Resource limitation prevents the emergence of drug resistance by intensifying within-host competition

Abstract

Slowing the evolution of antimicrobial resistance is essential if we are to continue to successfully treat infectious diseases. Whether a drug-resistant mutant grows to high densities, and so sickens the patient and spreads to new hosts, is determined by the competitive interactions it has with drug-susceptible pathogens within the host. Competitive interactions thus represent a good target for resistance management strategies. Using an in vivo model of malaria infection, we show that limiting a resource that is disproportionately required by resistant parasites retards the evolution of drug resistance by intensifying competitive interactions between susceptible and resistant parasites. Resource limitation prevented resistance emergence regardless of whether resistant mutants arose de novo or were experimentally added before drug treatment. Our work provides proof of principle that chemotherapy paired with an "ecological" intervention can slow the evolution of resistance to antimicrobial drugs, even when resistant pathogens are present at high frequencies. It also suggests that a broad range of previously untapped compounds could be used for treating infectious diseases.

Keywords: Plasmodium chabaudi; combination therapy; competition; drug resistance; evolutionary management.

Fig. 1.

Resource limitation prevents the emergence of drug resistance. (A) Mice were assigned to each resource treatment and inoculated with 10⁶ pyrimethamine-susceptible parasites. Rebounding parasite populations were genotyped (Fig. S3) and injected into a second drug-treated mouse to assess their phenotypic resistance. All parasites that were genotypically resistant and underwent phenotypic testing were found to be phenotypically resistant. (B) Proportion of mice in which parasites rebounded (light red, thin-lined bar) and which were confirmed to be either genetically or phenotypically resistant (dark red, thick-lined bar). Error bars represent the 95% confidence interval around the proportion as calculated from a binomial distribution. n, number of mice included in the analysis.

Fig. 2.

Resource limitation prevents the emergence of drug resistance by intensifying competitive suppression. Dynamics of pyrimethamine-susceptible (black, solid lines) and -resistant (red, dashed lines) parasites in mice given supplemental pABA (A and C; gray background) or not (B and D; white background) and infected with both susceptible and resistant parasites (A and B) or with only resistant parasites (C and D) are shown. Each line represents the dynamics of parasites in an individual mouse. The infection dynamics of each mouse are plotted in Figs. S4–S6. Stars indicate the number and timing at which parasites were inoculated; note that resistant parasites enter the host at a much greater frequency than a de novo mutant would. Dots indicate the density of parasites detected on a particular day in instances where parasites were not detected the day before or after, and the green bar represents the period of pyrimethamine treatment. n, number of mice included in the analysis. Susceptible and resistant parasites were of the AJ and AS genetic backgrounds (S_AJ and R_AS), respectively; only resistant parasites possess the S106N mutation associated with pyrimethamine resistance in P. chabaudi (Fig. S7). In the absence of drugs, resistant parasites were competitively suppressed by susceptible parasites and did not emerge in either resource treatment (Fig. S8).

Fig. 3.

Resource limitation improves host health and prevents the transmission of drug resistance in mice inoculated with both drug-susceptible and drug-resistant parasites. Red blood cell dynamics of individual mice (A and B; solid lines) and the probability of resistant parasites transmitting to mosquitoes (C and D; dashed lines) in mice infected with both susceptible and drug-resistant parasites (S_AJ and R_AS) and either supplemented with resources (A and C; gray background, pABA⁺) or not (B and D; white background, pABA⁻) are shown. All mice were administered pyrimethamine treatment (green bar). Limitation of pABA was associated with less anemia (total red blood cell density, resource treatment: F_1,17 = 24.3; P < 0.001, linear regression). Note that these data are from all (A and B) or a subset (C and D) of the infections shown in Fig. 2 A and B.

Fig. 4.

Resource limitation prevents the emergence of drug-resistant parasites, irrespective of the combination of strains with which the host is infected. Dynamics of susceptible (black, solid lines) and pyrimethamine-resistant (red, dashed lines) parasites in mice coinfected with S_AS and R_AJ (A and B), S_AT and R_CW (C and D), or only R_AJ (E) or R_CW (F), and either administered resources (A and C; gray background) or not (B and D–F; white background) are shown. The resistant strains (R_AJ and R_CW), but not the susceptible strains (S_AS and S_AT), possess the mutation associated with pyrimethamine resistance in this system (Fig. S7). Each line represents the dynamics of infections in an individual mouse; the infection dynamics of each mouse are plotted separately in Fig. S9. Stars represent the number of parasites inoculated and the time at which they were administered. Dots indicate the density of parasites detected on a particular day in instances where parasites were not detected the day before or after. The green bar indicates the duration and timing of pyrimethamine treatment. n, number of mice plotted and included in the analysis. Note that in these experiments, unlike those in Fig. 2, resistant parasites were not grown alone in pABA-supplemented mice.