Enhancing Limonene production by probing the metabolic network through time-series metabolomics data

Enhancing Limonene production by probing the metabolic network through time-series metabolomics data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Enhancing Limonene production by probing the metabolic network through time-series metabolomics data Jasmeet Kaur Khanijou, Clement P. M. Scipion, Shreyash Borkar, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4285213/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 07 May, 2025 Read the published version in Metabolomics → Version 1 posted 8 You are reading this latest preprint version Abstract Introduction Limonene is a monoterpene with diverse applications in food, medicine, fuel, and material science. Recently, engineered microbes have been used to biosynthesize target biochemicals such as limonene. Objective Metabolic engineering has shown that factors such as feedback inhibition, enzyme activity or abundance may contribute to the loss of target biochemicals. Incorporating a hypothesis driven experimental approach can help to streamline the process of improving target yield. Method In this work, time-series intracellular metabolomics data from Escherichia coli cultures of a wild-type strain engineered to overproduce limonene (EcoCTs3) was collected, where we hypothesized having more carbon flux towards the engineered mevalonate (MEV) pathway would increase limonene yield. Based on the topology of the metabolic network, the pathways involved in mixed fermentation were possibly causing carbon flux loss from the MEV pathway. To prove this, knockout strains of lactate dehydrogenase(LDH) and aldehyde dehydrogenase-alcohol dehydrogenase (ALDH-ADH) were created. Results The knockout strains showed 18 to 20 folds more intracellular mevalonate accumulation over time compared to the EcoCTs3 strain, thus indicating greater carbon flux directed towards the MEV pathway thereby increasing limonene yield by 8 to 9 folds. Conclusion Ensuring high intracellular mevalonate concentration is therefore a good strategy to enhance limonene yield and other target compounds using the MEV pathway. Once high intracellular mevalonate concentration has been achieved, the limonene producing strain can then be further modified through other strategies such as enzyme and protein engineering to ensure better conversion of mevalonate to downstream metabolites to produce the target product limonene. Intracellular Metabolomics Quantitative Metabolomics Metabolic Network Limonene Mevalonate Pathway Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Monoterpenoids are a class of structurally varied natural compounds with a strong presence in industrial applications ( 1 ). Limonene, a cyclized monoterpene, is assumed as GRAS (generally recognized as safe) by the US Food and Drug Administration ( 2 ). Limonene and its derivatives, such as carveol, menthol, and α-terpineol, have broad applications in food and beverage, pharmaceuticals, cosmetics, biomaterials, and biofuels due to their pleasant fragrance and physicochemical properties ( 3 – 5 ). Limonene has typically been extracted from plant biomass, where availability could be affected by variations in climate and agricultural land ( 6 ). As limonene’s structure has a chiral centre, it is found in nature as two enantiomers, (D)-limonene and (L)-limonene. (D)-limonene is usually obtained through cold pressing citrus peels and pulps ( 7 ) while (L)-limonene is found in essential oils of other plant species ( 8 ). With advancements in metabolic engineering and synthetic biology, microbes can be utilised as an alternative approach to sustainably produce high-value natural products, including limonene. However, the carbon yields for limonene biosynthesis are relatively low for economically feasible bioprocesses ( 9 ). The low productivity in limonene biosynthesis is due to various challenges such as: the inefficiency of enzymes for biosynthesis pathways, minimal metabolic fluxes directed towards limonene, intrusion of the native metabolism of microbes with complicated heterologous biosynthesis pathways, the absence of appropriate conditions for the heterologous expression of optimized enzymes and the cytotoxicity of limonene affecting microbial chassis ( 10 ). The biosynthesis of limonene in Escherichia coli can arise from two major terpenoid biosynthetic pathways: the native deoxyxylulose 5-phosphate (DXP) pathway and the heterologous mevalonate-dependent (MEV) pathway (Fig. 1 ). Both the DXP and MEV pathways produce two isomeric isoprene metabolites, namely isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), which are precursors for terpenoid production ( 11 ). As such, previous attempts to obtain high terpenoid yields, required a good balance of the availability of co-factors, energy demands, and carbon flux in the DXP and/or MEV pathways in E. coli ( 12 – 14 ). Furthermore, as the DXP and MEV pathways obtain their carbon from central carbon metabolism, these pathways compete with many other biochemical reactions to produce required precursors towards limonene production ( 15 ). The major biochemical reactions in E. coli leading towards eventual limonene production are illustrated in Fig. 1 , which include competing pathways for carbon flux such as the tricarboxylic acid (TCA) cycle, mixed fermentation pathways, and pentose phosphate pathway. By understanding such a metabolic network topology, in conjunction with intracellular metabolomics data, attempts could be made to improve limonene synthesis in bacterial bio-factories. Furthermore, as illustrated in Fig. 1 , there are various co-factors involved in the biochemical reactions in the metabolic network, which could also be sources of potential bottlenecks when the co-factors become limiting. Previous work has shown that limonene biosynthesis using the DXP pathway was limited by the low concentrations of terpenoid precursors produced, particularly geranyl pyrophosphate (GPP) ( 16 ). The overproduction of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) could be easily executed through the alteration of prenyl transferase and terpene synthase. However, the production of monoterpenes using microbes has been restricted as many monoterpenes are toxic and volatile ( 17 ), and the heterologous expression of GPP synthase and monoterpene synthases have been poor ( 16 , 18 ). In one study, a dedicated GPP synthase was unable to produce farnesyl diphosphate (FPP) that was instrumental in enhancing the yield of limonene ( 19 ). These strategies to improve limonene yield have focused mainly on the downstream enzymes or metabolites of the metabolic network. Previous literature focused more on downstream metabolites of the MEV pathway to improve limonene yield ( 16 , 18 , 19 ). There exists a gap in knowledge regarding the impact of upstream metabolites in improving limonene yield by understanding the network topology. In this study, we hypothesized the need to unblock carbon flux channelling towards MEV pathway by optimising upstream metabolic fluxes to enhance limonene yield in E. coli cell cultures. A wild-type E. coli strain (K-12 MG1655) was engineered to overproduce L-limonene by heterologous expression of the MEV pathway and a limonene synthase through plasmid pJBEI-6409 ( 19 ), resulting in an EcoCTs3 strain, based on which knockout strains were created. Following a rationale experimental design approach, we traced the metabolic topology of the key E. coli pathways involved from glycolysis, TCA cycle, pentose phosphate down to the MEV and DXP pathways, by measuring time-series concentrations of important intracellular and extracellular metabolites. A fast filtration and rapid quenching in liquid nitrogen sampling methodology was utilised coupled to C-18 liquid chromatography time of flight mass spectrometry (LC-TOF-MS) ( 20 , 21 ), whereby the column chemistry allowed both the hydrophilic intracellular metabolites upstream in central metabolism and the more hydrophobic secondary pathway intracellular metabolites to be analyzed simultaneously in negative mode. By studying the intracellular metabolomics data obtained from the EcoCTs3 strain and the metabolic network topology, attempts were made to improve the limonene yield through a systematic hypothesis driven approach. First, the carbon source was changed from glucose to fructose to improve intracellular metabolite concentrations in upstream fluxes. Second, to enhance carbon flux flow towards the MEV pathway to improve limonene production, two mutant strains were created, with the gene knockouts of adhe and ldh , which are genes involved in the mixed fermentation pathways. The mutant strains increased limonene yield by 8 to 9 folds through the elimination of pathways competing for the carbon flux. 2. Materials and Methods All chemicals, standards, solvents, and media components used in this study were obtained from Sigma-Aldrich (St. Louis, MO), Fisher Scientific (Fair Lawn, NJ, USA), or VWR (Radnor, PA, USA) unless otherwise noted. Water was deionised and filtered by Sartorius Arium Pro VF Type 1 water system (18.2 MΩ, 0.2 µm). 2.1. Strain construction The limonene producing engineered wild-type strain was created by transforming pJBEI-6409 plasmid which was a gift from Taek Soon Lee (Addgene plasmid #47048 ; http://n2t.net/addgene:47048 ; RRID:Addgene_47048) ( 19 ) by heat shock in E. coli K-12 MG1655 (F- lambda- ilvG - rfb-50 rph-1) forming the strain EcoCTs3. To create the ALDH-ADH and LDH knockouts, the deletion of adhE (UniProt P0A9Q7) and ldhA (UniProt P52643) genes from E. coli MG1655 strain was executed using CRISPR Cas9-assisted recombineering method as described previously ( 22 ). For the successful knockout strains, the ldhA and adhE regions were PCR amplified and sequenced to confirm the deletion of the genes. The strains were stored in 40% glycerol at -80°C before further use. For limonene production experiments, the pJBEI-6409 plasmid was transformed by heat shock in ΔldhA and ΔadhE strains. Competent cells were prepared using the Mix & Go! E. coli Transformation Kit (Zymo Chem) from the ldhA and adhE deletion strains, following manufacturer’s specifications. pJBEI-6409 transformed strains ( E. coli K-12 MG1655, ΔldhA and ΔadhE ) were each plated on a chloramphenicol selective Luria-Bertani agar plate and incubated overnight at 37°C. 2.2 Growth conditions After the prepared E. coli strains were streaked onto plates and incubated overnight, one colony was picked from each plate. Each colony was inoculated into 5 mL of Luria-Bertani medium consisting of 10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl, where 30 µg/mL chloramphenicol was added and grown overnight at 37 o C and 220 rpm. Each cell pellet was washed and re-suspended in 50 mL M9 medium adapted from a previous study ( 23 ), where the M9 medium consisted of 12.7 g/L Na 2 HPO 4 .7H 2 O, 3.1 g/L KH 2 PO 4 , 1 g/L NH 4 Cl, 0.5 g/L NaCl, 0.25 g/L MgSO 4 .7H 2 O, 15 mg/L CaCl 2 .2H 2 O, 8.1 mg/L FeCl 3, 0.89 mg/L MnCl 2 .4H 2 O, 1.7 mg/L ZnCl 2, 0.34 mg/L CuCl 2, 0.6 mg/L CoCl 2 .6H 2 O, 0.51 mg/L Na 2 MoO 4, 10 g/L glucose, and incubated overnight at 30 o C and 220 rpm. Each strain was stored in 40% glycerol at − 80 o C. Pre-cultures of each strain were then prepared by adding 100 µL of each strain stored in glycerol to 50 mL M9 medium in 250 mL flasks, which were left overnight at 30°C at 220 rpm. For experiments using mixtures of fructose and glucose as carbon sources, the carbon source was made up to 10 g/L using 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% fructose. To collect time-series metabolomics data, multiple cell culture flasks were prepared for duplicate biological samples for each time-point. Cell cultures of each strain were prepared by adding 100 µL of pre-cultures to 50 mL M9 medium in 250 mL flasks with incubation at 30°C and 220 rpm. Upon reaching an optical density of 1 at 600 nm, isopropyl β-d-1-thiogalactopyranoside (IPTG) was added, resulting in a concentration of 25 µM. A dodecane overlay of 5 mL was added to cell cultures to allow trapping of the volatile secreted limonene ( 19 ). The flasks of cell cultures were left at 30°C and 220 rpm. The flasks were sacrificed in duplicates at 2 h, 3 h, 6 h, 7 h, 24 h, and 25 h post-IPTG induction to profile the concentrations of the intracellular, extracellular, and secreted limonene. 