic properties. In addition, its efficacy as an inhibitor of insect mating behaviour has led to incorporation within synthetic pheromone formulations. With the development of synthetic biology, more terpenoids are synthesized utilizing microorganisms. In this study, lavandulol was biosynthesized in Saccharomyces cerevisiae for the first time. An initial screening of lavandulyl diphosphate candidates was performed to identify enzymes compatible with heterologous expression in yeast. To increase the supply of DMAPP, the key enzymes of the MVA pathway, IDI1 and tHMG, as well as IDI1, ERG12, and ERG8 were overexpressed. Furthermore, the metabolic flow loss of acetyl-CoA and DMAPP was also reduced by deleting the genes of MLS1 and CIT2, as well as replacing the ERG20 promoter. Through the modification of lavandulyl diphosphate synthase, the flask titer was increased to 136.68 mg/L. Finally, the highest reported level of lavandulol production (308.92 mg/L) in S. cerevisiae was achieved by fed-batch fermentation in a 5 L bioreactor. To our knowledge, this is the first report of heterologous biosynthesis of lavandulol in S. cerevisiae.
Keywords: Lavandulol, Irregular monoterpene, Heterologous biosynthesis, Saccharomyces cerevisiae
Graphical abstract
Abbreviations
- DMAPP
Dimethylallyl diphosphate
- MVA
Mevalonic acid
- IDI1
Isopentenyl diphosphate isomerase
- tHMG
Truncated 3-hydroxy-3-methylglutaryl-CoA reductase
- IDI1
Acetoacetyl-CoA thiolase
- ERG12
Mevalonate kinase
- ERG8
Phosphomevalonate kinase
- ERG20
Farnesyl pyrophosphate synthase
- MLS1
Malate synthase
- CIT2
Citrate synthase
1. Introduction
Terpenoids are complex natural substances in terms of their chemical structure and conformation [1]. To date, more than 50,000 terpenoids have been discovered, with a high prevalence observed in plants, fungi and bacteria [2,3]. Monoterpenes, which possess a C10 skeleton, are characterized by a pleasant aroma and are extensively employed in the production of perfumes and cosmetics [4]. Additionally, monoterpenes are employed in agricultural industries due to their ability to attract and repel pests [5,6].
Lavandulol, the primary chemical component of the lavender essential oil, is an irregular monoterpene present in Lavandula angustifolia [7]. It is widely used in the spice and cosmetics industries [8]. Moreover, lavandulol and its ester derivative are significant components of pheromones for major pests [9,10]. Consequently, they have been employed in artificial pheromone preparations to interfere with the mating behavior of pests. Studies have shown that lavandulol can be generated by regio-and stereoselective photocatalytic hydrogenolysis with (R)-(−)-carvone as the starting material under the action of palladium (II)-loaded titanium oxide [11]. In addition, proline catalyzed asymmetric alpha-aminooxylation and [3,3] Claisen rearrangement were also involved in the total synthesis of lavandulol [12]. However, natural extraction and chemical synthesis limit the scale of lavandulol production due to the complexity and contamination of the synthetic process.
The advent of metabolic engineering and synthetic biology has led to the utilization of microbial cell factories for the biosynthesis of natural products [13]. In nature, the biosynthesis of regular monoterpenes occurs through a process involving the condensation of one molecule of isopentenyl diphosphate (IPP) and one molecule o dimethylallyl diphosphate (DMAPP), resulting in the formation of the precursor geranyl diphosphate (GPP) or neryl diphosphate (NPP) with a C10 skeleton (Fig. 1A) [14,15]. This process is catalyzed by the respective terpene synthase, leading to the production of the specific monoterpene [16]. In contrast, the formation of irregular monoterpenes occurs through a distinct process involving the head-to-middle condensation of two molecules of DMAPP in different conformations [17]. In this process, two DMAPPs condense to form the lavandulyl diphosphate (LPP) and the chrysanthemyl diphosphate (CPP), which are catalyzed by the lavandulyl diphosphate synthase (LPPS) and the chrysanthemyl diphosphate synthase, respectively (Fig. 1B) [[18], [19], [20]]. Irregular monoterpenes represent a rare form of terpenoids, which have been found to occur in certain species of plants belonging to the tribe Anthemideae in the family Asteraceae [21] and were found to be vital precursors of insecticides and insect sex pheromones [22].
Fig. 1.
(A) Head-to-tail condensation of IPP and DMAPP to produce regular monoterpenes (taking the formation of GPP as an example). (B) Head-to-middle condensation of 2 molecules of DMAPP to produce irregular monoterpenes. GPP: geranyl diphosphate, IPP: isopentenyl diphosphate, DMAPP: dimethylallyl diphosphate, CPP: chrysanthemyl diphosphate, MPP: maconelliyl diphosphate, LPP: lavandulyl diphosphate, PPP: planococcyl diphosphate.
The formation of lavandulol begins with the condensation of two DMAPP molecules, accompanied by a rearrangement of double bonds to generate the highly reactive carbocation intermediate called the lavandulyl cation (L+). This intermediate then undergoes deprotonation to form LPP [18]. Finally, LPP is deprotonated and hydroxylated to form lavandulol.
In this study, we have successfully achieved the biosynthesis of lavandulol in Saccharomyces cerevisiae for the first time. A lavandulol synthase that could be expressed in yeast was obtained. To increase the DMAPP supply, the key enzymes of the MVA pathway, IDI1 and tHMG, as well as IDI1, ERG12, and ERG8 were overexpressed. The metabolic flux consumption of acetyl-CoA and DMAPP were also reduced by deleting the genes of MLS1 and CIT2, as well as replacing the ERG20 promoter. Besides, the truncation and modification of LILPPS enhanced the lavandulol production to 136.68 mg/L. Finally, the highest level of lavandulol production (308.92 mg/L) reported to date in S. cerevisiae was achieved by fed-batch fermentation in a 5 L bioreactor. To our knowledge, this is the first report of heterologous synthesis of lavandulol in S. cerevisiae.
