Improved cis-Abienol production through increasing precursor supply in Escherichia coli

cis-Abienol, a natural diterpene-diol isolated from balsam fir (Abies balsamea), can be employed as precursors for the semi-synthesis of amber compounds, which are sustainable replacement for ambergris and widely used in the fragmented industry. This study combinatorially co-expressed geranyl diphosphate synthase, geranylgeranyl diphosphate synthase, Labda-13-en-8-ol diphosphate synthase and diterpene synthase, with the best combination achieving ~ 0.3 mg/L of cis-abienol. An additional enhancement of cis-abienol production (up to 8.6 mg/L) was achieved by introducing an exogenous mevalonate pathway which was divided into the upper pathway containing acetyl-CoA acetyltransferase/HMG-CoA reductase and HMG-CoA synthase and the lower pathway containing mevalonate kinase, phosphomevalonate kinase, pyrophosphate mevalonate decarboxylase and isopentenyl pyrophosphate isomerase. The genetically modified strain carrying chromosomal copy of low genes of the mevalonate with the trc promoter accumulated cis-abienol up to 9.2 mg/L in shake flask. Finally, cis-abienol titers of ~ 220 mg/L could be achieved directly from glucose using this de novo cis-abienol-producing E. coli in high-cell-density fermentation. This study demonstrates a microbial process to apply the E. coli cell factory in the biosynthesis of cis-abienol.


Scientific RepoRtS
| (2020) 10:16791 | https://doi.org/10.1038/s41598-020-73934-z www.nature.com/scientificreports/ the biosynthesis of sclareol 17 . The second step of the pathway is catalyzed by cis-abienol synthase (Cas) which pertains to class I diTPS containing several types to produce a specific end product [18][19][20] . Based on this, the cisabienol was biosynthesized in Yarrowia lipolytica by overexpressing the class II diPTS and Cas from different species 21 . Many biotechnological studies have been focused on the efficient production of desirable isoprenoids using E. coli as a host 22 . In E. coli, the isoprenoid was naturally biosynthesized through the endogenous MEP pathway 23 . For isoprenoid biosynthesis, a sufficient supply of IPP and DMAPP is essential. This has been often achieved by the expression of the idi and dxs (1-deoxy-d-xylulose 5-phosphate reductoisomerase) gene in the MEP pathway 24 . However, it has been shown that the introduction of the MVA pathway is more successful than engineering endogenous MEP pathway in E. coli, as it may avoid the natural regulatory mechanisms associated with the MEP pathway 25,26 . Michel et al. overexpressed the MVA pathway from saccharomyces cerevisiae and the enzymes responsible for the biosynthesis of sclareol to construct the sclareol biosynthetic pathway in genetically engineered E. coli and reached sclareol titers of ~ 1.5 g/L in high-cell-density fermentation 17 . Therefore, increasing the efficient supply of precursors is important to enhance cis-abienol production.
In this study, based on our previous work 27,28 , cis-abienol was produced by assembling a biosynthetic pathway in an engineering E. coli, using the heterologous MVA pathway combined with the additional of CrtE gene, Labda-13-en-8-ol diphosphate synthase (LPPS) gene from Salvia. sclarea and Cas gene from A. balsamea. The cis-abienol production of the final genetically modified strain, CM2/pCLES/pLCEC, was finally investigated under fed-batch fermentation condition. This study has begun the necessary foundations for a more sustainable route of cis-abienol production.

