Modeling Native EHEC Outer Membrane Vesicles by Creating Synthetic Surrogates

Abstract

Enterohemorrhagic Escherichia coli (EHEC) is a zoonotic pathogen responsible for life-threating diseases such as hemolytic uremic syndrome. While its major virulence factor, the Shiga toxin (Stx), is known to exert its cytotoxic effect on various endothelial and epithelial cells when in its free, soluble form, Stx was also recently found to be associated with EHEC outer membrane vesicles (OMVs). However, depending on the strain background, other toxins can also be associated with native OMVs (nOMVs), and nOMVs are also made up of immunomodulatory agents such as lipopolysaccharides and flagellin. Thus, it is difficult to determine to which extent a single virulence factor in nOMVs, such as Stx, contributes to the molecular pathogenesis of EHEC. To reduce this complexity, we successfully developed a protocol for the preparation of synthetic OMVs (sOMVs) with a defined lipid composition resembling the E. coli outer membrane and loaded with specific proteins, i.e., bovine serum albumin (BSA) as a proxy for functional Stx2a. Using BSA for parameter evaluation, we found that (1) functional sOMVs can be prepared at room temperature instead of potentially detrimental higher temperatures (e.g., 45 °C), (2) a 1:10 ratio of protein to lipid, i.e., 100 µg protein with 1 mg of lipid mixture, yields homogenously sized sOMVs, and (3) long-term storage for up to one year at 4 °C is possible without losing structural integrity. Accordingly, we reproducibly generated Stx2a-loaded sOMVs with an average diameter of 132.4 ± 9.6 nm that preserve Stx2a's injuring activity, as determined by cytotoxicity assays with Vero cells. Overall, we successfully created sOMVs and loaded them with an EHEC toxin, which opens the door for future studies on the degree of virulence associated with individual toxins from EHEC and other bacterial pathogens.

Keywords: EHEC; OMV; Stx; liposome; toxin; vesicle.

Figure 1

Workflow for synthetic outer membrane vesicle (sOMV) preparation. Depicted is a combined process by which vesicles are homogenized through an extrusion step without proteins, and then the protein is added and the vesicles are dehydrated and rehydrated. The protein-loaded products are again homogenized by extrusion, and excess protein is removed by ultracentrifugation. Adapted from Walde and Ichikawa [30] and modified.

Figure 2

Long-term storage of sOMVs and properties of Stx2a-loaded sOMVs. Depicted in (A,B) is the size distribution as measured by dynamic light scattering (DLS). (A) Bovine serum albumin (BSA)-loaded sOMVs were stored at 4 °C and analyzed at the indicated time-points for structural integrity. A sample representative of three biological replicates is shown over time. (B) Size distribution of three biological replicates of Stx2a-loaded sOMVs (also stored at 4 °C) indicated by differently colored lines. (C) Cytotoxicity of free Stx2a, Stx2a-loaded sOMVs, and Stx2a-containing HUSEC029 native outer membrane vesicles (nOMVs) toward Vero cells, as determined by the crystal violet assay. The indicated concentrations refer to Stx2a for measurements with free Stx2a and Stx2a-loaded sOMVs or total protein for nOMVs of HUSEC029, respectively. Depicted is the mean ± standard deviation in relation to an untreated control of three biological replicates (n = 3) each performed in triplicates. Statistical analyses were performed by ANOVA with Bonferroni correction comparing free Stx2a with Stx2a-sOMVs (*) or with HUSEC029-nOMVs (‡), respectively, and significances are indicated as follows: */‡, p < 0.05; **/‡‡, p < 0.01; ***/‡‡‡, p < 0.001 (comparing Stx2a-sOMVs with HUSEC029-nOMVs yielded no significant differences).