A cocktail of three virulent bacteriophages prevents Vibrio cholerae infection in animal models

Abstract

Effective prevention strategies will be essential in reducing disease burden due to bacterial infections. Here we harness the specificity and rapid-acting properties of bacteriophages as a potential prophylaxis therapy for cholera, a severely dehydrating disease caused by Vibrio cholerae. To this end, we test a cocktail of three virulent phages in two animal models of cholera pathogenesis (infant mouse and rabbit models). Oral administration of the phages up to 24 h before V. cholerae challenge reduces colonization of the intestinal tract and prevents cholera-like diarrhea. None of the surviving V. cholerae colonies are resistant to all three phages. Genome sequencing and variant analysis of the surviving colonies indicate that resistance to the phages is largely conferred by mutations in genes required for the production of the phage receptors. For acute infections, such as cholera, phage prophylaxis could provide a strategy to limit the impact of bacterial disease on human health.

Figure 1. The ICP cocktail kills V. cholerae in vitro.

Growth curves of V. cholerae E7946 (AC53) in the absence and presence of phage predation. Cultures were grown in the absence of phage (black) or with ICP1 (blue), ICP2 (pink), ICP3 (green) or the ICP phage cocktail (orange) at an multiplicity of infection of 1. Samples of each culture were taken periodically and (a) enumerated for CFU per ml or (b) the phage fraction isolated and enumerated for PFU per ml by plaque assay. Data are plotted as the average of five independent replicates. Error bars represent the s.e.m. The dotted line represents the limit of detection.

Figure 2. Efficacy of the ICP cocktail and the ICP phages in reducing V. cholerae colonization of the infant mouse small intestine.

Mice were dosed with between 1 × 10⁶ and 1 × 10⁷ PFU of the indicated phages 3 h before infection with 5 × 10⁵ CFU V. cholerae E7946 (AC53). A control group did not receive phage (n=4). For each phage-treated group of animals, n=5. (a) V. cholerae that survived in the small intestine were enumerated as CFU per small intestine. (b) Phages that survived in the small intestine were enumerated as PFU per small intestine by plaque assay. The dotted line represents the limit of detection and the horizontal solid bar represents the median. Each square or circle represents one animal. Significance was calculated using the Kruskal–Wallis test with the Dunn's post-hoc multiple comparisons test. **P=0.001–0.01.

Figure 3. V. cholerae burden in the infant mouse small intestine is reduced when the ICP cocktail is administered up to 24 h before infection.

(a) Phages are retained in the infant mouse small intestine in the absence of the V. cholerae host. Mice were dosed with between 3 × 10⁷ and 3 × 10⁸ PFU of the indicated phage and killed after 3, 6, 12 or 24 h. Phages surviving in the small intestine were calculated by enumeration of PFU per small intestine from plaque assays. An orange circle represents ICP1, a blue square ICP2 and a green triangle ICP3. For each group, n=3. (b) Infant mice were dosed with between 2 × 10⁵ and 4.8 × 10⁵ PFU of the ICP cocktail either 6 h (n=7), 12 h (n=10) or 24 h (n=10) before infection with between 5 × 10⁵ and 9 × 10⁵ CFU V. cholerae E7946 (AC53). A control group did not receive phage (n=8). Twenty-four hours after infection, the mice were killed and V. cholerae surviving in the small intestine enumerated by calculation of CFU per small intestine. (c) Phages surviving in the small intestine after the challenge described in b were enumerated by calculation of PFU per small intestine from plaque assays. (d) Infant mice were dosed with between 1 × 10⁸ and 1.3 × 10⁸ PFU of the ICP cocktail 6 h (n=8), 12 h (n=8) or 24 h (n=8) before infection with a ‘high-challenge dose' of 1 × 10⁸ CFU V. cholerae E7946 (AC53). Twenty-four hours after infection, mice were killed and the V. cholerae surviving in the small intestine enumerated as CFU per small intestine, and phages enumerated as PFU per small intestine by plaque assay (e). The dotted line represents the limit of detection and the horizontal solid bar represents the median. Each square or circle represents one animal. Significance was calculated using the Kruskal–Wallis test with the Dunn's post-hoc multiple comparisons test. *P=0.01–0.05, ***P=0.0001–0.001 and ****P<0.0001.

Figure 4. Infant rabbits dosed with the ICP phage cocktail before V. cholerae infection have a reduced bacterial load in the intestine.

(a) Infant rabbits were dosed with 3 × 10⁸ PFU of the ICP phage cocktail and killed after 3 h (n=8) or 24 h (n=8). Phage retention in the intestine in the absence of the V. cholerae host was measured by enumeration of PFU per intestine. (b) Infant rabbits were dosed with between 4 × 10⁹ and 8 × 10⁹ PFU of the ICP phage cocktail either 3 h (n=7) or 24 h (n=8) before infection with 5 × 10⁸ CFU V. cholerae E7946 (AC53). A control group (n=8) did not receive phage. At the end of the infection period (either 12–14 h for the V. cholerae only group, or 20 h after V. cholerae infection for the phage-treated groups (see Methods for more details)), rabbits were killed and V. cholerae surviving in the intestine were enumerated as CFU per intestine. (c) Phages surviving in the intestine after the challenge described in b were enumerated as PFU per intestine. The dotted line represents the limit of detection and the horizontal solid bar represents the median. Each square or circle represents one animal. Significance was calculated using the Kruskal–Wallis test with the Dunn's post-hoc multiple comparisons test. *P=0.01–0.05 and ***P=0.0001–0.001.