Omics analysis coupled with gene editing revealed potential transporters and regulators related to levoglucosan metabolism efficiency of the engineered Escherichia coli

Background

Bioconversion of levoglucosan, a promising sugar derived from the pyrolysis of lignocellulose, into biofuels and chemicals can reduce our dependence on fossil-based raw materials. However, this bioconversion process in microbial strains is challenging due to the lack of catalytic enzyme relevant to levoglucosan metabolism, narrow production ranges of the native strains, poor cellular transport rate of levoglucosan, and inhibition of levoglucosan metabolism by other sugars co-existing in the lignocellulose pyrolysate. The heterologous expression of eukaryotic levoglucosan kinase gene in suitable microbial hosts like Escherichia coli could overcome the first two challenges to some extent; however, no research has been dedicated to resolving the last two issues till now.

Results

Aiming to resolve the two unsolved problems, we revealed that seven ABC transporters (XylF, MalE, UgpB, UgpC, YtfQ, YphF, and MglA), three MFS transporters (KgtP, GntT, and ActP), and seven regulatory proteins (GalS, MhpR, YkgD, Rsd, Ybl162, MalM, and IraP) in the previously engineered levoglucosan-utilizing and ethanol-producing E. coli LGE2 were induced upon exposure to levoglucosan using comparative proteomics technique, indicating these transporters and regulators were involved in the transport and metabolic regulation of levoglucosan. The proteomics results were further verified by transcriptional analysis of 16 randomly selected genes. Subsequent gene knockout and complementation tests revealed that ABC transporter XylF was likely to be a levoglucosan transporter. Molecular docking showed that levoglucosan can bind to the active pocket of XylF by seven H-bonds with relatively strong strength.

Conclusion

This study focusing on the omics discrepancies between the utilization of levoglucosan and non-levoglucosan sugar, could provide better understanding of levoglucosan transport and metabolism mechanisms by identifying the transporters and regulators related to the uptake and regulation of levoglucosan metabolism. The protein database generated from this study could be used for further screening and characterization of the transporter(s) and regulator(s) for downstream enzymatic/genetic engineering work, thereby facilitating more efficient microbial utilization of levoglucosan for biofuels and chemicals production in future.

Both levoglucosan kinase activity and transport capacity limit the utilization of levoglucosan in Saccharomyces cerevisiae

Manufacturing fuels and chemicals from cellulose materials is a promising strategy to achieve carbon neutralization goals. In addition to the commonly used enzymatic hydrolysis by cellulase, rapid pyrolysis is another way to degrade cellulose. The sugar obtained by fast pyrolysis is not glucose, but rather its isomer, levoglucosan (LG). Here, we revealed that both levoglucosan kinase activity and the transportation of levoglucosan are bottlenecks for LG utilization in Saccharomyces cerevisiae, a widely used cell factory. We revealed that among six heterologous proteins that had levoglucosan kinase activity, the 1,6-anhydro-N-acetylmuramic acid kinase from Rhodotorula toruloides was the best choice to construct levoglucosan-utilizing S. cerevisiae strain. Furthermore, we revealed that the amino acid residue Q341 and W455, which were located in the middle of the transport channel closer to the exit, are the sterically hindered barrier to levoglucosan transportation in Gal2p, a hexose transporter. The engineered yeast strain expressing the genes encoding the 1,6-anhydro-N-acetylmuramic acid kinase from R. toruloides and transporter mutant Gal2p^Q341A or Gal2p^W455A consumed ~ 4.2 g L⁻¹ LG in 48 h, which is the fastest LG-utilizing S. cerevisiae strain to date.

Mutations in adaptively evolved Escherichia coli LGE2 facilitated the cost-effective upgrading of undetoxified bio-oil to bioethanol fuel

Levoglucosan is a promising sugar present in the lignocellulose pyrolysis bio-oil, which is a renewable and environment-friendly source for various value-added productions. Although many microbial catalysts have been engineered to produce biofuels and chemicals from levoglucosan, the demerits that these biocatalysts can only utilize pure levoglucosan while inhibited by the inhibitors co-existing with levoglucosan in the bio-oil have greatly limited the industrial-scale application of these biocatalysts in lignocellulose biorefinery. In this study, the previously engineered Escherichia coli LGE2 was evolved for enhanced inhibitor tolerance using long-term adaptive evolution under the stress of multiple inhibitors and finally, a stable mutant E. coli-H was obtained after ~ 374 generations' evolution. In the bio-oil media with an extremely acidic pH of 3.1, E. coli-H with high inhibitor tolerance exhibited remarkable levoglucosan consumption and ethanol production abilities comparable to the control, while the growth of the non-evolved strain was completely blocked even when the pH was adjusted to 7.0. Finally, 8.4 g/L ethanol was achieved by E. coli-H in the undetoxified bio-oil media with ~ 2.0% (w/v) levoglucosan, reaching 82% of the theoretical yield. Whole-genome re-sequencing to monitor the acquisition of mutations identified 4 new mutations within the globally regulatory genes rssB, yqhA, and basR, and the - 10 box of the putative promoter of yqhD-dgkA operon. Especially, yqhA was the first time to be revealed as a gene responsible for inhibitor tolerance. The mutations were all responsible for improved fitness, while basR mutation greatly contributed to the fitness improvement of E. coli-H. This study, for the first time, generated an inhibitor-tolerant levoglucosan-utilizing strain that could produce cost-effective bioethanol from the toxic bio-oil without detoxification process, and provided important experimental evidence and valuable genetic/proteinic information for the development of other robust microbial platforms involved in lignocellulose biorefining processes.

