Metabolic engineering of a reduced-genome strain of Escherichia coli for L-threonine production
© Lee et al. 2009
Received: 22 October 2008
Accepted: 07 January 2009
Published: 07 January 2009
Skip to main content
© Lee et al. 2009
Received: 22 October 2008
Accepted: 07 January 2009
Published: 07 January 2009
Deletion of large blocks of nonessential genes that are not needed for metabolic pathways of interest can reduce the production of unwanted by-products, increase genome stability, and streamline metabolism without physiological compromise. Researchers have recently constructed a reduced-genome Escherichia coli strain MDS42 that lacks 14.3% of its chromosome.
Here we describe the reengineering of the MDS42 genome to increase the production of the essential amino acid L-threonine. To this end, we over-expressed a feedback-resistant threonine operon (thrA*BC), deleted the genes that encode threonine dehydrogenase (tdh) and threonine transporters (tdcC and sstT), and introduced a mutant threonine exporter (rhtA23) in MDS42. The resulting strain, MDS-205, shows an ~83% increase in L-threonine production when cells are grown by flask fermentation, compared to a wild-type E. coli strain MG1655 engineered with the same threonine-specific modifications described above. And transcriptional analysis revealed the effect of the deletion of non-essential genes on the central metabolism and threonine pathways in MDS-205.
This result demonstrates that the elimination of genes unnecessary for cell growth can increase the productivity of an industrial strain, most likely by reducing the metabolic burden and improving the metabolic efficiency of cells.
The vast increase in annotated genome information and high-throughput technologies has enabled a systematic improvement of industrial microbes through genome engineering. Restructuring of microbial genomes has been shown to have several advantages over conventional approaches for strain improvement [1–5]. Restructured genomes with the desired functionalities have served as customized industrial strains that display (i) streamlined metabolic pathways for the production of selected biomaterials, (ii) a reduced production of unwanted by-products, and (iii) increased genome stability [6–9].
Escherichia coli is the most commonly used microbe for both laboratory research and industrial-scale production of metabolites, such as amino acids and proteins, for therapeutic or commercial uses [10–14]. Because E. coli has adapted to a lifestyle that includes residence in animal intestines with frequent exposure to aqueous and soil environments, many of the genes in its genome are unnecessary for growth in the relatively simple environment within industrial fermentors. Further, the complete genome sequence of E. coli has revealed numerous genes whose products have no known functions and genetic materials that may have been acquired from other organisms in the recent past . In an effort to improve E. coli as an industrial host, many researchers have deleted or added limited numbers of selected genes to the genome or modified plasmids to complement the existing genome [16–24]. These efforts have helped researchers make significant progress in improving E. coli as a production host, but have not addressed the productivity problems caused by the numerous E. coli genes with potentially detrimental functions.
Recently, an E. coli genome was reduced by the precise deletion of nonessential genes and other DNA sequences-including all known recombinogenic and mobile DNA and cryptic virulence genes-to construct a genetically stable strain that displays robust metabolic performance . The resulting strain, E. coli MDS42, has a chromosome that is 14.3% smaller than that of its parental E. coli strain MG1655. MDS42 shows robust growth under normal laboratory conditions and even better growth in high-cell density fermentations, as well as increased transformation efficiency relative to MG1655 . Therefore, the elimination of unnecessary genes and sequences from an E. coli genome appears to have produced a stable reduced-genome strain without physiological compromise. Furthermore, the deletion of all insertion sequence (IS) elements from the genome means that the strain is free of IS-mediated mutagenesis and genomic rearrangements.
