Phenotyping antibiotic resistance with single-cell resolution for the detection of heteroresistance

Abstract

Antibiotic resistance has emerged as an imminent threat to public health. It is also increasingly recognized to be a heterogeneous phenomenon: a population of bacteria, found in vivo or in vitro, often consists of a mixture of cells that have different degrees of resistance to antibiotics. The conventional metric to measure antibiotic resistance based on an ensemble-average minimum inhibitory concentration (MIC) fails to characterize the heterogeneity present within a bacterial population. This work describes a droplet microfluidics method to encapsulate single cells from a population consisting of a mixture of antibiotic-sensitive and antibiotic-resistant bacteria. Co-encapsulating viability probe alamarBlue with the cells allows the use of fluorescent drops as a read-out for drops that contain live cells after their exposure to antibiotics. Enumerating the fluorescent drops thus gives the number of resistant cells in the population. Our method enables the quantitative phenotyping of heterogeneous resistance, or heteroresistance, with single-cell resolution. We show that it is possible to detect a resistant sub-population that comprises as low as 10−6 of the entire population of cells. Such high resolution further allows us to measure the evolution of heteroresistance arising from the exposure of a homogeneous isogenic culture of susceptible cells to sub-lethal dosages of antibiotics. We demonstrate an application of this system to characterize genetic determinants of antimicrobial resistance emergence. The high resolution of phenotypic detection and quantification of minority variants demonstrated in this work has the potential to facilitate the elucidation of the mechanisms underlying heteroresistance. Such understanding will in turn inform the best practices for antibiotic use and containment of antimicrobial resistance in a wide range of settings from agriculture and aquaculture to disease management. Clinically, the ability to quantify and track the composition of a bacterial population will benefit the decision-making process in both the diagnosis and the treatment of bacterial infections, and will ultimately improve patient outcome and avert the spread of resistant populations.