Using Vibrio natriegens for High-Yield Production of Challenging Expression Targets and for Protein Perdeuteration

Abstract

Production of soluble proteins is essential for structure/function studies; however, this usually requires milligram amounts of protein, which can be difficult to obtain with traditional expression systems. Recently, the Gram-negative bacterium Vibrio natriegens emerged as a novel and alternative host platform for production of proteins in high yields. Here, we used a commercial strain derived from V. natriegens (Vmax X2) to produce soluble bacterial and fungal proteins in milligram scale, which we struggled to achieve in Escherichia coli. These proteins include the cholera toxin (CT) and N-acetyl glucosamine-binding protein A (GbpA) from Vibrio cholerae, the heat-labile enterotoxin (LT) from E. coli and the fungal nematotoxin CCTX2 from Coprinopsis cinerea. CT, GbpA, and LT are secreted by the Type II secretion system in their natural hosts. When these three proteins were produced in Vmax, they were also secreted and could be recovered from the growth media. This simplified the downstream purification procedure and resulted in considerably higher protein yields compared to production in E. coli (6- to 26-fold increase). We also tested Vmax for protein perdeuteration using deuterated minimal media with deuterium oxide as solvent and achieved a 3-fold increase in yield compared to the equivalent protocol in E. coli. This is good news, since isotopic labeling is expensive and often ineffective but represents a necessary prerequisite for some structural biology techniques. Thus, Vmax represents a promising host for production of challenging expression targets and for protein perdeuteration in amounts suitable for structural biology studies.

Figure 1

CT production in E. coli and Vmax. (A) SDS-PAGE analysis of CT samples obtained by expression in E. coli and purification by IMAC and ion-exchange chromatography. Periplasmic: periplasmic fraction, FT: flow-through. (B) Cation-exchange chromatogram of CT produced in E. coli. At pH 8.0, the holotoxin is negatively charged and elutes in the flow-through and wash fractions, while (most) CTB binds to the column and elutes with a linear gradient of buffer B. (C) SEC chromatogram of CT produced in E. coli. SEC was performed on the holotoxin-containing flow-through and wash from the IEX step. SEC fractions F1–F3 (first half of peak in C) on native PAGE (inset) still contained considerable amounts of CTB in the purified holotoxin sample. (D) SDS-PAGE gel from CT expression in Vmax, where CT is secreted into the growth medium. (E) SEC chromatogram of CT produced in Vmax, and (F) SDS-PAGE gel of SEC peak (compared to the molecular mass marker).

Figure 2

Comparison of CT crystal structures. CT produced in Vmax (light green; PDB ID: 8QRE, this work), superimposed onto the published crystal structure of CT produced in E. coli (magenta; PDB ID: 1S5E). RMSD = 0.3 Å.

Figure 3

Toxicity and stoichiometry for commercial and Vmax preparations of CT. (A) CHO cells were incubated for 2 h with 10-fold dilutions of CT purchased from a commercial vendor (cv, filled squares) or purified from Vmax (Vmax, open squares). An ELISA was then used to quantify cAMP levels from the intoxicated cells. Background-subtracted data were expressed as percentages of the response elicited from cells challenged with 100 ng/mL of the commercial toxin and represent the means ± standard deviations of nine technical replicates from three independent experiments. (B) Samples of CT purchased from a commercial vendor (cv) or produced in Vmax (Vmax) were resolved by SDS-PAGE under reducing and nonreducing conditions. The samples (4 μg per lane) were visualized by Coomassie stain. The entire gel is shown, with the molecular mass marker from select protein standards shown on the left.

Figure 4

LT production in Vmax. (A) SDS-PAGE analysis of LT samples obtained for expression in Vmax and purification from the culture supernatant by galactose-affinity chromatography (GAC). (B) SEC chromatogram of LT captured by GAC, and SDS-PAGE of SEC fractions (inset).

Figure 5

GbpA production in E. coli and Vmax. (A) SEC chromatogram (Superdex 200 Increase 30/100 GL column) of hydrogenated (¹H) GbpA expressed in E. coli BL21 Star(DE3) (light orange) and in Vmax (green). Several SEC runs were performed for each purification batch; therefore, the intensity of absorbance is not correlated with yield. (B,C) SDS-PAGE analysis of GbpA hydrogenated (¹H) samples expressed in Vmax or E. coli BL21 Star(DE3) after the AEX and SEC purification steps. (D) SEC chromatogram (Superdex 75 Increase 30/100 GL column) of perdeuterated (²H) GbpA expressed in E. coli BL21 Star(DE3) (light orange) and in Vmax (green). Again, SEC was performed in several injections per batch, and the amounts shown do not correlate with yield. (E,F) SDS-PAGE analysis of GbpA perdeuterated (²H) samples expressed in Vmax or E. coli BL21 Star(DE3) after the AEX and SEC purification steps. Deuteration levels of the produced protein were determined to be 96 and 97% for the protein produced in E. coli and Vmax, respectively. ¹H refers to “common” hydrogen (also called protium); ²H refers to deuterium.

Figure 6

CCTX2 production in E. coli and Vmax. (A) Galactose-affinity chromatography of CCTX2 produced in 1 L cultures from E. coli (light orange) and V. natriegens (green). (B) Matching SEC chromatogram comparing production in E. coli (light orange) and V. natriegens (green) for 0.5 L cultures. (C,D) SDS-PAGE of CCTX2 samples from expression in E. coli C41(DE3) and Vmax, matching purification fractions labeled in panel B.