Patent Application
20110177570
The present invention relates to a method for obtaining enhanced metabolite production in micro-organisms, and to mutants and/or transformants obtained with said method. More particularly, it relates to bacterial mutants and/or transformants for enhanced succinate production, especially mutants and/or transformants that are affected in the import and export of succinate.
Most environments are substrate limiting for micro-organisms, which has lead to very diverse and efficient carbon uptake systems (1). On the other hand, the excretion of end or intermediate products is less limiting for a micro-organism. Unless the excretion product has a competitive advantage (e.g. acetate excretion for acidification of the environment), excretion of certain end or intermediate products never needed to be as efficient, which has lead to a diverse selection of transport mechanisms (2,3).
From an industrial biotechnological perspective efficient excretion of an end-product can be a great advantage. It can lead to lower by-product formation, since the metabolism will not redirect carbon towards other exportable compounds and thus will lead to more easy to purify end-products. Additionally feedback inhibition of the pathway towards the product will be lowered, which logically leads to higher production rates. Both these production parameters, product purity and production rate, have previously been referred to as key parameters next to production yield (4-6) and were linked to the economically feasibility of a production process. The rising interest in industrial biotechnology originates in the increased awareness of the environmental impact of the existing industrial processes, the limited availability of fossil resources and the increasing political unrest that accompanies these evolutions. Up to now only few biotechnological processes are truly competitive with their chemical counterparts. In order to develop novel competitive processes a whole set of new techniques had to be developed, grouped in the so called discipline of ‘metabolic engineering’. This has already led to many new processes, in particular the development of succinate-production. Recent years many E. coli strains have been genetically modified with success, parallel to strain-development of Actinobacillus succinogenes, Mannheimia succiniciproducens, Anaerobiospirillum succiniciproducens.
Succinate as base chemical has first been pointed out by Greg Zeikus and coworkers in 1999 (7), after which the US Department of Energy (DOE) marked it as one of the top added value chemicals from renewable resources (4). Based on the petrochemical analogue, maleic anhydride, they have set the production price at 0.45/kg. Nowadays, with the vastly increasing oil price, this analogue more than tripled in price. Herein lays the opportunity for bio-based chemicals to rise and become economical viable.
A second well defined parameter in the DOE report is the volumetric production rate, set at 2.5 g/l/h. These rates are not easily obtained. Low specific growth and production rates are thus far limiting to reach competitive succinic acid production, since high biomass concentrations are needed to obtain economical viable production rates.
A strategy that has never been tried before is pulling the metabolism towards a certain product instead of pushing it, leading to enhanced production rates. For this purpose the C4 transport systems lend themselves excellently.
A nice review on C4 dicarboxylic acid transport and sensors (8), groups the transporters in 5 large transporter families based on amino acid sequence similarities, the DctA family, the DcuAB family, DcuC family, CitT family and the TRAP family. This classification has been adopted and expanded by the transporter classification database, which summarizes all known transporters and membrane proteins (9) and has classified them in the class of the secondary transporters. All potential C4 dicarboxylic acid transporters are all then classified in 7 superfamilies: MFS, Dcu, DAACS, CSS, DASS, DcuC and AEC (3), of which the CSS superfamily does not have any representative in E. coli.
Looking more closely at the individual C4 dicarboxylic acid transport families, two main distinctions can be made, aerobic and anaerobic transport in Escherichia coli. While the DctA family mainly is operational in an aerobic environment, the DcuAB and DcuC family is operational in anaerobic conditions. Their function is closely related to the type of metabolism E. coli has in these conditions. Anaerobically, fumarate will function as a terminal electron acceptor, thus C4 dicarboxylic acids such as fumarate and malate will be interesting carbon sources for E. coli, while succinate is an end-product and will thus be preferably excreted (10). Transport in this condition will mainly be focussed on the import of fumarate, malate and other pathway intermediates and the export of succinate. Aerobically on the other hand, succinate is a crucial intermediate in the Krebs-cycle. It would thus be unfavourable for the cell to excrete succinate. In this case the cell is provided with a rather efficient succinate (C4-dicarboxylic acid) uptake system (DctA) which keeps the extracellular concentration low. It is also known that not only the DctA family, but a yet to be discovered carrier ensures the cell of succinate uptake (11). Enhancing succinate excretion would evidently mean, changing the whole expression scheme of these transporters.