2.3. Metabolomics extraction and analysis 2.3.1 Secreted Limonene and Extracellular Metabolites Dynamic sampling at time points 2 h, 3 h, 6 h, 7 h, 24 h and 25 h post IPTG induction was performed. Specifically, duplicate E. coli cell cultures of 50 mL were sacrificed and centrifuged for 10 mins at 3000 rpm. The dodecane layers containing secreted limonene were removed and stored at -80 o C prior to quantitative analysis. Limonene samples were diluted in ethyl acetate and run on an Agilent 7890B gas chromatography mass spectrometry (GC-MS) system with a DB-5 column (5%-phenyl-methylpolysiloxane, 30 m x 0.25 mm ID x 0.25 µm, Agilent Technologies, USA). For each run, 1 µL sample was injected with a 10:1 split ratio and 10 mL/min split flow. The GC oven temperature program was set at 40°C for 3 min, followed by a 10°C/min ramp to 100 o C and another 60°C/min ramp to 220°C with a hold time of 2 min. The injector and MS transfer line temperatures were at 250°C and 280°C, respectively. The MS was operated in selected ion-monitoring (SIM) mode using ions of m/z 136, 68 and 93, representing the molecular ion and two abundant fragmental ions of limonene. Calibration standards were run for limonene prepared in ethyl acetate from 0.05 µg/mL to 10 µg/mL. Limonene concentrations were determined using the Agilent Quantitative software. The media supernatants after centrifugation were stored at -80 o C prior to extracellular metabolites quantitative analysis. Subsequently after thawing these, 1 mL of each supernatant sample was filtered using a polyamide filter prior to analysis using an Agilent 1200 high performance liquid chromatography (HPLC) system with a Bio-rad Aminex HPX-87H column (300 x 7.8 mm) and 1260 Infinity II Refractive Index Detector (RID). For each run, 5 µL of sample was injected with an isocratic gradient using 0.01 N sulphuric acid and a 0.6 mL/min flow rate for 28 min. The column temperature was set at 35°C, while the RID temperature was at 30°C with positive polarity. Calibration mixtures of the following range: glucose (0.5 to 80 g/L), lactic acid (0.125 to 8 g/L), acetic acid (0.125 to 80 g/L), and ethanol (0.5 to 80 g/L) were used to construct calibration curves to establish linearity and the concentrations of each extracellular metabolite in the prepared samples. Cell pellets were dried in the oven overnight at 80°C prior to weighing with the Mettler Toledo XSE105 precision balance. 2.3.2 Intracellular Metabolites To measure intracellular metabolites, E. coli cell cultures were sacrificed in duplicates at 2 h, 3 h, 6 h, 7 h, 24 h, and 25 h after IPTG induction. Cell cultures were subjected to a modified protocol from the fast filtration and quenching in liquid nitrogen previously described ( 20 , 21 ). In brief, 10 mL of cell culture were filtered through a 0.2 µm polyamide membrane filter (Sartorius, Goettingen, Germany) and washed with 5 mL of wash solution (12.7 g/L Na 2 HPO 4 .7H 2 O, 3.1 g/L KH 2 PO 4 , 1 g/L NH 4 Cl, 0.5 g/L NaCl). The membrane filter containing the cells was folded in aluminum foil and rapidly plunged into liquid nitrogen for rapid quenching of biochemical reactions. The aluminum foil packages with membrane filters were stored at -80 o C until intracellular metabolites extraction. For the extraction of intracellular metabolites, membranes were removed from the aluminum foil packages and placed into 5 mL of extraction solvent consisting of methanol, acetonitrile, and water (4:4:2) and left in an ice bath for 10 mins. After this, the membranes were vortexed for 1 min followed by 3 rounds of sonication for 3 mins each time, with placement in an ice bath after each round of sonication for 1 min. The extracts were decanted into glass tubes and spiked with 20 µL internal standard mixture consisting of 50 µg/mL mevalonic acid-d3 (MVA-d3) and 10 µg/mL thymolphthalein monophosphate (TMP) prepared in methanol: 10 mM ammonium hydroxide mixture (7:3). The extracts were then subjected to a vacuum concentrator and reconstituted with 200 µL methanol: 10 mM ammonium hydroxide mixture (7:3) and filtered prior to analysis via liquid chromatography mass spectrometry. For quantitation of intracellular metabolites, an Agilent 6230 time of flight-mass spectrometer (TOF-MS) with a Dual Agilent Jet Stream (AJS) ion source was coupled with an Agilent ultra-performance liquid chromatography (UPLC) 1290 system and a Waters Acquity UPLC BEH C18 (2.1 x 150 mm, 1.7 µm) column with a VanGuard pre-column (2.1 x 5 mm). The chromatographic method used was adapted from previous studies ( 20 , 21 ), with appropriate modification. For each run, 2 µL of sample was injected and a dual mobile phase system was utilized with mobile phase A as 5 mM ammonium formate in water (pH 9.5) and mobile phase B as 5 mM ammonium formate (pH 9.5) in acetonitrile: water (9:1). The following solvent gradient was used: start with 100% mobile phase A from 0-3.5 min with 0.1 mL/min flow rate to 100% mobile phase B at 12 min and held for 8 min with increased flow rate to 0.5 mL/min. At 20 min, the system was recalibrated back to 100% mobile phase A for 5 min and held for another 5 min. The column temperature was kept constant at 35 o C. Negative electrospray ionisation was used with the following TOF settings: Gas temperature, 325°C; Gas flow, 11 L/min; Nebuliser pressure, 35 psi; Sheath gas temperature, 375°C; Sheath gas flow, 11 L/min; Vcap voltage, 3500 V; Nozzle voltage, 500 V; Skimmer, 65; OctopoleRFPeak, 750; Scan rate, 2 spectra/s. The fragmentor voltage was varied throughout each 35 min sample analysis: 2–7.5 min, 140 V; 7.5–15 min, 100 V, 140 V and 150 V. UPLC flow diversions were as follows: 0–2 min to waste, 2–15 min to TOF-MS, and 15–35 min to waste. Standards for quantitation of intracellular metabolites were prepared in methanol: 10 mM ammonium hydroxide mixture (7:3). The following intracellular metabolites were measured DXP (1-deoxy-D-xylulose-5-phosphate), DHAP + GAP (pool of dihydroxyacetone phosphate and glyceraldehyde-3-phosphate), F16BP (fructose-1,6-biphosphate), FPP (farnesyl diphosphate), GPP (geranyl diphosphate), MVA (mevalonate), MVAP (5-phosphomevalonate), PYR (pyruvate), R5P + Ru5P + X5P (pool of ribose-5-phosphate, ribulose-5-phosphate, and xylulose-5-phosphate) and G6P + F6P (pool of glucose-6-phosphate and fructose-6-phosphate). Different concentrations of intracellular intermediates were used to prepare calibration mixtures for construction of calibration curves: DHAP (DHAP + GAP pool) – 0.04 to 5 µg/mL; DXP – 0.04 to 10 µg/mL; F6P (F6P + G6P pool) – 0.04 to 10 µg/mL; F1,6BP – 0.04 to 6 µg/mL; MVA – 0.04 to 10 µg/mL; R5P (R5P + Ru5P + X5P pool) – 0.04 to 10 µg/mL; PYR – 0.04 to 1.5 µg/mL; MVAP – 0.01 to 0.3 µg/mL; GPP – 0.05 to 1.5 µg/mL; FPP – 0.05 to 1.5 µg/mL. Each calibration mixture was made up to 100 µL with 10 µL of internal standard mixture (50 µg/mL MVA-d3 and 10 µg/mL TMP) added. Metabolite quantitation was performed using the Agilent Masshunter Workstation Quantitative Analysis software for TOF. 3. Results and Discussion 3.1 Time-series intracellular metabolites from engineered wild-type strain EcoCTs3 The wild-type E. coli strain engineered to overproduce L-limonene (EcoCTs3) was cultured and harvested at aforesaid time-points after IPTG induction where the metabolomics data from their targeted intra-/ extra-cellular metabolite analyses via GC-MS and LC-ToF-MS were collated. Figures 2 A and 2 B show the data from the EcoCTs3 strain grown in glucose where it was observed that many of the intracellular metabolites such as G6P and F6P (pooled), R5P, Ru5P and X5P (pooled), DHAP and GAP (pooled), PYR, DXP, MVA, FPP were found to generally decrease from 2 h to 7 h post-IPTG induction followed by an accumulation 24 h to 25 h after IPTG induction. The decreased concentrations of such intracellular metabolites corresponded to increased limonene production from 2 h to 7 h post-IPTG induction as shown in Fig. 3 A. From 24 h to 25 h post-IPTG induction, the dry cell weights of the cell pellets from 50 mL cell cultures were relatively constant at 83.3 ± 2.2 mg, indicating the stationary phase with negligible culture growth (Figure S1 , Supplementary section). This could explain the observed accumulation of intracellular metabolites due to diminished cell growth and possibly reduced efficiencies of the biochemical reactions occurring in the metabolic network to produce limonene. The presence and quantitation of intracellular FPP (Fig. 2 B) showed that there was a loss of carbon flux towards limonene production, as this metabolite is formed from GPP and IPP (Fig. 1 ). With less carbon contribution to the immediate precursor metabolite to limonene, GPP, the limonene yield would not be optimal. The presence of both intracellular GPP and FPP as observed in Fig. 2 B could be due to their formations being catalyzed by the same prenyl diphosphate synthase, the enzyme farnesyl diphosphate synthase (IspA) ( 24 , 25 ). FPP production is favoured in E. coli as it is required in the formation of peptidoglycans which are important constituents of cell wall ( 26 ) and components of the respiratory chain ( 27 , 28 ), where its accumulation 24 h to 25 h after IPTG induction further indicates diminished cell growth in the cultures. To prevent the loss of carbon flux towards FPP, Alonso-Gutierrez et. al. ( 19 ) improved the titre of limonene through the expression of a heterologous dedicated GPP synthase which produced only GPP, unlike the native IspA, producing both GPP and FPP. The redirection of carbon flux towards GPP and preventing its loss to FPP helped in improving the yield of the target. Using this reported finding, it is therefore possible to redirect carbon flux towards the target by eliminating competing biochemical reactions, whether such reactions occur upstream or downstream of the MEV pathway. The metabolic network in Fig. 1 shows the importance of the metabolite pyruvate (PYR) in relation to the other biochemical pathways. PYR is formed as the end product of glycolysis, and is channelled into the fermentation fluxes, the DXP pathway, and it is the precursor to the intermediate metabolite, acetyl coenzyme A (AcCoA) involved in the TCA and MEV pathway. The cultures of the EcoCTs3 strain had a drastic decrease in the intracellular concentration of PYR by about 41 folds, when compared to the pool of the first two metabolites in the glycolysis pathway, i.e. G6P and F6P at the same time point of 2 h post IPTG induction (Figs. 2 A and 2 B). With such a significant concentration reduction from the G6P + F6P pool to PYR, we deduced that there would be a low amount of carbon flux entering the MEV pathway, resulting in the low production of limonene observed in the EcoCTs3 strain thus far. In principle, improvements in limonene production could be achieved by increasing metabolite concentrations upstream of the metabolite PYR in the metabolic network. Comparing F16BP at 2 h and 25 h post IPTG shows its concentration decreased by six times compared to the pool of G6P and F6P at 2 h point, and it steadily decreased until 25 h (Fig. 2 A). Therefore, we hypothesized increasing the concentrations of F16BP by channelling more flux from G6P + F6P could be a strategy to eventually improve the carbon flux going towards limonene production. 3.2 Attempting to improve intracellular F16BP concentration by changing carbon source In our attempts to increase the intracellular F16BP concentration, fructose was used as a possible carbon source alternative to glucose. From the metabolite network in Fig. 1 , and based on previous work ( 29 ), fructose can be converted to either F1P (fructose-1-phospahte) or F6P prior to forming F16BP. When varying proportions of fructose (0–100%) to glucose were tested in the culture medium, the highest amount of limonene was produced by growing the EcoCTs3 E. coli strain in 100% glucose (G100) followed by a mixture of 30% glucose and 70% fructose (G30F70), as shown in Fig. 3 A (Figure S2 represents the full dataset, Supplementary section). However, attempts with higher proportion of fructose to improve the initial intracellular F16BP concentration did not occur (Fig. 3 B), as glucose was still the main driver of F16BP production with possibly diauxic growth. Glucose passively diffuses into E. coli , where inner membrane transport followed by phosphorylation occurs ( 30 ). This process utilises the phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS), which consists of multiple proteins in the phospho-relay process that are involved in the import and concurrent phosphorylation of carbohydrates ( 31 ). With the presence of glucose in the cell culture medium, there is enhanced glucose transport activity by bacteria due to induced expression of ptsG from the phosphotransferase system ( 32 , 33 ). This corroborates with the findings in Fig. 3 , where glucose was still the best substrate for limonene production. With no improvement in intracellular F16BP and therefore limonene concentrations through mixtures of glucose and fructose as carbon source, we next hypothesized potential improvement of re-directing carbon flow towards MEV pathway by curtailing the loss of carbon flux to other pathways, and in so doing, enhance limonene yield. 