P. aeruginosa interactions with other microbes in biofilms during co-infection

1.   The concept of biofilm

In nature, microorganisms can appear in either a planktonic or sessile state. In the planktonic state, bacteria move freely, whereas in the sessile state, bacteria are frequently attached in multicellular aggregates, forming biofilms [1].

The first simple model of biofilm infection was proposed by Costerton and Stewart in 1999 [2], in which polymicrobial communities of bacteria are able to produce virulence factors leading to infection. The model puts an innovative bacterial strategy into evidence and allows the understanding of chronic infections at the biochemical and cellular levels, in which antibodies [3] and white blood cells [4] are ineffective in combating biofilms. Nowadays, the models are more comprehensive; however, a broader knowledge of how polymicrobial infections by biofilms progress will allow for the elaboration of more realistic models [5].

Biofilm formation is a multistep process whereby heterogeneous communities of microorganisms are embedded into a self-produced hydrated matrix. This matrix is formed by an extracellular polymeric substance often consisting of polysaccharides, proteins, glycoproteins, and extracellular DNA, conferring the capacity of microorganisms to adhere to either biotic or abiotic surfaces [6],[7]. A vast number of human polymicrobial diseases are spread through abiotic surfaces, such as intravenous and urinary catheters, ventilator tubes, pacemakers, and orthopedic devices [8]–[10], which can be colonized by potentially pathogenic microorganisms [11]. Bacteria are the main type of microorganisms involved in biofilm formation. However, many filamentous fungi and yeasts have been associated with biofilm formation, and Candida albicans, a commensal mucosal organism, remains the most widely studied biofilm-producing fungus. A few viruses, like the influenza A virus, and respiratory syncytial virus (RSV), have also been found in biofilm communities [12].

Biofilms are polymicrobial communities, organized in space. Traditionally, biofilms are considered to be formed by hundreds of thousands of cells encased in a matrix and attached to a surface; however, they can also be formed by dozens of cells simply arranged in small aggregates [1].

Polymicrobial biofilms are prevalent throughout the human body, both in health and disease status, and their exoproducts can help bacteria aggregate and therefore reduce the efficacy of high-dosage antimicrobial therapy [13],[14]. Broad-spectrum antibiotics are generally used to treat polymicrobial infections, however, the recovery rate of affected individuals is low [15],[16]. Cells within the biofilm may belong to different bacterial species but are functionally equivalent pathogroups [17], developing quorum sensing systems through which they are able to communicate [18], and controlling bacterial pathogenicity and mutual growth in a specific environment [11],[19].

The interactions between microorganisms within biofilms profoundly affect disease severity and progression [20]. Although some infections require colonization by multiple interacting microorganisms, as in the case of colonization of the oral cavity by Porphyromonas gingivalis and commensal oral microorganisms causing periodontal disease [21], other infections are more or less severe according to the presence of specific co-infecting species, such as Staphylococcus aureus and Pseudomonas aeruginosa in chronic wound infections and cystic fibrosis (CF) [22]. The timing of polymicrobial interactions can also vary. Co-infecting pathogens can appear stepwise, with one species followed by another, or they can appear concurrently in the host [23]. Occasionally, commensal strains have a more prominent role in the disease progression than a well-known pathogen, that can participate only as a minor player, as observed for Streptococcus mutans in dental caries [24].

2.   Interactions and spatial arrangement in polymicrobial biofilms

Bacteria within polymicrobial biofilms can benefit from each other and facilitate cohabitation on epithelial surfaces through the efficient use of metabolic by-products, and the production and secretion of beneficial molecules [25]. Beyond cooperating, they can also compete with each other for resources and space by producing toxins or other extracellular metabolites [26]. The mechanisms that allow these bacteria to interact in vivo, including cooperative and antagonistic behaviors between biofilm-specific communities, are not fully understood, but are known to be strongly influenced by the host and its immune system, and by chemical and physical interactions, including cell-cell communication via the bacteria's quorum-sensing cross talk [27]. The events associated with the establishment of these relationships include contact-dependent attachments, an increase in colonization, immunomodulation, and expression of virulence potential [11],[28].

Polymicrobial infections can be synergistic if the interactions between microbial strains or species result in an outcome that is greater than the sum of the one from individual microorganisms. Therefore, synergic infections are associated with more severe consequences for the host [29].