Surprisingly, we found that by overexpression of the dcuC exporter gene, preferably overexpression under aerobic conditions, and by the knock out of the dctA importer gene, the production of succinate can be enhanced, especially of mutants that do have already a slightly higher succinate production.
A first aspect of the invention is a mutant and/or recombinant micro-organism comprising a genetic change leading to increased succinate export activity and decreased succinate import activity. A mutant as used here can be obtained by any method known to the person skilled in the art, including but not limited to UV mutagenesis and chemical mutagenesis. Some features may be obtained by classical mutagenesis, while others may be obtained by genetic engineering. Preferably the mutant strain is a recombinant strain, where all mutations are obtained by site directed mutagenesis and/or transformation. Preferably said mutant and/or recombinant is selected from a genus known to produce succinic acid. Even more preferably, said mutant and/or recombinant is an Escherichia coli strain.
Preferably, the genetic change in said mutant and/or recombinant strain is affecting in the dcuC exporter gene and the dctA importer gene, or in the orthologues thereof. Orthologues, as used here are genes in other genera, with a certain percentage identity at amino acid level, and a similar function. Preferably, said percentage identity, as measured by a protein BLAST, is at least 40%, even more preferably at least 50%, most preferably at least 60%. Beside the dcuC exporter gene and the dctA importer genes other importer of exporter genes might be affected.
Preferably, said genetic change is the replacement of the promoter of the dcuC exporter gene, and the knock out of the dctA importer gene. Even more preferably, the promoter of the dcuC exporter gene is replaced by a strong promoter, most preferably by a strong promoter functioning under aerobic conditions.
Preferably, the mutant and/or recombinant micro-organism, according to the invention, further comprises a genetic change in one or more of the genes selected from the group consisting of ackA, poxB, pta, arcA, sdhA, sdhB, sdhC, sdhD, iclR, citD, citE, citF, pckA, maeA, maeB, eda, edd gltA, ppc, sstT, ydjN, ygjE, citT/ybdS, ybhI, yfbS, yhjE and ydfJ. Possibly, other genes may be selected on the base of their importance in the metabolic network (Table II).
Another aspect of the invention is the use of a mutant and/or recombinant micro-organism comprising a genetic change leading to increased succinate export activity and/or decreased succinate import activity, in combination with a genetic change leading to increased succinate production to produce succinate. Increased succinate production is defined here as an increase in succinate productivity per unit of biomass or per unit of volume, and/or an increased extracellular succinate concentration, and/or an increase in succinate yield per unit of substrate. Preferably, said genetic change leading to increased succinate production is a genetic change in one or more of the genes selected from the group consisting of ackA, poxB, pta, arcA, sdhA, sdhB, sdhC, sdhD, iclR, citD, citE, citF, pckA, maeA, maeB, eda, edd gltA, ppc, sstT, ydjN, ygjE, citT/ybdS, ybhI, yfbS, yhjE and ydfJ. Possibly, other genes may be selected on the base of their importance in the metabolic network (Table II). Preferably, said use is the use under aerobic conditions.
Still another aspect of the invention is a mutant and/or recombinant micro-organism comprising a genetic change leading to increased succinate export activity and decreased succinate import activity for the production of succinate. Preferably, said mutant and/or recombinant micro-organism, further comprises a genetic change in one or more of the genes selected from the group consisting of ackA, poxB, pta, arcA, sdhA, sdhB, sdhC, sdhD, iclR, citD, citE, citF, pckA, maeA, maeB, eda, edd gltA, ppc, sstT, ydjN, ygjE, citT/ybdS, ybhI, yfbS, yhjE and ydfJ. Possibly, other genes may be selected on the base of their importance in the metabolic network (Table II).
Materials and Methods to the Examples
Escherichia coli MG1655 [λ−, F−, rph-1] was obtained from the Coli Genetic Stock Center (CGSC). It was explicitly checked to not have the fnr deletion, as some strains with this name have it (12). The different strains were preserved in 50% glycerol-LB growth medium solution.
