Pseudomonas aeruginosa: ecology, evolution, pathogenesis and antimicrobial susceptibility

Abstract

Pseudomonas aeruginosa has long served as a model organism in microbiology, particularly for studies on gene expression, quorum sensing, antibiotic resistance, virulence and biofilm formation. Its genetic tractability has advanced the understanding of complex regulatory networks and experimental evolution. The versatility of this bacterium stems from its genomic variability, metabolic flexibility and phenotypic diversity, enabling it to thrive in diverse environments, both as a harmless saprophyte and an opportunistic human pathogen. P. aeruginosa can cause acute and chronic human infections, particularly in patients with underlying immune deficiencies. Its intrinsic antibiotic tolerance and resistance, together with its ability to produce multiple virulence factors while rapidly adapting to infection conditions, pose a major clinical challenge. In this Review, we explore key features contributing to the ecological and pathogenic versatility of P. aeruginosa. We examine the molecular mechanisms and ecological and evolutionary implications of quorum sensing and biofilm formation. We explore the virulence strategies and in vivo fitness determinants, as well as the evolutionary dynamics and global epidemiology of P. aeruginosa, with a focus on antimicrobial resistance. Finally, we discuss emerging strategies to control P. aeruginosa infections and address outstanding questions in the field.

Introduction

The Gram-negative bacterium Pseudomonas aeruginosa is a multifaceted organism that, for decades, has served as a model organism in molecular biology for understanding bacterial gene expression, regulation, cell–cell communication (quorum sensing), antibiotic resistance, virulence and biofilm formation. Its genetic tractability has allowed researchers to unravel complex gene regulatory networks, and it has also proved invaluable as an organism for experimental evolution studies. P. aeruginosa was first isolated in 1882 from bandages taken from the wound infections of soldiers, and was noted to exhibit a greenish-blue colour¹. This distinctive colour results from the combination of two pigmented metabolites, pyocyanin (blue) and pyoverdine (yellow–green), which defines the species term aeruginosa (verdigris: that is, aerūgō in Latin), as it resembles the blueish–green rust that forms on copper or brass by atmospheric oxidation.

P. aeruginosa is a functionally versatile organism owing to its tremendous genomic variability (Box 1), metabolic flexibility and phenotypic diversity (Box 2). These traits allow it to adapt to different habitats, resulting in a bacterium that can range from a harmless saprophyte and commensal organism to a deadly broad-host-range pathogen. Besides its environmental presence, where it is enriched in areas associated with human activity², P. aeruginosa has become a leading opportunistic human pathogen that rarely infects healthy individuals. In individuals with compromised immune defences, P. aeruginosa can cause both acute and chronic infections, including bloodstream³, urinary tract⁴, eye⁵ and soft-tissue wound infections⁶ (Fig. 1a). It is also highly problematic in lung infections, especially in people with the genetic disease cystic fibrosis⁷, chronic pulmonary disease⁸ and ventilator-associated pneumonia⁹. Recently, the oxygen-responsive small RNA inducer of chronic infection X (SicX) has been identified as a critical chronic-to-acute switch in P. aeruginosa during mammalian infection¹⁰.

Box 1 |. Genome and genetic variation of Pseudomonas aeruginosa.

The versatility of Pseudomonas aeruginosa in colonizing and thriving across various hosts and environments is often attributed to its relatively large genome (size varies between 5.5 and 7 megabases, guanine–cytosine content 63–65%) and ability to shape its genomic composition, resulting in a corresponding enhancement of its metabolic capacity. The first complete genome sequenced for this species was that of strain PAO1 (ref. 14), originally isolated from a chronic wound, followed by the highly virulent strain PA14 (ref. 166). PAO1 serves as the major reference strain for genetic and functional studies on P. aeruginosa. There are now hundreds of high-quality genome sequences available from environmental and clinical isolates of P. aeruginosa, whose comparative analysis has provided insights into the genetic diversity and phylogeny of this species. Pan-genome analyses had defined two major phylogroups, represented by PAO1 and PA14 strains. An additional minor group, represented by PA7, shows substantial genomic divergence from the dominant lineages. A five-phylogroups structure that includes two additional minor phylogroups, intermediate between the PAO1–PA14-like groups and the PA7-like group, has been identified¹⁶⁷. However, clade-2 strains, including PA7, have recently been proposed to constitute a novel species (Pseudomonas paraeruginosa)¹⁶⁸.

An examination of 5,098 genomes found the P. aeruginosa pan-genome to be composed of 126,951 non-redundant protein families, including 3,769 (3%) core gene clusters (present in at least 98% of genomes), 73,946 (58%) accessory gene clusters and 49,236 (39%) unique genes¹⁶⁹. The P. aeruginosa core genome is highly conserved and includes approximately 62% of the average gene content for the species (n = 6,103 protein genes using 5,098 genomes). This core genome encodes a set of metabolic and pathogenic factors shared by all P. aeruginosa strains, regardless of their origin. The core genome is interspersed with regions of genomic plasticity, which comprise accessory genes that encode niche adaptation properties, such as novel catabolic pathways and antimicrobial resistance. It is important to remember that although P. aeruginosa is a well-studied organism, one-third of its genes are still uncharacterized. Horizontal gene transfer, mediated by bacteriophages and plasmids, strongly influences the evolutionary trajectories of P. aeruginosa strains, contributing to the spread of virulence and antimicrobial resistance genes^167,169,170. Adaptation strategies in P. aeruginosa are not limited to the acquisition of new genes and metabolic functions. Rearrangements, mutations and genome reductive evolution are observed as active adaptation processes to niche-selective pressures, contributing to the genomic and phenotypic diversification of P. aeruginosa strains, especially among those obtained from chronic infections^{51,62,151,170–173}. Genetic and phenotypic variation have also been observed between P. aeruginosa PAO1 strains from different laboratories and strain banks worldwide. This variability probably arises from microevolution occurring during culture and storage, which may affect the reproducibility of experiments across different research groups.

Box 2 |. Basic microbiology, physiology and phenotypic diversity of Pseudomonas aeruginosa.

Pseudomonas aeruginosa is a Gram-negative, rod-shaped, non-sporing bacterium, approximately 1–5-μm long and 0.5–1.0-μm wide, with a single polar flagellum for motility (see the figure, part a; adapted from ref. 174, Springer Nature Limited). Its genome ranges from 5.5 to 7 million base pairs. It can grow in minimal medium with a single source of carbon and energy, and is capable of metabolizing hundreds of organic compounds. P. aeruginosa preferentially catabolizes small organic acids and amino acids¹⁷⁵. Although it can utilize carbohydrates such as glucose, it lacks a complete glycolytic pathway. As a facultative anaerobe, P. aeruginosa relies on five terminal oxidases for aerobic respiration and can use denitrification enzymes under anaerobic conditions. In the absence of terminal respiratory electron acceptors (oxygen and nitrogen oxides), P. aeruginosa can slowly grow anaerobically with arginine or utilize pyruvate fermentation¹⁷⁶. Carbon sources and nutritional cues strongly influence its physiology and pathogenesis^10,177,178. P. aeruginosa can survive in a broad range of temperatures (4–42 °C), with an optimal growth temperature of 37 °C. The production of 2-amino-acetophenone gives P. aeruginosa its characteristic sweet, grape-like odour¹⁷⁹. When grown on blood agar, it exhibits β-haemolysis with a greenish, metallic sheen (see the figure, part b; adapted from ref. 180, CC BY 4.0). Catalase positive, it has three main catalase genes (katA, katB and katE)¹⁸¹. P. aeruginosa produces numerous soluble pigments, including the siderophore pyoverdine (yellow–green and fluorescent)^182,183, the phenazine pyocyanin (blue–green)^182,184, pyomelanin (brown–black)^182,185 and pyorubin (red–brown)¹⁸⁵ (see the figure, part c; reprinted with permission from ref. 182, Elsevier). P. aeruginosa also synthesizes pyocins, which are bacteriocins that selectively kill other susceptible P. aeruginosa strains and contribute to intraspecies competition¹⁸⁶.

