Metabolic engineering with adaptive laboratory evolution for phenylalanine production by Corynebacterium glutamicum.

The mutants resistant to a phenylalanine analog, 4-fluorophenylalanine (4FP), were obtained for metabolic engineering of Corynebacterium glutamicum for producing aromatic amino acids synthesized through the shikimate pathway by adaptive laboratory evolution. Culture experiments of the C. glutamicum strains which carry the mutations found in the open reading frame from the 4FP-resistant mutants revealed that the mutations in the open reading frames of aroG (NCgl2098), pheA (NCgl2799) and aroP (NCgl1062) encoding 3-deoxy-d-arabino-heptulosonate-7-phosphate, prephenate dehydratase, and aromatic amino acid transporter are responsible for 4FP resistance and higher concentration of aromatic amino acids in their culture supernatants in the 4FP-resistant strains. It was expected that aroG and pheA mutations would release feedback inhibition of the enzymes involved in the shikimate pathway by phenylalanine and that aroP mutations would prevent intracellular uptake of aromatic amino acids. Therefore, we conducted metabolic engineering of the C. glutamicum wild-type strain for aromatic amino acid production and found that phenylalanine production at 6.11 ± 0.08 g L-1 was achieved by overexpressing the mutant pheA and aroG genes from the 4FP-resistant mutants and deleting aroP gene. This study demonstrates that adaptive laboratory evolution is an effective way to obtain useful mutant genes related to production of target material and to establish metabolic engineering strategies.

Loss of function of Hog1 improves glycerol assimilation in Saccharomyces cerevisiae.

We previously isolated a mutant of Saccharomyces cerevisiae strain 85_9 whose glycerol assimilation was improved through adaptive laboratory evolution. To investigate the mechanism for this improved glycerol assimilation, genome resequencing of the 85_9 strain was performed, and the mutations in the open reading frame of HOG1, SIR3, SSB2, and KGD2 genes were found. Among these, a frameshift mutation in the HOG1 open reading frame was responsible for the improved glycerol assimilation ability of the 85_9 strain. Moreover, the HOG1 gene disruption improved glycerol assimilation. As HOG1 encodes a mitogen-activated protein kinase (MAPK), which is responsible for the signal transduction cascade in response to osmotic stress, namely the high osmolarity glycerol (HOG) pathway, we investigated the effect of the disruption of PBS2 gene encoding MAPK kinase for Hog1 MAPK on glycerol assimilation, revealing that PBS2 disruption can increase glycerol assimilation. These results indicate that loss of function of Hog1 improves glycerol assimilation in S. cerevisiae. However, single disruption of the SSK2, SSK22 and STE11 genes encoding protein kinases responsible for Pbs2 phosphorylation in the HOG pathway did not increase glycerol assimilation, while their triple disruption partially improved glycerol assimilation in S. cerevisiae. In addition, the HOG1 frameshift mutation did not improve glycerol assimilation in the STL1-overexpressing RIM15 disruptant strain, which was previously constructed with high glycerol assimilation ability. Furthermore, the effectiveness of the HOG1 disruptant as a bioproduction host was validated, indicating that the HOG1 CYB2 double disruptant can produce L-lactic acid from glycerol.

Ergothioneine production by Corynebacterium glutamicum harboring heterologous biosynthesis pathways.

In this study, Corynebacterium glutamicum was engineered to produce ergothioneine, an amino acid derivative with high antioxidant activity. The ergothioneine biosynthesis genes, egtABCDE, from Mycolicibacterium smegmatis were introduced into wild-type and l-cysteine-producing strains of C. glutamicum to evaluate their ergothioneine production. In the l-cysteine-producing strain, ergothioneine production reached approximately 40 mg L-1 after 2 weeks, and the amount was higher than that in the wild-type strain. As C. glutamicum possesses an ortholog of M. smegmatis egtA, which encodes an enzyme responsible for γ-glutamyl-l-cysteine synthesis, the effect of introducing egtBCDE genes on ergothioneine production in the l-cysteine-producing strain was evaluated, revealing that a further increase to more than 70 mg L-1 was achieved. As EgtBs from Methylobacterium bacteria are reported to use l-cysteine as a sulfur donor in ergothioneine biosynthesis, egtB from Methylobacterium was expressed with M. smegmatis egtDE in the l-cysteine-producing strain. As a result, ergothioneine production was further improved to approximately 100 mg L-1. These results indicate that utilization of the l-cysteine-producing strain and introduction of heterologous biosynthesis pathways from M. smegmatis and Methylobacterium bacteria are effective for improved ergothioneine production by C. glutamicum.