3.3 Generating mutant strains to direct carbon flux towards MEV pathway The second hypothesis tested for upstream flux optimisation to improve limonene yield is to direct more carbon flux into the MEV pathway. One approach is to allow more carbon flux to flow towards PYR since it is a key metabolite involved in numerous pathways, and at the same time curtail PYR losses through pathways that do not lead to limonene production. Mixed fermentation is one common route where there is loss of carbon flux due to the formation of lactic acid and ethanol. To eliminate the mixed fermentation pathways, two mutant strains were prepared. One strain had the ldh gene removed, thereby representing a LDH knockout, while a second strain had the adhE gene removed, resulting in an ADH-ALDH knockout, as the adhE gene encodes for both enzymes ADH and ALDH ( 34 , 35 ). The outcome of limonene production of these mutant strains is shown in Fig. 4 . When compared to the EcoCTs3 strain, these knockout strains were found to substantially improve limonene yield by 8 to 9 folds (Fig. 4 A). This enhanced limonene production was correlated to the 18 to 20 fold increase in intracellular MVA measured in the knockout strains, as illustrated in Fig. 4 B. There was also greater accumulation of intracellular MVA compared to some other metabolites at 24 h to 25 h post-IPTG for these mutant strains, as observed in Figs. 5 A and 5 B for the LDH knockout strain and Figs. 6 A and 6 B for the ALDH-ADH knockout strain. MVA was observed at the later time points (Figs. 5 A and 6 A) likely because the enzyme mevalonate kinase (MK) could not efficiently convert MVA to MVAP. Since this biochemical reaction requires the presence of ATP (adenosine triphosphate), it could suggest the availability of ATP becomes a limiting factor in the biochemical synthesis of limonene. Such reduced ATP availability has also been observed in previous studies in various microbes, thereby affecting the productivity of the MEV pathway in generating isoprenoids ( 13 , 14 ). Furthermore, the observation of high levels of intracellular MVA could be explained by previous work, in which downstream intermediates such as IPP, DMAPP, FPP, and GPP have been found to inhibit MK activity through competitive binding to the ATP-binding site of MK ( 36 – 39 ). With the competitive inhibition of MK, there would be less MVAP produced and more intracellular MVA accumulated, as observed in Figs. 5 A and 5 B for the LDH knockout strain and Figs. 6 A and 6 B for the ALDH-ADH knockout strain. Furthermore, in another study using targeted proteomics, the production of enzymes MK and phosphomevalonate kinase (PMK) were found to be poor ( 40 ). These compounding factors make it difficult for the accumulation of intracellular metabolites downstream of MVA such as MVAP and GPP (Figs. 5 B and 6 B) and explains the poor conversion efficiency of MVA to limonene, as observed from the intracellular data, where the 18 to 20 fold increase in MVA did not result in the same fold increase of limonene (Fig. 4 ). The LDH and ALDH-ADH knockout strains also showed loss of flux to FPP (Figs. 5 B and 6 B) due to the native enzyme IspA catalyzing its formation. Redirecting the flux to GPP and preventing its loss to FPP can be executed through the expression of an improved heterologous GPP synthase ( 19 ) and improving limonene synthase activity, which could further improve limonene titres for both the knockout strains. 4. Conclusion When engineering bacterial strains to enhance the yield of target compounds, combining intracellular time-series metabolomics data together with the understanding of the metabolic network can be very useful. Incorporating a hypothesis-driven experimental approach with the idea of using different strategies such as changing the carbon source and eliminating competing pathways that result in flux loss can aid in achieving the desired goal of improving target yield whilst streamlining the workflow. From this study, enhancing the concentration of intracellular MVA is a key step in improving the yield of limonene and possibly other target compounds in the MEV pathway. Once a strain producing high intracellular MVA has been identified, further fine-tuning of the strain can be executed to improve the target yield. The presence of bottlenecks downstream of MVA in the MEV pathway could possibly be due to the lack of ATP to metabolize MVA to MVAP and the feedback-inhibition of MK. To further minimise the bottlenecks observed downstream of the metabolite MVA in the engineered MEV pathway, homologous enzymes from alternative organisms could be utilised ( 41 ). For instance, feedback-resistant MK homologs have been found to uphold high activity in the presence of DMAPP, IPP, GPP, FPP, and MVAPP ( 39 , 42 , 43 ). These homologs can reduce MVA accumulation, thereby improving titres. Furthermore, protein engineering can also enhance substrate affinity of feedback-resistant MK ( 44 ) and it has been used on enzyme isopentenyl diphosphate isomerase (IDI) to improve its catalytic activity and thereby increase titre levels ( 45 ). Declarations Author contributions J. K. K. carried out culturing and metabolomics experiments on the EcoCTS03 and modified strains. C. P. M. S., S. B., and X. C. designed and prepared the modified strains. J. K. K. and W. C. designed the study and wrote the paper. Conflict of interest The authors declare no competing interests. Acknowledgements This work was supported by the Intra-Create Thematic Grant “Cities” (grant number: NRF2019-THE001-0007) under the EcoCTs project. The EcoCTs research project is supported by the National Research Foundation, Prime Minister’s Office, Singapore, under its campus for Research Excellence and Technological Enterprise (CREATE) programme. In addition, we are thankful to Dr. Floriant Bellvert, Hanna Kulyk, and Cecilia Berges from MetaToul (Metabolomics & Fluxomics Facilities, Toulouse, France) for their experimental guidance and insights. We would also like to acknowledge Dr. Kumar Selvarajoo from the Bioinformatics Institute (BII), A*STAR, Singapore for technical advice and discussion. References Wojtunik-Kulesza KA, Kasprzak K, Oniszczuk T, Oniszczuk A. Natural Monoterpenes: Much More than Only a Scent. Chem Biodivers. 2019;16(12):e1900434. Mamidipally PK, Liu SX. First approach on rice bran oil extraction using limonene. European Journal of Lipid Science and Technology. 2004;106(2):122-5. Hąc-Wydro K, Flasiński M, Romańczuk K. Essential oils as food eco-preservatives: Model system studies on the effect of temperature on limonene antibacterial activity. Food Chem. 2017;235:127-35. Felipe LdO, Oliveira AMd, Bicas JL. Bioaromas – Perspectives for sustainable development. Trends in Food Science & Technology. 2017;62:141-53. Thomsett MR, Moore JC, Buchard A, Stockman RA, Howdle SM. New renewably-sourced polyesters from limonene-derived monomers. Green Chemistry. 2019;21(1):149-56. Ciriminna R, Lomeli-Rodriguez M, Demma Carà P, Lopez-Sanchez JA, Pagliaro M. Limonene: a versatile chemical of the bioeconomy. Chem Commun (Camb). 2014;50(97):15288-96. Ibáñez MD, Sanchez-Ballester NM, Blázquez MA. Encapsulated Limonene: A Pleasant Lemon-Like Aroma with Promising Application in the Agri-Food Industry. A Review. Molecules. 2020;25(11). Kvittingen L, Sjursnes BJ, Schmid R. Limonene in Citrus: A String of Unchecked Literature Citings? Journal of Chemical Education. 2021;98(11):3600-7. Sun C, Theodoropoulos C, Scrutton NS. Techno-economic assessment of microbial limonene production. Bioresource Technology. 2020;300:122666. Ren Y, Liu S, Jin G, Yang X, Zhou YJ. Microbial production of limonene and its derivatives: Achievements and perspectives. Biotechnology Advances. 2020;44:107628. Zhao L, Chang WC, Xiao Y, Liu HW, Liu P. Methylerythritol phosphate pathway of isoprenoid biosynthesis. Annu Rev Biochem. 2013;82:497-530. Ward VCA, Chatzivasileiou AO, Stephanopoulos G. Metabolic engineering of Escherichia coli for the production of isoprenoids. FEMS Microbiology Letters. 2018;365(10). Gruchattka E, Hädicke O, Klamt S, Schütz V, Kayser O. In silico profiling of Escherichia coli and Saccharomyces cerevisiae as terpenoid factories. Microb Cell Fact. 2013;12:84. Diner BA, Fan J, Scotcher MC, Wells DH, Whited GM. Synthesis of Heterologous Mevalonic Acid Pathway Enzymes in Clostridium ljungdahlii for the Conversion of Fructose and of Syngas to Mevalonate and Isoprene. Applied and Environmental Microbiology. 2018;84(1):e01723-17. Rohmer M, Seemann M, Horbach S, Bringer-Meyer S, Sahm H. Glyceraldehyde 3-Phosphate and Pyruvate as Precursors of Isoprenic Units in an Alternative Non-mevalonate Pathway for Terpenoid Biosynthesis. Journal of the American Chemical Society. 1996;118(11):2564-6. Carter OA, Peters RJ, Croteau R. Monoterpene biosynthesis pathway construction in Escherichia coli. Phytochemistry. 2003;64(2):425-33. Dunlop MJ, Dossani ZY, Szmidt HL, Chu HC, Lee TS, Keasling JD, et al. Engineering microbial biofuel tolerance and export using efflux pumps. Mol Syst Biol. 2011;7:487. Reiling KK, Yoshikuni Y, Martin VJ, Newman J, Bohlmann J, Keasling JD. Mono and diterpene production in Escherichia coli. Biotechnol Bioeng. 2004;87(2):200-12. Alonso-Gutierrez J, Chan R, Batth TS, Adams PD, Keasling JD, Petzold CJ, Lee TS. Metabolic engineering of Escherichia coli for limonene and perillyl alcohol production. Metab Eng. 2013;19:33-41. Ng P, Khoo LW, Thong A, Chew W. Optimization of extraction conditions for LC-ToF-MS analysis of mevalonate pathway metabolites in engineered E. coli strain via statistical experimental designs. Talanta. 2023;254:124182. Castaño-Cerezo S, Kulyk-Barbier H, Millard P, Portais JC, Heux S, Truan G, Bellvert F. Functional analysis of isoprenoid precursors biosynthesis by quantitative metabolomics and isotopologue profiling. Metabolomics. 2019;15(9):115. Shukal S, Lim XH, Zhang C, Chen X. Metabolic engineering of Escherichia coli BL21 strain using simplified CRISPR-Cas9 and asymmetric homology arms recombineering. Microbial Cell Factories. 2022;21(1):19. Wada K, Toya Y, Banno S, Yoshikawa K, Matsuda F, Shimizu H. 13C-metabolic flux analysis for mevalonate-producing strain of Escherichia coli. Journal of Bioscience and Bioengineering. 2017;123(2):177-82. Ogura K, Koyama T. Enzymatic Aspects of Isoprenoid Chain Elongation. Chem Rev. 1998;98(4):1263-76. Ku B, Jeong J-C, Mijts BN, Schmidt-Dannert C, Dordick JS. Preparation, Characterization, and Optimization of an In Vitro C 30 Carotenoid Pathway. Applied and Environmental Microbiology. 2005;71(11):6578-83. Apfel CM, Takács B, Fountoulakis M, Stieger M, Keck W. Use of genomics to identify bacterial undecaprenyl pyrophosphate synthetase: cloning, expression, and characterization of the essential uppS gene. J Bacteriol. 1999;181(2):483-92. Okada K, Minehira M, Zhu X, Suzuki K, Nakagawa T, Matsuda H, Kawamukai M. The ispB gene encoding octaprenyl diphosphate synthase is essential for growth of Escherichia coli. Journal of Bacteriology. 1997;179(9):3058-60. Saiki K, Mogi T, Anraku Y. Heme O biosynthesis in Escherichia coli: the cyoE gene in the cytochrome bo operon encodes a protoheme IX farnesyltransferase. Biochem Biophys Res Commun. 1992;189(3):1491-7. Kornberg HL. Routes for fructose utilization by Escherichia coli. J Mol Microbiol Biotechnol. 2001;3(3):355-9. Kotrba P, Inui M, Yukawa H. The ptsI Gene Encoding Enzyme I of the Phosphotransferase System of Corynebacterium glutamicum. Biochemical and Biophysical Research Communications. 2001;289(5):1307-13. Postma PW, Lengeler JW, Jacobson GR. Phosphoenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiological Reviews. 1993;57:543 - 94. Plumbridge J. Regulation of gene expression in the PTS in Escherichia coli: the role and interactions of Mlc. Current Opinion in Microbiology. 2002;5(2):187-93. Yao R, Hirose Y, Sarkar D, Nakahigashi K, Ye Q, Shimizu K. Catabolic regulation analysis of Escherichia coli and its crp, mlc, mgsA, pgi and ptsG mutants. Microbial Cell Factories. 2011;10(1):67. Bertsch J, Siemund AL, Kremp F, Müller V. A novel route for ethanol oxidation in the acetogenic bacterium Acetobacterium woodii: the acetaldehyde/ethanol dehydrogenase pathway. Environmental Microbiology. 2016;18(9):2913-22. Kim G, Yang J, Jang J, Choi J-S, Roe AJ, Byron O, et al. Aldehyde-alcohol dehydrogenase undergoes structural transition to form extended spirosomes for substrate channeling. Communications Biology. 2020;3(1):298. Voynova NE, Rios SE, Miziorko HM. Staphylococcus aureus Mevalonate Kinase: Isolation and Characterization of an Enzyme of the Isoprenoid Biosynthetic Pathway. Journal of Bacteriology. 2004;186(1):61-7. Dorsey JK, Porter JW. The Inhibition of Mevalonic Kinase by Geranyl and Farnesyl Pyrophosphates. Journal of Biological Chemistry. 1968;243(18):4667-70. Gray JC, Kekwick RGO. The inhibition of plant mevalonate kinase preparations by prenyl pyrophosphates. Biochimica et Biophysica Acta (BBA) - General Subjects. 