Results
Engineering a pathway for cis-Abienol biosynthesis. GGPP, generated from either MEP or MVA pathway, could be catalyzed by the Lpps and Cas into cis-abienol. In the previous study 28 , it has proved feasible that overexpressing the Lpps gene and sclareol synthase gene along with the native MEP pathway to produce sclareol in E. coli. In this study, the cis-abienol biosynthesis pathway was constructed by expressing the primary enzymes diphosphate synthase (IspA), CrtE, Lpps, and Cas. The biosynthesis genes (Lpps, Cas, CrtE, IspA) of cisabienol pathway were cloned into the plasmid pACYDuet-1 one by one to generate combination plasmid pLCCI. Accumulation of cis-abienol was not detected in the recombinant strain BL21 (DE3)/ pLCC for two days shake flask. We postulated that the low level of FPP may restrict the cis-abienol biosynthesis in the strain BL21(DE3)/ pLCC. Because FPP pool was required to synthesize trans-octaprenyl diphosphate (ODP) and cis-undecaprenyl trans-undecapreyl diphosphate (UDP) in E. coli, which were the precursors of ubiquinone and peptidoglycan, respectively [29][30][31] . To improve the accumulation of FPP, the diphosphate synthase (IspA) gene was added to the plasmid to enhance the efficiency of the conversion of IPP and DMAPP to FPP. Thus BL21 (DE3) carrying  www.nature.com/scientificreports/ pLCCI for two days shake flask resulted in one dominant peak in an LC-MS chromatogram ( Fig. 2A) with a concentration of 0.32 mg/L of cis-abienol (Fig. 3). Comparison of the retention time and mass-spectrum of this compound with authentic standard confirmed the production of cis-abienol ( Fig. 2A-D). The results showed that overexpressing the gene LPPS, Cas, CrtE, IspA together with the native MEP pathway was capable of synthesizing synthesize cis-abienol in E. coli.
cis-Abienol microbial toxicity. Oleoresin components, produced by conifers, most of which are a diverse of diterpenoids like cis-abienol, play a key role in the chemical defense against fungi 4 . Microbial production of cis-abienol in high titers may be restricted by its potential toxicity to the producing host. The effect of the concentration of cis-abienol on the E. coli strain was analyzed by measuring the OD 600 for 36 h in shake flasks containing 20 mL of LB medium, which showed that E. coli strains could grow well at the commercially available cis-abienol titer from 0 g/L to 2 g/L. As shown in Fig. 4, when added exogenously to the growth medium, cis-abienol shows low toxicity to BL21(DE3)/pLCCI, the lack toxicity to the producing host should enable the production of cis-abienol in higher titers.
Increasing the precursor supply for cis-Abienol biosynthesis. The low content of IPP and DMAPP also limits the efficient biosynthesis of terpenoids in E. coli. High-level polyprenyl pyrophosphate production may prove useful in producing a wide variety of compounds such as isoprene, pinene, longifolene, and sclareol 17,32-34 . One of the key strategies for constructing an efficient terpenoid biosynthesis host is to introduce a heterologous mevalonate pathway. Initially, our attempt to synthesize cis-abienol in E. coli by co-expressing genes LPPS, Cas, CrtE and IspA resulted in only a small amount of cis-abienol (~ 0.32 mg/L) in shake flask culture. To further improve production of cis-abienol, a metabolic engineering approach was adopted to increase the supply of precursor via overexpression of a heterologous mevalonate pathway into E. coli. The acetyl-CoA acetyltransferase/HMG-CoA reductase and HMG-CoA synthase from Enterococcus faecalis encoded by mvaE www.nature.com/scientificreports/ and mvaS were cloned into the low copy plasmid pCL1920 with medium-strength trc promoter to generate the recombinant plasmid pCLES. The plasmid pYJM14 containing the MVA lower pathway genes ERG12, ERG8, ERG19 and IDI1 from S. cerevisiae under the control of trc promoter (pTrcHis2B) has been constructed in the preliminary experiments in our laboratory 32 . The recombinant strain BL21 (DE3) (pCLES/pYJM14/pLCCI) carrying MVA pathway could accumulate 3.0 mg/L cis-abienol, which is about ten-fold of that (0.32 mg/L) pro-  www.nature.com/scientificreports/ duced by the