2. Materials and methods
2.1. Strains and medium
E.coli DH5α (Tsingke, Beijing, China) was used to genes cloning and plasmids construction, and cultivated at 37°C in LB media (0.5 % yeast extract, 1 % tryptone, and 1 % NaCl). S. cerevisiae CEN.PK2-1C was used as the origin strain to construct the cell factory of lavandulol in this study. S. cerevisiae was cultivated at 30°C in YPD medium (2 % peptone, 2 % glucose and 1 % yeast extract). The yeast cell containing recombinant plasmids was selected by the synthetic complete (SC) medium (2 % glucose, 0.67 % yeast nitrogen base and 0.2 % amino acid dropout powder).
The medium used in fed-batch fermentation contained glucose 20 g/L, peptone 20 g/L, yeast extract 10 g/L, KH2PO4 8 g/L, MgSO4 6.15 g/L, trace element solution, pH 7.0, 10 mL/L, vitamin solution 12 mL/L and antifoam 1 mL/L. The trace element solution was configured as follows: ZnSO4∙7H2O 10.2 g/L, EDTANa2∙2H2O 15 g/L, FeSO4∙7 H2O 5.12 g/L, CuSO4 0.5 g/L, MnCl2∙4H2O 0.5 g/L, CoCl2∙6H2O 0.86 g/L, CaCl2∙2H2O 3.84 g/L, Na2MoO4∙2 H2O 0.56 g/L. The vitamin solution was configured as follows: Inositol 25 g/L, biotin 0.05 g/L, niacin (nicotinic acid) 1 g/L, calcium pantothenate 1 g/L, pyridoxine hydrochloride 1 g/L, vitamin B1 1 g/L, p-aminobenzoic acid 0.2 g/L. 1 L feeding medium (contain 60 % glucose solution, trace element solution, pH 7.0, 10 mL/L, vitamin solution 12 mL/L) was used to provide carbon source.
2.2. Plasmid and strains construction
All the plasmids and strains constructed in this study are listed in Table 1, Table 2, respectively. The primers used are listed in Table S2.
Table 1.
Plasmids used in this study.
| Plasmid | Description | Source |
|---|---|---|
| pESC-URA | 2 μm URA3, Amp | stored in the lab |
| pESC-LiLPPS | pESC-URA vector containing codon-optimized LiLPPS | this study |
| pESC-SlNPPSN88H | pESC-URA vector containing codon-optimized PcPTS1 | this study |
| pESC-CDS | pESC-URA vector containing codon-optimized CDS | this study |
| pESC- LavPP1 | pESC-URA vector containing codon-optimized LavPP1 | this study |
| pESC-LavPP2 | pESC-URA vector containing codon-optimized LavPP2 | this study |
| pESC-LavPP3 | pESC-URA vector containing codon-optimized LavPP3 | this study |
| pWS158-Cas9 | KanR, Ura, Cas9 | stored in the lab |
| pWS158-GAL80 | pWS158-Cas9, GAL80-gRNA | this study |
| pWS158-YPL | pWS158-Cas9, YPL-gRNA | this study |
| pWS158-CIT2 | pWS158-Cas9, CIT2-gRNA | this study |
| pWS158-MLS | pWS158-Cas9, MLS-gRNA | this study |
| pWS158-PERG20 | pWS158-Cas9, PERG20-gRNA | this study |
| pESC-tLPPS1 | pESC-URA, containing tLPPS1 | this study |
| pESC-tLPPS2 | pESC-URA, containing tLPPS2 | this study |
| pESC-tLPPS3 | pESC-URA, containing tLPPS3 | this study |
| pESC-tLPPS4 | pESC-URA, containing tLPPS4 | this study |
| pESC-tLPPS5 | pESC-URA, containing tLPPS5 | this study |
| pESC-tLPPS4V89I | pESC-URA, containing tLPPS4V89I | this study |
| pESC-tLPPS4A119I | pESC-URA, containing tLPPS4A119I | this study |
| pESC-tLPPS4L120F | pESC-URA, containing tLPPS4L120F | this study |
| pESC-tLPPS4Q123D | pESC-URA, containing tLPPS4Q123D | this study |
| pESC-tLPPS4Q123E | pESC-URA, containing tLPPS4Q123E | this study |
| pESC-tLPPS4A249T | pESC-URA, containing tLPPS4A249T | this study |
| pESC-tLPPS4L255A | pESC-URA, containing tLPPS4L255A | this study |
| pESC-tLPPS4L255I | pESC-URA, containing tLPPS4L255I | this study |
| pESC-tLPPS4L255V | pESC-URA, containing tLPPS4L255V | this study |
| pESC-tLPPS4L274F | pESC-URA, containing tLPPS4L274F | this study |
| pESC-tLPPS4L274M | pESC-URA, containing tLPPS4L274M | this study |
| pESC-tLPPS4L274N | pESC-URA, containing tLPPS4L274N | this study |
Table 2.
Strains used in this study.