1972;279(2):290-6. Huang K-x, Scott AI, Bennett GN. Overexpression, Purification, and Characterization of the Thermostable Mevalonate Kinase from Methanococcus jannaschii. Protein Expression and Purification. 1999;17(1):33-40. Redding-Johanson AM, Batth TS, Chan R, Krupa R, Szmidt HL, Adams PD, et al. Targeted proteomics for metabolic pathway optimization: application to terpene production. Metab Eng. 2011;13(2):194-203. Rinaldi MA, Ferraz CA, Scrutton NS. Alternative metabolic pathways and strategies to high-titre terpenoid production in Escherichia coli. Natural Product Reports. 2022;39(1):90-118. Primak YA, Du M, Miller MC, Wells DH, Nielsen AT, Weyler W, Beck ZQ. Characterization of a feedback-resistant mevalonate kinase from the archaeon Methanosarcina mazei. Appl Environ Microbiol. 2011;77(21):7772-8. Kazieva E, Yamamoto Y, Tajima Y, Yokoyama K, Katashkina J, Nishio Y. Characterization of feedback-resistant mevalonate kinases from the methanogenic archaeons Methanosaeta concilii and Methanocella paludicola. Microbiology (Reading). 2017;163(9):1283-91. Chen H, Liu C, Li M, Zhang H, Xian M, Liu H. Directed evolution of mevalonate kinase in Escherichia coli by random mutagenesis for improved lycopene. RSC Advances. 2018;8(27):15021-8. Chen H, Li M, Liu C, Zhang H, Xian M, Liu H. Enhancement of the catalytic activity of Isopentenyl diphosphate isomerase (IDI) from Saccharomyces cerevisiae through random and site-directed mutagenesis. Microb Cell Fact. 2018;17(1):65. Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformation.docx Cite Share Download PDF Status: Published Journal Publication published 07 May, 2025 Read the published version in Metabolomics → Version 1 posted Editorial decision: Revision requested 16 Jan, 2025 Reviewers agreed at journal 13 Sep, 2024 Reviews received at journal 12 May, 2024 Reviewers agreed at journal 22 Apr, 2024 Reviewers invited by journal 20 Apr, 2024 Submission checks completed at journal 19 Apr, 2024 Editor assigned by journal 19 Apr, 2024 First submitted to journal 18 Apr, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4285213","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":294153556,"identity":"50f0eaa7-8935-4ae0-b244-d9d62bf4dc1e","order_by":0,"name":"Jasmeet Kaur Khanijou","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABA0lEQVRIiWNgGAWjYDACdjBpwcDAfLD9Q+IfGwYGCUJamMEkUBlbchvDx4Y0krSktzHObDhMWAt/M/MxiY97JOT42xjbHvPuOJ/YP7v54AOGGptoXFokDrOlSc54JmEscYyx3Zj3zO3EGXeOJRswHEvLbcChxYCZx9iY54BEYsP9xgZpHrbbiQ03cswkGBsO49fy54BE/fxjjCAt5xLnE6HF8DHDAYkEg2OMbZIz2w4kbiCkBeiXxIc9ByQMNx5jbDb4cCbZeOONtGSDBDx+4W9vPnDgxwEbeblj7A8fJFTYyc67kXzwwYcaG5xaMIAjWGUCscpBwJ4UxaNgFIyCUTAyAAAhpF4gRkD2eQAAAABJRU5ErkJggg==","orcid":"","institution":"Singapore Institute of Food and Biotechnology Innovation (SIFBI), Agency for Science, Technology and Research (A*STAR)","correspondingAuthor":true,"prefix":"","firstName":"Jasmeet","middleName":"Kaur","lastName":"Khanijou","suffix":""},{"id":294153557,"identity":"e182b60a-1693-4fc9-83ad-ec7a77208e2e","order_by":1,"name":"Clement P. M. Scipion","email":"","orcid":"","institution":"CNRS@CREATE","correspondingAuthor":false,"prefix":"","firstName":"Clement","middleName":"P. M.","lastName":"Scipion","suffix":""},{"id":294153558,"identity":"a47b263a-45bf-4fc9-a6cd-3e0dab6357f1","order_by":2,"name":"Shreyash Borkar","email":"","orcid":"","institution":"Singapore Institute of Food and Biotechnology Innovation (SIFBI), Agency for Science, Technology and Research (A*STAR)","correspondingAuthor":false,"prefix":"","firstName":"Shreyash","middleName":"","lastName":"Borkar","suffix":""},{"id":294153560,"identity":"26523ee5-78d1-4057-a71d-bc85a30b8bb7","order_by":3,"name":"Xixian Chen","email":"","orcid":"","institution":"Singapore Institute of Food and Biotechnology Innovation (SIFBI), Agency for Science, Technology and Research (A*STAR)","correspondingAuthor":false,"prefix":"","firstName":"Xixian","middleName":"","lastName":"Chen","suffix":""},{"id":294153563,"identity":"9a0a4102-7d68-449a-96e6-cfb48e846758","order_by":4,"name":"Wee Chew","email":"","orcid":"","institution":"Singapore Institute of Food and Biotechnology Innovation (SIFBI), Agency for Science, Technology and Research (A*STAR)","correspondingAuthor":false,"prefix":"","firstName":"Wee","middleName":"","lastName":"Chew","suffix":""}],"badges":[],"createdAt":"2024-04-18 05:21:01","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4285213/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4285213/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11306-025-02254-y","type":"published","date":"2025-05-07T15:57:09+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":55215221,"identity":"031ff5b8-d3fc-42a5-b09c-ecf377221af5","added_by":"auto","created_at":"2024-04-24 07:26:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":368773,"visible":true,"origin":"","legend":"\u003cp\u003eIllustration of various pathways involved in limonene production in \u003cem\u003eE. coli\u003c/em\u003e engineered with the mevalonate (MEV) pathway (in orange). The other pathways include the glycolytic, pentose-phosphate pathway (in blue), deoxyxylulose 5-phosphate (DXP) pathway (in green) and tricarboxylic acid (TCA) cycle (in grey). Cofactor consumption is represented by curved arrows. Intermediates: Glcex, glucose extracellular; Glc, glucose; Fruex, fructose extracellular; Fru, fructose; F1P, fructose-1-phosphate; G6P, glucose-6-phosphate; 6PG, 6-phosphogluconate; X5P, xylulose-5-phosphate; Ru5P, ribulose-5-phosphate; R5P, ribose-5-phosphate; F6P, fructose-6-phosphate; F16BP, fructose-1,6-biphosphate; GAP, glyceraldehyde-3-phosphate; DHAP, dihydroxyacetone phosphate; BPG, 1,3-bisphosphoglycerate; 3PG, 3-phosphoglycerate; PEP, phosphoenolpyruvate; PYR, pyruvate; DXP, 1-deoxy-D-xylulose-5-phosphate; B, flux to vitamin B6 pathway; MEP, 2-C-methylerythritol-4-phosphate; CDPME, 4-diphosphocytidyl-2-C-methylerythritol; CDPMEP, 4-diphosphocytidyl-2-C-methylerythritol-2-phosphate; MEcPP, 2-C-Methylerythritol-2,4-cyclodiphosphate; HMBPP, hydroxymethylbutenyl 4-diphosphate; IPP, isopentenyl diphosphate; DMAPP, dimethylallyl diphosphate; GPP, geranyl diphosphate; FPP, farnesyl diphosphate; LIM, limonene; LIMex, extracellular limonene; AcCoA, acetyl coenzyme A; AtAcCoA, acetoacetyl-CoA; HMGCoA, hydroxymethylglutaryl-CoA; MVA, mevalonate; MVAP, 5-phosphomevalonate; MVAPP, 5-diphosphomevalonate; ACE, acetic acid; ACEex, acetic acid extracellular; ACTLD, acetaldehyde; ETH, ethanol; ETHex, extracellular ethanol; LAC, lactic acid; LACex, extracellular lactic acid; AKG, α-ketoglutarate; SucCoA, succinyl CoA; SUC, succinate; SUCex, extracellular succinate; FUM, fumarate; OAA, oxaloacetate. Enzymes: PTS, phosphotransferase system; HK, hexokinase; FK, fructokinase; FruK, fructose-1-kinase; PFK, phosphofructokinase; G6PDH, lumped reactions of glucose-6-phosphate dehydrogenase and 6-phosphogluconolactonase; PGDH, 6-phosphogluconate dehydrogenase; Tkb, transketolase; PGI, phosphoglucose isomerase; PFK, phosphofructokinase; FBA, fructose-1,6-biphosphate aldolase; GDH, glutamate dehydrogenase; PGK, phosphoglycerate kinase; ENO, enolase; PYK, pyruvate kinase; DXS, DXP synthase; DXR, DXP reductase; ISPD, CDPME synthase; ISPE, CDPME kinase; ISPF, MecPP synthase; ISPG, HMBPP synthase; ISPH, HMBPP reductase; IDI, isopentenyl diphosphate isomerase; ISPA, farnesyl diphosphate synthase; LS, limonene synthase; PDH, pyruvate dehydrogenase; AtoB, acetyl-CoA acetyltransferase; HMGS, HMGCoA synthase; HMGR, HMGCoA reductase; MK, mevalonate kinase; PMK; phosphomevalonate kinase; PMD, diphosphate mevalonate decarboxylase; LDH, lactate dehydrogenase; PoxB, pyruvate oxidase; PCK, phosphoenolpyruvate carboxykinase; PPC, phosphoenolpyruvate carboxylase; ACS, acetyl-CoA synthetase; PTACK, lumped reactions of phosphate acetyltransferase and acetate kinase; ALDHB, aldehyde dehydrogenase B; ALDH, aldehyde dehydrogenase; ADH, alcohol dehydrogenase; CSICD, lumped enzymatic reactions of citrate synthase, aconitate hydratase A, aconitate hydratase B and isocitrate dehydrogenase; AKGDH, α-ketoglutarate dehydrogenase; SCS, succinyl-CoA synthetase; FRD, fumarate reductase; MDH, malate dehydrogenase.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/15973e7d879b52a4ad5eecaf.png"},{"id":55215220,"identity":"6c1c7355-8be5-4b3e-b30b-6052472b5432","added_by":"auto","created_at":"2024-04-24 07:26:27","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":110900,"visible":true,"origin":"","legend":"\u003cp\u003eTime-series concentrations of intracellular metabolites from wild-type \u003cem\u003eE. coli\u003c/em\u003e engineered to overproduce limonene (EcoCTs3) grown in glucose after IPTG induction. Figure 2A shows the intracellular metabolites of higher concentration DHAP+GAP (pool of dihydroxyacetone phosphate and glyceraldehyde-3-phosphate), F16BP (fructose-1,6-biphosphate), R5P+Ru5P+X5P (pool of ribose-5-phosphate, ribulose-5-phosphate, and xylulose-5-phosphate), and G6P+F6P (pool of glucose-6-phosphate and fructose-6-phosphate) while Figure 2B illustrates the measured intracellular metabolites \u0026nbsp;DXP (1-deoxy-D-xylulose-5-phosphate), FPP (farnesyl diphosphate), GPP (geranyl diphosphate), MVA (mevalonate), MVAP (5-phosphomevalonate), and PYR (pyruvate).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/07bc2bf00ecdc7a182273353.png"},{"id":55214554,"identity":"a0381d9b-52c7-4510-8794-92722eba471a","added_by":"auto","created_at":"2024-04-24 07:18:27","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":86184,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of metabolites from cell cultures of wild-type \u003cem\u003eE. coli\u003c/em\u003e engineered to overproduce limonene (EcoCTs3) grown on different substrates, 100% glucose (G100), and a mixture of 30% glucose and 70% fructose (G30F70). Figure 3A shows the concentration of secreted limonene while Figure 3B shows the concentration of intracellular fructose-1,6-biphosphate (F16BP) over the time points 2 h to 25 h post-IPTG induction.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/e34df1ffe93fa04f56e9f099.png"},{"id":55214558,"identity":"2090e221-e0a4-483b-a45a-86dd1cd681df","added_by":"auto","created_at":"2024-04-24 07:18:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":106192,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of metabolites from cell cultures containing different \u003cem\u003eE. coli\u003c/em\u003e strains: engineered wild-type, LDH (lactate dehydrogenase) knockout and ALDH-ADH (aldehyde dehydrogenase-alcohol dehydrogenase) knockout. Figure 4A shows the concentration of limonene produced while Figure 4B shows the concentration of intracellular mevalonate (MVA) over the time points 2 h to 25 h post-IPTG induction.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/f4cd1512c9c8c65a3dfe252c.png"},{"id":55214553,"identity":"b8d4beef-73e4-414e-993b-a9644070b7e3","added_by":"auto","created_at":"2024-04-24 07:18:27","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":112958,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentation of the time-series concentrations of intracellular metabolites from LDH (lactate dehydrogenase) knockout \u003cem\u003eE. coli\u003c/em\u003e strains grown in glucose after IPTG induction. Figure 5A represents the intracellular metabolites DHAP+GAP (pool of dihydroxyacetone phosphate and glyceraldehyde-3-phosphate), F16BP (fructose-1,6-biphosphate), MVA (mevalonate), R5P+Ru5P+X5P (pool of ribose-5-phosphate, ribulose-5-phosphate, and xylulose-5-phosphate), and G6P+F6P (pool of glucose-6-phosphate and fructose-6-phosphate), which were of higher concentration while Figure 5B displays intracellular metabolites DXP (1-deoxy-D-xylulose-5-phosphate), FPP (farnesyl diphosphate), GPP (geranyl diphosphate), MVAP (5-phosphomevalonate), and PYR (pyruvate).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/e0b79ed9dbb6000719d72f29.png"},{"id":55214556,"identity":"bcf0dcb5-d856-4954-9d49-d6d83fde65d3","added_by":"auto","created_at":"2024-04-24 07:18:27","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":114707,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentation of the intracellular concentrations of metabolites obtained at various time-points from ALDH-ADH (aldehyde dehydrogenase-alcohol dehydrogenase) knockout \u003cem\u003eE. coli\u003c/em\u003e strains grown in glucose after IPTG induction. Figure 6A illustrates the intracellular metabolites DHAP+GAP (pool of dihydroxyacetone phosphate and glyceraldehyde-3-phosphate), F16BP (fructose-1,6-biphosphate), MVA (mevalonate), R5P+Ru5P+X5P (pool of ribose-5-phosphate, ribulose-5-phosphate, and xylulose-5-phosphate), and G6P+F6P (pool of glucose-6-phosphate and fructose-6-phosphate), which were of higher concentration while Figure 6B shows intracellular metabolites DXP (1-deoxy-D-xylulose-5-phosphate), FPP (farnesyl diphosphate), GPP (geranyl diphosphate), MVAP (5-phosphomevalonate), and PYR (pyruvate).