DisCoTune: versatile auxiliary plasmids for the production of disulphide-containing proteins and peptides in the E. coli T7 system

Abstract

Secreted proteins and peptides hold large potential both as therapeutics and as enzyme catalysts in biotechnology. The high stability of many secreted proteins helps maintain functional integrity in changing chemical environments and is a contributing factor to their commercial potential. Disulphide bonds constitute an important post-translational modification that stabilizes many of these proteins and thus preserves the active state under chemically stressful conditions. Despite their importance, the discovery and applications within this group of proteins and peptides are limited by the availability of synthetic biology tools and heterologous production systems that allow for efficient formation of disulphide bonds. Here, we refine the design of two DisCoTune (Disulphide bond formation in E. coli with tunable expression) plasmids that enable the formation of disulphides in the highly popular Escherichia coli T7 protein production system. We show that this new system promotes significantly higher yield and activity of an industrial protease and a conotoxin, which belongs to a group of disulphide-rich venom peptides from cone snails with strong potential as research tools and pharmacological agents.

Fig. 1

Schematic illustration of the central elements in the CyDisCo and DisCoTune expression systems for heterologous expression of disulphide‐containing proteins. A. In the pLysS‐based CyDisCo plasmid, T7 lysozyme is controlled by the downstream T7 Pϕ3.8 promoter. The unknown effect this design has on the efficiency of the production system is illustrated by dashed arrows. B. On pDisCoTune, the CyDisCo system is introduced on a new backbone that allows for accurate titration of T7 lysozyme expression from PrhaB with activators RhaS and RhaR to control transcription of the target gene by T7RNApol. The chloramphenicol resistance gene (CmR) is indicated and genetic features like promoters, terminators and origins of replication are illustrated according to the SBOL standard (Madsen et al., 2019).

Fig. 2

Screening of proteinase K activity produced in E. coli facilitated by pDisCoTune. A. Proteinase K activity was screened based on casein cleavage either directly on skim milk agar plates by loss of opacity or based on activity in lysates measured by release of the fluorophore FITC. B. Colonies of BL21(DE3) harbouring either pLemo, pCyDisCo or pDisCoTune were stabbed on to skim milk agar plates with different concentrations of rhamnose (rha). C. Western blot against lysozyme showing expressing levels from pDisCoTune at the relevant rhamnose concentrations. An antibody (anti‐LepB) against E. coli leader peptidase was used as a loading control. D. Graph showing the protease activity in mU ml^‐1 of culture based on release of FITC. The activity was measured from lysates of E. coli carrying either pLemo (yellow), pCyDisCo (grey) or pDisCoTune (green). Proteinase K was expressed from pET28‐protK, and cultures were supplemented with relevant concentrations of rhamnose. Circles denote results obtained from three independent experiments.

Fig. 3

Functional assaying and quantitative proteomics analysis of proteinase K expressions. A. Schematic diagram of experimental procedure in the quantitative proteomics analysis. Expression of proteinase K was compared between pCyDisCo and pDisCoTune with or without 100 µM rhamnose (rha). Details on the experimental procedure of the quantitative proteomics are described in the experimental procedures. B. Graph showing the protease activity in mU ml⁻¹ of culture based on release of FITC. C–G. Quantification of peptide abundance for relevant proteins: Erv1p, hPDI, proteinase K(ProtK), T7 lysozyme and T7RNApol respectively. Individual samples from the combination of pCyDisCo and proteinase K are marked with C1‐4.

Fig. 4

Conk‐S3 constitutes a correctly folded and oxidized protein. A. Comparison of Ub–His₁₀–Conk‐S3 expression without or with the csCyDisCo or csDisCoTune system. 15% SDS‐PAGE gel stained with Coomassie Brilliant Blue showing the total cell extract (T), resuspended pellet after lysis and centrifugation (P), and the soluble fraction (S) from cells expressing Ub–His₁₀–Conk‐S3. The arrowhead denotes a semi‐oxidized form of Ub–His₁₀‐Conk‐S3 (see illustration above gel and Fig. S2B). Expression was performed as described under Experimental procedures. Protein levels are comparable between lanes, and the gel represents three independent experiments. B. Western blot against lysozyme in samples from Ub–His₁₀–Conk‐S3 expressions compared to expression level of lysozyme from pLysS with varying IPTG induction and in the absence and presence of rhamnose. The anti‐LepB signal was used as a loading control. C. 15% Tris‐Tricine SDS‐PAGE gel of Conk‐S3 in the oxidized (ox) and reduced (red; treated with 0.5 mM TECP and 5 mM DTT) state.

Fig. 5

Conk‐S3 constitutes a correctly folded peptide. A. CD spectra of Conk‐S3 produced in the presence of either pcsCyDisCo (grey) or pcsDisCoTune (green). B. Overlay of the experimental CD spectrum obtained for Conk‐S3 produced in the presence of pcsDisCoTune (green) with the deconvoluted spectrum (black). The secondary structure element contributions obtained from the deconvolution are listed along with those calculated from the crystal structure using the Define Secondary Structure of Proteins (DSSP) program (Kabsch and Sander, ; Touw et al., 2015). Deconvolution was performed using the BestSel web server (Micsonai et al., 2015).