Zei Wen Wang

Systematic Approach To Engineer Escherichia coli Pathways for Co-utilization of a Glucose–Xylose Mixture

Glucose and xylose are two major sugars of lignocellulosic hydrolysate. The regulatory program of catabolite repression in Escherichia coli dictates the preferred utilization of glucose over xylose, which handicaps the development of the lignocellulose-based fermentation process. To co-utilize a glucose-xylose mixture, the E. coli strain was manipulated by pathway engineering in a systematic way. The approach included (1) blocking catabolite repression, (2) enhancing glucose transport, (3) increasing the activity of the pentose phosphate pathway, and (4) eliminating undesirable pathways. Moreover, the ethanol synthetic pathway from Zymomonas mobilis was introduced into the engineered strain. As a consequence, the resulting strain was able to simultaneously metabolize glucose and xylose and consume all sugars (30 g/L each) in 16 h, leading to 97% of the theoretical ethanol yield. Overall, this indicates that this approach is effective and straightforward to engineer E. coli for the desired trait.

Metabolic engineering of Escherichia coli for production of butyric acid.

The butyrate production by engineered Escherichia coli is afflicted by both low titer and low selectivity (defined as the butyrate/acetate (B/A) ratio). To address this issue, a strategy for metabolic engineering of E. coli was implemented including (1) elimination of all major NADH-dependent reactions in the fermentation metabolism, (2) reconstruction of a heterologous pathway leading to butyryl-CoA, (3) recruitment of endogenous atoDA for conversion of butyryl-CoA to butyrate with acetate as a CoA acceptor, and (4) removal of the acetate-synthesis pathway. Grown on glucose (20 g/L) plus acetate (8 g/L), the engineered strain consumed almost all glucose and acetate and produced 10 g/L butyrate as a predominant product within 48 h. It leads to high butyrate selectivity with the B/A ratio reaching 143. The result shows that our proposed approach may open a new avenue in biotechnology for production of butyrate in E. coli.

Improvement of the thermoregulated T7 expression system by using the heat-sensitive lacI.

The thermoregulated T7 expression system was previously reported to be an effective way to produce massive amounts of recombinant proteins (Chao, Y. P.; Law, W. S.; Chen, P. T.; Hung, W. B. High production of heterologous proteins in Escherichia coli using the thermo‐regulated T7 expression system. Appl. Microbiol. Biotechnol. 2002b, 58, 446–453). To ensure its practical applicability, the system was improved for stringency with the construction of the T7 lac‐promoter‐containing plasmid associated with the thermolabile lacIgene ( lacIts). Owing to the recessive feature of lacIts, the wild‐type lacI was removed from the genome of the cell. Moreover, the cell was engineered to carry the chromosomal copy of the T7 gene 1 subject to the regulation of λPL and λPR promoters. To characterize the system, the lacZ gene was fused to the T7 lac promoter, and subsequent experiments showed that various amounts of LacZ could be synthesized in the plasmid‐bearing cell in response to heat. Among the producers, the cell with the plasmid containing lacIts (substitution of Gly265 with Asp in lacI) was able to produce the maximal LacZ, the production accounting for an amplification of more than 200‐fold over the uninduced level. A further demonstration was carried out to illustrate the practical usefulness of the developed system by producing carbamoylase on a 4000 L scale. Cultured to reach high cell density, the carbamolyase‐producing cell was shown to retain plasmids with 95% stability and to be capable of producing soluble protein equal to 13% of the total cell proteins. Overall, it illustrates the remarkable features of the developed system with tightness, high expression level, thermal inducibility, and high stability.

Applicability of new expression vectors for both engineering uses and biological studies.

To be applicable for both engineering and biological uses, the plasmid with the features of tight regulation, high‐level expression, and subtle modulation (or homogeneous induction) is required. IPTG‐inducible promoters are of particular interest since they acquire the latter two merits but usually lack stringency. To this end, two plamids have been developed to contain the T7 A1 promoter along with either lacIq or lacI gene. As a production system, the cells harboring the plamids with the lacZ gene clone enabled production of the maximal protein accounting for 35% total cell content upon induction by a saturating IPTG level. This protein yield is amplified over 700‐fold relative to that at the uninduced state. As a system for biological study, the ppc negative strain bearing the plasmid with the ppc gene clone failed to grow on glucose without IPTG induction but immediately resumed its growth in the presence of IPTG. Moreover, the level of the ppc gene product in the cell was varied by various IPTG, and the result revealed that the wild‐type ppc level was sufficient to support the saturated growth of the cell on glucose. Overall, it illustrates the applicability of these plasmids to needs in the post‐genome era.

Systematic Engineering of Escherichia coli for d-Lactate Production from Crude Glycerol.

Crude glycerol resulting from biodiesel production is an abundant and renewable resource. However, the impurities in crude glycerol usually make microbial fermentation problematic. This issue was addressed by systematic engineering of Escherichia coli for the production of d-lactate from crude glycerol. First, mgsA and the synthetic pathways of undesired products were eliminated in E. coli, rendering the strain capable of homofermentative production of optically pure d-lactate. To direct carbon flux toward d-lactate, the resulting strain was endowed with an enhanced expression of glpD-glpK in the glycerol catabolism and of a heterologous gene encoding d-lactate dehydrogenase. Moreover, the strain was evolved to improve its utilization of cruder glycerol and subsequently equipped with the FocA channel to export intracellular d-lactate. Finally, the fed-batch fermentation with two-phase culturing was carried out with a bioreactor. As a result, the engineered strain enabled production of 105 g/L d-lactate (99.9% optical purity) from 121 g/L crude glycerol at 40 h. The result indicates the feasibility of our approach to engineering E. coli for the crude glycerol-based fermentation.

A Glucose-Insensitive T7 Expression System for Fully-Induced Expression of Proteins at a Subsaturating Level of L-Arabinose

The L-arabinose (Ara)-controlled T7 expression system was previously constructed by creation of an Escherichia coli BL21(BAD) strain. The production of recombinant proteins in this strain was stringently regulated and reached a high level upon induction with Ara. Nevertheless, this system is still associated with inherent problems of interference with glucose and of the all-or-nothing induction profile at a subsaturating level of Ara. In this study, these problems were circumvented by modifying the physiological traits of BL21(BAD) strain. This was followed by deletion of ptsG gene and the araFGH and araBAD operon. The former encodes the glucose transporter while the latter two gene operons produce proteins responsible for Ara uptake and catabolism. In addition, the expression of genomic araE (encodes the Ara transporter) was constitutively enhanced. The resulting strain was designated BAD-5. By expression of the faster degrader GFP(LAA) at a subsaturating level of Ara, 80% of BAD-5 strain was found visually bright in the presence or absence of glucose. A further analysis by flow cytometry showed a uniform distribution of GFP expression for BAD-5 strain. In marked contrast, BL21(BAD) strain exhibiting visual brightness was less than 10% of the cell population and remained dark in the presence of glucose. Moreover, a saturated level of luciferase from Renilla reniformis (Rluc) could be readily obtained in BAD-5 strain at 20 μM Ara regardless of glucose. Rluc in BL21(BAD) strain was produced in an Ara dose-dependent manner, and the protein production became arrested when glucose was present. Overall, it illustrates the usefulness of the improved system for overproduction of recombinant proteins in an efficient, homogeneous, and glucose-insensitive way.