\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/921b6332b3a7e0ec2b70aad8.png"},{"id":82537815,"identity":"0d775aa3-5397-409f-a7ca-0b7858e62bb4","added_by":"auto","created_at":"2025-05-12 16:10:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1581749,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/79b24ed1-8ffe-41d0-81ab-458e52e547e6.pdf"},{"id":55214560,"identity":"d0ac97d9-9677-4eba-ba58-759d5ee7509a","added_by":"auto","created_at":"2024-04-24 07:18:28","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":76074,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-4285213/v1/74fbdd5f0ae8720234e8bb12.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Enhancing Limonene production by probing the metabolic network through time-series metabolomics data","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eMonoterpenoids are a class of structurally varied natural compounds with a strong presence in industrial applications (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Limonene, a cyclized monoterpene, is assumed as GRAS (generally recognized as safe) by the US Food and Drug Administration (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Limonene and its derivatives, such as carveol, menthol, and α-terpineol, have broad applications in food and beverage, pharmaceuticals, cosmetics, biomaterials, and biofuels due to their pleasant fragrance and physicochemical properties (\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Limonene has typically been extracted from plant biomass, where availability could be affected by variations in climate and agricultural land (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). As limonene\u0026rsquo;s structure has a chiral centre, it is found in nature as two enantiomers, (D)-limonene and (L)-limonene. (D)-limonene is usually obtained through cold pressing citrus peels and pulps (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e) while (L)-limonene is found in essential oils of other plant species (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). With advancements in metabolic engineering and synthetic biology, microbes can be utilised as an alternative approach to sustainably produce high-value natural products, including limonene. However, the carbon yields for limonene biosynthesis are relatively low for economically feasible bioprocesses (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). The low productivity in limonene biosynthesis is due to various challenges such as: the inefficiency of enzymes for biosynthesis pathways, minimal metabolic fluxes directed towards limonene, intrusion of the native metabolism of microbes with complicated heterologous biosynthesis pathways, the absence of appropriate conditions for the heterologous expression of optimized enzymes and the cytotoxicity of limonene affecting microbial chassis (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe biosynthesis of limonene in \u003cem\u003eEscherichia coli\u003c/em\u003e can arise from two major terpenoid biosynthetic pathways: the native deoxyxylulose 5-phosphate (DXP) pathway and the heterologous mevalonate-dependent (MEV) pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Both the DXP and MEV pathways produce two isomeric isoprene metabolites, namely isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), which are precursors for terpenoid production (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). As such, previous attempts to obtain high terpenoid yields, required a good balance of the availability of co-factors, energy demands, and carbon flux in the DXP and/or MEV pathways in \u003cem\u003eE. coli\u003c/em\u003e (\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Furthermore, as the DXP and MEV pathways obtain their carbon from central carbon metabolism, these pathways compete with many other biochemical reactions to produce required precursors towards limonene production (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe major biochemical reactions in \u003cem\u003eE. coli\u003c/em\u003e leading towards eventual limonene production are illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, which include competing pathways for carbon flux such as the tricarboxylic acid (TCA) cycle, mixed fermentation pathways, and pentose phosphate pathway. By understanding such a metabolic network topology, in conjunction with intracellular metabolomics data, attempts could be made to improve limonene synthesis in bacterial bio-factories. Furthermore, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, there are various co-factors involved in the biochemical reactions in the metabolic network, which could also be sources of potential bottlenecks when the co-factors become limiting. Previous work has shown that limonene biosynthesis using the DXP pathway was limited by the low concentrations of terpenoid precursors produced, particularly geranyl pyrophosphate (GPP) (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). The overproduction of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP) could be easily executed through the alteration of prenyl transferase and terpene synthase. However, the production of monoterpenes using microbes has been restricted as many monoterpenes are toxic and volatile (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e), and the heterologous expression of GPP synthase and monoterpene synthases have been poor (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). In one study, a dedicated GPP synthase was unable to produce farnesyl diphosphate (FPP) that was instrumental in enhancing the yield of limonene (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). These strategies to improve limonene yield have focused mainly on the downstream enzymes or metabolites of the metabolic network.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePrevious literature focused more on downstream metabolites of the MEV pathway to improve limonene yield (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). There exists a gap in knowledge regarding the impact of upstream metabolites in improving limonene yield by understanding the network topology. In this study, we hypothesized the need to unblock carbon flux channelling towards MEV pathway by optimising upstream metabolic fluxes to enhance limonene yield in \u003cem\u003eE. coli\u003c/em\u003e cell cultures. A wild-type \u003cem\u003eE. coli\u003c/em\u003e strain (K-12 MG1655) was engineered to overproduce L-limonene by heterologous expression of the MEV pathway and a limonene synthase through plasmid pJBEI-6409 (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e), resulting in an EcoCTs3 strain, based on which knockout strains were created. Following a rationale experimental design approach, we traced the metabolic topology of the key \u003cem\u003eE. coli\u003c/em\u003e pathways involved from glycolysis, TCA cycle, pentose phosphate down to the MEV and DXP pathways, by measuring time-series concentrations of important intracellular and extracellular metabolites. A fast filtration and rapid quenching in liquid nitrogen sampling methodology was utilised coupled to C-18 liquid chromatography time of flight mass spectrometry (LC-TOF-MS) (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e), whereby the column chemistry allowed both the hydrophilic intracellular metabolites upstream in central metabolism and the more hydrophobic secondary pathway intracellular metabolites to be analyzed simultaneously in negative mode. By studying the intracellular metabolomics data obtained from the EcoCTs3 strain and the metabolic network topology, attempts were made to improve the limonene yield through a systematic hypothesis driven approach. First, the carbon source was changed from glucose to fructose to improve intracellular metabolite concentrations in upstream fluxes. Second, to enhance carbon flux flow towards the MEV pathway to improve limonene production, two mutant strains were created, with the gene knockouts of \u003cem\u003eadhe\u003c/em\u003e and \u003cem\u003eldh\u003c/em\u003e, which are genes involved in the mixed fermentation pathways. The mutant strains increased limonene yield by 8 to 9 folds through the elimination of pathways competing for the carbon flux.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cp\u003eAll chemicals, standards, solvents, and media components used in this study were obtained from Sigma-Aldrich (St. Louis, MO), Fisher Scientific (Fair Lawn, NJ, USA), or VWR (Radnor, PA, USA) unless otherwise noted. Water was deionised and filtered by Sartorius Arium Pro VF Type 1 water system (18.2 MΩ, 0.2 \u0026micro;m).\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Strain construction\u003c/h2\u003e \u003cp\u003eThe limonene producing engineered wild-type strain was created by transforming pJBEI-6409 plasmid which was a gift from Taek Soon Lee (Addgene plasmid #47048 ; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://n2t.net/addgene:47048\u003c/span\u003e\u003cspan address=\"http://n2t.net/addgene:47048\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e ; RRID:Addgene_47048) (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e) by heat shock in \u003cem\u003eE. coli\u003c/em\u003e K-12 MG1655 (F- lambda- \u003cem\u003eilvG\u003c/em\u003e- rfb-50 rph-1) forming the strain EcoCTs3. To create the ALDH-ADH and LDH knockouts, the deletion of \u003cem\u003eadhE\u003c/em\u003e (UniProt P0A9Q7) and \u003cem\u003eldhA\u003c/em\u003e (UniProt P52643) genes from \u003cem\u003eE. coli\u003c/em\u003e MG1655 strain was executed using CRISPR Cas9-assisted recombineering method as described previously (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). For the successful knockout strains, the \u003cem\u003eldhA\u003c/em\u003e and \u003cem\u003eadhE\u003c/em\u003e regions were PCR amplified and sequenced to confirm the deletion of the genes. The strains were stored in 40% glycerol at -80\u0026deg;C before further use. For limonene production experiments, the pJBEI-6409 plasmid was transformed by heat shock in \u003cem\u003eΔldhA\u003c/em\u003e and \u003cem\u003eΔadhE\u003c/em\u003e strains. Competent cells were prepared using the Mix \u0026amp; Go! \u003cem\u003eE. coli\u003c/em\u003e Transformation Kit (Zymo Chem) from the \u003cem\u003eldhA\u003c/em\u003e and \u003cem\u003eadhE\u003c/em\u003e deletion strains, following manufacturer\u0026rsquo;s specifications. pJBEI-6409 transformed strains (\u003cem\u003eE. coli\u003c/em\u003e K-12 MG1655, \u003cem\u003eΔldhA\u003c/em\u003e and \u003cem\u003eΔadhE\u003c/em\u003e) were each plated on a chloramphenicol selective Luria-Bertani agar plate and incubated overnight at 37\u0026deg;C.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Growth conditions\u003c/h2\u003e \u003cp\u003eAfter the prepared \u003cem\u003eE. coli\u003c/em\u003e strains were streaked onto plates and incubated overnight, one colony was picked from each plate. Each colony was inoculated into 5 mL of Luria-Bertani medium consisting of 10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl, where 30 \u0026micro;g/mL chloramphenicol was added and grown overnight at 37\u003csup\u003eo\u003c/sup\u003eC and 220 rpm. Each cell pellet was washed and re-suspended in 50 mL M9 medium adapted from a previous study (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e), where the M9 medium consisted of 12.7 g/L Na\u003csub\u003e2\u003c/sub\u003eHPO\u003csub\u003e4\u003c/sub\u003e.7H\u003csub\u003e2\u003c/sub\u003eO, 3.1 g/L KH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, 1 g/L NH\u003csub\u003e4\u003c/sub\u003eCl, 0.5 g/L NaCl, 0.25 g/L MgSO\u003csub\u003e4\u003c/sub\u003e.7H\u003csub\u003e2\u003c/sub\u003eO, 15 mg/L CaCl\u003csub\u003e2\u003c/sub\u003e.2H\u003csub\u003e2\u003c/sub\u003eO, 8.1 mg/L FeCl\u003csub\u003e3,\u003c/sub\u003e 0.89 mg/L MnCl\u003csub\u003e2\u003c/sub\u003e.4H\u003csub\u003e2\u003c/sub\u003eO, 1.7 mg/L ZnCl\u003csub\u003e2,\u003c/sub\u003e 0.34 mg/L CuCl\u003csub\u003e2,\u003c/sub\u003e 0.6 mg/L CoCl\u003csub\u003e2\u003c/sub\u003e.6H\u003csub\u003e2\u003c/sub\u003eO, 0.51 mg/L Na\u003csub\u003e2\u003c/sub\u003eMoO\u003csub\u003e4,\u003c/sub\u003e 10 g/L glucose, and incubated overnight at 30 \u003csup\u003eo\u003c/sup\u003eC and 220 rpm. Each strain was stored in 40% glycerol at \u0026minus;\u0026thinsp;80 \u003csup\u003eo\u003c/sup\u003eC. Pre-cultures of each strain were then prepared by adding 100 \u0026micro;L of each strain stored in glycerol to 50 mL M9 medium in 250 mL flasks, which were left overnight at 30\u0026deg;C at 220 rpm. For experiments using mixtures of fructose and glucose as carbon sources, the carbon source was made up to 10 g/L using 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% fructose. To collect time-series metabolomics data, multiple cell culture flasks were prepared for duplicate biological samples for each time-point. Cell cultures of each strain were prepared by adding 100 \u0026micro;L of pre-cultures to 50 mL M9 medium in 250 mL flasks with incubation at 30\u0026deg;C and 220 rpm. Upon reaching an optical density of 1 at 600 nm, isopropyl β-d-1-thiogalactopyranoside (IPTG) was added, resulting in a concentration of 25 \u0026micro;M. A dodecane overlay of 5 mL was added to cell cultures to allow trapping of the volatile secreted limonene (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). The flasks of cell cultures were left at 30\u0026deg;C and 220 rpm. The flasks were sacrificed in duplicates at 2 h, 3 h, 6 h, 7 h, 24 h, and 25 h post-IPTG induction to profile the concentrations of the intracellular, extracellular, and secreted limonene.