control strain BL21 (DE3) (pLCCI) without MVA pathway (Fig. 3). According to the data obtained, increasing the supply of precursor of IPP and DMAPP by overexpression of MVA pathway is conducive to the cis-abienol biosynthesis. In order to investigate the effect of FPP synthase Erg20 of S. cerevisiae and different gene arrangement of Erg20 and CrtE on the synthesis of cis-abienol, the recombinant pLCCE and pLCEC were constructed and transformed into BL21(DE3)/pCLES/pYJM14 competent cells, respectively. The strain carrying FPP synthase of S. cerevisiae produced 6.1 mg/L of cis-abienol, which is twofold to the strain with native FPP synthase. Although prior reports have shown that native FPP synthase is more effective than S. cerevisiae to produce sesquiterpene in E. coli 34 , we observed that the FPP synthase from S. cerevisiae was advantageous for the synthesis of cis-abienol. The reason may be that a high concentration of FPP was toxic to E. coli cell growth, resulting in a lower level of cis-abienol in BL21(DE3)/pCLES/pYJM14/pLCCI. This result demonstrated that the FPP synthase of S. cerevisiae was more efficient than the native one for FPP supply. We also found that different order for gene Erg20 and CrtE leads to different production of cis-abienol. The strain BL21(DE3)/pCLES/pYJM14/pLCEC accumulated 8.6 mg/L of cisabienol, and the yields were 1.5 times that of strain BL21(DE3)/pCLES/pYJM14/pLCCE (Fig. 3). It is concluded that the FPP and CrtE genes order in the operon affected relative expression levels and, consequently, the yield of cis-abienol production.

Construction and characterization of chromosomal gene integration strain.
Compared with the over-expression of plasmids, the synthesis of desired products by chromosomal integrated metabolic pathways has considerable advantages in industrial production 35 . The chromosomal integration of low pathway of MVA may improve the stability of exogenous genes in cells and reduce the use of antibiotics. The lower genes of MVA pathway MVK, PMK, MVD, IDI with Gi1.2 constitutive promoter 36,37 and trc promoter were integrated into the E. coli chromosome to construct the strain CM1 carrying a chromosomal copy of Gi1.2-low genes and CM2 carrying a chromosomal copy of Trc-low genes, respectively. The integrated strain of CM1/pLCEC/pCLES resulted in a 30% decrease in cis-abienol production compared to the previous strain. In contrast, cis-abienol was accumulated up to 9.2 mg/L in the strain CM2/pLCEC/pCLES in shake flask (Fig. 3). It was 10% higher than the previous strain. The reason may be that CM1 with a synthetic promoter was too weaker to produce supplementary intermediate products IPP and DMAPP. Although a high copy plasmid (pYJM14) which contains the Erg8-ERg12-ERg19-IDI operon can provide a high copy number and stronger promoter, the replication of the plasmid and the expression of the genes may bring a heavy metabolic burden to the host. These indicate that only the suitable gene expression can increase the yield of the desired product. Due to less plasmid burden, the biomass of CM1 and CM2 was significantly higher than others strain.
Bioreactor production of cis-Abienol. In order to investigate the performance of the most efficient strain CM2/pLCEC/pCLES, fed-batch cultivation experiments were performed in a 5-L laboratory-scale bioreactor containing 3L of M9 medium plus 0.5 g/L yeast extract. Glucose was fed at an appropriate rate to maintain it lower than 1 g/L. The biomass, cis-abienol production, and by-product (acetate and mevalonate) during the fermentation process were shown in Fig. 5A,B. After 50 h of fermentation, maximum cell densities reached an OD 600 of 120, and cis-abienol titer reached a maximum of 220 ± 2 mg/L (Fig. 5A). The integration of the MVA downstream pathway into the genome may reduce the metabolic burden of the cells, obtaining higher biomass. Due to the low copy number (1 ~ 2) of plasmid (pCLES) in the upstream metabolic pathway of MVA in CM2/ pCLES/pCLEC, the accumulation of intermediate metabolite mevalonate in the culture broth was only 357 mg/L (Fig. 5B). We postulated that lower mevalonate supply may affect the biosynthesis of cis-abienol.