| Strains | Description | Source |
|---|---|---|
| CEN.PK2-1C | MATa; ura3-52; trp1289; leu2-3, 112; his3Δ 1; MAL2-8C; SUC2 | stored in the lab |
| CE1 | CEN.PK2-1C, containing pESC-LiLPPS | this study |
| CE2 | CEN.PK2-1C, containing pESC-SlNPPSN88H | this study |
| CE3 | CEN.PK2-1C, containing pESC-CDS | this study |
| CE4 | CEN.PK2-1C, containing pESC- LavPP1 | this study |
| CE5 | CEN.PK2-1C, containing pESC-LavPP2 | this study |
| CE6 | CEN.PK2-1C, containing pESC-LavPP3 | this study |
| Yl1 | Cen.pk, Δgal80::TIDI1-IDI1-GAL1,10-tHMG1-THMG | this study |
| Yl2 | Yl1, ΔYPL::TIDI1-IDI1-GAL1,10-tHMG1-THMG | this study |
| Yl3 | Yl2, ΔCIT2::TIDI1-IDI1-GAL1,10-tHMG1-THMG | this study |
| Yl4 | Yl3, ΔMLS::TERG8-EGR8-GAL1,10-EGR12-TEGR12 | this study |
| Yl5 | Yl4, ΔPERG20::PHXT1 | this study |
| Yl11 | Yl1, containing pESC-LiLPPS | this study |
| Yl21 | Yl2, containing pESC-LiLPPS | this study |
| Yl31 | Yl3, containing pESC-LiLPPS | this study |
| Yl41 | Y4, containing pESC-LiLPPS | this study |
| Yl51 | Yl5, containing pESC-LiLPPS | this study |
| Yl52 | Yl5, containing pESC-tLPPS1 | this study |
| Yl53 | Yl5, containing pESC-tLPPS2 | this study |
| Yl54 | Yl5, containing pESC-tLPPS3 | this study |
| Yl55 | Yl5, containing pESC-tLPPS4 | this study |
| Yl56 | Yl5, containing pESC-tLPPS5 | this study |
| Yl57 | Yl5, containing pESC-tLPPS4V89I | this study |
| Yl58 | Yl5, containing pESC-tLPPS4A119I | this study |
| Yl59 | Yl5, containing pESC-tLPPS4L120F | this study |
| Yl60 | Yl5, containing pESC-tLPPS4Q123D | this study |
| Yl61 | Yl5, containing pESC-tLPPS4Q123E | this study |
| Yl62 | Yl5, containing pESC-tLPPS4A249T | this study |
| Yl63 | Yl5, containing pESC-tLPPS4L255A | this study |
| Yl64 | Yl5, containing pESC-tLPPS4L255I | this study |
| Yl65 | Yl5, containing pESC-tLPPS4L255V | this study |
| Yl66 | Yl5, containing pESC-tLPPS4L274F | this study |
| Yl67 | Yl5, containing pESC-tLPPS4L274M | this study |
| Yl68 | Yl5, containing pESC-tLPPS4L274N | this study |
Three synthases with identified function in previous reports including lavandulyl diphosphate synthases (LILPPS, GenBank: AGH33890.1) from Lavandula x intermedia, the mutant of SlNPPS (GenBank: ACO56895.1) SlNPPSN88H from Solanum lycopersicum and CDS (GenBank: AAP74721.1) from Artemisia spiciformis were codon optimized for S. cerevisiae by GENEWIZ (Suzhou, China). Besides, three putative proteins including LavPP1 (GenBank: KAL1539878.1), LavPP2 (GenBank: QNC49773.1) and LavPP3 (GenBank: XP_042016829.1) were also selected as candidate proteins by sequence alignment. The genes above were induced to pESC-URA vector between BamHI and HindIII restriction sites to generate corresponding plasmids. The nucleotide sequence of the synthetic codon-optimized LILPPS was utilized as a template to generate mutants and truncated variants using the PCR-mediated method. The resultant plasmids shown in Table 1 were individually transformed into S. cerevisiae strains using the PEG/LiAc method.
The genes of truncated 3hydroxy-3-methylglutaryl-CoA reductase (tHMG), isopentenyl diphosphate isomerase (IDI1), IDI1, ERG8 and ERG12 were PCR-amplified from the genome of S. cerevisiae CEN.PK2-1C. Promoters and terminator for gene expression are also obtained from the genome by PCR-amplified. The individual expression cassettes and homology arms were linked together by over-lap PCR. CRISPR/Cas9 system was used to construct efficient yeast cells. Individual Cas9 plasmids containing the corresponding single guide RNA were constructed by the one-pot method using restriction endonuclease ESP3I and T4 ligase. The expression cassettes and Cas9 plasmids were transformed into S. cerevisiae strains using PEG/LiAc methods.
2.3. Structural modeling and molecular docking
Docking of the ligand DMAPP into the model of the LILPPS (PDB:5HC6) was performed by Autodock. The 3D structure of DMAPP was downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/), while visualization and labeling were done using the software PyMoL. The amino acid sequences were searched in the NCBI (https://blast.ncbi.nlm.nih.gov/) database and the UniProt (https://www.uniprot.org/) database. The multiple sequence alignments were performed with Jalview software. Sequence logos were created with the Weblogo (http://weblogo.berkeley.edu/).
2.4. Cultivation in shaking flask
For shake flasks fermentation, a single colony of S. cerevisiae strain was inoculated into 24-well plates containing 2 mL of SC medium and cultivated for 18 h at 30 °C, 220 rpm. Then the primary seed solution was inoculated into 250 mL flasks containing 50 mL SC medium to maintain an initial OD600 of 0.1 and cultivated at 30 °C and 220 rpm and 10 % (v/v) isopropyl myristate was added to reduce the toxicity of lavandulol to cells and the volatilisation. 500 μL Ethanol was added to the culture when 36 h and 60 h of cultivation, respectively. Fermentation ended at 96 h of cultivation.
2.5. Fed-batch fermentation
For fed-batch fermentation, a single colony of S. cerevisiae strain was inoculated into 250 mL flask containing 50 mL of SC medium and cultivated for 18 h at 30 °C, 220 rpm. 2 mL Primary seed solution was inoculated into 1 L flasks containing 200 mL SC medium and cultivated for 18 h at the same conditions. The secondary seed solution was inoculated into 5 L bioreactor containing 2 L YPD medium. The pH of the culture was adjusted to 5.5 by the automated addition of NH3·H2O and H2SO4. The dissolved oxygen level was kept above 30 % by adjusting the agitation speed and the airflow rate. After the initially added glucose was depleted, glucose concentration was maintained between 0 and 1 g/L by the automated addition of feeding medium. After the feeding medium was consumed completely, Ethanol was added to provide carbon source subsequently. Isopropyl myristate (10 % v/v) was added after 20 h for the two-phase fermentation.
2.6. Analysis of lavandulol
The products of two-phase fermentation were collected and centrifuged at 6000 rpm for 5 min to obtain the supernatant containing lavandulol. The samples were then dewatered with anhydrous sodium sulfate and properly diluted with hexane. Samples were then subjected to gas chromatography-mass spectrometry (GC-MS) to quantify the concentration of lavandulol using SHIMADZU GC-2030AM system equipped with autosampler.