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Metabolomics extraction and analysis\u003c/h2\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Secreted Limonene and Extracellular Metabolites\u003c/h2\u003e \u003cp\u003eDynamic sampling at time points 2 h, 3 h, 6 h, 7 h, 24 h and 25 h post IPTG induction was performed. Specifically, duplicate \u003cem\u003eE. coli\u003c/em\u003e cell cultures of 50 mL were sacrificed and centrifuged for 10 mins at 3000 rpm. The dodecane layers containing secreted limonene were removed and stored at -80 \u003csup\u003eo\u003c/sup\u003eC prior to quantitative analysis. Limonene samples were diluted in ethyl acetate and run on an Agilent 7890B gas chromatography mass spectrometry (GC-MS) system with a DB-5 column (5%-phenyl-methylpolysiloxane, 30 m x 0.25 mm ID x 0.25 \u0026micro;m, Agilent Technologies, USA). For each run, 1 \u0026micro;L sample was injected with a 10:1 split ratio and 10 mL/min split flow. The GC oven temperature program was set at 40\u0026deg;C for 3 min, followed by a 10\u0026deg;C/min ramp to 100 \u003csup\u003eo\u003c/sup\u003eC and another 60\u0026deg;C/min ramp to 220\u0026deg;C with a hold time of 2 min. The injector and MS transfer line temperatures were at 250\u0026deg;C and 280\u0026deg;C, respectively. The MS was operated in selected ion-monitoring (SIM) mode using ions of m/z 136, 68 and 93, representing the molecular ion and two abundant fragmental ions of limonene. Calibration standards were run for limonene prepared in ethyl acetate from 0.05 \u0026micro;g/mL to 10 \u0026micro;g/mL. Limonene concentrations were determined using the Agilent Quantitative software.\u003c/p\u003e \u003cp\u003eThe media supernatants after centrifugation were stored at -80 \u003csup\u003eo\u003c/sup\u003eC prior to extracellular metabolites quantitative analysis. Subsequently after thawing these, 1 mL of each supernatant sample was filtered using a polyamide filter prior to analysis using an Agilent 1200 high performance liquid chromatography (HPLC) system with a Bio-rad Aminex HPX-87H column (300 x 7.8 mm) and 1260 Infinity II Refractive Index Detector (RID). For each run, 5 \u0026micro;L of sample was injected with an isocratic gradient using 0.01 N sulphuric acid and a 0.6 mL/min flow rate for 28 min. The column temperature was set at 35\u0026deg;C, while the RID temperature was at 30\u0026deg;C with positive polarity. Calibration mixtures of the following range: glucose (0.5 to 80 g/L), lactic acid (0.125 to 8 g/L), acetic acid (0.125 to 80 g/L), and ethanol (0.5 to 80 g/L) were used to construct calibration curves to establish linearity and the concentrations of each extracellular metabolite in the prepared samples. Cell pellets were dried in the oven overnight at 80\u0026deg;C prior to weighing with the Mettler Toledo XSE105 precision balance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 Intracellular Metabolites\u003c/h2\u003e \u003cp\u003eTo measure intracellular metabolites, \u003cem\u003eE. coli\u003c/em\u003e cell cultures were sacrificed in duplicates at 2 h, 3 h, 6 h, 7 h, 24 h, and 25 h after IPTG induction. Cell cultures were subjected to a modified protocol from the fast filtration and quenching in liquid nitrogen previously described (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). In brief, 10 mL of cell culture were filtered through a 0.2 \u0026micro;m polyamide membrane filter (Sartorius, Goettingen, Germany) and washed with 5 mL of wash solution (12.7 g/L Na\u003csub\u003e2\u003c/sub\u003eHPO\u003csub\u003e4\u003c/sub\u003e.7H\u003csub\u003e2\u003c/sub\u003eO, 3.1 g/L KH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, 1 g/L NH\u003csub\u003e4\u003c/sub\u003eCl, 0.5 g/L NaCl). The membrane filter containing the cells was folded in aluminum foil and rapidly plunged into liquid nitrogen for rapid quenching of biochemical reactions. The aluminum foil packages with membrane filters were stored at -80 \u003csup\u003eo\u003c/sup\u003eC until intracellular metabolites extraction. For the extraction of intracellular metabolites, membranes were removed from the aluminum foil packages and placed into 5 mL of extraction solvent consisting of methanol, acetonitrile, and water (4:4:2) and left in an ice bath for 10 mins. After this, the membranes were vortexed for 1 min followed by 3 rounds of sonication for 3 mins each time, with placement in an ice bath after each round of sonication for 1 min. The extracts were decanted into glass tubes and spiked with 20 \u0026micro;L internal standard mixture consisting of 50 \u0026micro;g/mL mevalonic acid-d3 (MVA-d3) and 10 \u0026micro;g/mL thymolphthalein monophosphate (TMP) prepared in methanol: 10 mM ammonium hydroxide mixture (7:3). The extracts were then subjected to a vacuum concentrator and reconstituted with 200 \u0026micro;L methanol: 10 mM ammonium hydroxide mixture (7:3) and filtered prior to analysis via liquid chromatography mass spectrometry.\u003c/p\u003e \u003cp\u003eFor quantitation of intracellular metabolites, an Agilent 6230 time of flight-mass spectrometer (TOF-MS) with a Dual Agilent Jet Stream (AJS) ion source was coupled with an Agilent ultra-performance liquid chromatography (UPLC) 1290 system and a Waters Acquity UPLC BEH C18 (2.1 x 150 mm, 1.7 \u0026micro;m) column with a VanGuard pre-column (2.1 x 5 mm). The chromatographic method used was adapted from previous studies (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e), with appropriate modification. For each run, 2 \u0026micro;L of sample was injected and a dual mobile phase system was utilized with mobile phase A as 5 mM ammonium formate in water (pH 9.5) and mobile phase B as 5 mM ammonium formate (pH 9.5) in acetonitrile: water (9:1). The following solvent gradient was used: start with 100% mobile phase A from 0-3.5 min with 0.1 mL/min flow rate to 100% mobile phase B at 12 min and held for 8 min with increased flow rate to 0.5 mL/min. At 20 min, the system was recalibrated back to 100% mobile phase A for 5 min and held for another 5 min. The column temperature was kept constant at 35 \u003csup\u003eo\u003c/sup\u003eC. Negative electrospray ionisation was used with the following TOF settings: Gas temperature, 325\u0026deg;C; Gas flow, 11 L/min; Nebuliser pressure, 35 psi; Sheath gas temperature, 375\u0026deg;C; Sheath gas flow, 11 L/min; Vcap voltage, 3500 V; Nozzle voltage, 500 V; Skimmer, 65; OctopoleRFPeak, 750; Scan rate, 2 spectra/s. The fragmentor voltage was varied throughout each 35 min sample analysis: 2\u0026ndash;7.5 min, 140 V; 7.5\u0026ndash;15 min, 100 V, 140 V and 150 V. UPLC flow diversions were as follows: 0\u0026ndash;2 min to waste, 2\u0026ndash;15 min to TOF-MS, and 15\u0026ndash;35 min to waste.\u003c/p\u003e \u003cp\u003eStandards for quantitation of intracellular metabolites were prepared in methanol: 10 mM ammonium hydroxide mixture (7:3). The following intracellular metabolites were measured DXP (1-deoxy-D-xylulose-5-phosphate), DHAP\u0026thinsp;+\u0026thinsp;GAP (pool of dihydroxyacetone phosphate and glyceraldehyde-3-phosphate), F16BP (fructose-1,6-biphosphate), FPP (farnesyl diphosphate), GPP (geranyl diphosphate), MVA (mevalonate), MVAP (5-phosphomevalonate), PYR (pyruvate), R5P\u0026thinsp;+\u0026thinsp;Ru5P\u0026thinsp;+\u0026thinsp;X5P (pool of ribose-5-phosphate, ribulose-5-phosphate, and xylulose-5-phosphate) and G6P\u0026thinsp;+\u0026thinsp;F6P (pool of glucose-6-phosphate and fructose-6-phosphate). Different concentrations of intracellular intermediates were used to prepare calibration mixtures for construction of calibration curves: DHAP (DHAP\u0026thinsp;+\u0026thinsp;GAP pool) \u0026ndash; 0.04 to 5 \u0026micro;g/mL; DXP \u0026ndash; 0.04 to 10 \u0026micro;g/mL; F6P (F6P\u0026thinsp;+\u0026thinsp;G6P pool) \u0026ndash; 0.04 to 10 \u0026micro;g/mL; F1,6BP \u0026ndash; 0.04 to 6 \u0026micro;g/mL; MVA \u0026ndash; 0.04 to 10 \u0026micro;g/mL; R5P (R5P\u0026thinsp;+\u0026thinsp;Ru5P\u0026thinsp;+\u0026thinsp;X5P pool) \u0026ndash; 0.04 to 10 \u0026micro;g/mL; PYR \u0026ndash; 0.04 to 1.5 \u0026micro;g/mL; MVAP \u0026ndash; 0.01 to 0.3 \u0026micro;g/mL; GPP \u0026ndash; 0.05 to 1.5 \u0026micro;g/mL; FPP \u0026ndash; 0.05 to 1.5 \u0026micro;g/mL. Each calibration mixture was made up to 100 \u0026micro;L with 10 \u0026micro;L of internal standard mixture (50 \u0026micro;g/mL MVA-d3 and 10 \u0026micro;g/mL TMP) added. Metabolite quantitation was performed using the Agilent Masshunter Workstation Quantitative Analysis software for TOF.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"3. Results and Discussion","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Time-series intracellular metabolites from engineered wild-type strain EcoCTs3\u003c/h2\u003e \u003cp\u003eThe wild-type \u003cem\u003eE. coli\u003c/em\u003e strain engineered to overproduce L-limonene (EcoCTs3) was cultured and harvested at aforesaid time-points after IPTG induction where the metabolomics data from their targeted intra-/ extra-cellular metabolite analyses via GC-MS and LC-ToF-MS were collated. Figures\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB show the data from the EcoCTs3 strain grown in glucose where it was observed that many of the intracellular metabolites such as G6P and F6P (pooled), R5P, Ru5P and X5P (pooled), DHAP and GAP (pooled), PYR, DXP, MVA, FPP were found to generally decrease from 2 h to 7 h post-IPTG induction followed by an accumulation 24 h to 25 h after IPTG induction. The decreased concentrations of such intracellular metabolites corresponded to increased limonene production from 2 h to 7 h post-IPTG induction as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA. From 24 h to 25 h post-IPTG induction, the dry cell weights of the cell pellets from 50 mL cell cultures were relatively constant at 83.3\u0026thinsp;\u0026plusmn;\u0026thinsp;2.2 mg, indicating the stationary phase with negligible culture growth (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, Supplementary section). This could explain the observed accumulation of intracellular metabolites due to diminished cell growth and possibly reduced efficiencies of the biochemical reactions occurring in the metabolic network to produce limonene.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe presence and quantitation of intracellular FPP (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB) showed that there was a loss of carbon flux towards limonene production, as this metabolite is formed from GPP and IPP (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). With less carbon contribution to the immediate precursor metabolite to limonene, GPP, the limonene yield would not be optimal. The presence of both intracellular GPP and FPP as observed in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB could be due to their formations being catalyzed by the same prenyl diphosphate synthase, the enzyme farnesyl diphosphate synthase (IspA) (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). FPP production is favoured in \u003cem\u003eE. coli\u003c/em\u003e as it is required in the formation of peptidoglycans which are important constituents of cell wall (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e) and components of the respiratory chain (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e), where its accumulation 24 h to 25 h after IPTG induction further indicates diminished cell growth in the cultures. To prevent the loss of carbon flux towards FPP, Alonso-Gutierrez \u003cem\u003eet. al.\u003c/em\u003e (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e) improved the titre of limonene through the expression of a heterologous dedicated GPP synthase which produced only GPP, unlike the native IspA, producing both GPP and FPP. The redirection of carbon flux towards GPP and preventing its loss to FPP helped in improving the yield of the target. Using this reported finding, it is therefore possible to redirect carbon flux towards the target by eliminating competing biochemical reactions, whether such reactions occur upstream or downstream of the MEV pathway.