In this study, we report the results of a reengineering of MDS42 to increase production of the amino acid L-threonine, which is essential for growth and maintenance of commercial livestock . This reengineering included the overexpression of a feedback-resistant threonine operon (thrA*BC) under the control of a recombinant Tac promoter, deletion of the genes that encode threonine dehydrogenase (tdh) and threonine uptake proteins (tdcC and sstT), and introduction of a mutant threonine exporter gene (rhtA23). The reengineered strain, called MDS-205, shows an ~83% increase in threonine production by flask fermentation relative to the wild-type E. coli strain MG1655 that had been engineered to carry the same threonine-specific modifications. Minimization of an E. coli genome by the elimination of genes unnecessary for growth increases the productivity of the strain by reducing the metabolic burden caused by maintenance and expression of unnecessary genes and improving the metabolic efficiency of the cell.
E. coli strains used in this study
Source or reference
λ - F - ilvG rfb-50 rph-1
Blattner et al. (1997)
MG1655 P Tac -thrA*BC ΔlacI
MG1655 P Tac -thrA*BC ΔlacI Δtdh
MG1655 P Tac -thrA*BC ΔlacI Δtdh ΔtdcC::rhtA23
MG1655 P Tac -thrA*BC ΔlacI Δtdh ΔtdcC::rhtA23 ΔsstT::rhtA23
Reduced genome strain
Posfai et al. (2006)
MDS42 P Tac -thrA*BC ΔlacI
MDS42 P Tac -thrA*BC ΔlacI Δtdh
MDS42 P Tac -thrA*BC ΔlacI Δtdh ΔtdcC::rhtA23
MDS42 P Tac -thrA*BC ΔlacI Δtdh ΔtdcC::rhtA23 ΔsstT::rhtA23
Yu et al. (2002)
W3350 rpsL rhtA23 thr::Tn10
Livshits et al. (2003)
E. coli ATCC 21277
K-12 SupE relA+KmR-P TAC -thrA*BC ilvA422
Shiio et al. (1971)
The threonine auxotroph MG1655 thrB::Tn5  was inoculated into 3 ml of LB medium supplemented with kanamycin and grown at 37°C. When the OD600 reached 0.8, the threonine auxotrophic cells were harvested by centrifugation at 6,000 g for 10 min, resuspended in 3 ml of fresh M9 minimal medium, and cultivated for 5 h at 37°C to deplete any endogenous threonine available to the cells . The threonine auxotrophic culture was then diluted 1:100 into 3 ml of fresh M9 minimal medium to which 300 μl of the filterate of the culture broth of each threonine producing strains was added. After 24 h of cultivation at 37°C, growth of the threonine auxotroph was measured at OD600.
A standard growth curve was established in order to correlate the growth of the auxotroph to various concentrations of threonine in M9 minimal media. Linear regression analysis of the plotted data was considered a least-squares fit (R2) of the relationship between threonine supplementation and growth of the auxotroph, and was used as a standard curve when R2 was greater than 0.95. The R2 value was derived from the regression line of the resulting plots (OD600 vs. threonine concentration).
The threonine-producing strains were grown on LB plates overnight and then transferred to a 250 ml flask containing 50 ml of seed medium [32.5 g glucose, 24.35 g K2HPO4, 9.5 g KH2PO4, 15 g yeast extract, 5 g (NH4)2SO4, 1 g MgSO4·7H2O per liter at pH 7.0]. After growing the culture for 24 h at 37°C, an aliquot (1 ml) of the seed culture was transferred to 50 ml of fermentation medium-1 [2 g yeast extract, 2 g citric acid, 25 g (NH4)2SO4, 7.46 g KH2PO4, 40 g glucose, 2 g MgSO4·7 H2O, 5 mg FeSO4·7 H2O, 5 mg MnSO4·4 H2O, and 20 g CaCO3 per liter at pH 7.2]. The fermentation was run for 24 h at 37°C, with vigorous agitation of the culture on a shaker (300 rpm). After cultivation, the amount of threonine accumulated in the broth was analyzed with the bioassay system using the threonine auxotroph described above.