Gas chromatography was equipped with a 30 m × 0.25 mm × 0.25 μm HP-5MS column. Helium was used as the carrier gas at a flow rate of 1 mL/min. Temperature program was set as 70 °C for 1 min, then steps of 10 °C/min to 170 °C for 2 min and steps of 20 °C/min to 250 °C. The MS EI ion source temperature was 200 °C and the transfer line temperature was 230 °C. The MS scan range (m/z) was 40–300. The identification of components was based on the comparison of their RIs and MS spectra with those stored in the computer library NIST21. The percentage of components was computed by area normalization method. Residual sugar and ethanol concentrations were detected by an M-100 biosensor analyzer (SIEMAN, Shenzhen, China) according to the standard instructions. Cell growth was characterized by the optical density at 600 nm (OD600) or the biomass. The value of OD600 was measured with a spectrophotometer (UV1800PC, HangYi, Shanghai, China).
2.7. Lavandulol isolation
For isolation and purification, the fermentation broth with no extractant was collected and extracted with hexane after several extractions. The crude reaction mixture was mixed with silica gel (230–400 mesh particle size) according to 1:1.5 and added on top of the silica gel column (45 cm × 3 cm). The column was rinsed with 1 L pure petroleum ether and 3 L mixture of petroleum ether and ethyl acetate (80:1). The fraction was collected in tubes every 100 ml and checked for purity on a TLC plate. The resulting solution was filtered and solvent removed carefully under reduced pressure (−0.1 MPa, 25 °C) to give colorless oil. For NMR spectroscopic analysis, the sample was dissolved in CDCl3 and analyzed by 1H and 13C NMR spectroscopy.
3. Results
3.1. Screening for efficient lavandulol synthase
As the expression of lavandulol synthase in yeast has not been described, a literature review and database search were conducted to identify suitable candidates for further investigation. Three candidates with identified function in previous reports were selected including LILPPS, the mutation SlNPPSN88H and CDS. Besides, since that LILPPS exhibits less stringent requirements with respect to substrate concentration and type, three putative proteins with relatively high similarity to LILPPS, LavLPP1, LavLPP2 and LavLPP3 were selected as candidate proteins by sequence alignment.
The original strain CEN.PK2-1C was used for expressing candidate proteins. Plasmids containing each candidate were introduced into the chassis, and lavandulol was only detected in yeast containing LILPPS with a titer of 5.39 mg/L (Fig. S1). This may be since that the stringent catalytic conditions required for SlNPPSN88H and CDS in vitro cannot be achieved in yeast. LILPPS was selected for the synthesis of lavandulol in S. cerevisiae in this study.
3.2. Pathway modification for enhanced lavandulol production
To improve the gene expression in the optimized pathway, metabolic engineering was employed to construct chassis strains (Fig. 2A and B). DMAPP is the unit for the synthesis of lavandulol in yeast. Improving the flux of DMAPP in the MVA could be an efficient strategy to enhance the lavandulol production. In this study, the overexpression of IDI1 and tHMG increased the lavandulol production by 113.43 %. After IDI1 and tHMG were overexpressed twice in YI21, the lavandulol production was further increased by 42.76 %, which was much lower than the growth rate of Yl11. It may be caused by the imbalance in metabolic flux. To further enhance the flux of the MVA pathway, IDI1 and the third copy of tHMG were introduced at CIT2. ERG8 and ERG12 were constructed at MLS1. This strategy improved the production of lavandulol by 87.65 % than YI21 (Fig. 2C).
Fig. 2.
(A) The schematic overview of engineering pathway of lavandulol in S. cerevisiae. The overexpressed genes are marked in red, and the genes deleted or dynamic controlled are marked in green. HMG-CoA: (S)-3-hydroxy-3-methylglutaryl-CoA, FPP, farnesyl diphosphate; IDI1, acetoacetyl-CoA thiolase; tHMG, truncated 3-hydroxy-3-methylglutaryl-CoA reductase; ERG12, mevalonate kinase; ERG8, phosphomevalonate kinase; IDI1, isopentenyl pyrophosphate isomerase; ERG20, farnesyl pyrophosphate synthase; MLS1, malate synthase; CIT2, citrate synthase. (B) Chassis construction integration site schematic diagram in this study. (C) The lavandulol production of different strains overexpressing genes of the MVA pathway. Error bars represent standard deviations (n = 3). The one-way ANOVA was used for statistical analysis (∗P < 0.05, ∗∗∗P < 0.001).
In the biosynthesis of monoterpenes, weakening the expression of ERG20 is often used to increase the flux of direct precursors such as GPP [23]. To improve the utilization rate of DMAPP and reduce its metabolic flux loss for GPP, the promoter of ERG20 was replaced with PHXT1 for dynamic regulation in YI51. The lavandulol production was further increased to 34.81 mg/L (Fig. 2C).
3.3. Enzyme modification for enhanced lavandulol production
The N-terminal regions of monoterpene synthases from plants frequently contain signal peptides, which hinder their expression in yeast [24]. Uniprot and TargetP 2.0 were utilized to predict the signal peptide of LILPPS. Subsequently, the truncated LILPPSs (tLILPPSs) were generated by removing the first 20, 30, 40, 55, and 65 amino acids. The results demonstrated that after deleting the first 55 amino acids, the production of Yl55 was 2.91-fold of Yl51 which containing the full-length amino acids. As the truncated length was increased to 65 amino acids, the titer exhibited no significantly to that of Yl51 (Fig. 3).
Fig. 3.
Lavandulol titers of strains containing different lengths truncation of LILPPS. Error bars represent standard deviations (n = 3). The one-way ANOVA was used for statistical analysis (∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001).