Table 1 summarizes all used strains, with their respectively mutations
The Luria Broth (LB) medium consisted of 1% tryptone peptone (Difco, Erembodegem, Belgium), 0.5% yeast extract (Difco) and 0.5% sodium chloride (VWR, Leuven, Belgium). Shake flask medium contained 2 g/l NH4Cl, 5 g/l (NH4)2SO4, 2.993 g/l KH2PO4, 7.315 g/l K2HPO4, 8.372 g/l MOPS, 0.5 g/l NaCl, 0.5 g/l MgSO4.7H2O, 16.5 g/l glucose.H2O, 1 ml/l vitamin solution, 100 μl/l molybdate solution and 1 ml/l selenium solution. The medium was set to a pH of 7 with 1M of KH2PO4.
Vitamin solution consisted of 3.6 g/l FeCl2.4H2O, 5 g/l CaCl2.2H2O, 1.3 g/l MnCl2.2H2O, 0.38 g/l CuCl2.2H2O, 0.5 g/l CoCl2.6H2O, 0.94 g/l ZnCl2, 0.0311 g/l H3BO4, 0.4 g/l Na2EDTA.2H2O and 1.01 g/l thiamine.HCl. The molybdate solution contained 0.967 g/l Na2Moa4.2H2O. The selenium solution contained 42 g/l SeO2. The minimal medium during fermentations contained 6.75 g/l NH4Cl, 1.25 g/l (NH4)2SO4, 1.15 g/l KH2PO4, 0.5 g/l NaCl, 0.5 g/l MgSO4.7H2O, 16.5 g/l glucose.H2O, 1 ml/l vitamin solution, 100 μl/l molybdate solution and 1 ml/l selenium solution with the same composition as described above.
A preculture from a single colony on a LB-plate was started in 5 ml LB medium during 8 hours at 37° C. on an orbital shaker at 200 rpm. From this culture, 2 ml was transferred to 100 ml minimal medium in a 500 ml shake flask, and incubated for 16 hours at 37° C. on an orbital shaker at 200 rpm. 4% inoculum was used in a 2 l Biostat B culture vessel with 1.5 l working volume (Sartorius-Stedim Biotech SA, Melsungen, Germany). The culture conditions were: 37° C., stirring at 800 rpm, gas flow rate of 1.5 l/min. The pH was maintained at 7 with 0.5M H2SO4 and 4M KOH. The exhaust gas was cooled down to 4° C. by an exhaust cooler (Frigomix 1000, Sartorius-Stedim Biotech SA, Melsungen, Germany). 10% solution of silicone antifoaming agent (BDH 331512K, VWR Int Ltd., Poole, England) was added when foaming rised during the fermentation (approx 10 μl). The off-gas was measured with an EL3020 off-gas analyser (ABB Automation GmbH, 60488 Frankfurt am Main, Germany).
The bioreactor contains in its interior a harvest pipe (BD Spinal Needle, 1.2×152 mm (BDMedical Systems, Franklin Lakes, N.J.—USA) connected to a reactor port, linked outside to a Masterflex 14 tubing (Cole-Parmer, Antwerpen, Belgium) followed by a harvest port with a septum for sampling. The other side of this Masterflex 16 tubing is connected back to the reactor vessel. This system is referred to as the rapid sampling loop. During sampling, reactor broth is pumped around in the sampling loop. It has been estimated that, at a flow rate of 150 ml/min, the reactor broth needs 0.04 s to reach the harvest port and 3.2 s to re-enter the reactor. At a pO2 level of 50%, there is around 3 mg/l of oxygen in the liquid. The pO2 level should never go below 20%. Thus 1.8 mg/l of oxygen may be consumed during transit through the harvesting loop. Assuming an oxygen uptake rate of 0.4 g oxygen/g biomass/h (the maximal oxygen uptake rate found at μmax), this gives for 5 g/l biomass, an oxygen uptake rate of 2 g/l/h or 0.56 mg/l/s, which multiplied by 3.2 s (residence time in the loop) gives 1.8 mg/l oxygen consumption.