MDS-205 was grown in a 2-liter jar fermentor containing 1.5 liters of fermentation medium-2 [100 g glucose, 10 g (NH4)2SO4, 2 g KH2PO4, 0.5 g MgSO4,·7 H2O, 5 mg FeSO4·7 H2O, 5 mg MnSO4·4 H2O, and 3 g yeast extract per liter at pH 7.5]. A seed culture was grown for 12 h at 37°C in a 500 ml flask containing 75 ml of seed medium and then inoculated into the 2-liter jar fermentor. During batch phase fermentation, the pH was maintained at 7.5 with NH4OH, the temperature at 37°C, the aeration rate at 1 vvm (air volume·working volume-1·min-1), and the agitation speed at 800 rev/min. After 30 h of fermentation, the concentration of threonine was determined by pre-column derivation with OPA (o-phthaldehyde-thiol) using the method developed by Joseph and Marsden  with the following modifications. Threonine was analysed on a Micra NPS ODS-1 (33 mm × 4.6 mm) 1.5-μm column (Eichrom Technologies, IL, USA) in reversed phase with a concentration gradient of sodium acetate buffer. This gradient was formed from two buffers, 100 mM sodium acetate, pH 5.9 (adjusted with 1 M HCl; buffer A) and pure methanol (buffer B), with a flow rate of 0.5 ml/min. The time course of the gradient was as follows: the starting point, buffer A/buffer B (98:2, v/v); 1 min, A/B (85:15, v/v); 5 min, A/B (50:50, v/v); 10 min, A/B (30:70, v/v); 18 min, A/B (2:98, v/v). The retention times and response factors of the threonine were evaluated by injecting known amounts of L-threonine.
Strains MG1655, MDS42, MG-105, and MDS-205 were inoculated from single colonies into 5 ml of seed medium and grown at 37°C overnight. From each overnight culture, 500 μl was used to inoculate 50 ml of fresh fermentation medium-1. These cultures were grown at 37°C, and the cells were harvested at an early log phase corresponding to an OD600 of 0.4. Total RNA was extracted using the MasterPure™ RNA Purification Kit (Epicentre Technologies, Madison, WI, USA) from 1 ml of the early log phase culture. cDNA synthesis and labeling were performed as described in the Affymetrix GeneChip E. coli Antisense Genome Array Technical Manual . The resulting labeled cDNAs were hybridized to an Affymetrix E. coli antisense genome array. Patterns of hybridization were detected with an Affymetrix Genearray scanner 2500 (Affymetrix, Inc., Santa Clara, CA, USA). The raw data were analyzed using Microarray Analysis Suite version 5.0 (Affymetrix). Every E. coli open reading frame (ORF) was assayed by a set of perfect match (PM) and mismatch (MM) probe pairs. If the PM probe showed an intensity that was at least 200 U higher than that of the MM probe, the probe pair was considered to be present . An ORF was considered to be present with 95% confidence if neighboring probe pairs within an ORF were present.
Third, the tdh gene, which encodes threonine dehydrogenase, was deleted from MG-102 and MDS-202 to prevent the degradation of L-threonine, generating strains MG-103 and MDS-203, which produced 42.25 and 79.94 mg/L of L-threonine, respectively (Fig. 2). Finally, to enhance the export of L-threonine into the culture media and block re-uptake, we sequentially replaced the tdcC gene, which encodes the threonine STP importer , and the sstT gene, which encodes the DctA dicarboxylate (DAACS) importer , in MG-103 and MDS-203 with a mutant threonine exporter gene (rhtA23). The resulting strains, MG-105 and MDS-205, produced 165.54 and 303.23 mg/L of L-threonine, respectively (Fig. 2). These final strains showed an ~10- and ~15.5-fold increase in L-threonine production, compared to their parental strains, MG1655 and MDS42, respectively (Fig. 2).
When a batch-fermentation was carried out, the final strain MDS-205 produced 40.1 g/L of L-threonine (a yield of 0.401 g threonine/g glucose) after 30 h of fermentation.
