Determination of key enzymes for threonine synthesis through in vitro metabolic pathway analysis
© Zhang et al. 2015
Received: 5 February 2015
Accepted: 26 May 2015
Published: 13 June 2015
The overexpression of key enzymes in a metabolic pathway is a frequently used genetic engineering strategy for strain improvement. Metabolic control analysis has been proposed to quantitatively determine key enzymes. However, the lack of quality data often makes it difficult to correctly identify key enzymes through control analysis. Here, we proposed a method combining in vitro metabolic pathway analysis and proteomics measurement to find the key enzymes in threonine synthesis pathway.
All enzymes in the threonine synthesis pathway were purified for the reconstruction and perturbation of the in vitro pathway. Label-free proteomics technology combined with APEX (absolute protein expression measurements) data analysis method were employed to determine the absolute enzyme concentrations in the crude enzyme extract obtained from a threonine production strain during the fastest threonine production period. The flux control coefficient of each enzyme in the pathway was then calculated by measuring the flux changes after titration of the corresponding enzyme. The isoenzyme LysC catalyzing the first step in the pathway has the largest flux control coefficient, and thus its concentration change has the biggest impact on pathway flux. To verify that the key enzyme identified through in vitro pathway analysis is also the key enzyme in vivo, we overexpressed LysC in the original threonine production strain. Fermentation results showed that the threonine concentration was increased 30% and the yield was increased 20%.
In vitro metabolic pathways simulating in vivo cells can be built based on precise measurement of enzyme concentrations through proteomics technology and used for the determination of key enzymes through metabolic control analysis. This provides a new way to find gene overexpression targets for industrial strain improvement.
KeywordsKey enzyme Metabolic control analysis Threonine synthesis Proteomics Flux control coefficient
Experimental determination of FCC is possible by measuring the flux changes after perturbation on enzyme concentrations. A number of studies have dealt with the relationships between the flux of a metabolic system in vivo and the concentrations and activities of particular enzymes with the use of genetic means to alter the enzyme parameters [12, 13]. However, the in vivo studies usually confront difficulties in precise manipulation and measurement of enzyme concentrations. In contrast, alteration of enzyme concentrations in a cell free in vitro system is much easier if purified enzymes are available. The in vitro method bypasses cell walls and removes genetic regulation to enable direct access to the inner workings of the cell. Due to the unprecedented level of control and freedom of design, relative to in vivo systems, it has been widely studied for the production of biocommodities and bioelectricity in recent years [14–16]. In addition, in vitro analysis methods have also been used to guide the genetic engineering of strains for optimization of fatty acid, fatty alcohol and farnesene synthesis process [17–19].
One issue in using an in vitro system is that the enzyme concentrations must be similar with the in vivo system so that the key enzymes identified from in vitro analysis could also be effective overexpression targets. Intracellular protein concentrations could be measured through various proteomics technologies, which usually involve proteolysis of protein mixtures, followed by analysis of the peptides generated using chromatography and mass spectrometry (MS). However, due to the difference in size and amino acid composition of proteins, normal proteomics data is often suitable for comparing protein levels at different conditions but not for cross-protein comparison. To address this problem, Lu et al.  proposed a method called absolute protein expression measurements (APEX) to get the relative protein ratios from proteomics data. APEX relies upon correcting each protein’s mass spectrometry sampling depth (observed peptide count) by learned probabilities for identifying the peptides, like background expectation of observing each peptide in the experiment, the total sampling depth and the confidence in protein identification. And it is a robust and rapid method to quantify protein abundance without requiring construction of fusion protein libraries, labeling or internal standards. This method has been widely used for protein expression measurements [21–23].
In this research, we used label- free proteomics and APEX method to determine the absolute protein concentrations in a threonine production E. coli strain named Thr. Based on the data, we designed an in vitro multi-enzyme system to experimentally measure the FCCs of enzymes in threonine synthesis pathway. The results indicate that LysC, one of the isoenzymes of aspartate kinase, is the key enzyme for increasing threonine synthesis flux. We then overexpressed the lysC gene in the production strain and the threonine yield in the new strain was improved.