The phenotypic variability among P. aeruginosa isolates arises from genetic changes and correlates with variations in gene expression in response to environmental cues, with key levels of control for physiology and pathogenesis including gene induction, transcription, translation and small RNA regulation. P. aeruginosa exhibits a variety of colony morphologies. Small colony variants are slow-growing bacterial subpopulations that emerge under stress, such as low oxygen availability (see the figure, part d), and are correlated with the formation of a surface pellicle¹⁸⁷ (see the figure, part e; adapted from ref. 187, CC BY 4.0). The mucoid colony appearance, due to overproduction of alginate, is typical of chronic infections, especially in cystic fibrosis (see the figure, part d). Colony wrinkling or rugose morphology is a redox-driven morphological change that maximizes oxygen accessibility and requires the production of the exopolysaccharide Pel¹⁸⁸ (see the figure, part f; adapted with permission from ref. 188, ASM). Autolysis occurs due to the accumulation of 2-heptyl-4-hydroxyquinoline (HHQ)¹⁸⁹ (see the figure, part f; adapted from ref. 189, CC BY 4.0).

P. aeruginosa is highly motile, and utilizes flagellar swimming motility in liquid or in low agar concentrations¹⁹⁰, and type 4 pili-mediated twitching motility on hard and soft surfaces^190,191. It also exhibit collective swarming motility involving both pili and flagellum (see the figure, part g)¹⁹⁰. P. aeruginosa is an avid biofilm former (see the figure, part h; adapted with permission from ref. 95, Wiley, and ref. 104, Springer Nature Limited) and is a key model organism for studying biofilm development and bacterial cell–cell communication (quorum sensing)^30,32.

P. aeruginosa strains PAO1 (ref. 192) and UCBPP-PA14 (PA14)¹⁹³, isolated in 1954 and 1977 respectively from human wound infections, are the most studied. Owing to its genetic tractability and fast growth rate, P. aeruginosa is a model organism in microbiology, crucial for sociomicrobiology research¹⁹⁴, understanding microbial infections and advancing therapeutic strategies.

Fig. 1 |. Ecology of Pseudomonas aeruginosa.

a, Pseudomonas aeruginosa is responsible for several major infections, including respiratory tract infections (such as those in patients with cystic fibrosis, chronic obstructive pulmonary disease and ventilator-associated pneumonia), bloodstream infections, urinary tract infections (typically associated with catheters), eye infections (where the bacteria cause keratitis after ocular surgery or through contaminated contact lenses or eye drops), and soft-tissue infections (including acute burn wounds, chronic surgical wounds and diabetes foot infections). b, The environmental occurrence of P. aeruginosa: although P. aeruginosa is ubiquitous in nature, it is usually scarce in pristine environments and is predominantly found in locations associated with human activity. c, In health-care settings, infection and transmission of P. aeruginosa can occur through various routes, including contact with hospital staff or pre-colonized patients, contaminated ventilators, catheters, or other medical equipment and devices. The bacteria can also be found on the surfaces of sinks, drains and shower heads, and can spread through tap water. Additionally, arthropods, such as cockroaches, may serve as a potential source of transmission for P. aeruginosa.

P. aeruginosa infections are notably difficult to treat owing to their high intrinsic antibiotic tolerance or resistance, the production of multiple intra- and extracellular virulence factors, the prevalence of P. aeruginosa as a nosocomial pathogen and its rapid adaptation to infection. Therefore, P. aeruginosa, is considered a serious threat by the (CDC) Centers for Disease Control and Prevention¹¹, and the WHO (World Health Organization) has listed P. aeruginosa among the critical priority pathogens for which research and development of new antimicrobial compounds is urgently needed. P. aeruginosa has been associated with outbreaks containing the Verona integron-encoded metallo-β-lactamase (blaVIM) that provides resistance to carbapenem antibiotics, which are a key line of treatment for nosocomial P. aeruginosa infections. Multidrug-resistant, extensively drug-resistant and pandrug-resistant¹² strains are increasing and recognized to have higher mortality rates, and there are certain P. aeruginosa sequence types that are considered ‘high risk’ as they are disseminated worldwide and highly resistant to treatment¹³.

In this Review, we provide an overview of the key features that contribute to the remarkable versatility of P. aeruginosa, allowing it to adapt and thrive in diverse ecological niches. To this end, we detail well-characterized and recent findings that advance our knowledge of quorum sensing and biofilm formation, from molecular mechanisms to ecological and evolutionary perspectives. We also examine traditional virulence strategies that P. aeruginosa adopts to infect various hosts, along with metabolic and stress pathways important for in vivo fitness, which allow it to be an important pathogen. Additionally, we discuss the evolutionary dynamics of P. aeruginosa in nature and within the human host, evaluating its ‘ubiquitous’ presence, environmental incidence and epidemiology, with a special focus on antimicrobial resistance, the emergence and spread of highly virulent and resistant clones (Box 3), and possible routes of transmission. Finally, we explore emerging control strategies for P. aeruginosa infections.

Box 3 |. Pseudomonas aeruginosa as a model organism: research applications and considerations for selecting laboratory versus clinical strains in research.

Pseudomonas aeruginosa is one of the most extensively studied microorganisms, largely due to its genetic tractability and immense publicly available omics data. Recently, P. aeruginosa has been used as a model organism to tackle challenging problems in microbiology, including the development of a generalizable quantitative framework to validate and improve laboratory model systems^195,196 and the investigation of phage–host interactions and CRISPR–Cas biology to advance phage therapeutic applications¹⁹⁷.

The development of preclinical models that accurately replicate the natural environment is essential for understanding the function and resilience of natural microbial communities. However, quantifying ways in which a model does, and does not, recapitulate microbial physiology in natural environments has been difficult. A quantitative framework was developed to assess model accuracy by comparing P. aeruginosa gene expression during chronic human infection with that in laboratory models. The accuracy is measured by the degree in which the expression of each P. aeruginosa gene in the model deviates from that in chronic human infection^108,195,198. This approach has been used to improve in vitro and in vivo models of P. aeruginosa^108,196,199 and could be broadly applied to any microbial¹⁹⁸ or polymicrobial model, environmental setting and functional measurement.

P. aeruginosa also serves as a key model organism for studying phage–bacterial immune interactions and engineering phage genomes using CRISPR–Cas systems. Bioinformatic tools have uncovered new anti-phage immune systems, including two core defence hot spots in most P. aeruginosa genomes, which may represent novel anti-phage mechanisms. Jumbo phages, with large genomes, resist DNA-targeting immune systems by protecting their DNA with a ‘phage nucleus’ structure²⁰⁰. Although jumbo phages show potential for antimicrobial therapy, genetic tools for studying them were previously unavailable. Recent research on P. aeruginosa jumbo phage ΦKZ has developed a CRISPR–Cas13a-based system for precise genome editing¹⁹⁷, offering insights into phage biology. These studies highlight the role of P. aeruginosa in advancing bacterial physiology, microbial interactions and phage resistance research.

