Homology-dependent recombination of large synthetic pathways into E. coli genome via λ-Red and CRISPR/Cas9 dependent selection methodology

Abstract

Background: Metabolic engineering frequently needs genomic integration of many heterologous genes for biosynthetic pathway assembly. Despite great progresses in genome editing for the model microorganism Escherichia coli, the integration of large pathway into genome for stabilized chemical production is still challenging compared with small DNA integration.

Results: We have developed a λ-Red assisted homology-dependent recombination for large synthetic pathway integration in E. coli. With this approach, we can integrate as large as 12 kb DNA module into the chromosome of E. coli W3110 in a single step. The efficiency of this method can reach 100%, thus markedly improve the integration efficiency and overcome the limitation of the integration size adopted the common method. Furthermore, the limiting step in the methylerythritol 4-phosphate (MEP) pathway and lycopene synthetic pathway were integrated into the W3110 genome using our system. Subsequently, the yields of the final strain were increased 106 and 4.4-fold compared to the initial strain and the reference strain, respectively.

Conclusions: In addition to pre-existing method, our system presents an optional strategy for avoiding using plasmids and a valuable tool for large synthetic pathway assembly in E. coli.

Keywords: CRISPR-Cas9; Chromosomal integration; Escherichia coli; Lambda Red; Metabolic engineering.

Fig. 1

Outline of the λ-Red assisted homology-dependent recombination for large synthetic pathways integration in Escherichia coli. a Construction of plasmid pRC-IS5 with large synthetic pathways. pRC-IS5 (including R6K and a homologous region) replicates normally in E. coli with the expression of pir + protein and the plasmid replication was restricted in normal E. coli. b Single-crossover HDR assisted by λRed. The vector pRC-IS5 was introduced into the host which harbored pCas with the expression of Exo, Beta, and Gam, and then selection was conducted with the addition of chloramphenicol. Single crossover produced homology-dependent insertion events, where the entire vector pRC-IS5 was integrated into the chromosome at the target locus. A simple screening step by PCR diagnosis could identify the desired mutant. c Deleting redundant sequences assisted by λ-Red. The gRNA plasmid pTargetF-delete and the donor template were electroporated into the competent cells harbored plasmid pCas with the expression of Cas9 nuclease and λ-Red protein, and then the selection was carried out using kanamycin and spectinomycin. λ-Red mediated deletion at the lagging strand of the replication fork produced homologous recombination, where the redundant sequences were deleted

Fig. 2

a MEP pathway and related metabolism showing the major metabolic regulatory points. d-glyceraldehyde 3-phosphate (G-3P);1-deoxy-d-xylulose 5-phosphate (DXP); methylerythritol 4-phosphate (MEP); diphosphocytidyl methylerythritol (CDP-ME); diphosphocytidyl methylerythritol 2-phosphate (CDP-MEP); hydroxymethylbutenyl diphosphate (HMBPP); isopentenyl diphosphate (IPP); dimethylallyl diphosphate (DMAPP); farnesyl diphosphate (FPP). b The limiting step for lycopene production was divided into three modules. The feedforward module including dxs and dxr, feedback module including idi and crtE and lycopene synthetic module including crtI, crtE and crtB for lycopene production

Fig. 3

Promoters characterization of three modules for lycopene biosynthesis. a Selection promoters for lycopene synthetic modules (including crtI, crtE and crtB). Lycopene synthetic modules were overexpressed in E. coli W3110 with the native MEP pathway. trc, EC101; yciG, EC102; pstA, EC103; yodA, EC104; astC, EC105; ybiM, EC106. b Selection promoters for feedforward modules (including dxs and dxr). Feedforward modules were overexpressed along with lycopene synthetic modules. phnI, EC201; phoR, EC202; phnF, EC203; phnC, EC204; phnD, EC205. c Selection promoters for feedback modules (including idi and crtE). Feedback modules were overexpressed along with lycopene synthetic modules. yfiL, EC301; yijF, EC302; cysP, EC303; yeiG, EC304; yhcN, EC305. Each value represents the average ± SD of three biological replicates

Fig. 4

Integration of a 12 kb DNA module into E. coli W3110 genome. a The optimized lycopene synthetic pathways in pRC-IS5. b Colony forming unit (CFU, indicated the number of colonies on the selective plates with 34 μg mL⁻¹ chloramphenicol after one experiment of integrating optimized lycopene synthetic pathways into E. coli W3110) and integration efficiency with or without adding arabinose to induce λ-Red. c PCR confirmation of the integration of the optimized lycopene synthetic pathways using primers IS5-Q-P1 and dxs-dxr-P2 for feedforward module (6521 bp), IEB-P1 and IS5-Q-P2 for lycopene synthetic module (4811 bp), idi-crtE-P1 and idi-crtE-P2 for feedback module (3042 bp). M: DNA marker; CK: E. coli W3110; 1, 2, 3, 4, 5, 6: colonies from the plates after chromosomal integration

Fig. 5

Deletion of redundant sequences with CRISPR-Cas9 system. a PCR confirmation of the deletion of redundant sequences using primers IS5-check-P1 and IS5-check-P2. The decrescent bands indicated the successful deletion of redundant sequences. M: DNA marker; CK: strain without deletion of redundant sequences; 1,2,3,4,5,6,7,8,9: colonies from the plate after editing using CRISPR-Cas9 system. b The lycopene yields of shake flask fermentation of strain EC101, EC-IS5, EC401 and EC-IS5 (ΔCm). Each value represents the average ± SD of three biological replicates