Results and discussion
Characterization of the crude enzyme extract and enzyme purification
Determination of the absolute enzyme concentrations
Measured absolute enzyme concentrations in E. coli cells
Concentrations from this study (μM)
Concentrations from PaxDba (μM)
3.24 ± 0.32
13.09 ± 0.71
9.01 ± 0.65
6.27 ± 0.42
12.78 ± 0.61
Key enzymes in the threonine synthesis pathway
In vivo validation of the key enzyme
In this study, a method combining proteomics measurement and in vitro metabolic pathway analysis was proposed to determine the key enzymes in threonine synthesis pathway. The precise measurement of absolute enzyme concentrations in producing cells and the possibility of accurate modification of enzyme concentrations in the in vitro pathway make it possible to quantitatively evaluate the impact of different enzymes on the pathway flux based on metabolic control analysis. Subsequently key enzymes in the pathway can be determined and the overexpression of these enzymes is more likely to be an effective strain modification strategy to improve the pathway flux in vivo. The enzyme LysC (usually responsible for lysine synthesis) showed the largest impact on threonine synthesis in our threonine production strain Thr. A modified strain with overexpressed LysC produced 30% more threonine compared with the original strain. It should be noted that in a different E. coli strain other enzymes rather than LysC might be key enzymes in threonine synthesis. However, the method developed in this study is still useful for key enzyme determination in different cells and different pathways. With the construction of even complex in vitro pathways (e.g. the whole central pathway), the method can provide versatile insights for understanding the regulation and organization of metabolic pathways and subsequently discover unusual novel metabolic engineering strategies for artificial cell factory design.
The substrates aspartate, NADPH, ATP, PLP and the inhibitors for the enzymes reaction lysine, methionine, and threonine were purchased from Sigma (USA), the restriction enzymes and PCR mix were purchased from Fermentas (USA). Na2HPO4 and other organic solvents used for HPLC were purchased from Merck (German). Other chemicals used in this article otherwise demonstrated were purchased from Solarbio (Beijing, China). The kits and markers used for the construction of clones were from Transgen (Beijing, China). The protein markers and SDS-PAGE gels were purchased from life technologies (USA).
The fermentation of threonine production strains and sampling
The crude enzyme extract was prepared from threonine production strain Thr. The cells were grown 8 h in LB shake flask cultures and then were inoculated in 3.5 L fermentation medium (composed of 80 g glucose, 4 g (NH4)2SO4, 0.02 g Ile, 2 g KH2PO3, 1 g MgSO4·7H2O, 4 g yeast extract powder, 0.015 g FeSO4·7H2O, 0.015 g MnSO4·H2O and 1 g lycine of every 1 L medium). The fermentation temperature was 37°C, pH was controlled at 7.0 and glucose concentration in the fermentation was controlled between 0 and 5 g/L. The fermentation broth was sampled every 2 h for threonine production and cell density analysis. A sample obtained at exponential growth phase and with high threonine production rate was used for proteomics analysis and crude enzyme extract experiments. The cells were harvested by centrifugation at 6,000g for 30 min at 4°C. For the crude enzyme extract preparation, the cells were washed three times with enzyme assay buffer (23 mM Tes, 114 mM KCl, 6 mM MgCl2, pH 7.5) and were disrupted by high pressure homogenizer. The cells lysate was centrifuged at 12,000g for 30 min at 4°C to remove cell debris. Then the supernatant was dialyzed to remove small molecules. The final crude enzyme extracts were stored at −80°C after flash freezing with liquid nitrogen for further proteomics analysis and crude enzyme extract experiments.
The cells were the same as those for preparation of the cell crude extract. About 100 mg wet mass were processed through single tube whole cell lysis and protein digestion. Cells were frozen by Liquid nitrogen immediately after centrifugation and washed three times with phosphate buffer (pH 7.0). The cells were then resuspended in a lysis buffer (Tris–HCl 0.1 M, pH 7.6, DTT 0.1 M) and broken with sonication cracking on ice with the parameters set as follows: 5 s on, 5 s off, total 15 min using the sonification device made in Nanjing. After sonication the supernatant was centrifuged for 20 min at 16,000g. The supernatant was filtered through sterile membrane filter (0.22 μm) and then transferred into a new tube. After boiling for 10 min the supernatant was stored at −80°C prior to sample cleanup if not for immediate use. The protein pellet extracted from previous step was quantitated using 2D-Quant Kit (purchased from GE healthcare). And then protein pellet from previous step was washed by UA buffer [8 M Urea dissolved in 0.1 M Tris–HCl (pH 8.5)] and dissolved in digestion buffer [100 mM TEAB (triethylammonium bicarbonate)] to a final concentration of 1 mg/mL. Equal aliquots were then digested with trypsin overnight at 37°C (Promega).