To identify the key residue sites which effect the catalytic capability of LILPPS, the protein structure of LILPPS was docked with two DMAPPs (Fig. 4A). The docking results indicated that the substrate forms hydrogen bonds with LILPPS at the residues G77, R79, N124, R127, R248, R254, and S256 (Fig. S2). The sequences of LILPPS and proteins with high identity retrieved from the NCBI database were aligned (Fig. S3). As demonstrated in previous research, mutations in the proximity of the active pocket frequently result in a beneficial alteration of the catalytic activity of terpene synthase [25]. The preference of residue sites of difference sites within 4–6 Å from the ligand were analyzed (Fig. S4). Based on the above analysis, 12 mutants of tLILPPS were constructed for testing. The results showed that the production of most strains have decreased to varying degrees, among which the activity of Q123D, L255A, and L274 N mutants were almost completely lost (Fig. 4B). The lavandulol production of A249T mutant was increased by 34.97 % than the wide type, reaching 136.68 mg/L in flask fermentation. The simulation results of the bonding situation near A294 showed that when the residue at 294 was mutated from alanine to threonine, it formed a hydrogen bond with D76 (Fig. 4C). The hydrogen bond introduced by A294T was presumed to enhance the substrate binding and promote the diphosphate release by the influence on the D76-G77-H78 triplet.
Fig. 4.
The modification of LILPPS. (A) The molecular docking results of LILPPS and DMAPP, as well as the residues 4–6 Å from the active pocket. (B) The lavandulol production of strains containing single point mutant of LILPPS. (C) Comparison of bonding situation simulation when the residue 249 is alanine and threonine. Error bars represent standard deviations (n = 3). The one-way ANOVA was used for statistical analysis (∗∗∗P < 0.001).
3.4. Fed-batch fermentation
To test the industrial production potential of the strain containing tLILPPSA249T, fed-batch cultivation was performed with glucose as initial carbon source in a 5 L bioreactor. By optimizing the fermentation conditions, the optimum aeration rate of 5 L/h and the optimum fermentation pH of 5.5 were determined. Time courses of lavandulol production, cell growth, glucose and ethanol consumption during the fermentation are shown in Fig. 5. Glucose was drastically depleted within 24 h of inoculation. After that, a 0–1 g/L glucose concentration was maintained by the automated addition of the feeding medium. Isopropyl myristate was added after 20 h for the two-phase fermentation. Within 24 h, the accumulated ethanol concentration from fermentation peaked at 12 h, reaching 3.7 g/L. Ethanol was added to supplement the carbon source at 50 h and the concentration was maintained below 10 g/L by the automated addition of the feeding medium. The OD600 showed little variation between the 30 h and 66 h of cultivation, with the maximum OD600 value at the 48 h, reaching 163.06. The production rate of lavandulol was the fastest during 42 h–66 h. The accumulation of lavandulol reached 308.92 mg/L after 72 h of cultivation, which was the highest reported titer for lavandulol production to date.
Fig. 5.
Fed-batch cultivation of Yl62 in a 5 L bioreactor. Fermentation was conducted in a 5 L bioreactor with glucose as the main feedstock. Error bars represent standard deviations (n = 3).
3.5. Preparation of standard lavandolul
To determine the structure of product of tLILPPSA249T, 3 L solution obtained by fed-batch fermentation was collected. After chromatographic purification using deactivated silica with petroleum ether and ethyl acetate as eluent, a total of 154.67 mg of a colorless oil was isolated with >75 % purity, as judged by GC-MS (Fig. S5). The purified compound was analyzed by NMR spectroscopy. The characteristic peaks are as follows: 1H (400 MHz, CDCl3) δ 5.08 (t, 1H), 4.93 (s, 1H), 4.82 (s, 1H), 3.47–3.59 (m, 2H), 2.28 (m, 1H), 2.03–2.11 (m, 2H), 1.70 (s, 3H), 1.69 (s, 3H), 1.61 (s, 3H). 13C NMR (101 MHz, CDCl3) δ 145.22, 132.62, 121.83, 113.00, 63.43, 49.78, 28.18, 25.57, 19.31, 17.65 (Figure S6 and Figure S7).
The characteristic peaks of the product were consistent with those reported in the literature, and the product was identified as lavandulol [26].
4. Discussion
Lavandulol, an irregular monoterpene, is used in perfumery and cosmetic industry [27], as well as in artificial pheromone preparations to interfere with the mating behavior of economically important pests [28]. In this study, lavandulol was biosynthesized in S. cerevisiae successfully for the first time.
Three lavandulol synthases from different sources reported to possess lavandulol synthesis capacity in vitro were test in S. cerevisiae. LiLPPS expressed in E. coli was able to catalyze the substrate in vitro to obtain lavandulol specifically [18]. The mutant SlNPPSN88H exhibited elevated irregular activity, increasing from 0.4 % of wild-type SlNPPS to 13.8 % when utilizing solely DMAPP as the substrate [26]. CDS is a bifunctional enzyme with the capacity to synthesize CPP and LPP [21,29]. In this study, lavandulol was only detected in the fermentation broth of strains containing LILPPS. The harsh conditions required for the SlNPPSN88H in vitro may be cannot achieved in yeast and the catalytic promiscuity characteristic of CDS potentially constrains lavandulol biosynthetic efficiency through substrate channeling competition. In the future, machine learning can be used to study the genomes of more plants to find more lavandulol synthases with lower requirements for reaction conditions and promiscuity for expression in microbial cell factories. Besides, rational engineering of IDI1 to enhance DMAPP substrate specificity may constitute an effective strategy for improving the precursor pool of lavandulol.