The restructuring of microbial genomes by eliminating genes that are unnecessary for a cellular metabolism has received special attention as an important strategy for industrial strain improvement [1–5]. Recently, scientists reported on the advantages of the reduced-genome E. coli strain MDS42 and its applications for bioindustry [5, 38]. The MDS42 genome is 14.3% smaller than that of the wild-type strain MG1655 and has ~700 fewer genes.
In this study, we compared the metabolic efficiency of L-threonine production in the wild-type E. coli strain MG1655 and reduced-genome E. coli strain MDS42 by introducing into each of these strains a series of genetic modifications that altered L-threonine production, degradation, export into the media, and re-uptake from the media. These modifications gave rise finally to MG-105 (from MG1655) and MDS-205 (from MDS42). Although both MG1655 and MDS42 went through the same modifications, the threonine production of MDS-205 strain was an ~2-fold greater than that of the MG-105 strain.
Genome-scale gene expression analysis of MG-105 and MDS-205 revealed that the most of the genes involved in the central metabolism and L-threonine biosynthesis were up-regulated in MDS-205. Among the up-regulated genes, the expression level of the rhtA threonine exporter gene was increased by 4.79-fold in MDS-205. This result indicates that the maximization of L-threonine export combined with deletion of the tdcC and sstT gene involved in the re-uptake of L-threonine across the membrane is one of the important steps for the mass production of L-threonine from E. coli. An additional advantage of the overexpression of the rhtA gene is that it increases the tolerance of E. coli to L-threonine by an ~3-fold , which helps the E. coli to withstand high concentrations of L-threonine in the media. This adaptation also might contribute to the higher L-threonine production.
The up-regulation of the ATP-dependent glucose transport and phosphorylation system (mglABC and glk genes) and down-regulation of the PTS system (ptsG gene) in MDS-205 may increase the availability of the phosphoenolpyruvate (PEP) which is a precursor of oxaloacetate (OAA) and aspartate. This result is consistent with the prediction based on flux balance analyses of gene knock-outs in an E. coli metabolic model that the replacement of PTS activity by an ATP-dependent glucose transport system should increase aspartate-family amino acids [39, 40]. In addition, down-regulation of poxB, pta, and adhE genes, which are directing PEP to acetate and ethanol, may reduce the PEP flux to acetate and ethanol and further increase the PEP availability.
It is also reported that the up-regulation of pck gene involved in the carboxylation of PEP to OAA leads to increased cellular growth and biomolecular production, since the ppc reaction releases an inorganic phosphate, in contrast to the pck reaction, which produces a high-energy ATP [41–44]. The effect of the pentose phosphate shunt and glyoxylate bypass on the glycolytic flux to L-threonine production also has been reported . Therefore, the up-regulation of the pck gene involved in the PEP carboxylation and tktAB genes involved in the pentose phosphate shunt and the increased expression of the genes involved in TCA cycle and the glyoxylate bypass (mdh, fumABC, sdhABCD, and glcB genes) may increase the OAA synthesis in MDS-205, resulted in higher production of L-threonine in MDS-205 compared to MG-105. The increased PEP and OAA level of MDS-205 by increasing the non-PTS glucose uptake system, PEP carboxylation, the pentose phosphate pathway, and the glyoxylate shunt, and reducing the mixed acid fermentation may resulted in increased production of L-threonine.
In this study, we report the results of a reengineering of MDS42 to increase production of the amino acid L-threonine, which is essential for growth and maintenance of commercial livestock . Even though a series of systematic experiments is needed for better understanding of the mechanism underlying the higher L-threonine production in the reduced-genome E. coli MDS42, our results described herein clearly indicate that MDS42 can serve as an efficient host strain for the production of other useful biomaterials.
This work was supported in part by grants from 21C Frontier Program of Microbial Genomics and Applications (MG08-0204-1-0), the Research Program of New Drug Target Discovery (M10748222314-08N4800-31410) from the Ministry of Education, Science and Technology, the Korea Science and Engineering Foundation Grant (R01-2008-000-20559-0), and the Korea Research Foundation Grant (KRF-2004-042-D00072).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.