A NanoLC system (NanoLC 2D Ultra, Eksigent) equipped with Triple TOF 5600 mass spectrometer (AB SCIEX, USA) was used for analysis. Peptides were trapped on NanoLC pre-column (Chromxp C18, 3 μm, size 0.35 × 10 mm) and then eluted onto an analytical column (chromxp C18, 3 μm, size 0.075 × 150 mm) and separated by a 60 min gradient from 5 to 60% Buffer B (98% ACN, 2% H2O, 0.1% FA) at a flow rate of 300 nL/min. Full-scan MS was performed in positive ion mode with nano-ion spray voltage of 2.5 kv from 350 to 1,500 (m/z). For IDA, survey scans were acquired in 250 ms and as many as 30 product ion scans were collected if exceeding a threshold of 125 counts per second (counts/s) and with a + 2 to a + 5 charge-state.
PCR primers used in this study
Primer sequence (5′–3′)
CTAGCTAGCATGAAAAATGTTG GTT TTATCGGCT
CCG GAATTCATG ATGTCTGAAATTGTTGTCTCCAAAT
CGC GGATCCTTACTCAAACAAAT TACTATGC AGT
Strains and plasmids used in this research
Strains and plasmids
The genome of this strain was used as the template for amplification of threonine synthesis genes
E. coli DH5a
This strain was used for the clone of the plasmid
E. coli BL21
This strain was used for enzyme protein expression
N-terminal His-tagged thrA, inserted between NdeI and EcoRI sites of pET 28a (+)
N-terminal His-tagged lysC, inserted between NdeI and EcoRI sites of pET 28a (+)
N-terminal His-tagged asd, inserted between NheI and EcoRI sites of pET 28a (+)
N-terminal His-tagged thrB, inserted between NdeI and EcoRI sites of pET 28a (+)
N-terminal His-tagged thrC, inserted between NdeI and BamHI sites of pET 28a (+)
lysC, inserted between EcoRI and BamHI sites of pWSK29
Enzyme expression and purification
To purify the enzymes, the corresponding plasmids were introduced into E. coli BL21 (DE3). Single colonies were grown in LB medium with 50 μg/mL Kanamycin at 37°C until OD600 of the culture reached 0.6–0.8. Cultures were cooled to 16°C, and IPTG was added to a final concentration of 1 mM. After further growth at 16°C for 14–16 h, the cells were harvested by centrifugation at 6,000g for 30 min at 4°C, the supernatant was discarded. The pellet was suspended with buffer A (150 mM Tris–HCl, 150 mM NaCl and 20 mM imidazole, 1 mM DTT, pH 7.5) and the cells were disrupted by high pressure homogenizer. Then the cell lysate was centrifuged at 12,000g for 30 min at 4°C to remove cell debris. The supernatant was loaded onto Ni–NTA His-Bind column (GE Healthcare). The purification process was conducted on the protein purification machine AKTA purifier 10 with linear gradient imidazole from 20 mM to 500 mM in buffer. The fraction with higher OD280 was selected for SDS-PAGE. The fraction with pure protein was selected for dialysis in order to remove imidazole. After purification, the purified enzymes were quantitated using 2D-Quant Kit (purchased from GE healthcare) and purity was accessed by SDS-PAGE. The freshly purified proteins were stored at −80°C until use.
Assay of the enzyme activities
The activities of aspartate kinase in the crude enzyme extract were measured in a coupled assay with pyruvate kinase (PK) and lactate dehydrogenase (LDH) as described by Chassagnole and his colleagues  by following the NADH oxidation at 340 nm at 37°C. The reaction mixture contained in 200 μL of assay buffer, 5 mM aspartate, 5 mM ATP, 1.5 mM phosphoenolpyruvate, 0.3 mM NADH, 2.5 U PK, 2.5 U LDH and appropriate amount of crude enzyme extract. To investigate the contribution of different isoenzymes to the first reaction, 10 mM lysine and 10 mM methionine (enough to inhibit the corresponding isoenzyme) were added to the reaction system separately.
