Phage-induced protection against lethal bacterial reinfection

Abstract

Bacteriophages, or phages, are viruses that target and infect bacteria. Due to a worldwide rise in antimicrobial resistance (AMR), phages have been proposed as a promising alternative to antibiotics for the treatment of resistant bacterial infections. Up to this point in history, phage use in preclinical animal studies, clinical trials, and emergency-use compassionate care cases has centered around the original observation from 1915 showing phage as lytic agent, and thus a treatment that kills bacteria. Here, we describe an activity associated with phage therapy that extends beyond lytic activity that results in long-term protection against reinfection. This activity is potent, providing almost complete protection against a second lethal infection for animals treated with phage therapy. The activity also reduced infection burden an astounding billion-fold over the control. Reinfection protection requires phage lytic killing of its target bacterium but is independent of additional phage therapy. The effect is not driven by phage alone, lingering phage resistors, or a sublethal inoculum. In vitro phage-lysed bacteria provide partial protection, suggesting a combination of phage-induced lytic activity and immune stimulation by phage treatment is responsible for the effect. These observations imply certain phages may induce host adaptive responses following the lysis of the infecting bacteria. This work suggests phage therapy may contain a dual-action effect, an initial treatment efficacy followed by a long-term protection against reoccurring infection, a therapeutic-vaccination mechanism of action.

Keywords: ExPEC; immunity; phage therapy; sepsis.

Fig. 1.

Evaluation of the therapeutic and protective efficacy of ΦHP3 against lethal ExPEC infection in a murine model of bacteremia. (A) Schematic of the murine bacteremia model with ΦHP3 treatment (created with BioRender.com). Female BALB/cJ mice were infected intraperitoneally with a 1 × 10⁸ CFU of JJ2528. ΦHP3-treated mice received 1 × 10⁹ PFU of ΦHP3 phage per mouse, administered eight times at 12 h intervals. 4 wk posttreatment, mice were rechallenged with 1 × 10⁸ CFU of JJ2528. 2 wk following rechallenge, mice were euthanized, and the liver, kidneys, and spleen were homogenized and plated to measure bacterial burden. (B) Survival rates following the 1st infection and (C) 2nd infection were analyzed using the log-rank (Mantel–Cox) test. (D) Median CFU/mL of JJ2528 after 2nd infection, combining counts from all organs.

Fig. 2.

Evaluation of the protective efficacy of prior ΦHP3-resistant E. coli ExPEC JJ2528-8 infection against lethal ExPEC JJ2528 infection in a murine bacteremia model. (A) Schematic of the murine bacteremia model with the ΦHP3-resistant E. coli strain JJ2528-8 (created with BioRender.com). Female BALB/cJ mice were infected intraperitoneally with 1 × 10⁸ CFU of JJ2528. ΦHP3-treated mice received 1 × 10⁹ PFU of ΦHP3 phage per mouse, administered eight times at 12 h intervals. Mice in the JJ2528-8 group were infected intraperitoneally with 5 × 10⁷ CFU of the ΦHP3-resistant E. coli strain JJ2528-8 (LPS-truncated) instead. ΦHP3-treated mice received 1 × 10⁹ PFU of ΦHP3 phage per mouse, administered eight times at 12 h intervals. 4 wk posttreatment, all mice were challenged with 1 × 10⁸ CFU of JJ2528. 2 wk following challenge, mice were euthanized, and the liver, kidneys, and spleen were homogenized and plated to measure bacterial burden. (B) Survival rates following the 1st infection were analyzed using the log-rank (Mantel–Cox) test. (C) Median CFU/mL of JJ2528 after 1st, combining counts from all organs. (D) Survival rates following the 2nd infection were analyzed using the log-rank (Mantel-Cox) test. (E) Median CFU/mL of JJ2528 after 2nd infection, combining counts from all organs.

Fig. 3.

Evaluation of the protective efficacy of prior sublethal infection with ExPEC or ΦHP3-treated ExPEC sublethal infection against lethal ExPEC reinfection in a murine bacteremia model. (A) Schematic of the murine bacteremia model with the 1st infection using a sublethal dose of JJ2528, followed by a reinfection (2nd infection) with a lethal dose of JJ2528 (created with BioRender.com). Female BALB/cJ mice were infected intraperitoneally with 1 × 10⁶ CFU of JJ2528. ΦHP3-treated mice received 1 × 10⁹ PFU of ΦHP3 phage per mouse, administered eight times at 12 h intervals. 4 wk posttreatment, mice were rechallenged with 1 × 10⁸ CFU of JJ2528. 2 wk following rechallenge, mice were euthanized, and the liver, kidneys, and spleen were homogenized and plated to measure bacterial burden. (B) Survival rates following the 1st infection and (C) 2nd infection were analyzed using the log-rank (Mantel–Cox) test. (D) Median CFU/mL of JJ2528 after 2nd infection, combining counts from all organs.

Fig. 4.

Evaluation of the protective efficacy of ΦHP3-only and ΦHP3-lysate against lethal ExPEC infection in the murine model of bacteremia. (A) Schematic of the murine bacteremia model of ΦHP3-lysate and ΦHP3-only (created with BioRender.com). Female BALB/cJ mice were infected intraperitoneally with a 1 × 10⁸ CFU of JJ2528, except for the ΦHP3-only and ΦHP3-lysate groups. ΦHP3-treated and ΦHP3-only mice received either 1 × 10⁹ PFU or 3 × 10⁹ PFU of ΦHP3 phage per mouse, respectively, administered eight times at 12 h intervals. ΦHP3-lysate mice received 50 μL of ΦHP3-lysate (Materials and Methods) per mouse administered eight times at 12 h intervals. 4 wk posttreatment, all mice were challenged with 1 × 10⁸ CFU of JJ2528. 2 wk following challenge, mice were euthanized, and the liver, kidneys, and spleen were homogenized and plated to measure bacterial burden. (B) Survival rates following the 1st infection were analyzed using the log-rank (Mantel–Cox) test. (C) Median CFU/mL of JJ2528 after 1st infection, combining counts from all organs. (D) Survival rates following the 2nd infection were analyzed using the log-rank (Mantel-Cox) test. (E) Median CFU/mL of JJ2528 after 2nd infection, combining counts from all organs.