To provide sufficient mevalonate, the CM2/pETES/pCLEC strain containing stronger promoter (T7) and higher copy numbers (~ 50) was constructed to reinforce the upper pathway of MVA. Under the same fermentation conditions as CM2/pCLES/pCLEC, maximum cell densities reached an OD 600 of 70, cis-abienol titer reached a maximum of 200 mg/L (Fig. 5C) and the intermediate metabolite mevalonate accumulated in the culture broth and reached a maximum titer of 2200 mg/L (Fig. 5D). Due to the stronger promoter and higher copy numbers plasmid (pETES) in the upstream metabolic pathway of MVA in CM2/pCLES/pCLEC, the intermediate metabolite mevalonate increased in concentration by sixfold. However, cell biomass and the production of cis-abienol were significantly decreased. The excessive production of mevalonate in the upper pathway may exceed the catalytic capacity of the lower pathway, which leads to the accumulation of intermediate products. The metabolic unbalance between the upper and lower pathway resulted in miserable cell growth and decreased the production of cis-abienol. In our work, although CM2/pCLES/pCLEC strain enables the microbial biosynthesis of cis-abienol in high-cell-density fermentation, the cis-abienol production was relatively low. Future work should focus on optimizing expression levels of enzymes of Lpps and Cas to further increase cis-abienol production.

Discussion
Diterpenes have attracted much attention owing to their unique structures and diverse bioactivities in recent years 38 . As a typical diterpene, the expanding applications and growing demands of cis-abienol, which was raw material for Ambrox synthesis and a spice additive in cigarettes, have led to a widespread search for effective and economical manufacturing methods. Microorganisms provide a sustainable and environment-friendly alternative for the production of terpenoids, starting from simple carbon sources and, frequently, ensuring high product specificity 13 . Many biotechnological studies have focused on using E. coli as host for efficiently producing desired isoprenoids, in particular, the introduction of heterologous MVA pathway genes into E. coli has been very effective in improving productivity. In this study, E. coli was used as a host to preliminarily verify the feasibility of producing cis-abienol by microorganisms.

Scientific RepoRtS
| (2020) 10:16791 | https://doi.org/10.1038/s41598-020-73934-z www.nature.com/scientificreports/ The final reaction catalyzed by Cas should be the rate-limiting step in the biosynthesis of cis-abienol. Michel et al. overexpressed the enzymes responsible for the biosynthesis of sclareol and the MVA pathway from S. cerevisiae to construct a sclareol biosynthetic pathway in engineering E. coli, and achieved sclareol titers of about 1.5 g/L in high-cell-density fermentation. Due to the complex carbocation reaction to form bicyclic oxygencontaining diterpenoid, the activity of this enzyme is much lower than any other terpene synthase. However, in recent years, the advancements in comprehending the reaction mechanism of terpene synthase can guide the development of mutant cis-abienol enzymes with improved activity. Screening for effective cis-abienol synthase or using protein engineering strategies to improve its catalytic activity, which is currently being performed in our laboratories, have the potential to increase the productivity of this valuable diterpenes.
The low intracellular IPP and DMAPP content also restrict the efficient biosynthesis of terpenoids in E. coli. High-level yield of polyprenyl pyrophosphate has been demonstrated to be useful in the production of various compounds such as isoprene, pinene, longifolene, sclareol 17,32,33 . One of the Key strategies for constructing an efficient terpenoids biosynthesis host is introducing a heterologous mevalonate pathway. In this work, A hybrid MVA pathway was expressed in E. coli to promote the accumulation of IPP and DMAPP pools, thereby enhancing cis-abienol production to some extent. In order to improve the stability of foreign genes and reduce the use of the antibiotics, the lower genes of MVA pathway were integrated into the E. coli chromosome, and the production of cis-abienol has been further improved. It shows that further research on balancing the enzymes in MVA pathway will be beneficial for further increasing the production of cis-abienol. In the cis-abienol biosynthesis process, IPP and DMAPP are first condensed to form FPP, by FPP synthase, which is a critical enzyme in the mevalonate pathway, and then FPP condenses with IPP to form GGPP. After comparing two different FPP synthase, we established a complete pathway for the precursor biosynthesis, which further improved the yield of cis-abienol.
