Persister Cells May Explain Why Urinary Tract Infections Keep Coming Back

Urinary tract infections are among the most common bacterial infections in the world, and yet a frustrating number of them refuse to go away. A course of antibiotics is prescribed, symptoms fade, and then weeks or months later the infection returns, often with the same strain of bacteria that was supposedly eradicated the first time. A new Perspective article published in Nature Reviews Urology argues that the explanation for these treatment failures may not lie in the mechanisms that scientists and clinicians usually blame, such as antibiotic resistance, but in a far less visible phenomenon: the ability of bacteria to survive antibiotic exposure without ever acquiring resistance genes. The authors, Ariana Costas, Bill Söderström and Molly A. Ingersoll, propose that uropathogenic Escherichia coli, the dominant cause of urinary tract infections, may be exploiting antibiotic tolerance and persistence to outlast treatment and regrow once the drugs are gone.

The article begins by drawing a careful distinction between three related but fundamentally different ways in which bacteria can evade antibiotic killing, collectively described as the antibiotic non-response. Resistance is the most familiar of these. It arises when bacteria carry genetic determinants, whether encoded on chromosomes or on mobile elements such as plasmids, that allow them to keep growing even at antibiotic concentrations above the minimum inhibitory concentration. Resistance mechanisms include enzymes that degrade or modify drugs, efflux pumps that expel antibiotics from the cell, altered drug targets, and reduced permeability of the outer membrane. Resistance is well characterized across bacterial species and is strongly implicated in poor clinical outcomes, and its global burden has been documented in detail by large-scale analyses of antimicrobial resistance published in recent years.

Tolerance, by contrast, requires no resistance genes at all. In response to environmental conditions, an entire bacterial population can slow its growth rate, rendering it less susceptible to bactericidal antibiotics, which typically kill only actively growing cells. The minimum inhibitory concentration of the population remains unchanged, but the time needed to kill the bacteria increases substantially. Tolerance can be triggered by nutrient limitation, host stress signals, and metabolic slowdown, and it has been documented in a range of infection contexts. In the bladder, where nutrient availability fluctuates and host immune responses impose oxidative and nitrosative stress, the conditions favoring tolerance are likely to be common. The Perspective emphasizes that tolerance is a population-wide, often reversible phenotype, distinct from the genetic changes that underpin classical resistance.

Persistence is a subtler phenomenon still. Rather than the whole population slowing down, a small subpopulation of cells, known as persister cells, enters a state in which they survive antibiotic concentrations that kill the rest of their kin. Persisters were first described in the 1940s, when researchers observed that penicillin failed to sterilize bacterial cultures completely, and the phenomenon was later linked to the hipA gene of Escherichia coli K-12 in the early 1980s. Modern single-cell studies have shown that persister formation can arise through stochastic phenotypic switching, toxin-antitoxin systems, the stringent response mediated by the signaling nucleotide (p)ppGpp, and the SOS response triggered by DNA damage. Crucially, persisters are not resistant: their minimum inhibitory concentration is unchanged, and once the antibiotic is removed, they can resume growth and re-establish an infection.

The authors argue that both tolerance and persistence have been largely overlooked in the context of urinary tract infection, both in the clinic and in research. Standard susceptibility testing measures the minimum inhibitory concentration, which captures resistance but is blind to tolerance and persistence, since neither changes that value. A urine isolate may therefore be reported as fully susceptible, and the patient treated with an appropriate antibiotic, while a subpopulation of tolerant or persistent cells survives the course of therapy. When treatment stops, these surviving cells can regrow, causing recurrent infection with the same strain. This scenario could help explain why recurrent urinary tract infections are so common, why they often involve the same bacterial lineage as the initial episode, and why antibiotic failure rates in young women have been documented at clinically significant levels even when the infecting strain tests as susceptible.

Uropathogenic E. coli has several life history traits that make it a plausible candidate for persistence during infection. The bacterium uses type 1 pili to invade bladder epithelial cells, where it can form intracellular bacterial communities that resemble biofilms. Within these communities, bacteria adopt heterogeneous growth states, including coccoid morphotypes and filamentous forms, and some cells may enter quiescent, metabolically reduced states. Intracellular bacterial communities can persist after the acute phase of infection and serve as reservoirs from which bacteria re-emerge to cause recurrent episodes. Studies in mouse models have shown that bacteria surviving multiple courses of different antibiotics display characteristics consistent with persistence, and toxin-antitoxin systems have been shown to contribute to niche-specific colonization and stress resistance in these strains.