Constitutive expression of the global regulator AbrB restores the growth defect of a genome-reduced Bacillus subtilis strain and improves its metabolite production.

Partial bacterial genome reduction by genome engineering can improve the productivity of various metabolites, possibly via deletion of non-essential genome regions involved in undesirable metabolic pathways competing with pathways for the desired end products. However, such reduction may cause growth defects. Genome reduction of Bacillus subtilis MGB874 increases the productivity of cellulases and proteases but reduces their growth rate. Here, we show that this growth defect could be restored by silencing redundant or less important genes affecting exponential growth by manipulating the global transcription factor AbrB. Comparative transcriptome analysis revealed that AbrB-regulated genes were upregulated and those involved in central metabolic pathway and synthetic pathways of amino acids and purine/pyrimidine nucleotides were downregulated in MGB874 compared with the wild-type strain, which we speculated were the cause of the growth defects. By constitutively expressing high levels of AbrB, AbrB regulon genes were repressed, while glycolytic flux increased, thereby restoring the growth rate to wild-type levels. This manipulation also enhanced the productivity of metabolites including γ-polyglutamic acid. This study provides the first evidence that undesired features induced by genome reduction can be relieved, at least partly, by manipulating a global transcription regulation system. A similar strategy could be applied to other genome engineering-based challenges aiming toward efficient material production in bacteria.

Effectiveness of the Bacillus subtilis genome-reduced strain as an ethanol production host.

We investigated the performance of a genome-reduced strain of Bacillus subtilis MGB874, whose 0.87 Mbp of genomic DNA was cumulatively deleted, as an ethanol production host. A recombinant strain A267_EtOH was constructed by introducing the pdc and adhB genes from Zymomonas mobilis, both of which were expressed from an isopropyl-ß-d-1-thiogalactopyranoside-inducible spac promoter, into the A267 strain, a tryptophan prototrophic derivative of the MGB874 with disruption of metabolic pathways for producing lactic acid, acetic acid, and acetoin. Focusing on the stationary phase in fed-batch fermentation, 1.6 g L-1 ethanol was produced by the A267_EtOH strain after 144 h. Moreover, its ethanol production further increased by approximately 3.7-fold (5.9 g L-1) at 80 h through replacing the spac promoter for expressing pdc and adhB genes with the lytR promoter and the yield was about 112%. These results indicate that the MGB874 is an effective host for ethanol production during the stationary phase.

Adaptive Laboratory Evolution of Microorganisms: Methodology and Application for Bioproduction.

Adaptive laboratory evolution (ALE) is a useful experimental methodology for fundamental scientific research and industrial applications to create microbial cell factories. By using ALE, cells are adapted to the environment that researchers set based on their objectives through the serial transfer of cell populations in batch cultivations or continuous cultures and the fitness of the cells (i.e., cell growth) under such an environment increases. Then, omics analyses of the evolved mutants, including genome sequencing, transcriptome, proteome and metabolome analyses, are performed. It is expected that researchers can understand the evolutionary adaptation processes, and for industrial applications, researchers can create useful microorganisms that exhibit increased carbon source availability, stress tolerance, and production of target compounds based on omics analysis data. In this review article, the methodologies for ALE in microorganisms are introduced. Moreover, the application of ALE for the creation of useful microorganisms as cell factories has also been introduced.

