Localized pmrB hypermutation drives the evolution of colistin heteroresistance

Abstract

Colistin has emerged as an important last line of defense for the treatment of infections caused by antibiotic-resistant gram-negative pathogens, but colistin resistance remains poorly understood. Here, we investigate the responses of ≈1,000 populations of a multi-drug-resistant (MDR) strain of P. aeruginosa to a high dose of colistin. Colistin exposure causes rapid cell death, but some populations eventually recover due to the growth of sub-populations of heteroresistant cells. Heteroresistance is unstable, and resistance is rapidly lost under culture in colistin-free medium. The evolution of heteroresistance is primarily driven by selection for heteroresistance at two hotspot sites in the PmrAB regulatory system. Localized hypermutation of pmrB generates colistin resistance at 10³-10⁴ times the background resistance mutation rate (≈2 × 10^-5 per cell division). PmrAB provides resistance to antimicrobial peptides that are involved in host immunity, suggesting that this pathogen may have evolved a highly mutable pmrB as an adaptation to host immunity.

Keywords: AMR; CP: Microbiology; Pseudomonas aeruginosa; antibiotic resistance; colistin; evolutionary rescue; experimental evolution; heteroresistance; hypermutability; mutation rate.

Graphical abstract

Figure 1

Population responses to colistin treatment Independent cultures of strain EP67 (n = 100) were inoculated into culture medium supplemented with colistin (2 mg/L). (A) Viable cell titer (n = 100; ±95% confidence interval [CI]) over the first 8 h of incubation. (B) Proportion of cultures (n = 100; ±95% CI) showing detectable growth over time, with a minimal limit of detection of 400 CFUs/culture. (C) Distribution of viable cell titer in cultures showing growth at 8, 24, and 48 h. The limits of detection in this assay were 400–8000 CFUs.

Figure 2

Testing the stability of colistin resistance Independent cultures of EP67 were incubated with colistin for 24 h, to select for resistance (n = 171). Cultures were then transferred to LB to allow resistant populations to expand. Recovered populations were then passaged into medium with LB or LB + colistin to test the stability of colistin resistance. After each passage, cultures diluted 10-fold and spotted out on LB agar plates to determine population viability, as determined by confluent colony growth.