As antibiotic resistance increases, and ‘high-risk’ strains spread globally, a question to carefully consider is which bacterial strains we should use in future research? The choice is often between clinical isolates and traditional laboratory strains such as PAO1, PA14 and PAK, and there are advantages and disadvantages to both. Laboratory strains, although initially isolated from human infections, have been adapted to laboratory conditions over the course of many decades. These strains are well characterized and standardized, which makes them valuable for reproducible experiments. Their genetic makeup is relatively stable, reducing experimental variables and allowing researchers to focus on specific aspects of bacterial behaviour. Additionally, extensive background information about these strains, including their genetic sequences and metabolic pathways, supports the interpretation of experimental results across different laboratories around the world. They are also genetically tractable, making them excellent strain choices to perform detailed molecular studies on systems such as quorum sensing and biofilm formation.

Conversely, clinical isolates, which are obtained directly from recent infections or clinical settings, provide insights into real-world bacterial behaviour. These strains often exhibit unique or emerging antibiotic resistance patterns, offering valuable information for developing new strategies to combat resistant bacteria. They also display a range of pathogenic traits, which can reveal how different strains cause disease and interact with host organisms. Clinical isolates also typically present a broader genetic diversity than laboratory strains, shedding light on the evolutionary dynamics of bacterial populations and the genetic basis of various traits. Many of these isolates have adapted to form enhanced biofilms or have gained or lost certain traits that allow them to persist in the human body, enhancing our understanding of chronic infections and informing the development of new treatments or preventive measures. Research on clinical isolates could, therefore, directly impact treatment strategies and improve clinical outcomes by identifying more effective antibiotics or alternative therapies. A major disadvantage is that clinical isolates of P. aeruginosa are often phenotypically and genomically diverse, even within a single patient^{151,171–173}. This raises important questions, such as which isolate to select for experimentation or whether we should consider working with more heterogeneous populations of P. aeruginosa rather than just single isolates. In summary, whereas laboratory strains offer controlled and reproducible research conditions, clinical isolates can provide a more accurate reflection of real-world bacterial behaviours and resistance patterns. Ultimately, the choice of strain should be guided by the context of the study and the specific research goals.

Lifestyle and ecology of P. aeruginosa

P. aeruginosa is a highly adaptable bacterium that can thrive in a range of environments, particularly environments with high organic content. This success hinges on its ability to continuously monitor its surroundings and adjust gene expression accordingly in response to external cues (for example, osmolarity, pH, temperature, iron and oxygen levels). Indeed, approximately 9% of the P. aeruginosa genome encodes either transcriptional regulators or two-component regulatory system proteins¹⁴, and so far, about 550 small RNA have been predicted in P. aeruginosa, with more than 200 characterized^15–17. These regulatory mechanisms ultimately drive phenotypic changes and enable P. aeruginosa to activate substrate-specific metabolic pathways, facilitating the uptake and utilization of various carbon and nitrogen sources, as well as many other essential nutrients such as iron.

P. aeruginosa has been isolated from diverse habitats ranging from natural ecosystems^18–20 to human hosts²¹ and even distilled water²², earning it the label of a ‘ubiquitous’ microorganism. However, recent investigations suggest that the generalized definition of P. aeruginosa as being widespread in nature may need to be reassessed². Studies indicate that P. aeruginosa is generally rare in pristine environments but shows a higher prevalence in areas impacted by human activities (Fig. 1b). For instance, in 2020, a rigorous study evaluated environmental samples collected in three countries across three continents, alongside 16S rRNA sequence data from public databases, and performed a systematic review and meta-data analysis of published data. P. aeruginosa was found in only 12% of uncontaminated samples, with higher distribution in animal faeces and compost (32%) compared with soil and rhizosphere (7%) or environmental water (0%). Overall, the occurrence of P. aeruginosa in the environment is closely associated with high human activity compared with unpolluted habitats². These conclusions align with older reports^23,24 stating that waters polluted by humans or animals, such as sewages or urban rivers, are the most frequent sources of P. aeruginosa, suggesting that environments containing human and animal faeces are important natural reservoirs for P. aeruginosa².

P. aeruginosa is a major pathogen in health-care settings, frequently found in hospitals — on medical devices and in moist areas — where it causes severe infections, particularly in immunocompromised patients. A key factor in its ecological versatility and adaptability is quorum system, a mechanism that coordinates gene expression across populations, regulating crucial processes for pathogenesis, including virulence and antibiotic resistance^25,26. Additionally, quorum sensing enhances biofilm formation, allowing P. aeruginosa to persist on surfaces, evade the immune response and resist antimicrobial treatments, which makes these infections particularly challenging to treat²⁷.

Quorum sensing

To adapt to diverse and dynamic environments, P. aeruginosa finely tunes gene expression at the population level through quorum sensing, which is a form of cell–cell communication used by many bacterial species to coordinate behaviours in response to cell density²⁸. A high density of bacterial cells leads to increased concentrations of signal molecule(s) that bind(s) to and activate(s) a cognate receptor. Once activated, the receptor acts as a transcriptional regulator controlling the expression of quorum sensing target genes. Quorum sensing-dependent signal molecules have been termed autoinducers as these molecules often promote the establishment of a self-induction mechanism by which they increase their own synthesis²⁹.

P. aeruginosa possesses a complex quorum sensing network, controlling transcription of approximately 10% of all its genes (Fig. 2a). Many of these genes encode virulence factors such as pyocyanin, hydrogen cyanide (HCN) and proteases, as well as genes encoding proteins critical for biofilm formation and antibiotic susceptibility^30–34. The P. aeruginosa quorum sensing network is composed of two N-acyl-homoserine lactone-dependent systems, las and rhl, and a 2-alkyl-4(1H)-quinolone-dependent system termed here as the PQS (Pseudomonas quinolone signal) system. The las and rhl circuits consist of a LuxR-type receptor, LasR and RhlR, along with a LuxI-type synthase, LasI and RhlI, which produce the signals N-(3-oxododecanoyl)-l-homoserine lactone (3OC₁₂-HSL) and N-butanoyl-homoserine lactone (C₄-HSL), respectively³⁵. The main signalling molecule of the pqs system is 2-heptyl-3-hydroxy-4-quinolone (PQS), which interacts with the LysR-type transcriptional regulator PqsR (also called MvfR) to modulate gene expression. The PQS synthase enzymes, encoded within the pqsABCDE operon, are responsible for generating multiple 2-alkyl-4-quinolones, including 2-heptyl-4-hydroxyquinoline (HHQ), which is converted into PQS by the monooxygenase PqsH. Although not as potent as PQS, HHQ also serves as a low-affinity signal that activates PqsR. PQS also promotes the expression of genes involved in the iron-starvation response and virulence factor production via PqsR-independent pathways³⁴. Another key effector of the PQS system is the PqsE protein, which is required for full virulence in P. aeruginosa via a mechanism probably involving direct interaction with RhlR^33,36,37. Importantly, P. aeruginosa quorum sensing mutants have been demonstrated to be less virulent than the parental strain in various plant and animal infection models^38–40.