Conversion of levoglucosan into glucose by the coordination of four enzymes through oxidation, elimination, hydration, and reduction

Levoglucosan (LG) is an anhydrosugar produced through glucan pyrolysis and is widely found in nature. We previously isolated an LG-utilizing thermophile, Bacillus smithii S-2701M, and suggested that this bacterium may have a metabolic pathway from LG to glucose, initiated by LG dehydrogenase (LGDH). Here, we completely elucidated the metabolic pathway of LG involving three novel enzymes in addition to LGDH. In the S-2701M genome, three genes expected to be involved in the LG metabolism were found in the vicinity of the LGDH gene locus. These four genes including LGDH gene (lgdA, lgdB1, lgdB2, and lgdC) were expressed in Escherichia coli and purified to obtain functional recombinant proteins. Thin layer chromatography analyses of the reactions with the combination of the four enzymes elucidated the following metabolic pathway: LgdA (LGDH) catalyzes 3-dehydrogenation of LG to produce 3-keto-LG, which undergoes β-elimination of 3-keto-LG by LgdB1, followed by hydration to produce 3-keto-D-glucose by LgdB2; next, LgdC reduces 3-keto-D-glucose to glucose. This sequential reaction mechanism resembles that proposed for an enzyme belonging to glycoside hydrolase family 4, and results in the observational hydrolysis of LG into glucose with coordination of the four enzymes.

Heterologous Production of Squalene from Glucose in Engineered Corynebacterium glutamicum Using Multiplex CRISPR Interference and High-Throughput Fermentation

The sustainable production of squalene has driven the development of microbial cell factories due to the limitation of low-yielding bioprocesses from plants and illegal harvesting shark liver. We report the metabolic engineering of Corynebacterium glutamicum to produce squalene from glucose. Combinatorial metabolic engineering strategies for precursor rebalancing, redox balancing, and blocking the competing pathway for the isopentenyl diphosphate availabilities were applied by repressing the target genes using the CRISPR interference. The best engineered strain using high-throughput fermentation produced squalene from glucose at 5.4 ± 0.3 mg/g dry cell weight (DCW) and 105.3 ± 3.0 mg/L, which was a 5.2-fold increase over the parental strain. In addition, flask cultivation of C. glutamicum overexpressing the dxs and idi genes with squalene synthase gene and repressing the idsA gene resulted in production of squalene at 5.8 ± 0.4 mg/g DCW and 82.8 ± 6.2 mg/L, which was a 3.4-fold increase over the parental strain.

Rapid identification of unknown carboxyl esterase activity in Corynebacterium glutamicum using RNA-guided CRISPR interference

RNA-guided genome engineering technologies have been developed for the advanced metabolic engineering of microbial cells to enhance production of value-added chemicals in Corynebacterium glutamicum as an industrial host. In this study, the RNA-guided CRISPR interference (CRISPRi) was applied to rapidly identify of unknown genes for native esterase activity in C. glutamicum. Combining with the carboxyl esterase (MekB) protein sequence alignment, two target genes (the cg0961 and cg0754) were selected for the CRISPRi application to investigate the possible native esterase in C. glutamicum. The recombinant strain with repressed expression of the cg0961 gene exhibited almost no capability on degradation of methyl acetate as a substrate of carboxyl esterase. This result was also confirmed in the cg0961 gene deletion mutant. Thus, we concluded that Cg0961 plays a major role of the native carboxyl esterase activity in C. glutamicum. In addition, CRISPRi demonstrated an application for gene identification and its function as another genetic tool for metabolic engineering in C. glutamicum.

RNA-guided single/double gene repressions in Corynebacterium glutamicum using an efficient CRISPR interference and its application to industrial strain

BACKGROUND

The construction of microbial cell factories requires cost-effective and rapid strain development through metabolic engineering. Recently, RNA-guided CRISPR technologies have been developed for metabolic engineering of industrially-relevant host.

RESULTS

To demonstrate the application of the CRISPR interference (CRISPRi), we developed two-plasmid CRISPRi vectors and applied the CRISPRi in Corynebacterium glutamicum to repress single target genes and double target genes simultaneously. Four-different single genes (the pyc, gltA, idsA, and glgC genes) repressions were successfully performed using the CRISPRi vectors, resulting significant mRNA reductions of the targets compared to a control. Subsequently, the phenotypes for the target gene-repressed strains were analyzed, showing the expected cell growth behaviors with different carbon sources. In addition, double gene repression (the idsA and glgC genes in a different order) by the CRISPRi resulted in an independent gene repression to each target gene simultaneously. To demonstrate an industrial application of the CRISPRi, citrate synthase (CS)-targeting DM1919 (L-lysine producer) strains with a sgRNA-gltA-r showed reduced CS activity, resulting in the improvement of L-lysine yield by 1.39-fold than the parental DM1919 (a lysine producer).

