Tuning Escherichia coli for membrane protein overexpression

Abstract

A simple generic method for optimizing membrane protein overexpression in Escherichia coli is still lacking. We have studied the physiological response of the widely used "Walker strains" C41(DE3) and C43(DE3), which are derived from BL21(DE3), to membrane protein overexpression. For unknown reasons, overexpression of many membrane proteins in these strains is hardly toxic, often resulting in high overexpression yields. By using a combination of physiological, proteomic, and genetic techniques we have shown that mutations in the lacUV5 promoter governing expression of T7 RNA polymerase are key to the improved membrane protein overexpression characteristics of the Walker strains. Based on this observation, we have engineered a derivative strain of E. coli BL21(DE3), termed Lemo21(DE3), in which the activity of the T7 RNA polymerase can be precisely controlled by its natural inhibitor T7 lysozyme (T7Lys). Lemo21(DE3) is tunable for membrane protein overexpression and conveniently allows optimizing overexpression of any given membrane protein by using only a single strain rather than a multitude of different strains. The generality and simplicity of our approach make it ideal for high-throughput applications.

Fig. 1.

Analysis of growth, protein expression, morphology, and respiration of BL21(DE3)pLysS, C41(DE3), and C43(DE3) overexpressing YidC-GFP. (A and B) Growth (A) and protein expression of cells overexpressing YidC-GFP (B) were monitored by measuring the A₆₀₀ and GFP fluorescence, respectively, every 30 min. (C–F) The following parameters were monitored by flow cytometry: forward and side scatter (C–E), which provide information about cell size and granularity, and GFP fusion protein expression (F). For C–F, cells were harvested 4 h after induction with IPTG. (G) Oxygen consumption was measured in whole cells every hour. Experiments were done in triplicate. Respiratory activities of control cells were set to 100.

Fig. 2.

Analysis of subproteomes of BL21(DE3)pLysS, C41(DE3), and C43(DE3) overexpressing YidC-GFP. (A) Proteins of whole cell lysates of cells overexpressing YidC-GFP fusions for 4 h were separated by means of SDS/PAGE and subsequently subjected to Western blotting with antibodies to ClpB and IbpA/B. (B) Proteins of whole cell lysates of cells overexpressing YidC-GFP fusions for 4 h were separated by means of 2D IEF/SDS/PAGE. Shown is the relative quantification of protein spots representing ClpB, IbpA, HslV, and HslU. (C) Protein aggregates were isolated from cells overexpressing YidC-GFP fusions as described in the Materials and Methods section. The aggregates were analyzed by 1D SDS/PAGE, and gels were stained with colloidal Coomassie.

Fig. 3.

Mutations in the lacUV5 promoter controlling T7RNAP transcription lead to reduced T7RNAP expression levels in C41(DE3) and C43(DE3). (A) YidC-GFP overexpression was monitored on-line by measuring GFP fluorescence every 30 sec in cells cultured in a 96-well plate in a spectrofluorometer. (B) Quantification of mRNA levels of YidC-GFP by RT PCR. Cells overexpressing YidC-GFP were harvested at indicated timepoints after induction of expression with IPTG. Subsequently, mRNA was isolated and cDNA transcribed as described in the Materials and Methods section. Experiments were done in triplicate. (C) Accumulation levels of LacY in the membrane and T7RNAP in whole cells over time as monitored by Western blotting. (D) Quantification of mRNA levels of T7 RNA polymerase was done by RT PCR. Experiments were done in triplicate. (E) Sequences of the promoters controlling transcription of T7 RNA polymerase in BL21(DE3), C41(DE3), and C43(DE3) were determined as described in Materials and Methods. Mutations are shaded gray.

Fig. 4.

Tuning T7 RNA polymerase activity with T7Lys. (A) Model of tuning T7 RNA polymerase activity with T7Lys. Expression of T7Lys is under the control of the rhaBAD promoter and is induced by the addition of L-rhamnose. The amidase activity deficient T7Lys mutant LysY was used (Fig. S5). (B) Expression of YidC-GFP in Lemo21(DE3), Lemo41(DE3) [C41(DE3) with pLemo], and Lemo43(DE3) [C43(DE3) with pLemo] at different L-rhamnose concentrations. C41(DE3) and C43(DE3) were included as controls (white bars). Experiments were done in triplicate. (C) Cells overexpressing YidC-GFP were harvested at indicated timepoints after induction with IPTG and mRNA levels of YidC-GFP were determined by RT PCR. Experiments were done in triplicate.

Fig. 5.

Characterization of Lemo21(DE3). In all experiments, Lemo21(DE3) overexpressed YidC-GFP in the presence of the indicated amounts of L-rhamnose. (A–C) Growth (A) and protein expression of Lemo21(DE3) (B) were monitored by measuring the A₆₀₀ and GFP fluorescence, respectively, every 1–2 h. Fraction of cells not expressing YidC-GFP 4 and 18 h after induction as monitored by flow cytometry (C). Experiments were done in triplicate. (D) Oxygen consumption was measured in whole cells four hours after induction. Experiments were done in triplicate. Respiratory activities of control cells were set to 100.

Fig. 6.

Converting BL21(DE3) into C43(DE3) and vice versa, and overexpression screen. (A and B) Promoters controlling T7RNAP expression in BL21(DE3) and C43(DE3) were swapped, and YidC-GFP expression in the strain with swapped promoters in the absence and presence of pLemo was monitored by GFP-fluorescence (A), and T7RNAP, ClpB, and IbpA/B accumulation levels were monitored in the strains with swapped promoters and their controls by Western blotting (B). (C) Overexpression of membrane protein GFP-fusions, and GFP in different strains was monitored by measuring GFP fluorescence 8 h after induction. For graphical reasons, fluorescence values of TSpA and B, and NTR were multiplied by 10 and fluorescence values of GFP were divided by 50.