Fig. 2 |. The quorum sensing circuitry and biofilm formation of Pseudomonas aeruginosa.

a, Together, the las, rhl and Pseudomonas Quinolone Signal (PQS) quorum sensing (QS) regulatory systems of Pseudomonas aeruginosa control hundreds of genes in response to increasing cell density. The las and rhl systems utilize the N-(3-oxododecanoyl)-l-homoserine lactone (3OC₁₂-HSL) and N-butanoyl-homoserine lactone (C₄-HSL) as autoinducers for LasR and RhlR, respectively; the PQS system uses the 2-heptyl-4-quinolone (HHQ; low-affinity signal) or 2-heptyl-3-hydroxy-4-quinolone (PQS; high-affinity signal) to activate PqsR. RhlR activity is further influenced by association with PqsE. LasR, RsaL, RhlR, PqsR and PqsE are homodimeric proteins. The regulatory systems interact with each other, as shown by the connecting bold arrows in colour. Another regulator, QscR, binds the LasI-generated 3OC₁₂-HSL signal and activates a single linked operon, PA1891–PA1897. The headed arrows represent positive regulation and blunt-headed arrows represent negative regulation. b, Surface-associated biofilm. c, Free-floating biofilm aggregates. P. aeruginosa biofilms grown in the laboratory on solid surfaces form high-density communities that develop elaborate, mushroom-like towers. The extracellular matrix is mainly composed of exopolysaccharides (alginate, Pel and Psl), extracellular DNA (eDNA) and proteins (structural proteins, enzymes and adhesive proteins). Alginate is a negatively charged polymer of β-1,4-linked d-mannuronic acid and l-guluronic acid. Overproduction of alginate is characteristic of mucoid P. aeruginosa isolates in chronic infections, particularly in cystic fibrosis. Psl is a neutral charged, branched pentasaccharide consisting of d-mannose, l-rhamnose and d-glucose, and is the primary exopolysaccharide of the P. aeruginosa PAO1 biofilm matrix. Pel is a positively charged polymer of α-1,4-linked galactosamine and N-acetylgalactosamine. P. aeruginosa PA14 strain exclusively produces Pel. Psl and Pel both contribute to protection against aminoglycoside antibiotics and interact with eDNA to provide structural stability to biofilms. Bacterial cells exhibit heterogeneity in relation to their biogeography. During infection in humans, P. aeruginosa can form biofilm aggregates, with an extracellular matrix primarily composed of bacterial exopolysaccharides and eDNA, along with host polymers such as mucin and host-derived eDNA. Host immune cells surround the biofilm aggregates.

Seminal studies, primarily conducted with P. aeruginosa PAO1 and PA14 strains grown under standard conditions, have shown that the three quorum sensing systems are arranged in a hierarchical organization. Positioned at the top of the regulatory cascade, the las system activates the rhl system^31,41 and the PQS system^42–44, whereas the rhl system negatively affects the expression of the PQS system^42,43. In turn, the PQS system positively modulates the activation of the rhl circuit⁴⁴, ultimately revealing a strong interdependence between the three quorum sensing systems in these model strains. However, further studies with clinical P. aeruginosa strains revealed that this hierarchical organization is not ubiquitous. Indeed, it was discovered that some strains, both inside and outside clinical settings, carry inactivating mutations in the las quorum sensing system, primarily in LasR. In the absence of LasR activity, many of these strains have rewired the quorum sensing circuitry in a manner that results in LasR-independent activation of the rhl system^45–49. Thus, these LasR-defective isolates continue to express genes controlled by the rhl system, including those encoding proteins responsible for producing important virulence factors such as rhamnolipids, HCN, elastase and pyocyanin. These observations have spurred interest in targeting RhlR, rather than LasR, as a potential target for therapeutic development⁵⁰.

Although it is now clear that there is rewiring in many LasR mutants, their emergence, particularly in chronic clinical infections such as cystic fibrosis^{6,39,45,51–54}, remains poorly understood. Experimental evolution studies have tried to understand how lasR mutants evolve, while ecological and evolutionary considerations have been made^46,49. Several hypotheses as to why P. aeruginosa would lose lasR have been investigated^52,55,56. A popular explanation is that lasR mutants arise as social cheaters that exploit quorum sensing-regulated extracellular public goods produced by quorum sensing-cooperating cells, without investing in the costly production of these factors. This exploitation allows lasR mutants to increase their own fitness within the population. This behaviour has been supported both under specific in vitro conditions^56–58 and in vivo in murine models of acute and chronic infection⁵⁹, but has never been demonstrated in human infection. Another hypothesis is that lasR mutations confer adaptive growth advantages in specific conditions and environmental settings. Increased fitness of lasR mutants has been observed in the presence of specific amino acids, low oxygen levels, high nitrogen levels, alkaline stress and antibiotics, all conditions characteristic of cystic fibrosis infections^60–65. Nonetheless, these conditions could probably reflect other environmental settings, as lasR mutants have been found in diverse infection and environmental contexts apart from the cystic fibrosis lung⁵¹. It is also possible that quorum sensing may impose a fitness cost in certain environments, such as during infection, leading to the loss or rewiring of the quorum sensing hierarchy.

P. aeruginosa has an additional orphan N-acyl-homoserine lactone-responsive transcription factor, called QscR, which lacks a paired signal synthase. QscR mutants have demonstrated to be hyper-virulent in a Drosophila infection model⁶⁶. Interestingly, QscR acts as a quorum sensing anti-activator, delaying the activation of the las system through an unknown mechanism, which seems to be mediated by the single set of genes regulated by QscR⁶⁷. P. aeruginosa has two additional anti-activator proteins, QteE and QslA, that, unlike QscR, are not homologous to LasR and RhlR. These prevent premature activation of quorum sensing by sequestering LasR, therefore delaying the activation of quorum sensing-regulated genes. Recently, these anti-activators were demonstrated to prevent self-sensing in P. aeruginosa quorum sensing⁶⁸.

From a single-cell level perspective, P. aeruginosa appears to have a graded quorum sensing response to changes in population density, with substantial heterogeneity in quorum sensing activation even at high-cell densities^69,70. This cell-to-cell variation in the quorum sensing activation state has been attributed to the RsaL-driven negative regulation of the lasI transcription⁷⁰. RsaL is a transcriptional regulator that represses both the lasI gene and its own expression, establishing a negative feedback loop that limits the accumulation of 3OC₁₂-HSL beyond physiologically beneficial levels⁷¹. From an ecological and evolutionary perspective, phenotypic heterogeneity represents an adaptable trait that can confer advantages through specialization and division of labour or serve as a bet hedging strategy, allowing populations to persist in dynamic environments⁷². Quorum sensing in P. aeruginosa also plays a crucial role in biofilm formation by regulating the expression of genes involved in surface attachment, extracellular matrix (ECM) production and antibiotic resistance, enabling the bacteria to coordinate their behaviour and establish resilient, structured communities^25–27.

Surface-associated biofilms and free-floating biofilm aggregates

Besides growing as planktonic free-swimming cells, P. aeruginosa grows as biofilms attached to biotic and abiotic surfaces (Fig. 2b) and as free-floating biofilm aggregates (Fig. 2c). Biofilms are communities of microorganisms embedded in an ECM, which can be self-produced and/or recruited from the environment. P. aeruginosa growth in biofilms increases its tolerance to antimicrobials (reviewed in ref. 73) (Fig. 3b), and biofilms are particularly problematic in chronic infections, including those in cystic fibrosis lungs, chronic wound infections and on medical devices such as catheters. Moreover, the cells within a biofilm are phenotypically diverse⁷⁴, and this heterogeneity has been proposed to promote antibiotic tolerance through slow growth rates and unstable heteroresistance (Figs. 3b,c).