Induction of glutamic acid production by copper in Corynebacterium glutamicum.

From the previous transcriptome analysis (Hirasawa et al. Biotechnol J 13:e1700612, 2018), it was found that expression of genes whose expression is regulated by stress-responsive transcriptional regulators was altered during penicillin-induced glutamic acid production in Corynebacterium glutamicum. Therefore, we investigated whether stress treatments, such as copper and iron addition, could induce glutamic acid production in C. glutamicum and found that the addition of copper did induce glutamic acid production in this species. Moreover, we also determined that glutamic acid production levels upon copper addition in a gain-of-function mutant strain of the mechanosensitive channel, NCgl1221, involved in glutamic acid export, were comparable to glutamic acid levels produced upon penicillin addition and biotin limitation in the wild-type strain. Furthermore, disruption of the odhI gene, which encodes a protein responsible for the decreased activity of the 2-oxoglutarate dehydrogenase complex during glutamic acid production, significantly diminished glutamic acid production induced by copper. These results indicate that copper can induce glutamic acid production and this induction requires OdhI like biotin limitation and penicillin addition, but a gain-of-function mutation in the NCgl1221 mechanosensitive channel is necessary for its high-level glutamic acid production. However, a significant increase in odhI transcription was not observed with copper addition in both wild-type and NCgl1221 gain-of-function mutant strains. In addition, disruption of the csoR gene encoding a copper-responsive transcriptional repressor enhanced copper-induced glutamic acid production in the NCgl1221 gain-of-function mutant, indicating that unidentified CsoR-regulated genes may contribute to copper-induced glutamic acid production in C. glutamicum. KEY POINTS: • Copper can induce glutamic acid production by Corynebacterium glutamicum. • Copper-induced glutamic acid production requires OdhI protein. • Copper-induced glutamic acid production requires a gain-of-function mutation in the mechanosensitive channel NCgl1221, which is responsible for the production of glutamic acid.

13C-metabolic flux analysis in glycerol-assimilating strains of Saccharomyces cerevisiae.

Glycerol is an attractive raw material for the production of useful chemicals using microbial cells. We previously identified metabolic engineering targets for the improvement of glycerol assimilation ability in Saccharomyces cerevisiae based on adaptive laboratory evolution (ALE) and transcriptome analysis of the evolved cells. We also successfully improved glycerol assimilation ability by the disruption of the RIM15 gene encoding a Greatwall protein kinase together with overexpression of the STL1 gene encoding the glycerol/H+ symporter. To understand glycerol assimilation metabolism in the evolved glycerol-assimilating strains and STL1-overexpressing RIM15 disruptant, we performed metabolic flux analysis using 13C-labeled glycerol. Significant differences in metabolic flux distributions between the strains obtained from the culture after 35 and 85 generations in ALE were not found, indicating that metabolic flux changes might occur in the early phase of ALE (i.e., before 35 generations at least). Similarly, metabolic flux distribution was not significantly changed by RIM15 gene disruption. However, fluxes for the lower part of glycolysis and the TCA cycle were larger and, as a result, flux for the pentose phosphate pathway was smaller in the STL1-overexpressing RIM15 disruptant than in the strain obtained from the culture after 85 generations in ALE. It could be effective to increase flux for the pentose phosphate pathway to improve the glycerol assimilation ability in S. cerevisiae.

Requirement of the LtsA Protein for Formation of the Mycolic Acid-Containing Layer on the Cell Surface of Corynebacterium glutamicum.