An example of a polymicrobial synergistic environment is the oral cavity, since oral microbes generally only exist there, and so the interactions between them have positively evolved [30]. On the other hand, wounds are polymicrobial synergistic environments where microbes, such as P. aeruginosa and S. aureus, have evolved in different habitats interacting over time with other microbes. P. aeruginosa main habitats are soil and water, while for S. aureus is the respiratory tract and the skin of humans and animals [22]. However, over time, P. aeruginosa started to colonize the same habitats as S. aureus [31]. The interaction between these bacterial species is sometimes cooperative but mostly antagonistic, and the outcome of their interactions is synergistic, leading to chronic infections resistant to antimicrobial therapy [32].

It is known that microorganisms are not accidentally distributed throughout the body; for example, the endemic Helicobacter pylori only resides in the stomach. Additionally, the same body sites of different individuals are usually more similar in microbial composition than those from different sites of the same individual [33]. These examples are studied by biogeography, which uncovers the distribution of species through space and time, including the spatial organization of microbial communities in biofilms, that depends on different ecological and evolutionary forces [23],[34],[35].

Microbial community organization can be classified as either spatially mixed or segregated, with the former being associated with enhanced interactions between microbes and the latter with attenuated interactions. Both types of spatial organization and their associated polymicrobial relations can influence virulence and infection, and according to Stacy et al. [23], they are divided into four categories: two types of cooperative interactions, 1. physical with a mixed organization, and 2. chemical with a mixed organization; and two types of antagonist interactions, 3. chemical with a segregated interaction, and 4. physical with a segregated interaction (Figure 1).

One example of physical interaction resulting in a mixed organization is co-aggregation [23]. Co-aggregation is prevalent in the oral cavity, where intercellular binding between distinct bacterial taxa occurs. Most oral bacterial species need at least one partner to co-aggregate with in order to multiply [36], allowing for the formation of human dental plaque [37]. For example, Veillonella atypica can only grow in saliva if co-aggregated with streptococci [38], and Streptococcus gordonii can only multiply in low-arginine media if co-aggregated with Actinomyces naeslundii [39]. Most co-aggregation cases are mutualistic, though there are also examples of parasitic interactions [40].

Local growth promotion is an example of a chemical interaction resulting in a mixed organization [23]. Chemical interactions generally occur when bacteria are located near each other and when the protector pathogen is at a high density. The level of mixing in biofilms formed by strongly interdependent species is higher than in those formed by weakly interdependent species, and even more so than in biofilms formed by competing species [41],[42].

These interactions can promote an observed antimicrobial cross-protection, for example, when S. aureus is closely surrounded by a highly dense layer of ampicillin-resistant P. aeruginosa, shielding S. aureus from ampicillin action [43]. In addition, these biofilms have the capability to facilitate metabolic cross-feeding. For instance, Aggregatibacter actinomycetemcomitans utilizes the lactate produced by S. gordonii. This relationship is synergistic because, in turn, A. actinomycetemcomitans clears peroxide, a by-product of S. gordonii metabolism, maintaining a gap larger than 4 µm from S. gordonii to avoid the growth inhibition by peroxide, resulting in higher microbial burdens in the presence of oral abscesses and periodontitis [23],[44]. Other synergistic oral pathogens also show identical patterns [45],[46].

Local growth inhibition by a chemical antagonist interaction causes spatial segregation [23]. Chemical signals act at a very short-range (smaller than 10 µm), generating fine-scale spatial organization. Generally, signals with no need for a specific receptor are more widespread. For example, many streptococci produce lactate and hydrogen peroxide as waste products that concentrate in the surrounding area. Hydrogen peroxide can reach concentrations in the millimolar range at a distance of 100 µm from S. gordonii biofilm [47]. These products act as broad-spectrum toxins, eliminating local competitors and consequently regulating fine-scale segregation patterns.

Another example is the production of pyocyanin by P. aeruginosa. In chronic wounds, this virulence factor senses cell wall fragments discarded by S. aureus [22] and contributes to the elimination of this bacterial species when located in the proximity. The spatial organization of these bacteria is highly segregated, with S. aureus generally present only at a distance of 20–30 µm from the wound surface, and P. aeruginosa in deeper depths, at around 55 µm from the wound surface [48]. Because the wound environment is highly viscous and restrictive for cell migration, it avoids species mixing and further killing, allowing S. aureus and P. aeruginosa to co-infect wounds [26].