CONCLUSIONS

Single or double gene repression were successfully performed using the CRISPRi vectors and sequence specific sgRNAs. The CRISPRi can be applied for multiplex metabolic engineering to enhanced lysine production and it will promote the further rapid development of microbial cell factories of C. glutamicum.

Carotenoid Production by Recombinant Corynebacterium glutamicum: Strain Construction, Cultivation, Extraction, and Quantification of Carotenoids and Terpenes

Corynebacterium glutamicum is a workhorse of industrial amino acid production employed for more than five decades for the million-ton-scale production of L-glutamate and L-lysine. This bacterium is pigmented due to the biosynthesis of the carotenoid decaprenoxanthin. Decaprenoxanthin is a carotenoid with 50 carbon atoms, and, thus, C. glutamicum belongs to the rare group of bacteria that produce long-chain C50 carotenoids. C50 carotenoids have been mainly isolated from extremely halophilic archaea (Kelly and Jensen, Acta Chem Scand 21:2578, 1967; Pfander, Pure Appl Chem 66:2369-2374, 1994) and from Gram-positive bacteria of the order Actinomycetales (Netzer et al., J Bacteriol 192:5688-5699, 2010). The characteristic yellow phenotype of C. glutamicum is due to the cyclic C50 carotenoid decaprenoxanthin and its glycosides. Decaprenoxanthin production has been improved by plasmid-borne overexpression of endogenous genes of carotenogenesis. Gene deletion resulted in the production of the C40 carotenoid lycopene, an intermediate of decaprenoxanthin biosynthesis. Heterologous gene expression was required to develop strains overproducing nonnative carotenoids and terpenes, such as astaxanthin (Henke et al., Mar Drugs 14:E124, 2016) and (+)-valencene (Frohwitter et al., J Biotechnol 191:205-213, 2014). Integration of additional copies of endogenous genes expressed from strong promoters improved isoprenoid biosynthesis. Here, we describe C. glutamicum strains, plasmids, and methods for overexpression of endogenous and heterologous genes, gene deletion, replacement, and genomic integration. Moreover, strain cultivation as well as extraction, identification, and quantitative determination of terpenes and carotenoids produced by C. glutamicum is detailed.

Modular pathway engineering of Corynebacterium glutamicum to improve xylose utilization and succinate production

Xylose-negative Corynebacterium glutamicum has been engineered to utilize xylose as the sole carbon source via either the xylose isomerase (XI) pathway or the Weimberg pathway. Heterologous expression of xylose isomerase and overexpression of a gene encoding for xylulose kinase enabled efficient xylose utilization. In this study, we show that two functionally-redundant transcriptional regulators (GntR1 and GntR2) present on xylose repress the pentose phosphate pathway genes. For efficient xylose utilization, pentose phosphate pathway genes and a phosphoketolase gene were overexpressed with the XI pathway in C. glutamicum. Overexpression of the genes encoding for transaldolase (Tal), 6-phosphogluconate dehydrogenase (Gnd), or phosphoketolase (XpkA) enhanced the growth and xylose consumption rates compared to the wild-type with the XI pathway alone. However, co-expression of these genes did not have a synergetic effect on xylose utilization. For the succinate production from xylose, overexpression of the tal gene with the XI pathway in a succinate-producing strain improved xylose utilization and increased the specific succinate production rate by 2.5-fold compared to wild-type with the XI pathway alone. Thus, overexpression of the tal, gnd, or xpkA gene could be helpful for engineering C. glutamicum toward production of value-added chemicals with efficient xylose utilization.

Bioconversion of anhydrosugars: Emerging concepts and strategies

As methods for the use of anhydrosugars in chemical and biofuel production continue to develop, our collective knowledge of anhydrosugar processing enzymes continues to improve, including their mechanistic details, structural dynamics and modes of substrate binding. Of particular interest, anhydrosugar kinases, such as levoglucosan kinase (LGK) and 1,6-anhydro-N-acetylmuramic acid kinase (AnmK), utilize an unusual mechanism whereby the sugar substrate is both cleaved and phosphorylated. The phosphorylated sugar can then be routed to other metabolic pathways, thereby allowing its further bioconversion. Advanced engineering efforts to improve the catalytic efficiency and stability of LGK have been steadily progressing. Other enzymes that cleave the glycosidic bond of disaccharide sugars containing an anhydrosugar component are also being identified and characterized. Accordingly, the potential future use of these enzymes in large-scale production strategies is becoming increasingly viable. Here, a mini-review of the observed characteristics of anhydrosugar processing enzymes is presented along with recent developments in the bioconversion of these sugars. © 2016 IUBMB Life 68(9):700-708, 2016.