Fig. 3 |. Antimicrobial resistance, antimicrobial tolerance and heteroresistance in Pseudomonas aeruginosa.

a, The intrinsic, acquired and adaptive mechanisms of antimicrobial resistance in Pseudomonas aeruginosa are illustrated, along with examples of associated genes and proteins. Intrinsic mechanisms include low outer membrane permeability, AmpC β-lactamase production, and intrinsic or induced expression of efflux pumps (particularly MexAB–OprM and MexXY–OprM). Acquired mechanisms comprise the acquisition of antibiotic resistance genes through horizontal gene transfer or mutations leading to efflux pumps overexpression, reduced antibiotic uptake resulting from decreased levels of OprD porin, modifications in the lipopolysaccharide (LPS) (for example, modification of lipid A and modification or loss of O antigen side chains), hyperproduction of β-lactamases, and alterations to antibiotic targets (DNA gyrase, topoisomerase IV and ribosome). Adaptive mechanisms involve a reversible, temporary increase in resistance triggered by specific environmental signals or antibiotic pressure. b, The antimicrobial tolerance conferred by P. aeruginosa biofilm formation involves a combination of physical, physiological and genetic determinants. The physical and chemical interaction of antibiotics with components of the biofilm matrix, such as exopolysaccharides and extracellular DNA, limit their penetration. Antibiotic efficacy can be altered by several factors, including nutrients and oxygen gradients that create a heterogeneous bacterial population of slow-growing or non-dividing cells, a low metabolic activity that leads to the inactivation of the major antibiotic targets, and adaptive responses to stress — nutrient and iron starvation, oxygen limitation or oxidative stress (SOS response, stringent response) — that lead to reduced bacterial growth rates, making biofilms more tolerant to antibiotics. Oxygen molecules are required for the formation of reactive oxygen species (ROS), which are important for the bactericidal effect of antibiotics. The expression of specific genes such as the biofilm resistance locus regulator gene (brlR) and efflux pumps genes, together with high levels of the secondary messenger bis-(3′-5′)-cyclic dimeric guanosine monophosphate (c-di-GMP), play a role in antibiotic tolerance mediated by P. aeruginosa biofilms. c, Examples of the main mechanisms of heteroresistance in P. aeruginosa. Heteroresistance is difficult to detect and highly unstable; the surviving subpopulation often converts to a susceptible one after the removal of the antibiotic.

The ECM, composed of both self-produced and environmental components, is primarily made up of exopolysaccharides (EPS), extracellular DNA (eDNA) and proteins. The ECM can promote attachment to both abiotic and biotic surfaces, attachment between P. aeruginosa cells, and protection from harsh environmental conditions, including the host immune response. P. aeruginosa produces at least three EPS, namely alginate, Psl and Pel, which can be cell associated or released. Generally, individual strains predominantly produce one of these EPS, with each playing a distinct role in biofilm development, maintaining cell–cell interactions and providing different physiological properties to the biofilm matrix^75,76. Alginate is a negatively charged polymer of β-1,4-linked mannuronic acid and guluronic acid⁷⁵, and its overproduction is a hallmark of mucoid P. aeruginosa strains isolated from infections in lungs with chronic cystic fibrosis. The mucoid phenotype in vivo is frequently associated with mutations in the anti-sigma factor-encoding mucA gene⁷⁷. Biofilms composed of cells overproducing alginate are highly structured and exhibit increased resistance to tobramycin⁷⁸. Alginate is not an essential component of the matrix in biofilms produced by non-mucoid strains, where Pel and/or Psl serve as the primary structural scaffold. In P. aeruginosa PAO1, the matrix is mainly composed of the EPS Psl, whereas Pel is produced at low levels, and deletion of pel genes does not significantly affect biofilm development^76,79. On the other hand, the PA14 strain exclusively produces Pel⁷⁶. Psl is a neutral, branched pentasaccharide consisting of glucose, mannose and rhamnose, whereas Pel is a cationic, partially de-N-acetylated linear polymer of α-1,4-N-acetylgalactosamine comprised predominantly of dimeric repeats of galactosamine and N-acetylgalactosamine⁸⁰. Both Psl and Pel confer protection against aminoglycoside antibiotics^79,81 and have been shown to interact with eDNA, providing structural stability to biofilms^82,83.

Along with EPS, the ECM also contains eDNA and proteins. eDNA, released by cellular lysis^84,85, promotes biofilm formation, stability and tolerance towards positively charged antibiotics such as aminoglycosides and antimicrobial peptides. The ECM of P. aeruginosa biofilms contains structural proteins, enzymes and adhesive proteins^86,87. Among the most studied are lectins (for example, LecA⁸⁸ and LecB⁸⁹) and adhesins (for example, CdrA⁹⁰), which contribute to biofilm stability and mediate cell–cell, cell–surface and cell–host interactions.

The secondary messenger bis-(3′-5′)-cyclic dimeric guanosine monophosphate (c-di-GMP) also plays a pivotal role in the formation and regulation of P. aeruginosa biofilms. In the strain PAO1, over 40 putative diguanylate cyclases (DGCs) and phosphodiesterases (PDEs) associated with c-di-GMP turnover have been identified, several of which have been functionally characterized. Environmental cues can modulate biofilm formation by regulating the activity of DGCs and PDEs^74,91,92. These cues act directly on DGCs or PDEs through their sensory domains (such as PAS, GAF or HAMP domains) or indirectly via signal transduction pathways, such as two-component systems. Changes in c-di-GMP levels directly affect its binding to effector proteins, driving phenotypes critical to biofilm dynamics, including biofilm formation, motility reduction, enhanced antibiotic resistance and increased virulence⁹¹.

Biofilms vary in size and shape, as well as their matrix components and spatial organization. They can exist either attached to a surface or as suspended aggregates, and often contain functionally heterogeneous subpopulations of cells. During human infection, P. aeruginosa often grow in biofilms, and we now recognize there are both spatial and functional differences between in vitro biofilms and in vivo biofilms^93,94. P. aeruginosa biofilms grown in the laboratory on solid surfaces such as plastic or glass result in high-cell-density communities that form either flat structures or elaborate mushroom-like towers⁹⁵. The inactivation of large numbers of genes has been shown to impact the structure and function of these in vitro biofilms, leading to the proposal of a developmental model for biofilm formation. The clinical relevance of these in vitro biofilm studies remains unclear, although there is reasonable evidence suggesting that they offer valuable insights into P. aeruginosa biofilms found on the surface of medical devices, such as catheters and implants^96,97. Biofilm formation in vitro can also be impacted by non-biological processes such as depletion aggregation^98–100. Although the importance of this aggregation mechanism in vivo is unknown, biofilm aggregation through this process can enhance antibiotic tolerance and is impacted by the chemical composition of the outer surface of P. aeruginosa^98–100.