Biofilm remodeling is an example of physical interactions resulting in a segregated organization [23]. This is due to either the production or breakdown of extracellular matrix components, which contributes to the segregation of microbial members. For example, when growing on an agar surface, Pseudomonas fluorescens regularly produces mucoid variants that overproduce exopolysaccharides. Interestingly, these variants can move to the top of the biofilm, gaining access to oxygen and restraining their competitors [49]. In CF patients, P. aeruginosa also produces mucoid variants that are able to overproduce the polysaccharide alginate, interfering with the presence of S. aureus and Burkholderia cenocepacia [50],[51], creating a highly marked organization [52]. Other major P. aeruginosa polysaccharides, such as Pel and Psl, also have different roles in the spatial organization of biofilms [53]. The re-positioning of polymicrobial members in biofilms can influence community interactions and disease progression [23].

It has been observed that cells within a biofilm can detach and disseminate into the surrounding environment [12],[54]. It was suggested that biofilm dispersion is an effective mechanism for disseminating bacteria with reinforced colonization properties and pathogenic potential in the host environment. The factors that promote biofilm formation and dispersal have been studied mostly in Gram-negative bacteria and involve the degradation of matrix components, such as polysaccharides, proteins, and extracellular DNA, the occurrence of local cell lysis, and the production of surfactant molecules that reduce surface tension. The dispersion of biofilm cells is essential to maintaining the infection, promoting its recidivism and chronicity [12],[54].

Interestingly, biofilm assembly can be driven by two processes that are independent of the bacterial genes present, and intrinsic to the physical properties of the host-derived macromolecules such as mucus: depletion aggregation and bridging aggregation. In both mechanisms, the bacterial functions usually required for biofilm formation, including surface adherence, synthesis of the extracellular matrix, cell-to-cell communication, and motility, are not involved [126]–[128]. In lung infections, biofilm assembly by P. aeruginosa colonizing the mucus is driven by mucin, which is abundant in chronic infections, through depletion aggregation. This non-adsorbing polymer creates mutual attraction forces between neighboring cells, creating an unbalanced osmotic pressure that pushes them together into forming ordered aggregates, which lead bacteria to become less susceptible to antibiotics [126],[127]. The higher the cell number and polymer concentration are, the greater the aggregation is [126],[127]. However, aggregation can also occur at low cell numbers when two or more non-adsorbing polymers are mixed [126],[128], as observed in the early development of CF, in which mixtures of DNA, F-actin, and mucin found in CF secretions can aggregate P. aeruginosa at low concentrations (10⁵ CFU/mL) [126],[128]. These studies suggest that the physical mechanisms associated with mucus contribute to biofilm assembly, and that further studies aiming to target depletion-mediated antibiotic tolerance for the treatment of chronic infections, such as CF, would be most valuable [128].

3.   Polymicrobial infections in humans

There is a substantial diversity and concentration of microorganisms in the limited space of the human body, with evidence of their co-evolution over thousands of years. For a disease to occur, the microbial populations colonizing the host cells generate a state of dysbiosis that shifts the microbial composition from healthy to unhealthy [11],[55]. For example, P. gingivalis is a secondary colonizer of the gingival plaque, which harbors more than 500 bacterial species that colonize the oral cavity either permanently or transiently, producing a set of virulence factors that will benefit this species and other oral microorganisms; it is a major etiologic agent driving dysbiosis and contributing to chronic periodontitis [56]. Furthermore, bacteria forming biofilms are present in higher concentrations in disease cases. For example, biofilms are at least one hundred-fold more concentrated in biopsies from patients with inflammatory bowel disease compared with those from healthy individuals [57]. Therefore, in diseases caused by polymicrobial biofilms, the composition and concentration of the microbial community can predict the disease severity and outcome [11].