In this study, the best-performing strain CM2/pLCEC/pCLES was assessed in a shake flask with the M9 minimal medium and a 3L fed-batch fermentation with M9 fermentation medium. Cell density reached an optical density 2.3 and the cis-abienol concentration reached 9.2 mg/L (Fig. 3) in shake flask with M9 minimal medium, which was lower than 10.1 (cell density) and 41.1 mg/L (cis-abienol) with TB medium 39 . While the titer of cis-abienol in this study is lower than the previous report, the cell density in this study is much lower than in the last report 39 . Therefore, increasing cell density through optimization might lead to higher titers of cis-abienol. The maximum production of cis-abienol reached only 220 mg/L under fed-batch fermentation in the growth medium which was lower than 634.7 mg/L reported with two-phase cultivation in a fed-batch bioreactor. It may be that the higher content of cis-abienol in the broth hinders the spread out of the cell, thus exerting a negative www.nature.com/scientificreports/ feedback inhibition to the biosynthesis of cis-abienol. Therefore, the method of two-phase cultivation will be employed to improve the production of cis-abienol in future work. In summary, we established a complete pathway to enhance the precursor supply of cis-abienol from simple building blocks. cis-Abienol production was achieved by assembling biosynthetic genes encoding a heterologous MVA pathway, an FPP synthase from S. cerevisiae, a GGPP synthase from P. agglomerans, Lpps and Cas from A. balsamea in an engineered E. coli strain. The yield of cis-abienol was further improved by fed-batch fermentation of the engineered strain. The best-performing strain CM2/pCLES/PLCEC accumulated cis-abienol up to 220 mg/L under fed-batch fermentation conditions. Although the production of cis-abienol in this study was not as high as previously reported, our study still shows a good strategy to use E. coli cell factory as a robust and competitive platform for the synthesis of cis-abienol.

Methods
Strains, media and culture conditions. Strains and plasmids used in this study were listed in Table 1.
S. cerevisiae was used for gene cloning. The E. coli DH5α strain was used for the plasmid construction. The E. coli χ7213 was used for the chromosomal integration as the donor strain 40 , and E. coli BL21(DE3) was used for production of cis-abienol. E. coli DH5α and E. coli BL21(DE3) were cultured in Luria-Bertani (LB) broth with appropriate antibiotics added to the broth when necessary (50 μg/mL for kanamycin, 34 μg/mL for chloramphenicol, 100 μg/mL for ampicillin and 100 μg/mL for spectinomycin). S. cerevisiae was cultured in YPD medium. M9 minimal medium (15.3 g/L NaH 2 PO 4 ·12H 2 O, 3 g/L KH 2 PO 4 , 1 g/L NH 4 Cl, 0.5 g/L NaCl) was used for shake flask fermentation. Growth medium (9.8 g/L K 2 HPO 4 ·3H 2 O, 0.5 g/L yeast extract, 2 g/L MgSO 4 ·7H 2 O, 2.1 g/L citric acid monohydrate and 0.3 g/L ferric ammonium citrate) supplemented with 10 g/L glucose, and 1 mL/L trace elements (3.7 g/L (NH 4 ) 6 Mo 7 O 24 ·4H 2 O, 2.9 g/L ZnSO 4 ·7H 2 O, 24.7 g/L H 3 BO 3 , 2.5 g/L CuSO 4 ·5H 2 O, 15.8 g/L MnCl 2 ·4H 2 O) was used for fed-batch fermentation as previously described 41 . Plasmid and strain construction. All PCRs were done using PrimerSTAR Max DNA polymerase (TAKARA, Dalian, China). All of the primers used for DNA manipulation were listed in Table 2. The Labda-13-en-8-ol diphosphate synthase (LPPS, GenBank Accession No.: JQ478434.1) from S. sclarea and diterpene synthase TPS4 (Cas, GenBank Accession No.: JN254808.1) from A. balsamea were code optimized and synthesized by BGI. The CrtE gene were obtained from pAC-lyc 42 . The geranyl diphosphate/farnesyl pyrophosphate synthase gene ispA and Erg20 was obtained from E. coli and S. cerevisiae, respectively. Four genes of the enzymes LPPS, CAS, CrtE, ispA were amplified and cloned into the plasmid pACYDuet-1. The gene for LPPS was amplified by PCR with primers and cloned into BamHI and SacI sites of vector pACYDuet-1and the resultant plasmid was named pACY-Lpps. The gene for CAS