Although biofilm formation is well studied in vitro, this process is less understood during human infection. Imaging studies in human-derived samples, including chronic and acute lung infections, chronic wounds and otitis media, revealed bacteria present as biofilm aggregates that are not attached to surfaces or epithelia^101–104. In human infections, aggregates vary in size, with more than 50% of the bacterial biomass present in aggregates smaller than 100 μm³ (ref. 105), in both acute and chronic human lung infections. The mechanisms producing aggregation in vivo remain unclear, although it is unlikely that the depletion aggregation mechanism plays a critical role, as the specific aggregate structures created by this mechanism have not, so far, been observed in vivo^105,106. Instead, a bridging aggregation mechanism has been proposed, where the biofilm aggregate matrix extends beyond the cell to promote aggregation¹⁰⁶. An alternative scenario suggests that P. aeruginosa actively contracts the mucus surface using retractile filaments called type IV pili (T4P), drawing aggregate cells closer¹⁰⁷. Host polymers such as mucin and eDNA also appear to play a role in biofilm aggregate formation in vivo, speculated to be due to association with the biofilm EPS and/or receptors on the bacterial surface. Finally, biofilm aggregates in human chronic infections are surrounded by host immune cells¹⁰⁶, which, along with the lack of planktonic cell biomass in these infections, lead to the hypothesis that biofilm aggregates are critical for surviving the robust immune response observed in many human infections⁷⁴. Future studies investigating the mechanisms in vivo are essential and will require preclinical models that accurately replicate human infection biology¹⁰⁸.

Virulence factors and in vivo fitness determinants

P. aeruginosa can infect and cause serious diseases in a range of multicellular organisms, from nematodes to humans. Its virulence is driven by a complex array of factors that enable it to infect, persist and cause damage to host tissues. As traditional P. aeruginosa virulence factors have been reviewed in depth recently^25,109, here we will briefly outline some of the key virulence factors relevant to host infection (Fig. 4), then discuss the metabolic pathways important for in vivo fitness.

Fig. 4 |. Virulence factors of Pseudomonas aeruginosa.

Pseudomonas aeruginosa employs a diverse array of virulence factors for establish infection and promote pathogenesis. The type III secretion system (T3SS) injects toxins such as ExoU, ExoT, ExoS and ExoY directly into host cells, which impair cellular homeostasis, cellular signalling pathways, cause membrane disruption and induce a pro-inflammatory response. Exotoxin A (ETA), a potent toxin secreted by the type II secretion system (T2SS), inhibits protein synthesis, leading to cell death. Additional virulence factors include proteases (AprA, PrpL, LasA and LasB) and rhamnolipids, which contribute to tissue invasion and immune evasion. Other factors, such as pyoverdine, hydrogen cyanide (HCN) and pyocyanin, promote iron chelation, oxidative stress, mitochondrial dysfunction and apoptosis. AMP, adenosine monophosphate; EF2, elongation factor 2; PMN, polymorphonuclear leukocyte; ROS, reactive oxygen species.

Traditional virulence factors

P. aeruginosa employs several secretion systems, including the type III secretion system (T3SS), which injects toxins directly into host cells^110,111. These toxins, including ExoU, ExoT, ExoS and ExoY, have distinct functions that disrupt normal cellular processes¹¹⁰. ExoU is a potent phospholipase that causes rapid cell death through membrane disruption, leading to tissue damage and severe inflammatory responses. ExoT and ExoS are bifunctional proteins with GTPase-activating and ADP-ribosyltransferase activities. These enzymes interfere with the host’s cytoskeleton, inhibit phagocytosis and disrupt cellular signalling pathways, thereby preventing the immune system from effectively clearing the infection. ExoY, an adenylate cyclase, disrupts cellular homeostasis by increasing cyclic adenosine monophosphate (cyclic-AMP) levels, leading to impaired barrier function and vascular leakage. The expression of the T3SS is tightly regulated by environmental signals^110,111, allowing P. aeruginosa to deploy this system when in contact with host cells. Once activated, the T3SS facilitates rapid bacterial invasion and colonization, contributing to the establishment of infection and the progression of disease.

Exotoxin A (ETA) is encoded by the toxA gene and is one of the most potent virulence factors produced by P. aeruginosa^112,113. ETA is secreted into the extracellular environment via the general secretory pathway (type II secretion system) where it can interact with and enter host cells. Once inside cells, ETA is an ADP-ribosylating enzyme that targets the elongation factor 2, a critical component of the protein synthesis machinery in eukaryotic cells^114,115. Thus, ETA functions by inhibiting protein synthesis in host cells, leading to cell death and contributing to the bacterium’s ability to cause severe infections. ETA is particularly associated with tissue damage and necrosis, making it a major factor in the severity of acute infections. The expression of ETA is regulated by environmental factors and is typically upregulated during infection, particularly in the iron-depleted environments of infected tissues¹¹⁶. Owing to its potent effects and significant role in disease progression, ETA is a major target for therapeutic interventions and vaccine development in the fight against P. aeruginosa infections¹¹⁷.

P. aeruginosa is highly proteolytic, producing extracellular proteases such as LasB (elastase), LasA (staphylolysin), alkaline protease (AprA) and protease IV (PrpL) that degrade host proteins and promote tissue damage¹¹⁸. LasB degrades host ECM proteins, such as elastin and collagen, facilitating tissue invasion and bacterial dissemination. It also targets immune components, such as complement proteins and immunoglobulins, undermining host defences. LasA exhibits a limited elastolytic activity per se, but enhances the elastolytic activity of other proteases, including LasB and host elastolytic proteases, and is better known as staphylolysin, due to its ability to cleave the pentaglycine bonds in Staphylococcus aureus peptidoglycan. Alkaline protease contributes to tissue damage by degrading important components of the basal lamina and endothelium, such as fibronectin, and aids in immune evasion by disrupting host complement proteins and inflammatory responses. PrpL contributes to the degradation of the coagulation factor fibrinogen, participating in tissue invasion and vascular damaging processes during P. aeruginosa infection¹¹⁸.

P. aeruginosa produces an array of additional cytotoxic secreted factors that can impact disease progression, including rhamnolipids, phenazines and HCN. Rhamnolipids are biosurfactants that solubilize and promote the uptake of hydrophobic substrates, disrupt the chemotactic responses of polymorphonuclear leukocytes, and inhibit the mucociliary transport and ciliary function of the human respiratory epithelium, contributing to tissue invasion and immune evasion. These molecules also play a key role in biofilm formation and motility, and display antibacterial activity, enhancing the bacterium’s ability to colonize and persist in various environments¹¹⁹. Phenazines are highly diffusible toxic secondary metabolites, of which pyocyanin is the most well studied. Beyond its role in giving the bacterium its distinct blueish–green colour, pyocyanin is a redox-active compound that generates reactive oxygen species, leading to oxidative stress in host cells¹²⁰. Pyocyanin has been shown to suppress ciliary function in respiratory epithelial cells, and to affect host immune responses, interfering with immune signalling pathways, disrupting macrophage activity and inducing neutrophil apoptosis^120,121. HCN is a volatile cyanogenic compound produced by P. aeruginosa through the action of the hcnABC-encoded hydrogen cyanide synthase. HCN is highly toxic to host cells, primarily due to its ability to inhibit cytochrome c oxidase in the mitochondrial electron transport chain^122,123. This inhibition disrupts cellular respiration, leading to energy depletion and cell death.

Nutritional adaptation and metabolic requirements for virulence

The successful colonization and persistence of P. aeruginosa in diverse host environments depend on its ability to acquire essential nutrients despite host-imposed nutritional immunity. Nutritional immunity is a host innate immune defence strategy in which essential nutrients (such as iron, zinc and manganese) are sequestered by the host, thereby limiting the ability of pathogens to grow and cause infection. P. aeruginosa possesses an array of mechanisms for acquiring host sequestered metals, with the most elaborate and well studied being iron acquisition. Although iron is an essential cofactor for numerous P. aeruginosa enzymes, free iron is scarce in the host environment due to sequestration by host proteins such as transferrin and lactoferrin. Aside from scavenging iron directly from host haem-containing proteins, P. aeruginosa also synthesizes siderophores. The two siderophores, pyochelin and pyoverdine, are characterized by low and high affinities for iron, respectively, and enable P. aeruginosa to thrive in iron-limited environments, such as those encountered during infection¹²⁴. Beyond its role in iron acquisition, pyoverdine contributes to P. aeruginosa virulence by inducing oxidative stress in host tissues and regulating the production of other virulence factors, including exotoxin A, the endoprotease PrpL and additional pyoverdine. This regulation occurs via a transmembrane signalling cascade that ultimately activates the alternative sigma factor PvdS, which controls the expression of these genes¹²⁵.

