Dynamics of success and failure in phage and antibiotic therapy in experimental infections

Abstract

Background: In 1982 Smith and Huggins showed that bacteriophages could be at least as effective as antibiotics in preventing mortality from experimental infections with a capsulated E. coli (K1) in mice. Phages that required the K1 capsule for infection were more effective than phages that did not require this capsule, but the efficacies of phages and antibiotics in preventing mortality both declined with time between infection and treatment, becoming virtually ineffective within 16 hours.

Results: We develop quantitative microbiological procedures that (1) explore the in vivo processes responsible for the efficacy of phage and antibiotic treatment protocols in experimental infections (the Resistance Competition Assay, or RCA), and (2) survey the therapeutic potential of phages in vitro (the Phage Replication Assay or PRA). We illustrate the application and utility of these methods in a repetition of Smith and Huggins' experiments, using the E. coli K1 mouse thigh infection model, and applying treatments of phages or streptomycin.

Conclusions: 1) The Smith and Huggins phage and antibiotic therapy results are quantitatively and qualitatively robust. (2) Our RCA values reflect the microbiological efficacies of the different phages and of streptomycin in preventing mortality, and reflect the decline in their efficacy with a delay in treatment. These results show specifically that bacteria become refractory to treatment over the term of infection. (3) The K1-specific and non-specific phages had similar replication rates on bacteria grown in broth (based on the PRA), but the K1-specific phage had markedly greater replication rates in mouse serum.

Figure 1

Survival of CAB1 infected mice: (Top) Immediate Treatment – Mice were inoculated with 10⁸ CAB1 in the right thigh and within one minute treated by inoculation in the left thigh with (i) a control (saline or a CAB1-pseudolysate) (ii) 10⁸ non-K1 specific phage φLW, (iii) 10⁸ K1-specific phage φLH, or (iv) 60 μg/ml streptomycin. (Bottom) Delayed Treatment: – Mice were inoculated with 10⁸ CAB1 in the right thigh and, 8 hours later, treated by inoculation in the left thigh with (i) 0.85% saline (control), (ii) 10⁸ K1-specific, φLH, phage, (iii) 60 μg/gm streptomycin, or (iv) 100 μg/ml streptomycin.

Figure 2

Phage Replication Assay: Rate of replication of φLH and φLW on CAB1 in LB and in mouse serum. In LB, the growth rates of the K1 phage (φLH, squares connected by the solid line) are similar to those of the non-K1 phage (φLW, solid circles connected by the dashed line) across a range of densities, with both phages growing more rapidly on cells at low density than on cells at high density, near saturation. However, on cells grown in mouse serum, the growth rate is much higher for the K1-specific phage, φLH, (gray triangles) than for the non-K1 specific phage, φLW (gray circles). Growth rate is presented as doublings of phage concentration per hour; standard error bars are given for each black point (on the lines), but most are so short as to be obscured by the point. Each gray point represents one assay in serum.