was amplified by PCR with primers and cloned into BamHI and SacI sites of vector pACY-LPPS to generate recombinant plasmid pACY-LPPS-Cas. The gene crtE and ispA were amplified with primers and ligated using overlapping PCR to generate an engineering fragment, which was cloned to the plasmid pACY-LPPS-Cas to generate recombinant plasmid pLCCI. The crtE and Erg20 gene were amplified with primers and ligated using overlapping PCR to generate an engineering fragment, which www.nature.com/scientificreports/ was cloned to the plasmid pACY-LPPS-Cas to generate recombinant plasmid pLCCE. To construct pLCEC, Erg20 and CrtE were amplified with primers and ligated using overlapping PCR to generate an engineering fragment and cloned into the plasmid pLC at AatII and XhoI. To construct pTES, mvaE-mvaS was amplified from pACY-mvaE-mvaS which was constructed in our previous work and cloned into pTrchis2B at SacI and NcoI. To construct pLCES, trc-mvaE-mvaS was amplified from pTES and cloned into pCL1920 using an In-Fusion HD Cloning Kit (Takara-Clontech, Japan). To construct pETES, mvaE-mvaS was amplified from pACY-mvaE-mvaS and cloned into pET28 at SacI and NcoI.
Construction of BL21(DE3) strains with chromosomal mutation. The mutant strain was construction using suicide plasmid pRE112 as previously described 40 . To insert the lower pathway of MVA into E. coli BL21 (DE3) chromosome, a set of suicide plasmid was constructed based on the vector pRE112. For example, the flanking regions of glmS and glmU gene were amplified and linked up with each other by overlap extension PCR, and the restriction sites of NotI and XhoI were introduced at the connection point by primer design. This fragment was cloned into the vector pRE112 to generate pRE112-ΔSU. Then a NotI-KpnI fragment from pYJM14, encoding lower pathway of MVA with Trc promoter or Gi1.2 constitutive promoter, was inserted into the corresponding site of pRE112-ΔSU, and the resulting plasmid was defined as pRE112-ΔSU-Trc-low or pRE112-ΔSU-Gi1.2-low, which was used to mediate the allelic exchange. After two rounds of selection, based on the positive marker chloramphenicol resistance gene cat and negative marker levan-sucrose gene sacB from Bacillus spp. 43 , we obtained the strain CM1 carrying chromosomal copy of low genes with gi1.2 promoter and the strain CM2 carrying chromosomal copy of low genes with Trc promoter.
Shake flask cultivation. A single clone was inoculated into 3 mL of LB medium containing appropriate antibiotic and cultivated at 37 °C, 180 rpm for about 6 h. Then 1 mL of culture was used to inoculate into 500 mL saline bottle containing 100 mL M9 minimal medium with 2.0% (w/v) glucose as the main carbon source and cultivated at 37 °C, 180 rpm. 0.5 mM of IPTG was added to induce recombinant proteins expression at an OD 600 reached about 0.6. Then the temperature was shifted to 30 °C for cis-abienol production. The cell mass and cisabienol were determined after 48 h culturing. Table 2. Primers used in this study for plasmid and mutant strain construction. The restriction sites in the primers were underlined. Lpps_F_BamHI  CGC GGA TCC GAT GAC CAG CGT GAA TTT A   Lpps_R_SacI  CGC GAG CTC TTA CAC CAC CGG GCG AAA C   Cas_F_SacI  CGC GAG CTC CTT AAG GTA GCT GCA TGC A   Cas_R_HindIII  CCC AAG CTT TTA GGT CGC CGG TTC AAA C   CrtE_F_XhoI  CTC GAG ATG GTG AGT GGC AGT AAA GC   CrtE_R  AGA TCT GCC ATA TGT ATA TCT CCT TTC AGG CGA TTT TCA TGA CCG   IspA_F1  AAG GAG ATA TAC ATA TGG CAG ATC TCA TGG ACT TTC  www.nature.com/scientificreports/ cis-Abienol microbial toxicity. To investigate the cytotoxicity of cis-abienol to the producing E. coli strain, LB medium was adjusted to the six different cis-abienol concentrations ranging from 0 to 2 g/L by adding various amounts of a highly concentrated cis-abienol micellar solution (50 g/L). Subsequently, the medium (10 mL in 100 mL flask) was inoculated with 5% of preculture grown overnight in LB and incubate at 37 °C under agitation (200 rpm) for 24 h. For each concentration of abieonl, the maximum specific growth rate was determined.