Beyond metal acquisition, results from transposon sequencing (Tn-seq) studies have identified key metabolic pathways that are critical for the growth and virulence of P. aeruginosa across different infection sites.

Tn-seq studies in murine models of chronic and acute wounds revealed that the synthesis of several nutrients, including purines, branch chain amino acids and vitamins (for example, riboflavin and p-aminobenzoate), are critical for P. aeruginosa virulence in these infection models⁶. Although P. aeruginosa can utilize a wide range of molecules as sources of carbon and energy, this study also revealed that the catabolism of long-chain fatty acids is crucial for P. aeruginosa fitness in murine wound models, with the inactivation of the fatty acid oxidation genes faoAB significantly reducing fitness and virulence in these models⁶. This suggests that long-chain fatty acids (C12 or greater) are important sources of carbon and/or energy during both chronic and acute murine wound infection.

Ex vivo Tn-seq studies have provided insights into functions crucial for P. aeruginosa fitness in human blood and secretions. A Tn-seq screen conducted in human plasma examined how P. aeruginosa evades the complement system¹²⁶. This study employed a gain-of-function approach in a clinical bloodstream isolate to identify mutants that were able to persist in plasma, surviving complement attack. They identified a novel three-gene operon (srgABC) that enhances serum resistance. In particular, the overexpression of srgA, encoding a small periplasmic protein, increased survival in plasma by up to 100-fold, making it a potent complement-evasion factor. Additionally, mutants with insertions in purine and biotin biosynthesis genes showed higher tolerance or persistence in plasma. This suggests that metabolic down-shifts (for example, purine auxotrophy or biotin limitation) might protect bacteria from complement, possibly by slowing growth or altering surface targets.

Ex vivo growth of P. aeruginosa Tn-seq libraries in mucus (sputum) from the cystic fibrosis lung revealed that no carbon catabolic pathways were critical for growth in this secretion, suggesting the existence of a number of carbon and energy sources in the lung with cystic fibrosis¹²⁷. However, this study revealed that although sputum from patients with cystic fibrosis is an environment rich in carbon and amino acids, it is relatively poor in enzyme cofactors. As in murine wound infections, the biosynthesis of diaminopimelate, chorismate, purines, pantothenate, pyridoxal phosphate, polyamines and riboflavin is critical for P. aeruginosa fitness during ex vivo growth in sputum from individuals with cystic fibrosis. These data suggest that metabolic inhibitors developed to target P. aeruginosa cofactor biosynthesis, rather than carbon catabolic pathways, may be effective in treating P. aeruginosa chronic infections.

Epidemiology and antimicrobial resistance

P. aeruginosa is a resilient pathogen with a widespread presence in hospital environments² (Fig. 1c), particularly in intensive care units, where it is a significant cause of nosocomial infections¹²⁸. Its presence in health-care settings is often linked to contaminated moist environments and medical equipment, such as humidifiers, sinks, shower heads, ventilators and catheters², which serve as potential sources of infection^129–131. Cross-infection is frequently documented, with genotypically identical strains isolated from multiple patients^132,133. Cross-transmission can be mediated by hospital personnel^129,134, and possibly by arthropods such as cockroaches¹³⁵. Moreover, P. aeruginosa infections in hospitals may also originate from patients who are pre-colonized with the bacterium, as intestinal carriage can lead to colonization of other body sites, especially in immunocompromised patients, potentially disseminating infection to others¹³⁶.

Clinical and environmental isolates appear to be indistinguishable, with no strong correlation between specific clones and habitats. However, hospital transmission can increase the prevalence of adapted clones, many of which have high resistance to clinically relevant antibiotics. P. aeruginosa exhibits a non-clonal epidemic population structure, comprising a limited number of widely distributed clones that arise through selection from a background of relatively rare and unrelated genotypes, which often undergo recombination^137,138. Notable sequence types (STs), classified based on multilocus sequence typing¹³⁸, include widespread clones such as clone C (ST17) and PA14 (ST253), alongside epidemic clones associated with cystic fibrosis, such as the Liverpool Epidemic Strain (LES, ST146)^139–143. Certain STs are globally disseminated and are frequently linked to outbreaks and have multidrug-resistant, extensively drug-resistant and pandrug-resistant phenotypes¹². ‘High-risk’ clones¹³, such as ST235, ST111 and ST175 (refs. 144–146), are particularly concerning owing to their worldwide prevalence and the dissemination of resistance genes, including carbapenemases.

The antimicrobial resistance mechanisms of P. aeruginosa are multifaceted, encompassing intrinsic, adaptive and acquired forms^139,147,148 (Fig. 3a). Intrinsic resistance in most strains arises from low outer membrane permeability, the production of enzymes that inactivate antimicrobials such as the AmpC β-lactamase, and antimicrobial efflux pumps^139,147,149. Adaptive resistance is mediated by defined changes in the gene expression pattern of P. aeruginosa occurring in response to environmental cues or the presence of antibiotics. Acquired resistance is gained through horizontal gene transfer (HGT) or mutations that lead to enhanced efflux pump activity¹⁴⁹, reduced antibiotic uptake, hyperproduction of β-lactamases and alterations to antibiotic targets^139,147. Although mutations and HGT are considered the dominant drivers of resistance emergence within the host, recent investigations provided new insights into how antimicrobial resistance arises in individuals with P. aeruginosa infections¹⁵⁰, also considering that multiple strains of P. aeruginosa can colonize the host at the same time^148,150,151. Interestingly, research has shown that although de novo mutations occur at similar rates in both single-strain and mixed-strain populations, the incidence of acquiring new resistance genes through HGT is relatively low. In diverse P. aeruginosa populations, resistance evolves rapidly, primarily driven by natural selection for pre-existing resistant strains, emphasizing the link between resistance and pathogen diversity¹⁵⁰. Furthermore, within mixed-strain populations, there are notable trade-offs between resistance and growth rate, suggesting that diversity can also lead to a loss of resistance when antibiotics are not present^150,152. Metabolic and biophysical fitness trade-offs have also been investigated as P. aeruginosa forms biofilms under the pressures of mucosal growth and antibiotic treatment. These trade-offs probably promote phenotypic heterogeneity and flexibility, with antibiotic-resistant biofilms shielding less tolerant but more cytotoxic cells, enabling P. aeruginosa to persist through different stages of infection¹⁵³.