Enhancing MVA flux has been widely applied in the efficient biosynthesis of various terpenoids [5,30]. In Yl11 and Yl21 which possess different copies of the rate-limiting enzymes tHMG and IDI1, the lavandulol productions were both increased. After the overexpression of IDI1, ERG8 and ERG12, the lavandulol production was enhanced by 5.46 folds than the starting strain. Simultaneously, CIT2 and MLS1 were deleted to reduce the competitive consumption of upstream acetyl-CoA in the MVA pathway. CIT2 catalyzes the formation of citrate from acetyl-CoA and oxaloacetate in peroxisomes, while MLS1 catalyzes the synthesis of malate from acetyl-CoA and glyoxylate in the cytoplasm [31]. To reduce competitive consumption of DMAPP, the expression of ERG20 was dynamically regulated by PHXT1. Typically, IPP and DMAPP are catalyzed by ERG20 to generate GPP or NPP, which in turn are catalyzed by ERG20 to generate some sesquiterpene which play an important role in maintaining normal cell growth [32,33]. The glucose-sensing promoter, PHXT1, exhibits a stronger response to glucose and a weaker response to its absence. Glucose is utilized in the initial stage of cell growth, resulting in the concurrent production of ethanol which is then employed in the subsequent stage of cell growth and the synthesis of the desired product [34]. In the future, various metabolic engineering strategies targeting acetyl-CoA, cofactors, and transcription regulatory factors can be employed to further enhance lavandulol production in S. cerevisiae. Although the lavandulol production was enhanced through metabolic engineering approaches, it was remained at a relatively low level compared to other monoterpenes [[35], [36], [37]], which is likely attributed to the catalytic specificity of LILPPS. In the head-to-middle condensation of two DMAPPs, LILPPS catalyzes the generation of LPP and is responsible for the deprotonation of LPP. The complexity of such cascade reactions often limits the catalytic capacity of enzymes.
The appropriate truncation of the N-terminal region has been shown to be beneficial to the increase of monoterpene and some diterpene synthase activities [38,39]. The N-terminal region of LILPPS was truncated with varying lengths in this study. The truncation of the first 55 amino acids demonstrated the best performance, increasing lavandulol production by 1.91-fold. It is noteworthy that the presence of the first 55 amino acids has been shown to impede successful crystallization of LILPPS [17]. This region may be a lipid (membrane) targeting domain that can influence the solubility of protein. Besides, the mutant A294T which improved lavandulol production by 34.97 % was obtained through molecular docking and sequence alignment. The structure of LILPPS was found to be similar to undecaprenyl diphosphate synthase (UPPS). H78 (Asn in UPPS) has been determined to facilitate diphosphate release in allylic site [40] and G77 forms hydrogen bonds with the substrate (Fig. S2). The hydrogen bond introduced by A294T may potentially enhance the substrate binding and promote the diphosphate release by the influence on the D76-G77-H78 triplet. This provides important insights for enhancing the enzymatic activity of lavandulol and other irregular monoterpenes.
5. Conclusion
In summary, we have achieved the sustainable production of lavandulol for the first time in S. cerevisiae. Through functional screening, LiLPPS was identified to exhibit heterologous expression potential in S. cerevisiae. The titer in shake flasks was increased to 136.68 mg/L through metabolic pathway regulation and protein engineering. Finally, the titer reached 308.92 mg/L in a 5 L bioreactor. This study lays the foundation for the subsequent biosynthesis of other irregular terpenoids in S. cerevisiae.
CRediT authorship contribution statement
Shengxin Nie: Writing – original draft. Ruiqi Chen: Methodology. Mingyue Ge: Validation. Yue Qu: Project administration. Xiaofeng Liu: Software. Chaofeng Ruan: Methodology. Xiaoguang Yan: Writing – review & editing. Jianjun Qiao: Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was financially supported by the "Pioneer" and "Leading Goose" R&D Program of Zhejiang (2025C01102), the National Key R&D Program of China (2020YFA0907900) and National Natural Science Foundation of China (32300055).
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.synbio.2025.06.001.
Contributor Information
Xiaoguang Yan, Email: yanxiaoguang@tju.edu.cn.
Jianjun Qiao, Email: jianjunq@tju.edu.cn.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
The protein sequences of candidate lavandulol synthases used in this study were provided in Table S1.
The primers used in this study were provided in Table S2.
The lavandulol production of strains containing different candidate lavandulol synthase were provided in Fig. S1.
Amino acid residues that forming hydrogen bonds with the substrate DMAPP were provided in Fig. S2.
The multiple sequence alignment result of LILPPS and similar proteins with high identity retrieved from the NCBI database was provided in Fig. S3.
Amino acid preference of LILPPS at site 89, 119, 120, 123, 249, 255, 274 was provided in Fig. S4.
Total ion chromatogram of lavandulol in final purity sample was provided in Figure S5.
1H NMR and 13C spectrum of lavandulol in CDCl3 were provided in Figs. S6 and S7.
