From In silico Protein Epitope Density Prediction to Testing Escherichia coli O157:H7 Vaccine Candidates in a Murine Model of Colonization

Abstract

Enterohemorrhagic Escherichia coli (EHEC) O157:H7 is a leading cause of foodborne illnesses worldwide and is a common serotype linked to hemorrhagic colitis and an important cause of hemolytic uremic syndrome (HUS). Treatment of EHEC O157:H7 infections is complicated, as antibiotics can exacerbate Shiga toxin (Stx) production and lead to more severe symptoms including HUS. To date, no vaccines have been approved for human use, exposing a void in both treatment and prevention of EHEC O157:H7 infections. Previously, our lab has shown success in identifying novel vaccine candidates via bio- and immunoinformatics approaches, which are capable of reducing bacterial colonization in an in vivo model of intestinal colonization. In this study, we further characterized 17 of the identified vaccine candidates at the bioinformatics level and evaluated the protective capacity of the top three candidates when administered as DNA vaccines in our murine model of EHEC O157:H7 colonization. Based on further immunoinformatic predictions, these vaccine candidates were expected to induce neutralizing antibodies in a Th2-skewed immunological response. Immunization of BALB/c mice with two of these candidates resulted in reduced bacterial colonization following EHEC O157:H7 challenge. Additionally, immune sera was shown to prevent bacterial adhesion in vitro to Caco-2 cells. Together, this study provides further validation of our immunoinformatic analyses and identifies promising vaccine candidates against EHEC O157:H7.

Keywords: Escherichia coli O157:H7; bioinformatics; immunoinformatics; type III secretion system; vaccine.

Figure 1

Classification of identified candidates. (A) Schematic representing prediction strategy for candidate selection of HP candidates. (B) Table showing individual candidates along with their compiled MED_Th2 score and allele coverage across 17 high-priority candidates. The input to the equation was MHC-II prediction results (NetMHCII) only. Hence, it is expected to provide us with Th2-oriented inference.

Figure 2

Dendrogram of vaccine candidates. Clustering of candidates based on Mature Epitope Density (MED) Score. Selected candidates (green), their compiled MED_Th2 score (red) and the number of HLA alleles covered based (black) are highlighted.

Figure 3

Bacterial counts in infected mice with EHEC O157:H7. Bacterial colonization of large intestine (A) and cecum (B) segments as collected from mice vaccinated with pVAX1, lomW (pVAX-10), escJ (pVAX-41), escC (pVAX-56), and pComb followed by challenge with 5 × 10⁹ CFU of EHEC O157:H7. Bacterial counts are represented as CFU per gram of tissue. Means ± the SEM of the CFU/g from 10 mice presented and an asterisk (^*) indicates statistical significance as defined (p < 0.05).

Figure 4

Immune response from mice immunized with pVAX candidates. Graphs show secreted immunoglobulin A (A) and IgG (B) total levels, 2 weeks after last immunization. Mean IgA levels were measured from fecal samples of three immunized mice with lomW (pVAX-10), escJ (pVAX-41), escC (pVAX-56), or pComb. Feces collected prior to immunization (baseline) and of mice immunized with pVAX1 were used as controls. The results are expressed as means ± the SEM of triplicate values obtained from three mice from each group. Statistical significance was defined as (p < 0.05). (B) Sera collected from mice immunized with vaccine candidates was used to measure total IgG antibodies by ELISA. The results are expressed as means ± of the SEM of triplicate values from three mice in each group.

Figure 5

Bacterial adhesion reduction by immune sera from vaccinated mice. EHEC O157:H7 serotype EDL933 was incubated with 10% (A) or 5% (B) pooled sera (n = 3) from immunized mice with lomW (pVAX-10), escJ (pVAX-41), escC (pVAX-56), and –pComb in PBS and further incubated with Caco-2 cells at an MOI of 1:100 for 3 h to allow adherence. Sera from pVAX1 immunized mice as well as bacteria alone served as control groups. Bacterial adherence is shown as a percentage of bacteria recovered after incubation. Results are shown as percent adherence and as means ± of the SEM of triplicate values obtained from individual incubation well of bacteria with Caco-2 cells, and an asterisk (^*) indicates statistical significance as defined (p < 0.05), ^**p < 0.005.