Maternal vaccination with a fimbrial tip adhesin and passive protection of neonatal mice against lethal human enterotoxigenic Escherichia coli challenge

Abstract

Globally, enterotoxigenic Escherichia coli (ETEC) is a leading cause of childhood and travelers' diarrhea, for which an effective vaccine is needed. Prevalent intestinal colonization factors (CFs) such as CFA/I fimbriae and heat-labile enterotoxin (LT) are important virulence factors and protective antigens. We tested the hypothesis that donor strand-complemented CfaE (dscCfaE), a stabilized form of the CFA/I fimbrial tip adhesin, is a protective antigen, using a lethal neonatal mouse ETEC challenge model and passive dam vaccination. For CFA/I-ETEC strain H10407, which has been extensively studied in volunteers, an inoculum of 2 × 10(7) bacteria resulted in 50% lethal doses (LD50) in neonatal DBA/2 mice. Vaccination of female DBA/2 mice with CFA/I fimbriae or dscCfaE, each given with a genetically attenuated LT adjuvant (LTK63) by intranasal or orogastric delivery, induced high antigen-specific serum IgG and fecal IgA titers and detectable milk IgA responses. Neonates born to and suckled by dams antenatally vaccinated with each of these four regimens showed 78 to 93% survival after a 20× LD50 challenge with H10407, compared to 100% mortality in pups from dams vaccinated with sham vaccine or LTK63 only. Crossover experiments showed that high pup survival rates after ETEC challenge were associated with suckling but not birthing from vaccinated dams, suggesting that vaccine-specific milk antibodies are protective. In corroboration, preincubation of the ETEC inoculum with antiadhesin and antifimbrial bovine colostral antibodies conferred a dose-dependent increase in pup survival after challenge. These findings indicate that the dscCfaE fimbrial tip adhesin serves as a protective passive vaccine antigen in this small animal model and merits further evaluation.

FIG 1

CFA/I- and CfaE-specific antibody responses elicited in female DBA/2 mice exposed to the vaccine regimen. Serum IgG (A and B) and fecal IgA (C and D) responses were measured in samples collected from mouse groups vaccinated via the i.n. or o.g. route with three doses of CFA/I fimbriae (A and C) or dscCfaE (B and D) admixed or not with LTK63. (A to D) Blood and fecal samples were collected 2 weeks after the first (□), second (), and third (■) vaccinations. (E and F) Gastric samples of milk were collected from suckling pups born from dams vaccinated via the i.n. or o.g. route with CFA/I fimbriae (E) or dscCfaE (F) and pooled before testing (IgG, □; IgA, ■). Experiments to determine the CFA/I- and CfaE-specific antibody responses were independently performed between three and six times. Serum samples were processed and tested individually. Values represent means of endpoint titers ± the SD of three independent measurements. Statistically significant differences are indicated with brackets (one-way ANOVA with Bonferroni's test). P values are indicated by asterisks (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

FIG 2

Determination of the survival curves of newborn DBA/2 mice after challenge with ETEC strains H10407 and 258909-3. Different numbers of viable cells of H10407 (A), 258909-3 (B), and 258909-3M (mETEC) (B) were inoculated into the milk filled stomachs of neonatal mice up to 48 h after birth. The y axis values correspond to the percentage of surviving animals over an observation period of 7 days. Deaths recorded within the first 24 h after the challenge were not included in the results.

FIG 3

Mucosal immunization of female DBA/2 mice confers passive protection to newborns to lethal challenge with the CFA/I-ETEC strain H10407. (A and B) Survival curves of newborn mice challenged with 4 × 10⁸ CFU (20× LD₅₀) of H10407 and born from dams vaccinated via the i.n. (A) or o.g. (B) route. Dams were mated during the vaccination period and allowed to feed their offspring during the observation period. (C and D) Neonatal mice born from nonvaccinated dams were transferred to foster dams vaccinated with different immunization regimens, and neonatal mice born to vaccinated dams were transferred to nonvaccinated dams, as indicated in the figure. Pup survival was monitored over a total period of 8 days. Mice that died in the first 24 h after challenge were not included in the results. The numbers of animals challenged in each vaccination group are indicated (in parentheses) at the bottom of the panels. P values were calculated by two-tailed Fisher exact test comparing survival rate to the referent PBS-treated control group (PBS) (A and B) or between groups fostered by vaccinated (closed symbols) or nonvaccinated (open symbols) dams treated with the same vaccine formulation (C and D). P values are indicated for each comparison by asterisks (*, P < 0.05; **, P < 0.001; ***, P < 0.001). In addition, a Mantel-Cox test was applied to the same groups (not indicated in the figures), taking into account the complete curves leading to a P value of <0.0001.

FIG 4

Anti-CFA/I and anti-CfaE bovine colostral IgG preparations protected newborn mice against a lethal challenge with the CFA/I-ETEC strain 258909-3. Groups of DBA/2 mice were inoculated with 2 × 10⁷ CFU (20× LD₅₀ for 258909-3) previously incubated with different amounts of the anti-CFA/I (A) or the anti-CfaE (B) bIgG preparation. The survival curves were established over a period of 1 week. Deaths recorded in the first 24 h after the challenges were not considered. The number of pups used to test each bIgG preparation ranged from 21 to 37. P values were calculated by two-tailed Fisher exact test in which the last values of the survival curves were compared to those corresponding to the reference group (skim milk) and are indicated by asterisks (*, P < 0.05; **, P < 0.001; ***, P < 0.001). In addition, a Mantel-Cox test was applied to the same groups (not indicated in the figures), taking into account the complete curves leading to a P value of <0.0001.