For the establishment of a disease, the inhibition of DNA repair mechanisms must occur [58],[59]; alternatively, the damage of the host cells and their DNA can occur, either indirectly through oxidative stress induced by the activation of the innate host response [60], or directly via the presence of microbial virulence factors, such as genotoxins produced by several enteric pathogens that have a strong immunomodulatory effect on the intestinal mucosa [61]–[63]. Chronic induction of severe DNA damage can lead to the activation of cellular senescence [64], by which host cells enter a permanent cell cycle arrest but maintain a high metabolic state, named the senescence-associated secretory phenotype. In this state, they are able to secrete a huge number of inflammatory cytokines, immune modulators, growth factors, and proteases [63],[65]. Cellular senescence can increase host susceptibility to genotoxin-producing organisms, such as typhoid and non-typhoidal Salmonella, Escherichia coli phylogroup B2, Shigella spp., and Campylobacter spp. [61],[62], as well as to the action of other microorganisms like Fusobacterium nucleatum, Streptococcus pneumoniae, influenza virus, and varicella zoster virus [66]–[68].

Oral cavity diseases, otitis media, keratitis, diabetic foot wound infections, chronic infection in the CF lung, pneumonia, burn wounds, osteomyelitis, urinary tract infections, vaginitis, medical device-related infections, percutaneous endoscopic gastrostomy tube infections, peritonitis [11], and certain types of cancer [63] are examples of human diseases caused by polymicrobial biofilm communities (Table 1).

In the next chapter, some examples of polymicrobial biofilm-mediated infections occurring in different regions of the human body will be discussed, mainly focusing on biofilm formation, including otitis media, colorectal cancer, diabetic foot wound infections and CF.

4.   Examples of polymicrobial biofilm-mediated infections and diseases

4.1.   Otitis media and related nasopharyngeal infections

Nasopharyngeal infections are related to otitis media. The eustachian tube connects the nasopharynx with the middle ear cavity; as such, an infection in the nasopharynx may spread through this tube to the middle ear, prompting otitis media [92]. Frequently, most cases of otitis media are spontaneously resolved within a few weeks, but chronic infections can lead to partial or total hearing loss [93].

The microbial community is established in the nasopharynx in the first year of life, varying during individual's lifetime [94]. The microbial species responsible for otitis media are the commensal bacteria normally found colonizing this region, such as S. pneumoniae, non-typeable H. influenzae, and Moraxella catarrhalis, along with upper respiratory viruses, including influenza A virus, RSV, human rhinovirus, and adenoviruses (Table 1) [69]. Pre-infection with upper respiratory tract viruses predisposes the host to disease development and enhances the otitis media progress [95] by changing the physiological properties of infected airways and modulating the immune response [11],[96]. Bacteria shift from colonizers to invaders causing infection, often as polymicrobial biofilms, when there is either an epithelial rupture or immune dysfunction [97].

Concerning biofilm formation by S. pneumoniae, this species starts by attaching to the substratum through the mediation of pneumococcal adhesion factors, such as PspA, CbpA, and PcpA [98], followed by pneumococcal aggregation and matrix formation and maturation [99],[100]. Events like cellular autolysis, genetic competence, and biofilm formation are regulated by two quorum sensing systems, Com and LuxS/AI-2, in an intercrossed way [101], through the secretion of strain-specific competence-stimulating peptides, which allow for the release and uptake of DNA from other streptococci present in the biofilm [102]. Competence is highly efficient for spreading β-lactam resistance determinants between the streptococcal species [100],[102]. In polymicrobial biofilms combining S. pneumoniae and H. influenzae, it has been found that susceptible S. pneumoniae strains can gain protection against antibiotics through the passive transference of β-lactamase genes from resistant H. influenzae [103]. A few other streptococcal species that colonize the pharynx, such as S. intermedius, S. oralis, S. gordonii, and S. mutans, use LuxS/AI-2 to regulate competition or cooperation mechanisms among species present in the pharynx and to control biofilms [104].

4.2.   Colorectal cancer

Polymicrobial biofilms have been described in proximal colorectal cancer and in polyps of patients with familial adenomatous polyposis, a hereditary condition caused by mutations in the tumor suppressor gene, adenomatous polyposis coli [86]. Of note, patients with biofilm tumors in the colon possess analogous structures in distal regions that are tumor-free, though they present significant pro-tumorigenic changes on the underlying colonic epithelium [105]. As such, it can be inferred that the alteration of the colon microbiota precedes tumor formation, and that colon mucosal biofilm detection may predict an increased risk for the development of colorectal cancer and adenomas [63],[86].

Biofilms in familial adenomatous polyposis are mainly formed by E. coli, Bacteroides fragilis, and other Proteobacteria and Bacteroidetes, whereas biofilms in colorectal cancer are composed of B. fragilis and oral pathogens, including F. nucleatum, Peptostreptococcus stomatis, and Parvimonas micra (Table 1) [86],[87].