The activities of purified enzymes were assayed with detection of threonine after all five enzymes added. The reaction systems were as follows: 5 mM ASP, 5 mM ATP, 10 mM NADPH, 0.5 mM PLP, the concentrations of LysC, ThrA, Asd, ThrB, ThrC were all 100 nM. The reaction was conducted at 37°C for 0.5 h. The reaction liquid was heated at 100°C for 10 min to inactivate the enzymes when the reaction finished. After inactivation the liquid was centrifuged at 12,000g for 20 min and used for the assay of threonine by HPLC using the method provided by Agilent Company. The mobile phase A was 40 mM Na2HPO4, pH 7.8, mobile B was ACN:MeOH:water (45:45:10, v/v/v). The flow rate was 2 mL/min. The same procedure was used for crude enzyme extract activity analysis. There were three replicates of each assay.
In vivo validation of the key enzyme
For the in vivo validation of the key enzyme identified from the in vitro enzyme reactions, the overexpression plasmid was constructed using pWSK29 and primers named as lysC_F2 and lysC_R2 (shown in Table 2) in order to distinguish them from those used for construction of LysC expression plasmid. Thr strain containing the LysC overexpression plasmid was fermented and the fermentation process was the same as those in the fermentation of Thr strain for samples of crude enzyme extracts preparation.
ZY carried out the sequencing and cloning steps, protein purification and the writing of the manuscript. MQ and ZY participated in enzyme analyses, assay of the threonine and the FCCs analysis. MH supervised the design of the project and critically revised the manuscript. MY, ZP and SJ coordinated the whole project. LY and ZD conducted the proteomics analysis. ZX and JW performed the fermentation experiments. CG constructed the new strain and validated the key enzyme in vivo. All authors read and approved the final manuscript.
This work was supported by the National Key Basic Research Program of China (973 Program) (Nos. 2012CB725203, 2011CBA00804), National High Technology Research and Development Program of China (863 Program) (No. 2012AA022103), the Key Projects in the Tianjin Science & Technology Pillar Program (No. 11ZCZDSY08600, No.11ZCZDSY08400). We are thankful to Technology Support center of Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences for providing the use of experimental instruments and materials.
Compliance with ethical guidelines
Competing interests The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Lee KH, Park JH, Kim TY, Kim HU, Lee SY (2007) System metabolic engineering of Escherichia coli for l-threonine production. Mol Syst Biol 3:149View ArticleGoogle Scholar
- Lee JW, Na D, Park JM, Lee J, Choi S, Lee SY (2012) Systems metabolic engineering of microorganisms for natural and non-natural chemicals. Nat Chem Biol 8:536–546View ArticleGoogle Scholar
- Kim B, Kim WJ, Kim DI, Lee SY (2015) Application of genome-scale metabolic network model in metabolic engineering. J Ind Microbiol Biotechnol 42:339–348View ArticleGoogle Scholar
- King ZA, Lloyd CJ, Feist AM, Palsson BO (2015) Next-generation genome-scale models for metabolic engineering. Curr Opin Biotechnol 35C:23–29View ArticleGoogle Scholar
- McCloskey D, Palsson BØ, Feist AM (2013) Basic and applied uses of genome-scale metabolic network reconstructions of Escherichia coli. Mol Syst Biol 9:661View ArticleGoogle Scholar
- Park JM, Park HM, Kim WJ, Kim HU, Kim TY, Lee SY (2012) Flux variability scanning based on enforced objective flux for identifying gene amplification targets. BMC Syst Biol 6:106View ArticleGoogle Scholar
- Kacser H, Burns JA (1995) The control of flux. Biochem Soc Trans 23:341–366View ArticleGoogle Scholar
- Small JR, Kacser H (1993) Responses of metabolic systems to large changes in enzyme activities and effectors. Eur J Biochem 213:625–640View ArticleGoogle Scholar
- Raïs B, Chassagnole C, Letellier T, Fell DA, Mazat JP (2001) Threonine synthesis from aspartate in Escherichia coli cell-free extracts: pathway dynamics. Biochem J 356:425–432View ArticleGoogle Scholar
