Exploring the performance of Escherichia coli outer membrane vesicles as a tool for vaccine development against Chagas disease

Abstract

Background: Vaccine development is a laborious craftwork in which at least two main components must be defined: a highly immunogenic antigen and a suitable delivery method. Hence, the interplay of these elements could elicit the required immune response to cope with the targeted pathogen with a long-lasting protective capacity.

Objectives: Here we evaluate the properties of Escherichia coli spherical proteoliposomes - known as outer membrane vesicles (OMVs) - as particles with natural adjuvant capacities and as antigen-carrier structures to assemble an innovative prophylactic vaccine for Chagas disease.

Methods: To achieve this, genetic manipulation was carried out on E. coli using an engineered plasmid containing the Tc24 Trypanosoma cruzi antigen. The goal was to induce the release of OMVs displaying the parasite protein on their surface.

Findings: As a proof of principle, we observed that native OMVs - as well as those carrying the T. cruzi antigen - were able to trigger a slight, but functional humoral response at low immunization doses. Of note, compared to the non-immunized group, native OMVs-vaccinated animals survived the lethal challenge and showed minor parasitemia values, suggesting a possible involvement of innate trained immunity mechanism.

Main conclusion: These results open the range for further research on the design of new carrier strategies focused on innate immunity activation as an additional immunization target and venture to seek for alternative forms in which OMVs could be used for optimizing vaccine development.

Fig. 1:. genetically engineered outer membrane vesicles (OMVs) are uniform vesicles able to display Trypanosoma cruzi Tc24 antigen efficiently. (A) Scheme of a nascent OMV from the outermost membrane of Gram-negative bacteria such as Escherichia coli. A recombinant gene circuit encoding a stimuli-responsive promoter (here via IPTG), lipoylation sequence (lpp’), membrane spanning domain (OmpA) and Tc24 was assembled and introduced to a non-pathogenic strain of E. coli. Activation of the gene circuit leads to protein synthesis, periplasmic targeting, and subsequent outer membrane insertion and OMV loading. (B) Particle intensity (in arbitrary units, a.u.) and size data was captured in triplicate over 90 s intervals and then converted to particle concentration and size. (C) Histogram demonstrating relative size and abundance of nOMVs and OMVs-Tc24 samples. (D) A representative Western blot image confirming the presence of Tc24 within OMVs-Tc24 content (4 µg per lane of total protein were loaded for both vesicle samples). Purified rTc24 protein (100 ng per lane) was used as a positive control. As expected, no presence of antigen Tc24 was detected in the nOMVs sample. Coomassie stained gel (cutout) is presented in the lower panel; a non-related vesicle protein band (~35 kDa) is shown as a loading reference.

Fig. 2:. native outer membrane vesicles (nOMVs) and OMVs-Tc24 vaccinated animals elicited anti-OMVs specific humoral response. (A) Graphical representation of the immunization and challenge scheme. Animals were primed and boosted with nOMVs, OMVs-Tc24 or phosphate buffered saline (PBS) (non-vaccinated) three weeks apart. Twenty days after the second immunization, vaccine-induced IgG1 (solid bars) and IgG2c (striped bars) specificity was determined by enzyme-linked immunosorbent assay (ELISA) after incubating the serum samples with (B) nOMVs, (C) OMVs-Tc24 soluble lysates or (D) rTc24, respectively. Serum samples from non-vaccinated animals were used as negative controls. Data (mean ± SEM) are representative of three independent experiments (n = 4 mice per experimental group, duplicate observations per sample); significance is presented as # (non-vaccinated vs. vaccinated groups) or * (comparison between IgG1 and IgG2c within experimental groups). The p values of p ≤ 0.05, p ≤ 0.01, p ≤ 0.001 are annotated with one, two, and three symbols, respectively and were determined by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test (comparison of multiple groups).

Fig. 3:. native outer membrane vesicles (nOMVs) and OMVs-Tc24 immunizations generate functional antibodies. (A) Activation of complement-mediated lysis of trypomastigote forms. Lysis percentage (%) was estimated by comparing the remaining number of motile trypomastigotes in samples pretreated with sera obtained from nOMVs and OMVs-Tc24 vaccinated vs. non-vaccinated animals (considered as 0%). (B) Percentage of in vitro infected cells after incubation with infective Trypanosoma cruzi forms pretreated with experimental sera [mice administered with nOMVs, OMVs or phosphate buffered saline (PBS)]. Data (mean ± SEM) are representative of two independent experiments with triplicate samples; each sera pool was composed of n = 4 animals per experimental group. Significance is presented as # (vaccinated groups vs. non-vaccinated). The values of p ≤ 0.05, p ≤ 0.01, p ≤ 0.001 are annotated with one, two, and three symbols, respectively and determined by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test (comparison of multiple groups). ns = no significant differences among vaccinated groups (p > 0.05). (C) Representative images of Giemsa-stained infected cells after trypomastigote-pretreatment with serum from PBS-inoculated mice (left) and from OMVs-Tc24-administered animals (right) are shown; micrographs were obtained under a ×400 magnification factor.

Fig. 4:. T cells profile after prime and boost with native outer membrane vesicles (nOMVs) and OMVs-Tc24. Nineteen days after boost, vaccinated mice were euthanized and freshly collected splenocytes were used for CD4⁺ and CD8⁺ T cell determination. Splenocytes were labeled with fluorescent-conjugated antibodies and analyzed by flow cytometry. (A, C) Splenic percentages of CD4⁺ and CD8⁺ T cells respectively. (B, D) CD4⁺ as well as CD8⁺ T cells were analyzed to distinguish naïve (CD44^lo CD62L^hi), effector memory (T_EM, CD44^hi CD62L^lo), and central memory (T_CM, CD44^hi CD62L^hi) phenotypes. Data (mean ± SEM) are representative of duplicate observations per sample (n = 4 mice per group). Significance was determined by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test (comparison of multiple groups). ns = no significant differences among groups, p > 0.05.

Fig. 5:. native outer membrane vesicles (nOMVs) and OMVs-Tc24 confer partial protection against Trypanosoma cruzi virulent infection. Animals inoculated with nOMVs, OMVs-Tc24, or non-vaccinated (n = 4) were challenged with 500 trypomastigotes from the Tulahuen strain, through intraperitoneal injection, 21 days after the last immunization dose. (A) Parasitemia curve obtained during the acute phase of the infection from 10 μL of blood taken twice a week. (B) Concentration-time curve (AUC) of the parasitemia curve (in relative units, RU). (C) Parasite equivalents in 100 ng of total DNA estimated by real-time qPCR amplification of SAT sequence at day 19 post-challenge. (D) Survival rates were monitored daily during the acute phase until animals from the control group started to die. Data are expressed as the mean ± SEM of three independent experiments. Significance was determined by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test (comparison of multiple groups). Numeral represent statistical significance with respect to the non-vaccinated group. ns = no significant differences among groups, p > 0.05.