The ltsA gene of Corynebacterium glutamicum encodes a purF-type glutamine-dependent amidotransferase, and mutations in this gene result in increased susceptibility to lysozyme. Recently, it was shown that the LtsA protein catalyzes the amidation of diaminopimelate residues in the lipid intermediates of peptidoglycan biosynthesis. In this study, intracellular localization of wild-type and mutant LtsA proteins fused with green fluorescent protein (GFP) was investigated. The GFP-fused wild-type LtsA protein showed a peripheral localization pattern characteristic of membrane-associated proteins. The GFP-fusions with a mutation in the N-terminal domain of LtsA, which is necessary for the glutamine amido transfer reaction, exhibited a similar localization to the wild type, whereas those with a mutation or a truncation in the C-terminal domain, which is not conserved among the purF-type glutamine-dependent amidotransferases, did not. These results suggest that the C-terminal domain is required for peripheral localization. Differential staining of cell wall structures with fluorescent dyes revealed that formation of the mycolic acid-containing layer at the cell division planes was affected in the ltsA mutant cells. This was also confirmed by observation that bulge formation was induced at the cell division planes in the ltsA mutant cells upon lysozyme treatment. These results suggest that the LtsA protein function is required for the formation of a mycolic acid-containing layer at the cell division planes and that this impairment results in increased susceptibility to lysozyme.

Growth promotion in Corynebacterium glutamicum by overexpression of the NCgl2986 gene encoding a protein homologous to peptidoglycan amidases.

We previously reported the extracellular production of antibody fragment Fab by Corynebacterium glutamicum. In the course of searching for genes which improve the secretion efficiency of Fab, we coincidentally found that the final growth increased significantly when the NCgl2986 gene encoding an amidase-like protein was overexpressed. This effect was observed when cells were grown on the production medium MMTG, which contains high concentrations of glucose and neutralizing agent CaCO3, but not on MMTG without CaCO3 or Lennox medium. Not only turbidity but also dry cell weight was increased by NCgl2986 overexpression, although the growth rate was not affected. It was recently reported that the Mycobacterium tuberculosis homolog Rv3915 functions as an activator of MurA protein, which catalyzes the initial step of peptidoglycan synthesis. Growth promotion was also observed when the MurA protein was overproduced. His-tagged NCgl2986 protein was purified, but its peptidoglycan hydrolyzing activity could not be detected. These results suggest that NCgl2986 promotes cell growth by activating the peptidoglycan synthetic pathway.

Enhanced L-cysteine production by overexpressing potential L-cysteine exporter genes in an L-cysteine-producing recombinant strain of Corynebacterium glutamicum.

We identified L-cysteine exporter candidates of Corynebacterium glutamicum and investigated the effect of overexpression of the potential L-cysteine exporter genes on L-cysteine production in a recombinant strain of C. glutamicum. Overexpression of NCgl2566 and NCgl0580 resulted in enhanced L-cysteine production in an L-cysteine-producing recombinant strain of C. glutamicum.

Identification of metabolic engineering targets for improving glycerol assimilation ability of Saccharomyces cerevisiae based on adaptive laboratory evolution and transcriptome analysis.

Glycerol, a by-product of biodiesel production, has been utilized as a raw material for bioproduction. Saccharomyces cerevisiae, which has been used as a host microorganism for bioproduction, possesses the metabolic pathways for glycerol assimilation, but it cannot grow on glycerol as a carbon source. In this study, we identified metabolic engineering targets to improve the glycerol assimilation ability of S. cerevisiae based on adaptive laboratory evolution experiments using serial transfer of culture on glycerol and transcriptome analysis of the evolved cells using RNA-sequencing. The transcriptome data revealed that the upregulation of genes related to the tricarboxylic acid (TCA) cycle and oxidative phosphorylation contributed to the increased specific growth rate on glycerol during adaptive evolution. Furthermore, genes related to the pentose phosphate pathway were downregulated. Based on these observations, we identified metabolic engineering targets for improving glycerol assimilation. Overexpression of HAP4, which encodes one of the subunits of the Hap2p/3p/4p/5p transcription factor complex involved in the upregulation of the TCA cycle genes, or disruption of RIM15, which encodes a protein kinase related to the transcription regulator Gis1p, as well as overexpression of STL1, which encodes the glycerol/H+ symporter, improved the growth of S. cerevisiae on glycerol as the main carbon source. Our results indicate that the engineering targets can be identified based on adaptive laboratory evolution and transcriptome analysis of the evolved cells, and that the glycerol assimilation ability of S. cerevisiae is indeed improved by engineering the identified targets.