References
- 1.Bernard A., Cha S., Shin H., Lee D., Hahn J.S. Efficient production of (S)-limonene and geraniol in Saccharomyces cerevisiae through the utilization of an ERG20 mutant with enhanced Gpp accumulation capability. Metab Eng. 2024;83:183–192. doi: 10.1016/j.ymben.2024.04.003. [DOI] [PubMed] [Google Scholar]
- 2.Belcher M.S., Mahinthakumar J., Keasling J.D. New frontiers: harnessing pivotal advances in microbial engineering for the biosynthesis of plant-derived terpenoids. Curr Opin Biotechnol. 2020;65:88–93. doi: 10.1016/j.copbio.2020.02.001. [DOI] [PubMed] [Google Scholar]
- 3.Sun Z., Meng H., Li J., Wang J., Li Q., Wang Y., Zhang Y. Identification of novel knockout targets for improving terpenoids biosynthesis in Saccharomyces cerevisiae. PLoS One. 2014;9 doi: 10.1371/journal.pone.0112615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chen L., Liao P. Current insights into plant volatile organic compound biosynthesis. Curr Opin Plant Biol. 2025;85 doi: 10.1016/j.pbi.2025.102708. [DOI] [PubMed] [Google Scholar]
- 5.Li D.X., Guo Q., Yang Y.X., Jiang S.J., Ji X.J., Ye C., et al. Recent advances and multiple strategies of monoterpenoid overproduction in Saccharomyces cerevisiae and Yarrowia lipolytica. ACS Synth Biol. 2024;13:1647–1662. doi: 10.1021/acssynbio.4c00297. [DOI] [PubMed] [Google Scholar]
- 6.Leferink N.G.H., Dunstan M.S., Hollywood K.A., Swainston N., Currin A., Jervis A.J., et al. An automated pipeline for the screening of diverse monoterpene synthase libraries. Sci Rep. 2019;9 doi: 10.1038/s41598-019-48452-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Landmann C., Fink B., Festner M., Dregus M., Engel K.H., Schwab W. Cloning and functional characterization of three terpene synthases from Lavender (Lavandula angustifolia) Arch Biochem Biophys. 2007;465:417–429. doi: 10.1016/j.abb.2007.06.011. [DOI] [PubMed] [Google Scholar]
- 8.Daramwar P.P., Rincy R., Niloferjahan S., Krithika R., Gulati A., Yadav A., et al. Transformation of (+/-)-lavandulol and (+/-)-tetrahydrolavandulol by a fungal strain Rhizopus oryzae. Bioresour Technol. 2012;115:70–74. doi: 10.1016/j.biortech.2011.11.038. [DOI] [PubMed] [Google Scholar]
- 9.Tabata J., Ichiki R.T.A. New lavandulol-related monoterpene in the sex pheromone of the grey pineapple mealybug, Dysmicoccus neobrevipes. J Chem Ecol. 2015;41:194–201. doi: 10.1007/s10886-015-0545-2. [DOI] [PubMed] [Google Scholar]
- 10.McGlinchey M.J. Organotransition metal chemistry of terpenes: syntheses, structures, reactivity and molecular rearrangements. Molecules. 2024;29:1409. doi: 10.3390/molecules29061409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Takada Y., Caner J., Kaliyamoorthy S., Naka H., Saito S. Photocatalytic transfer hydrogenolysis of allylic alcohols on Pd/TiO2: a shortcut to (S)-(+)-lavandulol. Chem Eur J. 2017;23:18025–18032. doi: 10.1002/chem.201704099. [DOI] [PubMed] [Google Scholar]
- 12.Bhosale V.A., Waghmode S.B. An efficient total synthesis of (-)-(R), (+)-(S)-Lavandulol pheromones and their derivatives through proline catalyzed asymmetric α-aminooxylation and [3,3] claisen rearrangement. ChemistrySelect. 2017;2:1262–1266. doi: 10.1002/slct.201601890. [DOI] [Google Scholar]
- 13.Bureau J.A., Oliva M.E., Dong Y., Ignea C. Engineering yeast for the production of plant terpenoids using synthetic biology approaches. Nat Prod Rep. 2023;40:1822–1848. doi: 10.1039/d3np00005b. [DOI] [PubMed] [Google Scholar]
- 14.Yang D.Q., Liang H., Li X.X., Zhang C.Y., Lu Z.W., Ma X.Q. Unleashing the potential of microbial biosynthesis of monoterpenes via enzyme and metabolic engineering. Biomater Adv. 2025;79 doi: 10.1016/j.biotechadv.2025.108525. [DOI] [PubMed] [Google Scholar]
- 15.Jongedijk E., Cankar K., Ranzijn J., van der Krol S., Bouwmeester H., Beekwilder J. Capturing of the monoterpene olefin limonene produced in Saccharomyces cerevisiae. Yeast. 2015;32:159–171. doi: 10.1002/yea.3038. [DOI] [PubMed] [Google Scholar]
- 16.Davis E.M., Croteau R. Cyclization enzymes in the biosynthesis of monoterpenes, sesquiterpenes, and diterpenes. Top Curr Chem. 2000;209:53–95. [Google Scholar]
- 17.Liu M., Chen C.C., Chen L., Xiao X., Zheng Y., Huang J., et al. Structure and function of a "Head-to-Middle" prenyltransferase: lavandulyl diphosphate synthase. Angew Chem Int Ed Engl. 2016;55:4721–4724. doi: 10.1002/anie.201600656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Demissie Z.A., Erland L.A., Rheault M.R., Mahmoud S.S. The biosynthetic origin of irregular monoterpenes in lavandula: isolation and biochemical characterization of a novel cis-prenyl diphosphate synthase gene, lavandulyl diphosphate synthase. J Biol Chem. 2013;288:6333–6341. doi: 10.1074/jbc.M112.431171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Thulasiram H.V., Erickson H.K., Poulter C.D. Chimeras of two isoprenoid synthases catalyze all four coupling reactions in isoprenoid biosynthesis. Science. 2007;316:73–76. doi: 10.1126/science.1137786. [DOI] [PubMed] [Google Scholar]
- 20.Lee J.S., Pan J., Ramamoorthy G., Poulter C.D. Structure-function studies of Artemisia tridentata farnesyl diphosphate synthase and chrysanthemyl diphosphate synthase by site-directed mutagenesis and morphogenesis. J Am Chem Soc. 2017;139:14556–14567. doi: 10.1021/jacs.7b07608. [DOI] [PubMed] [Google Scholar]
- 21.Liu P.L., Wan J.N., Guo Y.P., Ge S., Rao G.Y. Adaptive evolution of the chrysanthemyl diphosphate synthase gene involved in irregular monoterpene metabolism. BMC Evol Biol. 2012;12:214. doi: 10.1186/1471-2148-12-214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Matáeos-Fernández R., Vacas S., Navarro-Fuertes I., Navarro-Llopis V., Orzáez D., Gianoglio S. Assessment of tobacco (Nicotiana tabacum) and N. benthamiana as biofactories of irregular monoterpenes for sustainable crop rrotection. Ind Crop Prod. 2023;206 doi: 10.1016/j.indcrop.2023.117634. [DOI] [Google Scholar]