In order to stop the metabolism of cells during the sampling, reactor broth was sucked through the harvest port in a syringe filled with 62 g stainless steel beads precooled at −20° C., to cool down 5 ml broth immediately to 4° C.). Sampling was immediately followed by cold centrifugation (15000 g, 5 min, 4° C.). In the batch experiments, a sample for OD600 and extracellular measurements was taken each hour using the rapid sampling loop and the cold stainless bead sampling method. When exponential growth was reached, the sampling frequency was increased to every 20 minutes.
Cell density of the culture was frequently monitored by measuring optical density at 600 nm (Uvikom 922 spectrophotometer, BRS, Brussel, Belgium). Cell dry weight was obtained by centrifugation (15 min, 5000 g, GSA rotor, Sorvall RC-5B, Goffin Meyvis, Kapellen, Belgium) of 20 g reactor broth in pre-dried and weighted falcons. The pellets were subsequently washed once with 20 ml physiological solution (9 g/l NaCl) and dried at 70° C. to a constant weight. To be able to convert OD measurements to biomass concentrations, a correlation curve of the OD to the biomass concentration was made.
The concentrations of glucose and organic acids were determined on a Varian Prostar HPLC system (Varian, Sint-Katelijne-Waver, Belgium), using an Aminex HPX-87H column (Bio-Rad, Eke, Belgium) heated at 65° C., equipped with a 1 cm precolumn, using 5 mM H2SO4 (0.6 ml/min) as mobile phase. Detection was done by a dual-wave UV-VIS (210 nm and 265 nm) detector (Varian Prostar 325) and a differential refractive index detector (Merck LaChrom L-7490, Merck, Leuven, Belgium). Peak identification was done by dividing the absorptions of the peaks in both 265 and 210 nm, which results in a constant value, typical for a certain compound (formula of Beer-Lambert).
Plasmids were maintained in the host E. coli. DH5α (F−, φ80dlacZΔM15, Δ(lacZYA-argF)U169, deoR, recA1, endA1, hsdR17(rk−, mk+), phoA, supE44, λ−, thi-1, gyrA96, relA1), pKD46 (Red helper plasmid, Ampicillin resistance), pKD3 (contain an FRT-flanked chloramphenicol resistance (cat) gene), pKD4 (contains an FRT-flanked kanamycin resistance (kan) gene), and pCP20 (expresses FLP recombinase activity) plasmids were obtained from Prof. Dr. J-P Hernalsteens (Vrije Universiteit Brussel, Belgium). The plasmid pBluescript (Fermentas, St. Leon-Rot, Germany) was used to construct the derivates of pKD3 and pKD4 with a promoter library, or with alleles carrying a point mutation.
Mutations. The mutations consisted in gene disruption (knock-out, KO), replacement of an endogenous promoter by an artificial promoter (knock-in, KI), and point mutation (PM) (
Transformants carrying a Red helper plasmid were grown in 10-ml LB media with ampicillin (100 mg/L) and L-arabinose (10 mM) at 30° C. to an OD600 of 0.6. The cells were made electrocompetent by washing them with 50 ml of ice-cold water, a first time, and with 1 ml ice-cold water, a second time. Then, the cells were resuspended in 50 μl of ice-cold water.
Electroporation was done with 50 μl of cells and 10-100 ng of linear double-stranded-DNA product by using a Gene Pulser™ (BioRad) (600 OHMS, 25 μFD, and 250 volts). After electroporation, cells were added to 1-ml LB media incubated 1 h at 37° C., and finally spread onto LB-agar containing 25 mg/L of chloramphenicol or 50 mg/L of kanamycin to select antibiotic resistant transformants. The selected mutants were verified by PCR with primers upstream and downstream of the modified region and were grown in LB-agar at 42° C. for the loss of the helper plasmid. The mutants were tested for ampicillin sensitivity.
Linear double-stranded-DNA. The linear ds-DNA amplicons were obtained by PCR using pKD3, pKD4 and their derivates as template. The primers used had a part of the sequence complementary to the template and another part complementary to the side on the chromosomal DNA where the recombination has to take place. For the KO, the region of homology was designed 50-nt upstream and 50-nt downstream of the start and stop codon of the gene of interest. For the KI, the transcriptional starting point (+1) had to be respected. The PM were generated with primers that contained the mutation. PCR products were PCR-purified, digested with DpnI, repurified from an agarose gel, and suspended in elution buffer (5 mM Tris, pH 8.0).