\u003c/p\u003e \u003cp\u003eThe metabolic network in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the importance of the metabolite pyruvate (PYR) in relation to the other biochemical pathways. PYR is formed as the end product of glycolysis, and is channelled into the fermentation fluxes, the DXP pathway, and it is the precursor to the intermediate metabolite, acetyl coenzyme A (AcCoA) involved in the TCA and MEV pathway. The cultures of the EcoCTs3 strain had a drastic decrease in the intracellular concentration of PYR by about 41 folds, when compared to the pool of the first two metabolites in the glycolysis pathway, \u003cem\u003ei.e.\u003c/em\u003e G6P and F6P at the same time point of 2 h post IPTG induction (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). With such a significant concentration reduction from the G6P\u0026thinsp;+\u0026thinsp;F6P pool to PYR, we deduced that there would be a low amount of carbon flux entering the MEV pathway, resulting in the low production of limonene observed in the EcoCTs3 strain thus far. In principle, improvements in limonene production could be achieved by increasing metabolite concentrations upstream of the metabolite PYR in the metabolic network. Comparing F16BP at 2 h and 25 h post IPTG shows its concentration decreased by six times compared to the pool of G6P and F6P at 2 h point, and it steadily decreased until 25 h (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Therefore, we hypothesized increasing the concentrations of F16BP by channelling more flux from G6P\u0026thinsp;+\u0026thinsp;F6P could be a strategy to eventually improve the carbon flux going towards limonene production.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Attempting to improve intracellular F16BP concentration by changing carbon source\u003c/h2\u003e \u003cp\u003eIn our attempts to increase the intracellular F16BP concentration, fructose was used as a possible carbon source alternative to glucose. From the metabolite network in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, and based on previous work (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e), fructose can be converted to either F1P (fructose-1-phospahte) or F6P prior to forming F16BP. When varying proportions of fructose (0\u0026ndash;100%) to glucose were tested in the culture medium, the highest amount of limonene was produced by growing the EcoCTs3 \u003cem\u003eE. coli\u003c/em\u003e strain in 100% glucose (G100) followed by a mixture of 30% glucose and 70% fructose (G30F70), as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA (Figure S2 represents the full dataset, Supplementary section). However, attempts with higher proportion of fructose to improve the initial intracellular F16BP concentration did not occur (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), as glucose was still the main driver of F16BP production with possibly diauxic growth. Glucose passively diffuses into \u003cem\u003eE. coli\u003c/em\u003e, where inner membrane transport followed by phosphorylation occurs (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). This process utilises the phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS), which consists of multiple proteins in the phospho-relay process that are involved in the import and concurrent phosphorylation of carbohydrates (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). With the presence of glucose in the cell culture medium, there is enhanced glucose transport activity by bacteria due to induced expression of \u003cem\u003eptsG\u003c/em\u003e from the phosphotransferase system (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). This corroborates with the findings in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, where glucose was still the best substrate for limonene production. With no improvement in intracellular F16BP and therefore limonene concentrations through mixtures of glucose and fructose as carbon source, we next hypothesized potential improvement of re-directing carbon flow towards MEV pathway by curtailing the loss of carbon flux to other pathways, and in so doing, enhance limonene yield.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Generating mutant strains to direct carbon flux towards MEV pathway\u003c/h2\u003e \u003cp\u003eThe second hypothesis tested for upstream flux optimisation to improve limonene yield is to direct more carbon flux into the MEV pathway. One approach is to allow more carbon flux to flow towards PYR since it is a key metabolite involved in numerous pathways, and at the same time curtail PYR losses through pathways that do not lead to limonene production. Mixed fermentation is one common route where there is loss of carbon flux due to the formation of lactic acid and ethanol. To eliminate the mixed fermentation pathways, two mutant strains were prepared. One strain had the \u003cem\u003eldh\u003c/em\u003e gene removed, thereby representing a LDH knockout, while a second strain had the \u003cem\u003eadhE\u003c/em\u003e gene removed, resulting in an ADH-ALDH knockout, as the \u003cem\u003eadhE\u003c/em\u003e gene encodes for both enzymes ADH and ALDH (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). The outcome of limonene production of these mutant strains is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhen compared to the EcoCTs3 strain, these knockout strains were found to substantially improve limonene yield by 8 to 9 folds (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). This enhanced limonene production was correlated to the 18 to 20 fold increase in intracellular MVA measured in the knockout strains, as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB. There was also greater accumulation of intracellular MVA compared to some other metabolites at 24 h to 25 h post-IPTG for these mutant strains, as observed in Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB for the LDH knockout strain and Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB for the ALDH-ADH knockout strain. MVA was observed at the later time points (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA) likely because the enzyme mevalonate kinase (MK) could not efficiently convert MVA to MVAP. Since this biochemical reaction requires the presence of ATP (adenosine triphosphate), it could suggest the availability of ATP becomes a limiting factor in the biochemical synthesis of limonene. Such reduced ATP availability has also been observed in previous studies in various microbes, thereby affecting the productivity of the MEV pathway in generating isoprenoids (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Furthermore, the observation of high levels of intracellular MVA could be explained by previous work, in which downstream intermediates such as IPP, DMAPP, FPP, and GPP have been found to inhibit MK activity through competitive binding to the ATP-binding site of MK (\u003cspan additionalcitationids=\"CR37 CR38\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). With the competitive inhibition of MK, there would be less MVAP produced and more intracellular MVA accumulated, as observed in Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB for the LDH knockout strain and Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB for the ALDH-ADH knockout strain. Furthermore, in another study using targeted proteomics, the production of enzymes MK and phosphomevalonate kinase (PMK) were found to be poor (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). These compounding factors make it difficult for the accumulation of intracellular metabolites downstream of MVA such as MVAP and GPP (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB) and explains the poor conversion efficiency of MVA to limonene, as observed from the intracellular data, where the 18 to 20 fold increase in MVA did not result in the same fold increase of limonene (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The LDH and ALDH-ADH knockout strains also showed loss of flux to FPP (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB) due to the native enzyme IspA catalyzing its formation. Redirecting the flux to GPP and preventing its loss to FPP can be executed through the expression of an improved heterologous GPP synthase (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e) and improving limonene synthase activity, which could further improve limonene titres for both the knockout strains.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eWhen engineering bacterial strains to enhance the yield of target compounds, combining intracellular time-series metabolomics data together with the understanding of the metabolic network can be very useful. Incorporating a hypothesis-driven experimental approach with the idea of using different strategies such as changing the carbon source and eliminating competing pathways that result in flux loss can aid in achieving the desired goal of improving target yield whilst streamlining the workflow.\u003c/p\u003e \u003cp\u003eFrom this study, enhancing the concentration of intracellular MVA is a key step in improving the yield of limonene and possibly other target compounds in the MEV pathway. Once a strain producing high intracellular MVA has been identified, further fine-tuning of the strain can be executed to improve the target yield. The presence of bottlenecks downstream of MVA in the MEV pathway could possibly be due to the lack of ATP to metabolize MVA to MVAP and the feedback-inhibition of MK. To further minimise the bottlenecks observed downstream of the metabolite MVA in the engineered MEV pathway, homologous enzymes from alternative organisms could be utilised (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). For instance, feedback-resistant MK homologs have been found to uphold high activity in the presence of DMAPP, IPP, GPP, FPP, and MVAPP (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). These homologs can reduce MVA accumulation, thereby improving titres. Furthermore, protein engineering can also enhance substrate affinity of feedback-resistant MK (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e) and it has been used on enzyme isopentenyl diphosphate isomerase (IDI) to improve its catalytic activity and thereby increase titre levels (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJ. K. K. carried out culturing and metabolomics experiments on the EcoCTS03 and modified strains. C. P. M. S., S. B., and X. C. designed and prepared the modified strains. J. K. K. and W. C. designed the study and wrote the paper.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Intra-Create Thematic Grant \u0026ldquo;Cities\u0026rdquo; (grant number: NRF2019-THE001-0007) under the EcoCTs project. The EcoCTs research project is supported by the National Research Foundation, Prime Minister\u0026rsquo;s Office, Singapore, under its campus for Research Excellence and Technological Enterprise (CREATE) programme. In addition, we are thankful to Dr. Floriant Bellvert, Hanna Kulyk, and Cecilia Berges from MetaToul (Metabolomics \u0026amp; Fluxomics Facilities, Toulouse, France) for their experimental guidance and insights. We would also like to acknowledge Dr. Kumar Selvarajoo from the Bioinformatics Institute (BII), A*STAR, Singapore for technical advice and discussion.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eWojtunik-Kulesza KA, Kasprzak K, Oniszczuk T, Oniszczuk A. Natural Monoterpenes: Much More than Only a Scent. Chem Biodivers. 2019;16(12):e1900434.\u003c/li\u003e\n \u003cli\u003eMamidipally PK, Liu SX. First approach on rice bran oil extraction using limonene. European Journal of Lipid Science and Technology. 2004;106(2):122-5.\u003c/li\u003e\n \u003cli\u003eHąc-Wydro K, Flasiński M, Romańczuk K. Essential oils as food eco-preservatives: Model system studies on the effect of temperature on limonene antibacterial activity. Food Chem. 2017;235:127-35.\u003c/li\u003e\n \u003cli\u003eFelipe LdO, Oliveira AMd, Bicas JL. Bioaromas \u0026ndash; Perspectives for sustainable development. Trends in Food Science \u0026amp; Technology. 2017;62:141-53.\u003c/li\u003e\n \u003cli\u003eThomsett MR, Moore JC, Buchard A, Stockman RA, Howdle SM. New renewably-sourced polyesters from limonene-derived monomers. Green Chemistry. 2019;21(1):149-56.\u003c/li\u003e\n \u003cli\u003eCiriminna R, Lomeli-Rodriguez M, Demma Car\u0026agrave; P, Lopez-Sanchez JA, Pagliaro M. Limonene: a versatile chemical of the bioeconomy. Chem Commun (Camb). 2014;50(97):15288-96.\u003c/li\u003e\n \u003cli\u003eIb\u0026aacute;\u0026ntilde;ez MD, Sanchez-Ballester NM, Bl\u0026aacute;zquez MA. Encapsulated Limonene: A Pleasant Lemon-Like Aroma with Promising Application in the Agri-Food Industry. A Review. Molecules. 2020;25(11).\u003c/li\u003e\n \u003cli\u003eKvittingen L, Sjursnes BJ, Schmid R. Limonene in Citrus: A String of Unchecked Literature Citings? Journal of Chemical Education. 2021;98(11):3600-7.\u003c/li\u003e\n \u003cli\u003eSun C, Theodoropoulos C, Scrutton NS. Techno-economic assessment of microbial limonene production. Bioresource Technology. 