L-Cysteine production by metabolically engineered Corynebacterium glutamicum.

L-Cysteine is a commercially important amino acid. Here, we report the construction of L-cysteine-producing Corynebacterium glutamicum using a metabolic engineering approach. L-Serine O-acetyltransferase (SAT), encoded by cysE gene, is a key enzyme of L-cysteine biosynthesis, because of its feedback inhibition by L-cysteine. Therefore, we introduced a mutation into the C. glutamicum cysE gene, which appeared to desensitize SAT against feedback inhibition by L-cysteine. We successfully produced L-cysteine by overexpressing this mutant cysE gene in C. glutamicum, while the wild-type strain scarcely produced L-cysteine. To enhance the biosynthesis of L-serine (a substrate for SAT), a mutant serA gene, encoding D-3-phosphoglycerate dehydrogenase to desensitize it against feedback inhibition by L-serine, was additionally overexpressed in the mutant cysE-overexpressing strain and its L-cysteine production was indeed improved. Moreover, we disrupted the ldh gene encoding L-lactate dehydrogenase and the aecD gene encoding cysteine desulfhydrase to prevent the formation of lactic acid as a by-product and degradation of L-cysteine produced at the stationary phase, respectively, which resulted in enhanced L-cysteine production. However, since the concentration of L-cysteine produced still decreased at the stationary phase despite the aecD disruption, NCgl2463 encoding a possible cystine importer protein was further disrupted to prevent cystine import, because the produced L-cysteine is immediately oxidized to cystine. As a result, the time before the start of the decrease in L-cysteine concentration was successfully prolonged. Approximately 200 mg/L of L-cysteine production was achieved by overexpression of mutant cysE and serA genes and disruption of aecD and NCgl2463 genes in C. glutamicum.

Integrated Analysis of the Transcriptome and Metabolome of Corynebacterium glutamicum during Penicillin-Induced Glutamic Acid Production.

Corynebacterium glutamicum is known for its ability to produce glutamic acid and has been utilized for the fermentative production of various amino acids. Glutamic acid production in C. glutamicum is induced by penicillin. In this study, the transcriptome and metabolome of C. glutamicum is analyzed to understand the mechanism of penicillin-induced glutamic acid production. Transcriptomic analysis with DNA microarray revealed that expression of some glycolysis- and TCA cycle-related genes, which include those encoding the enzymes involved in conversion of glucose to 2-oxoglutaric acid, is upregulated after penicillin addition. Meanwhile, expression of some TCA cycle-related genes, encoding the enzymes for conversion of 2-oxoglutaric acid to oxaloacetic acid, and the anaplerotic reactions decreased. In addition, expression of NCgl1221 and odhI, encoding proteins involved in glutamic acid excretion and inhibition of the 2-oxoglutarate dehydrogenase, respectively, is upregulated. Functional category enrichment analysis of genes upregulated and downregulated after penicillin addition revealed that genes for signal transduction systems are enriched among upregulated genes, whereas those for energy production and carbohydrate and amino acid metabolisms are enriched among the downregulated genes. As for the metabolomic analysis using capillary electrophoresis time-of-flight mass spectrometry, the intracellular content of most metabolites of the glycolysis and the TCA cycle decreased dramatically after penicillin addition. Overall, these results indicate that the cellular metabolism and glutamic acid excretion are mainly optimized at the transcription level during penicillin-induced glutamic acid production by C. glutamicum.

Glutamate Fermentation-2: Mechanism of L-Glutamate Overproduction in Corynebacterium glutamicum.