Elimination of the Antibiotic Resistance Gene. The selected mutants (chloramphenicol or kanamycin resistant) were transformed with pCP20 plasmid, which is an ampicillin and chloramphenicol resistant plasmid that shows temperature-sensitive replication and thermal induction of FLP synthesis. The ampicillin-resistant transformants were selected at 30° C., after which a few were colony purified in LB at 42° C. and then tested for loss of all antibiotic resistances and of the FLP helper plasmid.
Point Mutations. The strategy consisted in two-steps, first a KO of the gene of interest and second to introduce the mutated gene in the same chromosomal location (
The metabolic network model of Lequeux et al. (2005) (14) was used. It includes glycolysis, with glucose transport by the phosphotransferase system (PTS), the pentose phosphate pathway, the Krebs cycle, and overflow metabolism. For each amino acid and nucleotide the anabolic reactions were included. Biosynthesis of lipopolysaccharides (LPS), lipid A, peptidoglycane, and the lipid bilayer are incorporated as well. The oxidative phosphorylation ratio (P/O) was set to 1.33 (15,16). The reactions and metabolites considered in the model are depicted in Tables 2 and 3 respectively.
Partial Least Squares (PLS) regression has been performed in the software package R (17). This generalization of multiple linear regression is able to analyze data with strongly collinear and numerous independent variables as is the case for the elementary flux modes under study. Partial least squares regression is a statistical method that links a matrix of independent variables X with a matrix of dependent variables Y, i.e., the flux ratios and the succinate yield, respectively. Therefore, the multivariate spaces of X and Y are transformed to new matrices of lower dimensionality that are correlated to each other. This reduction of dimensionality is accomplished by principal component analysis like decompositions that are slightly titled to achieve maximum correlation between the latent variables of X and Y (18).
The elementary flux modes of the stoichiometric E. coli model of Lequeux et al (2005) (14) were calculated by using Metatool 5 (19).
Three different promoters, P8, P37 and P55 were selected from a promoter bank. These P8, P37 and P55 are ranked from weak to strong. By evaluating in a chemostat, peculiarly enough higher acetate production rates were found in the strain with dcuC constitutively expressed with promoter P55 in comparison with the other promoters. Moreover, inclusion bodies were observed at the cellular poles of the dcuC-P55 strain. This leads to the conclusion that P55 is too strong as promoter, and the weaker P37 was used for further experiments.
The effect of the transporters was tested in an sdhAB knock out strain, which produces already some succinic acid. Neither enhanced production rate nor higher yield could be observed in strains in which solely DctA or DcuC activity was altered. The combination of altered import and export increased the specific production rate with about 55% and the yield with approximately 53% (
Further investigation of the dctA single knock out has led to the conclusion that this strain grows faster on succinic acid than the wild type strain (
Different mutants affecting the succinate pathway have been constructed, as shown in Table I. These mutations were combined with the DctA knock out and the ΔFNR-pro37-dcuC overproducing construction. The results on the succinate yield are shown in
F6P
G3P + DHAP
G3P
NADH + H + BPG
ATP + 3PG
2PG
H2O + PEP
iCit
NADH + H + CO2 + aKGA
ATP + CoA + Suc
Mal
NADH − H + OAA
NAD + Lac
2NAD + CoA + Eth
ATP + CoA + Ac
NAD + CoA + Acdh
NAD | Eth
H2CO3
R5P
Xu5P
G3P + S7P
F6P + E4P
F6P + G3P
R1P
NADP + H2O + Glu
aKGA + Asp
aKGA + Ala
aKIV + Ala
aKGA + Val
H2O + Dhs
NADP + Shi
NADP + HSer
NADP + ThioredH2
NADH − FAD + H
NAD + CoQH2
NAD + NADPH
2ADP
ADP + CTP
NADH + H + CO2 + NH3 + MeTHF
ADP + PiOH + MalCoA
CoA + MalACP
AcCoA + ACP
NADP + Go3P
PPiOH + CDPDGo
GA1P
| Number | Date | Country | Kind |
|---|---|---|---|
| 08159442.6 | Jul 2008 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2009/058173 | 6/30/2009 | WO | 00 | 3/28/2011 |
