Re-directing bacterial microcompartment systems to enhance recombinant expression of lysis protein E from bacteriophage ϕX174 in Escherichia coli

Abstract

Background: Recombinant expression of toxic proteins remains a challenging problem. One potential method to shield toxicity and thus improve expression of these proteins is to encapsulate them within protein compartments to sequester them away from their targets. Many bacteria naturally produce so-called bacterial microcompartments (BMCs) in which enzymes comprising a biosynthetic pathway are encapsulated in a proteinaeous shell, which is in part thought to shield the cells from the toxicity of reaction intermediates. As a proof-of-concept, we attempted to encapsulate toxic, lysis protein E (E) from bacteriophage ϕX174 inside recombinant BMCs to enhance its expression and achieve higher yields during downstream purification.

Results: E was fused with various N-terminal BMC targeting tags (PduP-, PduD-, and EutC-tags, 18-20 amino acids) and co-expressed with appropriate BMC shell proteins that associate with the tags and are required to form BMCs. Only BMC targeted E fusions, but not non-tagged E, could be successfully cloned, suggesting that the BMC tags reduce the toxicity of E. A PduP-tagged E system appeared to achieve the highest expression of E. Co-expression of Pdu BMC shell proteins with PduP-E increased its expression by 20-50%. Affinity purification of PduP-E via Ni-NTA in the presence of Empigen BB detergent yielded 270 µg of PduP-E per L of induced culture. Removal of the PduP-tag via proteolysis resulted in a final yield of 200 µg of E per L of induced culture, a nearly order of magnitude (~sevenfold) improvement compared to prior reports.

Conclusions: These results demonstrate improved expression of ϕX174 lysis protein E via re-directed BMC systems and ultimately higher E purification yields. Similar strategies can be used to enhance expression of other toxic proteins in recombinant Escherichia coli systems.

Keywords: BMC; Bacterial microcompartment; Bacteriophage phiX174; Lysis protein E; Toxic protein expression.

Fig. 1

Growth and E expression in different BMC systems. a Genetic scheme for the different BMC systems. Black circles represent ribosome binding/re-initiation sites. b Growth curves for each strain under different induction conditions. Induction with 0.5 mM IPTG only at t = −3 h (open circle); induction with 0.1 mM rhamnose only at t = 0 (open square); and co-induction with both IPTG and rhamnose (filled triangle). Error bars represent data from 3 replicates. c SDS-PAGE and anti-His₆ Western blot at 4 h post-rhamnose induction from 1 representative growth. The equivalent of 50 µL and 2 µL of original culture were analyzed by SDS-PAGE and Western blot, respectively. Left red arrows EutL (top), EutS/M/N (bottom). Right red arrows PduB (top), PduB′ (middle), and PduA/J (bottom). Sizes of molecular weight standards in kDa are shown on the left. d Relative Western blot intensities compared to Ec2985, Rha only. Gray bars rhamnose only induction; striped bars IPTG and rhamnose co-induction. Error bars represent data from 3 replicates from the different growths in b (see Additional file 1: Figure S2 for raw data)

Fig. 2

Growth and PduP-E expression in Ec3087 and Ec0087 under different induction conditions. Ec0087 induced with rhamnose only (open circle); Ec3087 induced with rhamnose only (open square); and Ec3087 co-induced with rhamnose and IPTG (filled triangle). IPTG was added at t = −3 h to 0.5 mM. Rhamnose was added at t = 0 to 0.1 mM in a, d, g, to 0.2 mM in b, e, h, and to 0.5 mM in c, f, i. Error bars represent data from 3 biological replicates. a–c Growth of cells. d–f Amounts of PduP-E expressed per L of culture determined by quantitative Western blot (see Additional file 1: Figure S5 for raw data). g–i Amounts of PduP-E expressed per L of culture normalized to OD₆₀₀

Fig. 3

Co-expression of PduP-E with mCherry (Ec10187) versus with the Pdu BMC system (Ec3087). a Genetic scheme for the different strains. Black circles represent ribosome binding/re-initiation sites. b Growth curves for each strain under different induction conditions. Open shapes induction with 0.1 mM rhamnose only at t = 0; closed shapes co-induction with 0.5 mM IPTG at t = −3 h and 0.1 mM rhamnose at t = 0. Error bars represent data from 3 replicates. c SDS-PAGE and anti-His₆ Western blot at 4 h post-rhamnose induction from 1 representative growth. The equivalent of 50 and 2 µL of original culture were analyzed by SDS-PAGE and Western blot, respectively. Sizes of molecular weight standards in kDa are shown on the left. d Relative Western blot intensities compared to Ec3087, Rha only. Gray bars rhamnose only induction; striped bars IPTG and rhamnose co-induction. Error bars represent data from 3 replicates from the different growths in b (see Additional file 1: Figure S6 for raw data)

Fig. 4

Co-expression of PduP-E with a truncated (Ec9987) versus full (Ec3087) Pdu BMC system. a Genetic scheme for the different strains. Black circles represent ribosome binding/re-initiation sites. b Growth curves for each cell strain under different induction conditions. Open shapes induction with 0.1 mM rhamnose only at t = 0; closed shapes co-induction with 0.5 mM IPTG at t = −3 h and 0.1 mM rhamnose at t = 0. Error bars represent data from 3 replicates. c SDS-PAGE and anti-His₆ Western blot at 4 h post-rhamnose induction from 1 representative growth. The equivalent of 50 and 2 µL of original culture were analyzed by SDS-PAGE and Western blot, respectively. Sizes of molecular weight standards in kDa are shown on the left. d Relative Western blot intensities compared to Ec3087, Rha only. Gray bars rhamnose only induction; striped bars IPTG and rhamnose co-induction. Error bars represent data from 3 replicates from the different growths in b (see Additional file 1: Figure S8 for raw data)

Fig. 5

Co-isolation of PduP-tagged protein with purified BMCs. a SDS-PAGE and anti-His₆ Western blot of purified BMCs from Ec3087. As negative and positive controls, BMCs were also purified from Ec0087 (no BMCs) and Ec3090 (PduP-mCherry + full Pdu), respectively. BMCs were purified from cells at 3 h post-rhamnose induction. Induction conditions are identified at the top of the gel. 10 µg of BMCs from Ec3087 and Ec3090 and 1 µg of mock BMCs from Ec0087 were loaded on the SDS-PAGE. 1 µg of each BMC sample was analyzed by Western blot. Sizes of molecular weight standards in kDa are shown on the left. b TEM analysis of isolated BMCs from Ec3087 and Ec3090. Scale bars are 50 nm

Fig. 6

Purified PduP-E and non-tagged E from whole cells and purified BMCs. a SDS-PAGE analysis of PduP-E purified from whole cells and non-tagged E after Factor Xa proteolysis. Lane 1 molecular weight markers with sizes of standards in kDa on the left. Lane 2 PduP-E purified from Ec3087 induced with 0.1 mM rhamnose only. Lane 3 PduP-E purified from Ec3087 co-induced with 0.1 mM rhamnose and 0.5 mM IPTG. Lane 4 non-tagged E after Factor Xa proteolysis. Lanes 2–4 were loaded with 1 µg of protein. b SDS-PAGE and anti-His₆ Western blot analysis during purification of PduP-E from purified BMCs. Lane 1 molecular weight markers. Lane 2 purified BMCs from Ec3087 co-induced with rhamnose and IPTG (starting material). Lane 3 flow through after binding to the Ni–NTA column. Lanes 4–6 sequential wash fractions containing 20 mM imidazole. Lanes 7–15 sequential elution fractions containing 200 mM imidazole