Experimental evidence supporting the tolerance and persistence hypothesis in urinary tract infection is accumulating. In vitro models of intracellular infection have shown that uropathogenic E. coli exhibits antibiotic tolerance and growth heterogeneity inside host cells. A human bladder-chip model revealed dynamic persistence within intracellular bacterial communities. Metabolite-dependent quiescence has been proposed as a mechanism explaining antibiotic tolerance during infection, and studies of bacterial division across different anatomical sites of the urinary tract have shown that variability in cell division shapes antibiotic survival in mouse models. Work on filamentation, in which bacteria elongate without dividing in response to host and antibiotic stress, has implicated the cell division machinery in beta-lactam tolerance. Meanwhile, a 2026 study found that survival of ampicillin-treated uropathogenic E. coli is independent of single-cell growth rates, suggesting that the classical picture of persisters as uniformly dormant cells may be too simple.

The Perspective also considers how host factors might shape the antibiotic non-response in the bladder. Research in other infection models has shown that host stresses, including glucose limitation induced by inflammasome activation, reactive oxygen species, and macrophage-produced peroxynitrite, can induce antibiotic tolerance in bacteria such as Staphylococcus aureus and Salmonella. In Salmonella, uptake by macrophages induces the formation of non-replicating persisters, and the decline of nitrosative stress in host tissues drives persister regrowth during infection. The bladder has its own resident macrophage populations that shape responses to infection, and the interplay between immune stress and bacterial survival states remains poorly understood for uropathogenic E. coli. The authors suggest that the urothelium, intracellular bacterial communities, and possibly L-form switching, in which bacteria shed their cell walls and become intrinsically resistant to beta-lactams, represent candidate sites and mechanisms of tolerance or persistence during infection.

The clinical implications of this framework are considerable. If tolerance and persistence contribute to treatment failure and recurrence, then current diagnostic practice, which relies on minimum inhibitory concentration breakpoints, systematically misses the relevant phenotypes. Modified assays, such as a disk-diffusion test designed to detect tolerance and persistence in clinical isolates, and high-throughput approaches for detecting hidden resistance, could help identify patients at risk of failure. Therapeutically, combinations of strongly and weakly metabolism-dependent antibiotics have been proposed as a strategy to eradicate persisters, and studies have shown that modulating the evolutionary trajectory of tolerance is possible by choosing antibiotics with different metabolic dependencies. There is also a warning embedded in the science: tolerance can facilitate the evolution of resistance, and persisters exposed to fluoroquinolones show enhanced development of resistance, meaning that the overlooked phenotypes may actively seed the emergence of the better-known one.

The authors conclude that two bacterial phenotypes capable of driving poor outcomes are currently overlooked both in the clinic and in research, adding to the challenges of treatment failure and recurrent urinary tract infection. Whether uropathogenic E. coli truly behaves as a persister during human infection remains an open question, and answering it will require models that capture the host environment of the urinary tract, single-cell approaches that resolve heterogeneous bacterial states, and clinical studies that track tolerance and persistence in patient isolates over time. What is clear is that the antibiotic non-response is broader than resistance alone, and that understanding the full repertoire of bacterial survival strategies during urinary tract infection may be essential if the cycle of treatment, apparent cure, and recurrence is ever to be broken for the millions of patients affected worldwide each year.

Subject of Research: Antibiotic tolerance and persistence in uropathogenic Escherichia coli during urinary tract infection

Article Title: Bacterial strategies for surviving antibiotics during urinary tract infection: could uropathogenic Escherichia coli be persisters?

Article References: Bacterial strategies for surviving antibiotics during urinary tract infection: could uropathogenic Escherichia coli be persisters?. (n.d.). https://doi.org/10.1038/s41585-026-01191-2

Image Credits: AI Generated

DOI: 10.1038/s41585-026-01191-2

Keywords: urinary tract infection, uropathogenic Escherichia coli, antibiotic tolerance, persister cells, antibiotic resistance, antibiotic non-response, intracellular bacterial communities, recurrent UTI, toxin-antitoxin systems, minimum inhibitory concentration, stringent response, bladder epithelium