2020;300:122666.\u003c/li\u003e\n \u003cli\u003eRen Y, Liu S, Jin G, Yang X, Zhou YJ. Microbial production of limonene and its derivatives: Achievements and perspectives. Biotechnology Advances. 2020;44:107628.\u003c/li\u003e\n \u003cli\u003eZhao L, Chang WC, Xiao Y, Liu HW, Liu P. Methylerythritol phosphate pathway of isoprenoid biosynthesis. Annu Rev Biochem. 2013;82:497-530.\u003c/li\u003e\n \u003cli\u003eWard VCA, Chatzivasileiou AO, Stephanopoulos G. Metabolic engineering of Escherichia coli for the production of isoprenoids. FEMS Microbiology Letters. 2018;365(10).\u003c/li\u003e\n \u003cli\u003eGruchattka E, H\u0026auml;dicke O, Klamt S, Sch\u0026uuml;tz V, Kayser O. In silico profiling of Escherichia coli and Saccharomyces cerevisiae as terpenoid factories. Microb Cell Fact. 2013;12:84.\u003c/li\u003e\n \u003cli\u003eDiner BA, Fan J, Scotcher MC, Wells DH, Whited GM. Synthesis of Heterologous Mevalonic Acid Pathway Enzymes in Clostridium ljungdahlii for the Conversion of Fructose and of Syngas to Mevalonate and Isoprene. Applied and Environmental Microbiology. 2018;84(1):e01723-17.\u003c/li\u003e\n \u003cli\u003eRohmer M, Seemann M, Horbach S, Bringer-Meyer S, Sahm H. Glyceraldehyde 3-Phosphate and Pyruvate as Precursors of Isoprenic Units in an Alternative Non-mevalonate Pathway for Terpenoid Biosynthesis. Journal of the American Chemical Society. 1996;118(11):2564-6.\u003c/li\u003e\n \u003cli\u003eCarter OA, Peters RJ, Croteau R. Monoterpene biosynthesis pathway construction in Escherichia coli. Phytochemistry. 2003;64(2):425-33.\u003c/li\u003e\n \u003cli\u003eDunlop MJ, Dossani ZY, Szmidt HL, Chu HC, Lee TS, Keasling JD, et al. Engineering microbial biofuel tolerance and export using efflux pumps. Mol Syst Biol. 2011;7:487.\u003c/li\u003e\n \u003cli\u003eReiling KK, Yoshikuni Y, Martin VJ, Newman J, Bohlmann J, Keasling JD. Mono and diterpene production in Escherichia coli. Biotechnol Bioeng. 2004;87(2):200-12.\u003c/li\u003e\n \u003cli\u003eAlonso-Gutierrez J, Chan R, Batth TS, Adams PD, Keasling JD, Petzold CJ, Lee TS. Metabolic engineering of Escherichia coli for limonene and perillyl alcohol production. Metab Eng. 2013;19:33-41.\u003c/li\u003e\n \u003cli\u003eNg P, Khoo LW, Thong A, Chew W. Optimization of extraction conditions for LC-ToF-MS analysis of mevalonate pathway metabolites in engineered E. coli strain via statistical experimental designs. Talanta. 2023;254:124182.\u003c/li\u003e\n \u003cli\u003eCasta\u0026ntilde;o-Cerezo S, Kulyk-Barbier H, Millard P, Portais JC, Heux S, Truan G, Bellvert F. Functional analysis of isoprenoid precursors biosynthesis by quantitative metabolomics and isotopologue profiling. Metabolomics. 2019;15(9):115.\u003c/li\u003e\n \u003cli\u003eShukal S, Lim XH, Zhang C, Chen X. Metabolic engineering of Escherichia coli BL21 strain using simplified CRISPR-Cas9 and asymmetric homology arms recombineering. Microbial Cell Factories. 2022;21(1):19.\u003c/li\u003e\n \u003cli\u003eWada K, Toya Y, Banno S, Yoshikawa K, Matsuda F, Shimizu H. 13C-metabolic flux analysis for mevalonate-producing strain of Escherichia coli. Journal of Bioscience and Bioengineering. 2017;123(2):177-82.\u003c/li\u003e\n \u003cli\u003eOgura K, Koyama T. Enzymatic Aspects of Isoprenoid Chain Elongation. Chem Rev. 1998;98(4):1263-76.\u003c/li\u003e\n \u003cli\u003eKu B, Jeong J-C, Mijts BN, Schmidt-Dannert C, Dordick JS. Preparation, Characterization, and Optimization of an In Vitro C\u003csub\u003e30\u003c/sub\u003e Carotenoid Pathway. Applied and Environmental Microbiology. 2005;71(11):6578-83.\u003c/li\u003e\n \u003cli\u003eApfel CM, Tak\u0026aacute;cs B, Fountoulakis M, Stieger M, Keck W. Use of genomics to identify bacterial undecaprenyl pyrophosphate synthetase: cloning, expression, and characterization of the essential uppS gene. J Bacteriol. 1999;181(2):483-92.\u003c/li\u003e\n \u003cli\u003eOkada K, Minehira M, Zhu X, Suzuki K, Nakagawa T, Matsuda H, Kawamukai M. The ispB gene encoding octaprenyl diphosphate synthase is essential for growth of Escherichia coli. Journal of Bacteriology. 1997;179(9):3058-60.\u003c/li\u003e\n \u003cli\u003eSaiki K, Mogi T, Anraku Y. Heme O biosynthesis in Escherichia coli: the cyoE gene in the cytochrome bo operon encodes a protoheme IX farnesyltransferase. Biochem Biophys Res Commun. 1992;189(3):1491-7.\u003c/li\u003e\n \u003cli\u003eKornberg HL. Routes for fructose utilization by Escherichia coli. J Mol Microbiol Biotechnol. 2001;3(3):355-9.\u003c/li\u003e\n \u003cli\u003eKotrba P, Inui M, Yukawa H. The ptsI Gene Encoding Enzyme I of the Phosphotransferase System of Corynebacterium glutamicum. Biochemical and Biophysical Research Communications. 2001;289(5):1307-13.\u003c/li\u003e\n \u003cli\u003ePostma PW, Lengeler JW, Jacobson GR. Phosphoenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiological Reviews. 1993;57:543 - 94.\u003c/li\u003e\n \u003cli\u003ePlumbridge J. Regulation of gene expression in the PTS in Escherichia coli: the role and interactions of Mlc. Current Opinion in Microbiology. 2002;5(2):187-93.\u003c/li\u003e\n \u003cli\u003eYao R, Hirose Y, Sarkar D, Nakahigashi K, Ye Q, Shimizu K. Catabolic regulation analysis of Escherichia coli and its crp, mlc, mgsA, pgi and ptsG mutants. Microbial Cell Factories. 2011;10(1):67.\u003c/li\u003e\n \u003cli\u003eBertsch J, Siemund AL, Kremp F, M\u0026uuml;ller V. A novel route for ethanol oxidation in the acetogenic bacterium Acetobacterium woodii: the acetaldehyde/ethanol dehydrogenase pathway. Environmental Microbiology. 2016;18(9):2913-22.\u003c/li\u003e\n \u003cli\u003eKim G, Yang J, Jang J, Choi J-S, Roe AJ, Byron O, et al. Aldehyde-alcohol dehydrogenase undergoes structural transition to form extended spirosomes for substrate channeling. Communications Biology. 2020;3(1):298.\u003c/li\u003e\n \u003cli\u003eVoynova NE, Rios SE, Miziorko HM. \u0026lt;i\u0026gt;Staphylococcus aureus\u0026lt;/i\u0026gt; Mevalonate Kinase: Isolation and Characterization of an Enzyme of the Isoprenoid Biosynthetic Pathway. Journal of Bacteriology. 2004;186(1):61-7.\u003c/li\u003e\n \u003cli\u003eDorsey JK, Porter JW. The Inhibition of Mevalonic Kinase by Geranyl and Farnesyl Pyrophosphates. Journal of Biological Chemistry. 1968;243(18):4667-70.\u003c/li\u003e\n \u003cli\u003eGray JC, Kekwick RGO. The inhibition of plant mevalonate kinase preparations by prenyl pyrophosphates. Biochimica et Biophysica Acta (BBA) - General Subjects. 1972;279(2):290-6.\u003c/li\u003e\n \u003cli\u003eHuang K-x, Scott AI, Bennett GN. Overexpression, Purification, and Characterization of the Thermostable Mevalonate Kinase from Methanococcus jannaschii. Protein Expression and Purification. 1999;17(1):33-40.\u003c/li\u003e\n \u003cli\u003eRedding-Johanson AM, Batth TS, Chan R, Krupa R, Szmidt HL, Adams PD, et al. Targeted proteomics for metabolic pathway optimization: application to terpene production. Metab Eng. 2011;13(2):194-203.\u003c/li\u003e\n \u003cli\u003eRinaldi MA, Ferraz CA, Scrutton NS. Alternative metabolic pathways and strategies to high-titre terpenoid production in Escherichia coli. Natural Product Reports. 2022;39(1):90-118.\u003c/li\u003e\n \u003cli\u003ePrimak YA, Du M, Miller MC, Wells DH, Nielsen AT, Weyler W, Beck ZQ. Characterization of a feedback-resistant mevalonate kinase from the archaeon Methanosarcina mazei. Appl Environ Microbiol. 2011;77(21):7772-8.\u003c/li\u003e\n \u003cli\u003eKazieva E, Yamamoto Y, Tajima Y, Yokoyama K, Katashkina J, Nishio Y. Characterization of feedback-resistant mevalonate kinases from the methanogenic archaeons Methanosaeta concilii and Methanocella paludicola. Microbiology (Reading). 2017;163(9):1283-91.\u003c/li\u003e\n \u003cli\u003eChen H, Liu C, Li M, Zhang H, Xian M, Liu H. Directed evolution of mevalonate kinase in Escherichia coli by random mutagenesis for improved lycopene. RSC Advances. 2018;8(27):15021-8.\u003c/li\u003e\n \u003cli\u003eChen H, Li M, Liu C, Zhang H, Xian M, Liu H. Enhancement of the catalytic activity of Isopentenyl diphosphate isomerase (IDI) from Saccharomyces cerevisiae through random and site-directed mutagenesis. Microb Cell Fact. 2018;17(1):65.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"metabolomics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mebo","sideBox":"Learn more about [Metabolomics](http://link.springer.com/journal/11306)","snPcode":"11306","submissionUrl":"https://submission.nature.com/new-submission/11306/3","title":"Metabolomics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Intracellular Metabolomics, Quantitative Metabolomics, Metabolic Network, Limonene, Mevalonate Pathway","lastPublishedDoi":"10.21203/rs.3.rs-4285213/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4285213/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIntroduction\u003c/p\u003e\n\u003cp\u003eLimonene is a monoterpene with diverse applications in food, medicine, fuel, and material science. Recently, engineered microbes have been used to biosynthesize target biochemicals such as limonene.\u003c/p\u003e\n\u003cp\u003eObjective\u003c/p\u003e\n\u003cp\u003eMetabolic engineering has shown that factors such as feedback inhibition, enzyme activity or abundance may contribute to the loss of target biochemicals. Incorporating a hypothesis driven experimental approach can help to streamline the process of improving target yield.\u003c/p\u003e\n\u003cp\u003eMethod\u003c/p\u003e\n\u003cp\u003eIn this work, time-series intracellular metabolomics data from \u003cem\u003eEscherichia coli\u003c/em\u003e cultures of a wild-type strain engineered to overproduce limonene (EcoCTs3) was collected, where we hypothesized having more carbon flux towards the engineered mevalonate (MEV) pathway would increase limonene yield. Based on the topology of the metabolic network, the pathways involved in mixed fermentation were possibly causing carbon flux loss from the MEV pathway. To prove this, knockout strains of lactate dehydrogenase(LDH) and aldehyde dehydrogenase-alcohol dehydrogenase (ALDH-ADH) were created.\u003c/p\u003e\n\u003cp\u003eResults\u003c/p\u003e\n\u003cp\u003eThe knockout strains showed 18 to 20 folds more intracellular mevalonate accumulation over time compared to the EcoCTs3 strain, thus indicating greater carbon flux directed towards the MEV pathway thereby increasing limonene yield by 8 to 9 folds.\u003c/p\u003e\n\u003cp\u003eConclusion\u003c/p\u003e\n\u003cp\u003eEnsuring high intracellular mevalonate concentration is therefore a good strategy to enhance limonene yield and other target compounds using the MEV pathway. Once high intracellular mevalonate concentration has been achieved, the limonene producing strain can then be further modified through other strategies such as enzyme and protein engineering to ensure better conversion of mevalonate to downstream metabolites to produce the target product limonene.\u003c/p\u003e","manuscriptTitle":"Enhancing Limonene production by probing the metabolic network through time-series metabolomics data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-24 07:18:22","doi":"10.21203/rs.3.rs-4285213/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-01-16T12:52:39+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"287829020366947451344322737848453481361","date":"2024-09-13T14:43:03+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-05-12T17:18:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"184732ca-cbbf-4f13-a011-c6c6cec1ced8","date":"2024-04-22T07:19:18+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-04-20T14:47:38+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-04-19T12:03:17+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-04-19T12:03:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Metabolomics","date":"2024-04-18T05:19:50+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"metabolomics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mebo","sideBox":"Learn more about [Metabolomics](http://link.springer.com/journal/11306)","snPcode":"11306","submissionUrl":"https://submission.nature.com/new-submission/11306/3","title":"Metabolomics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"f91ca0c2-08ea-4d06-8816-cf40a2aeacff","owner":[],"postedDate":"April 24th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-05-12T16:06:59+00:00","versionOfRecord":{"articleIdentity":"rs-4285213","link":"https://doi.org/10.1007/s11306-025-02254-y","journal":{"identity":"metabolomics","isVorOnly":false,"title":"Metabolomics"},"publishedOn":"2025-05-07 15:57:09","publishedOnDateReadable":"May 7th, 2025"},"versionCreatedAt":"2024-04-24 07:18:22","video":"","vorDoi":"10.1007/s11306-025-02254-y","vorDoiUrl":"https://doi.org/10.1007/s11306-025-02254-y","workflowStages":[]},"version":"v1","identity":"rs-4285213","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4285213","identity":"rs-4285213","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}