- 23.Ren Y., Liu S., Jin G., Yang X., Zhou Y.J. Microbial production of limonene and its derivatives: achievements and perspectives. Biotechnol Adv. 2020;44 doi: 10.1016/j.biotechadv.2020.107628. [DOI] [PubMed] [Google Scholar]
- 24.Christian R.W., Hewitt S.L., Nelson G., Roalson E.H., Dhingra A. Plastid transit peptides-where do they come from and where do they all belong? Multi-genome and pan-genomic assessment of chloroplast transit peptide evolution. PeerJ. 2020;8 doi: 10.7717/peerj.9772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nie S.X., Wang S.L., Chen R.Q., Ge M.Y., Yan X.G., Qiao J.J. Catalytic mechanism and heterologous biosynthesis application of sesquiterpene synthases. J Agric Food Chem. 2024;72:6871–6888. doi: 10.1021/acs.jafc.4c00030. [DOI] [PubMed] [Google Scholar]
- 26.Gerasymenko I., Sheludko Y.V., Navarro Fuertes I., Schmidts V., Steinel L., Haumann E., et al. Engineering of a plant isoprenyl diphosphate synthase for development of irregular coupling activity. Chembiochem. 2021;22:1–6. doi: 10.1002/cbic.202100465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Seino Y., Nakatani A., Nishikawa T., Yoshida C., Suematsu A., Higashimura K., et al. Evaluation of lavender varietal characteristics using three content ratios between alcohol-based fragrance ingredients and their derivatives. J Jpn Soc Hortic Sci. 2008;77:304–311. doi: 10.2503/jjshs1.77.304. [DOI] [Google Scholar]
- 28.Zada A., Dunkelblum E., Assael F., Franco J.C., da Silva E.B., Protasov A., et al. Attraction of Planococcus ficus males to racemic and chiral pheromone baits: flight activity and bait longevity. J Appl Entomol. 2008;132:480–489. doi: 10.1111/j.1439-0418.2008.01277.x. [DOI] [Google Scholar]
- 29.Hemmerlin A., Rivera S.B., Erickson H.K., Poulter C.D. Enzymes encoded by the farnesyl diphosphate synthase gene family in the big Sagebrush Artemisia tridentata ssp. spiciformis. J Biol Chem. 2003;278:32132–32140. doi: 10.1074/jbc.M213045200. [DOI] [PubMed] [Google Scholar]
- 30.Li W., Cu L., Mai J., Shi T., Lin L., Zhang Z., et al. Advances in metabolic engineering paving the way for the efficient of terpenes in yeasts. J Agric Food Chem. 2022;70:9246–9261. doi: 10.1021/acs.jafc.2c03917. [DOI] [PubMed] [Google Scholar]
- 31.Chen Y., Daviet L., Schalk M., Siewers V., Nielsen J. Establishing a platform cell factory through engineering of yeast acetyl-CoA metabolism. Metab Eng. 2013;15:48–54. doi: 10.1016/j.ymben.2012.11.002. [DOI] [PubMed] [Google Scholar]
- 32.Mai J., Li W.J., Ledesma-Amaro R., Ji X.J. Engineering plant sesquiterpene synthesis into yeasts: a review. J Agric Food Chem. 2021;69:9498–9510. doi: 10.1021/acs.jafc.1c03864. [DOI] [PubMed] [Google Scholar]
- 33.Lu Z., Peng B., Ebert B.E., Dumsday G., Vickers C.E. Auxin-mediated protein depletion for metabolic engineering in terpene-producing yeast. Nat Commun. 2021;12:1051. doi: 10.1038/s41467-021-21313-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Callari R., Meier Y., Ravasio D., Heider H. Dynamic control of ERG20 and Erg9 expression for improved casbene production in Saccharomyces cerevisiae. Front Bioeng Biotechnol. 2018;6:160. doi: 10.3389/fbioe.2018.00160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Zhou P.P., Du Y., Xu N.N., Yue C.L., Ye L.D. Improved linalool production in Saccharomyces cerevisiae by combining directed evolution of linalool synthase and overexpression of the complete mevalonate pathway. Biochem Eng J. 2020;161 doi: 10.1016/j.bej.2020.107655. [DOI] [Google Scholar]
- 36.Jiang G.Z., Yao M.D., Wang Y., Zhou L., Song T.Q., Liu H., et al. Manipulation of GES and ERG20 for geraniol overproduction in Saccharomyces cerevisiae. Metab Eng. 2017;41:57–66. doi: 10.1016/j.ymben.2017.03.005. [DOI] [PubMed] [Google Scholar]
- 37.Lv X., Zhou X., Ma J., Tao M., Liu Y., Li J., et al. Engineered Saccharomyces cerevisiae for the De Novo biosynthesis of (-)-menthol. J Fungi (Basel) 2022;8:982. doi: 10.3390/jof8090982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Zhao J., Bao X., Li C., Shen Y., Hou J. Improving monoterpene geraniol production through geranyl diphosphate synthesis regulation in Saccharomyces cerevisiae. Appl Microbiol Biotechnol. 2016;100:4561–4571. doi: 10.1007/s00253-016-7375-1. [DOI] [PubMed] [Google Scholar]
- 39.Bai X., Wang S., Zhang Q., Hu Y., Zhou J., Men L., et al. Reprogramming the metabolism of yeast for high-level production of miltiradiene. J Agric Food Chem. 2024;72:8704–8714. doi: 10.1021/acs.jafc.4c01203. [DOI] [PubMed] [Google Scholar]
- 40.Köksal M., Jin Y.H., Coate R.M., Croteau R., Christianson D.W. Taxadiene synthase structure and evolution of modular architecture in terpene biosynthesis. Nature. 2011;469:116–138. doi: 10.1038/nature09628. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
The protein sequences of candidate lavandulol synthases used in this study were provided in Table S1.
The primers used in this study were provided in Table S2.
The lavandulol production of strains containing different candidate lavandulol synthase were provided in Fig. S1.
Amino acid residues that forming hydrogen bonds with the substrate DMAPP were provided in Fig. S2.
The multiple sequence alignment result of LILPPS and similar proteins with high identity retrieved from the NCBI database was provided in Fig. S3.
Amino acid preference of LILPPS at site 89, 119, 120, 123, 249, 255, 274 was provided in Fig. S4.
Total ion chromatogram of lavandulol in final purity sample was provided in Figure S5.
1H NMR and 13C spectrum of lavandulol in CDCl3 were provided in Figs. S6 and S7.