Production of cis-Abienol by fed-batch.
A single colony was picked and used to inoculate 3 mL LB medium supplemented with 34 μg/mL of chloramphenicol and 100 μg/mL of spectinomycin and cultivated at 37 °C, 180 rpm for about 6 h. Afterwards, 1 mL of the culture was used to inoculate a 100 mL M9 minimal medium containing appropriate antibiotics in a 500 mL shake flask and cultivated at 37 °C, 180 rpm for about 8 h. The seed culture was used in a ratio of 3% (v/v) to inoculate a 5-L fermenter containing 3L M9 fermentation medium. Sparger aeration was employed using filtered air to maintain the dissolved oxygen (DO) concentration The fermentation process was operated under the following conditions: temperature 32 °C, pH was controlled at 7.0 ± 0.1 by automatic addition of 25% ammonia water, agitation rate at 400 rpm and airflow at 1 vvm. The agitation was associated with DO to maintain a DO concentration above 30% saturation. When the initial glucose was depleted, 60% (v/v) concentrated glucose intermittently fed into the fermenter to maintain the residual glucose below 1 g/L to control the formation of acetic acid. The culture was induced when the OD 600 was 12 by the addition of 0.25 mM IPTG unless specified according to different purposes. Samples were collected at certain intervals for cell density, residual glucose, acetic acid, mevalonate and cis-abienol analysis.
Analytical methods. Cell densities of the cultures were determined by measuring optical density at 600 nm using a spectrophotometer (Cary 50 UV-vis, Varian). Cell density samples were diluted as necessary so as to fall within the linear range. The residual glucose in the culture broth was determined using a Biochemistry Analyzer (YSI 2950D, YSI Life sciences, USA).
The concentration of acetic acid and mevalonate were determined in an Agilent 1200 Infinity series HPLC system (Agilent, Santa Clara, CA), coupled with an Aminex HPX-87H column (Bio-Rad, Hercules, CA) heated at 50 °C and a 1 cm precolumn. The mobile phase of 5 mM H 2 SO 4 was run at 0.5 mL/min. A differential refractive index detector (Agilent, Santa Clara, CA) was used for analyte detection and quantification.
cis-Abienol was determined using an Ultimate 3000 UHPLC (Thermo, USA) quipped with an ultraviolet (UV, 237) detector and a Thermo Acclaim RSLC C18 column (2.1 mm × 100 mm, 2.2 μm) based on a method adapted from Minghong Gao et al. 44 Measurement were performed at a column temperature of 30 °C and a flow rate of 0.2 mL/min. A solvent gradient of 0.1% of formic acid in water (A) and ACN (B) with 0.1% formic acid starting at 2% B to 100% B in 15 min and hold for 10 min, returning to 2% B at 25.1 min, holding these conditions at 30 min and stopping the controller. Mass spectrometry (Q-TOF) system (Bruker Daltonics, Billerica, USA) was performed in positive mode measuring ESI ionization, using the following operation parameters: capillary voltage 4500 V, dry temperature 200 °C, nebulizing gas of 1.5 bar, drying gas (N 2 , purity 99.999%) flowing of 4.5L/ min. High resolution MS and MS/MS spectra were acquired in the range 50-1300 m/z. Otof Control software was used to carry out mass spectrometer control and data acquisition and Compass Data Analysis software was applied for data analysis.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.