- Chassagnole C, Fell DA, Raïs B, Kudla B, Mazat JP (2001) Control of the threonine synthesis pathway in Escherichia coli: a theoretical and experimental approach. Biochem J 356:433–444View ArticleGoogle Scholar
- Rodríguez-Prados JC, de Atauri P, Maury J, Ortega F, Portais JC, Chassagnole C et al (2009) In silico strategy to rationally engineer metabolite production: a case study for threonine in Escherichia coli. Biotechnol Bioeng 103:609–620View ArticleGoogle Scholar
- Moreno-Sánchez R, Saavedra E, Rodríguez-Enríquez S, Olín-Sandoval V (2008) Metabolic control analysis: a tool for designing strategies to manipulate metabolic pathways. J Biomed Biotechnol 2008:597913View ArticleGoogle Scholar
- Morandini P (2013) Control limits for accumulation of plant metabolites: brute force is no substitute for understanding. Plant Biotechnol J 11:253–267View ArticleGoogle Scholar
- Zhang YH (2010) Production of biocommodities and bioelectricity by cell-free synthetic enzymatic pathway biotransformations: challenges and opportunities. Biotechnol Bioeng 105:663–677Google Scholar
- Zhang YH Production of biofuels and biochemicals by in vitro synthetic biosystems: opportunities and challenges. Biotechnol Adv, in pressGoogle Scholar
- Dudley QM, Karim AS, Jewett MC (2015) Cell-free metabolic engineering: biomanufacturing beyond the cell. Biotechnol J 10:69–82View ArticleGoogle Scholar
- Yu X, Liu T, Zhu F, Khosla C (2011) In vitro reconstitution and steady state analysis of the fatty acid synthase from Escherichia coli. Proc Natl Acad Sci USA 108:18643–18648View ArticleGoogle Scholar
- Liu R, Zhu F, Lu L, Fu A, Lu J, Deng Z et al (2014) Metabolic engineering of fatty acyl-ACP reductase-dependent pathway to improve fatty alcohol production in Escherichia coli. Metab Eng 22:10–21View ArticleGoogle Scholar
- Zhu F, Zhong X, Hu M, Lu L, Deng Z, Liu T (2014) In vitro reconstitution of mevalonate pathway and targeted engineering of farnesene overproduction in Escherichia coli. Biotechnol Bioeng 111:1396–1405View ArticleGoogle Scholar
- Lu P, Vogel C, Wang R, Yao X, Marcotte EM (2007) Absolute protein expression profiling estimates the relative contributions of transcriptional and translational regulation. Nat Biotechnol 25:117–124View ArticleGoogle Scholar
- Choi DS, Choi DY, Hong BS, Jang SC, Kim DK, Lee J et al (2012) Quantitative proteomics of extracellular vesicles derived from human primary and metastatic colorectal cancer cells. J Extracell Vesicles 1:18704Google Scholar
- Zhang Y, Fonslow BR, Shan B, Baek MC, Yates JR (2013) Protein analysis by shotgun/bottom-up proteomics. Chem Rev 113:2343–2394View ArticleGoogle Scholar
- Chen ZF, Zhang H, Wang H, Matsumura K, Wong YH, Ravasi T et al (2014) Quantitative proteomics study of larval settlement in the Barnacle Balanus amphitrite. PLoS One 9:e88744View ArticleGoogle Scholar
- Chassagnole C, Raïs B, Quentin E, Fell DA, Mazat JP (2001) An integreted study of threonine pathway enzyme kinetics in Escherichia coli. Biochem J 356:415–423View ArticleGoogle Scholar
- Protein abundance across organisms database. http://pax-db.org/#!Species/511145. Accessed 26 May 2015
- Bennett BD, Kimball EH, Gao M, Osterhout R, Van Dien SJ, Rabinowitz JD (2009) Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat Chem Biol 5:593–599View ArticleGoogle Scholar
- Lewis NE, Hixson KK, Conrad TM, Lerman JA, Charusanti P, Polpitiya AD et al (2010) Omic data from evolved E. coli are consistent with computed optimal growth from genome-scale models. Mol Syst Biol 6:690View ArticleGoogle Scholar
- Mancuso F, Bunkenborg J, Wierer M, Molina H (2012) Data extraction from proteomics raw data: an evaluation of nine tandem MS tools using a large Orbitrap data set. J Proteomics. 75:5293–5303View ArticleGoogle Scholar
- Wright JC, Collins MO, Yu L, Käll L, Brosch M, Choudhary JS (2012) Enhanced peptide identification by electron transfer dissociation using an improved Mascot percolator. Mol Cell Proteomics 11:478–491View ArticleGoogle Scholar
- Wikipedia. http://en.wikipedia.org/wiki/Escherichiacoli. Accessed 26 May 2015