In addition to conventional resistance mechanisms such as those observed in the ‘high-risk’ sequence types, P. aeruginosa can also display heteroresistance. Heteroresistance is a phenotype in which a bacterial strain contains subpopulations of cells with increased antibiotic resistance when compared with the susceptible main population¹⁵⁴ (Fig. 3c). This subpopulation, often undetected by standard susceptibility testing methods, can survive antibiotic treatment, leading to treatment failure and chronic infections. Heteroresistance is highly unstable, and the surviving subpopulation often converts to susceptible after the removal of the antibiotic. The mechanisms controlling unstable heteroresistance in P. aeruginosa are varied and include overexpression of efflux pumps^155–157, decreased levels of the OprD porin^155,157 and increased expression of genes involved in DNA replication and repair¹⁵⁸, among others. Heteroresistance is particularly concerning in P. aeruginosa owing to the large number of intrinsic resistance mechanisms and ability to adapt to chronic infection environments, such as within the lung with cystic fibrosis. The existence of heteroresistant subpopulations complicates clinical management, requiring more precise diagnostic methods and tailored treatment strategies to effectively eradicate the infection and prevent resistance development.

Treatment of P. aeruginosa infections

The treatment of P. aeruginosa infections can be extremely challenging due to its intrinsic resistance and in vivo tolerance to many antibiotics, unstable heteroresistance and its ability to acquire additional resistance mechanisms. The treatment approach depends on the type and severity of the infection, as well as the specific resistance profile of the strain. Antibiotic therapy remains the primary treatment strategy. Broad-spectrum antibiotics such as β-lactams (piperacillin–tazobactam, ceftazidime and cefepime) and carbapenems (imipenem and meropenem) are routinely used. Aminoglycosides such as gentamicin and tobramycin can also be effective and often used in combination with other antibiotics for enhanced efficacy. Fluoroquinolones (ciprofloxacin and levofloxacin) and polymyxins (colistin) are also part of the treatment toolkit, with colistin often used as a last-resort treatment for multidrug-resistant strains. Cefiderocol is a relatively new antibiotic that is effective against multidrug-resistant P. aeruginosa strains¹⁵⁹. It is a siderophore cephalosporin that utilizes iron uptake mechanisms to enter host cells. Surgical intervention for infections involving abscesses, prosthetic devices, severe tissue damage, surgical debridement or removal of infected devices may be necessary in addition to antibiotic therapy.

As antibiotic resistance continues to rise, the search for innovative alternatives to traditional treatments has become increasingly urgent. Among the most promising solutions is bacteriophage (phage) therapy^160,161. Unlike antibiotics, phages are living viruses that evolve alongside bacteria, allowing them to adapt and remain effective even in the face of bacterial resistance. Their precision is a key advantage, as they can specifically target P. aeruginosa strains without harming beneficial microbiota. However, this specificity comes with challenges. Each infection often requires a personalized selection of phages, making large-scale application complex. Additionally, although phages can, at least in theory, outpace bacterial resistance, their success depends on rapid identification of effective phages and navigating regulatory hurdles. Antimicrobial peptides are another well-studied alternative. These small molecules, such as polymyxins, attack bacterial membranes and are effective against multidrug-resistant strains, including those protected by biofilms. However, their therapeutic potential is tempered by risks of toxicity to human cells and rapid degradation in the body, necessitating further refinement in their design and delivery.

Biofilms pose a critical challenge to most treatments. Biofilm-disrupting agents such as DNase enzymes aim to break down these barriers, allowing antibiotics to penetrate¹⁶². Although this approach shows promise, biofilms are complex structures that often require combination therapies for complete eradication. Another strategy involves depriving P. aeruginosa of essential nutrients, such as iron. Iron-sequestering therapies such as gallium nitrate mimic iron to disrupt bacterial metabolism, effectively starving the pathogen¹⁶³. However, this approach risks unintended effects on host cells, raising safety concerns. Other strategies, such as monoclonal antibodies, have been developed to neutralize bacterial toxins such as exotoxin A¹¹⁷. Although highly specific and immune boosting, these therapies face challenges related to high costs and potential immune escape. Emerging technologies have introduced nanotechnology-based drug delivery, which targets pathogens more precisely, and quorum sensing inhibitors¹⁶⁴, which disrupt bacterial communication to help prevent things such as biofilm formation. Finally, pyocins are protein-based bacteriocins produced naturally by P. aeruginosa¹⁶⁵. These highly specific agents can target rival strains without damaging the microbiota. However, scaling up their production and precisely matching them to bacterial strains remains a challenge. Despite the obstacles, these alternative therapies offer hope for combating P. aeruginosa multidrug-resistant infections.

Conclusions and future questions

P. aeruginosa exemplifies bacterial adaptability and resilience, demonstrating a remarkable capacity to thrive across a spectrum of environments — from natural ecosystems to complex clinical settings. This versatility is fundamentally driven by the bacterium’s extensive genomic variability, metabolic flexibility and sophisticated array of virulence factors, which collectively enable it to establish both acute and chronic infections in humans. One of the key aspects of the pathogenicity of P. aeruginosa is its ability to form biofilms and engage in quorum sensing, which enhances its virulence and complicates treatment efforts. These characteristics contribute to its prominence as a leading nosocomial pathogen, often presenting challenges in infection control and management.

Despite substantial progress in understanding the biology and pathogenic mechanisms of P. aeruginosa, several critical questions remain unanswered. For instance, more research is needed to elucidate the epidemiological mechanisms that facilitate the spread and fitness of the most prevalent, antibiotic-resistant and virulent STs in clinical settings. Understanding the factors that drive the success of these strains, including genetic determinants, transmission dynamics and environmental influences, is crucial for developing targeted interventions and improving infection control measures. Although extensive research has been conducted using laboratory strains of P. aeruginosa, there is a pressing need to bridge the gap between these model systems and the behaviour of the bacterium in natural and clinical environments. Integrating our current knowledge with real-world scenarios will enhance our understanding of P. aeruginosa biology and inform strategies for prevention and treatment.

The role of quorum sensing circuits in both acute and chronic human infections requires deeper investigation. Detailed studies on how each quorum sensing circuit contributes to pathogenicity, including biofilm formation, virulence factor production and antibiotic resistance, will be critical in identifying potential therapeutic targets. Exploring ways to disrupt quorum sensing signalling could offer novel approaches for managing P. aeruginosa infections. In relation to biofilm formation, further exploration is needed to understand its contribution to antibiotic tolerance in both acute and chronic infections. Uncovering the mechanisms by which biofilms confer resistance and assessing the feasibility of biofilm dispersal as an antimicrobial strategy could provide new avenues for treatment.

Last, further investigation into single-cell heterogeneity in P. aeruginosa virulence and antibiotic tolerance is needed. The variability among individual cells within a population could influence the overall pathogenic potential and resistance profile of the bacterium. Understanding this heterogeneity could lead to more effective treatment strategies and enhance our ability to combat infections.

Overcoming these challenges requires a multifaceted approach, including the development of combination therapies that target biofilms, quorum sensing and antibiotic resistance mechanisms. Improved infection control, rapid diagnostics and personalized medicine will help mitigate transmission and resistance, optimizing patient outcomes.

Glossary

Bet hedging

Stochastic switching between phenotypic states to enhance population long-term fitness in fluctuating environmental conditions

Biogeography

The spatial assembly and distribution of various organisms in an environment through time

Bridging aggregation

A mechanism of bacterial aggregation driven by electrostatic interactions between bacterial cell surfaces and polymers present in the environment

Depletion aggregation

A process in which the reduction of free energy through increased entropy of the whole system induces the stacked aggregation of bacterial cells in polymer-rich environments

Division of labour

Cooperating individuals specialize in carrying out specific tasks, providing an inclusive fitness benefit to all individuals involved

Self-sensing

Cell-autonomous and density-independent reception of signals produced by the same cell

Sociomicrobiology

Studies on the group behaviours of microorganisms