The nonpathogenic coryneform bacterium, Corynebacterium glutamicum, was isolated as an L-glutamate-overproducing microorganism by Japanese researchers and is currently utilized in various amino acid fermentation processes. L-Glutamate production by C. glutamicum is induced by limitation of biotin and addition of fatty acid ester surfactants and ß-lactam antibiotics. These treatments affect the cell surface structures of C. glutamicum. After the discovery of C. glutamicum, many researchers have investigated the underlying mechanism of L-glutamate overproduction with respect to the cell surface structures of this organism. Furthermore, metabolic regulation during L-glutamate overproduction by C. glutamicum, particularly, the relationship between central carbon metabolism and L-glutamate biosynthesis, has been investigated. Recently, the role of a mechanosensitive channel protein in L-glutamate overproduction has been reported. In this chapter, mechanisms of L-glutamate overproduction by C. glutamicum have been reviewed.

Recent advances in amino acid production by microbial cells.

Amino acids have been utilized for the production of foods, animal feeds and pharmaceuticals. After the discovery of the glutamic acid-producing bacterium Corynebacterium glutamicum by Japanese researchers, the production of amino acids, which are primary metabolites, has been achieved using various microbial cells as hosts. Recently, metabolic engineering studies on the rational design of amino acid-producing microbial cells have been successfully conducted. Moreover, the technology of systems biology has been applied to metabolic engineering for the creation of amino acid-producing microbial cells. Currently, new technologies including synthetic biology, single-cell analysis, and evolutionary engineering have been utilized to create amino acid-producing microbial cells. In addition, useful compounds from amino acids have been produced by microbial cells. Here, current researches into the metabolic engineering of microbial cells toward production of amino acids and amino acid-related compounds are reviewed.

Phenotypic convergence in bacterial adaptive evolution to ethanol stress.

BACKGROUND: Bacterial cells have a remarkable ability to adapt to environmental changes, a phenomenon known as adaptive evolution. During adaptive evolution, phenotype and genotype dynamically changes; however, the relationship between these changes and associated constraints is yet to be fully elucidated. RESULTS: In this study, we analyzed phenotypic and genotypic changes in Escherichia coli cells during adaptive evolution to ethanol stress. Phenotypic changes were quantified by transcriptome and metabolome analyses and were similar among independently evolved ethanol tolerant populations, which indicate the existence of evolutionary constraints in the dynamics of adaptive evolution. Furthermore, the contribution of identified mutations in one of the tolerant strains was evaluated using site-directed mutagenesis. The result demonstrated that the introduction of all identified mutations cannot fully explain the observed tolerance in the tolerant strain. CONCLUSIONS: The results demonstrated that the convergence of adaptive phenotypic changes and diverse genotypic changes, which suggested that the phenotype-genotype mapping is complex. The integration of transcriptome and genome data provides a quantitative understanding of evolutionary constraints.

Enhanced dipicolinic acid production during the stationary phase in Bacillus subtilis by blocking acetoin synthesis.

Bacterial bio-production during the stationary phase is expected to lead to a high target yield because the cells do not consume the substrate for growth. Bacillus subtilis is widely used for bio-production, but little is known about the metabolism during the stationary phase. In this study, we focused on the dipicolinic acid (DPA) production by B. subtilis and investigated the metabolism. We found that DPA production competes with acetoin synthesis and that acetoin synthesis genes (alsSD) deletion increases DPA productivity by 1.4-fold. The mutant showed interesting features where the glucose uptake was inhibited, whereas the cell density increased by approximately 50%, resulting in similar volumetric glucose consumption to that of the parental strain. The metabolic profiles revealed accumulation of pyruvate, acetyl-CoA, and the TCA cycle intermediates in the alsSD mutant. Our results indicate that alsSD-deleted B. subtilis has potential as an effective host for stationary-phase production of compounds synthesized from these intermediates.

Effect of malic enzyme on ethanol production by Synechocystis sp. PCC 6803.

We investigated effects of malic enzyme on ethanol production by Synechocystis sp. PCC 6803 under autotrophic conditions. Deletion of me, which encodes malic enzyme, decreased ethanol production, whereas its overexpression had no effect. Our results suggest that maintaining optimal malic enzyme activity controls ethanol production by Synechocystis sp. PCC 6803.