In familial adenomatous polyposis, there is a synergistic effect when E. coli and B. fragilis biofilms co-colonize the colon, which is associated with an enhanced morbidity and mortality [63],[86]. In this case, the carcinogenic properties are related to two key virulence factors. The first is a zinc-dependent metalloprotease toxin with a mucolytic effect secreted by some enterotoxigenic B. fragilis, which promotes a colonic inflammatory condition [106], creating favorable conditions for the formation of specific polymicrobial biofilms. The second is the expression of a functional psk island of E. coli, which encodes the genotoxin colibactin, which promotes DNA damage and contributes to disease progression [59].

4.3.   Diabetic foot wound infections

Diabetes mellitus is characterized by a defective physical response to insulin. It affects nearly 6.4% of the worldwide population, and since this number is estimated to double by the year 2030, the Centers for Disease Control and Prevention classified diabetes as a current epidemic [107]. People with diabetes mellitus have a 15–25% estimated probability of developing a diabetic foot ulcer (DFU) [108], and among DFU patients, 80% will suffer lower-limb amputations due to an ulcer infection [109]. Generally, DFU are infected with polymicrobial biofilms [17],[110],[111], including opportunistic pathogens [111] and anaerobic bacteria [112],[113], that are associated with delayed healing of these chronic wounds [114] and an increased risk of death within 18 months [115]. There are several studies based on different methods, from conventional culture data to genomic approaches, aiming at establishing the microbe population present in infected DFU. According to Citron et al. [91], among positive cultures, approximately 16% contain one bacterial species, 20% contain two species, 20% contain three species, 13% contain four species, and 30% present five or more bacterial species. Additionally, among the microbes identified, 49% are anaerobes, with Fusobacterium spp., Porphyromonas spp., Prevotella spp., Bacteroides spp., and Clostridium spp. being the most frequently detected. Most anaerobic species can be found alongside aerobic organisms [91],[116], suggesting that the former may play a significant role in the etiology of chronic wound infections, helping to control bacterial pathogenicity and biofilm formation as functionally equivalent pathogroups [1],[17],[112]. The most predominant aerobic and facultative anaerobic species identified in different studies include Pseudomonas spp. (16%), E. coli (14.6%), methicillin-susceptible S. aureus (13.3%), and Streptococcus pyogenes (10.6%) [112],[117].

In DFU, the hypoxic environment deeply influences bacterial diversity and localization, being observed at a higher occurrence of strictly anaerobic bacteria and some Proteobacteria in deeper ulcers [48],[118], making each DFU a unique microbial complex [119]. DFU duration is directly correlated with the bacterial diversity present in the wounds and with the Proteobacteria frequency, and is indirectly correlated with the frequency of staphylococci [120]. Additionally, the presence of fungi in polymicrobial biofilms is associated with a poor diagnosis and delayed healing of chronic wounds [121].

In DFU, some polymicrobial communities have a higher ability to produce biofilms than others. According to Mottola et al. [122] biofilms formed by P. aeruginosa plus Enterococcus spp., Acinetobacter spp. plus Staphylococcus spp., and Corynebacterium spp. plus Staphylococcus spp., have a higher biofilm-producing ability than those formed by E. faecalis plus Staphylococcus spp. and E. faecalis plus Corynebacterium spp. The authors have also demonstrated that these polymicrobial biofilms are able to synergistically produce higher biofilm concentrations and virulent factors than individual species [122].

4.4.   Cystic fibrosis of the lung

CF is an autosomal recessive genetic disorder that is most common among Caucasians. The majority of CF cases are due to an inherited mutation in phenylalanine residue 508 of a specific chloride ion channel, named the cystic fibrosis transmembrane conductance regulator (CFTR) [123]. However, there are approximately 1,900 known mutations associated with CF (http://genet.sickkids.on.ca/app). In CF, there is a Na⁺ and Cl⁻ ion imbalance that causes water to be retained inside the cells, leading to dehydration of the extracellular space [124]. This creates a viscous and thick mucus layer that cannot be easily cleared by cilia and that traps bacteria, resulting in extensive mucous accumulation and severe lung infection [125]. Lung infection caused by polymicrobial biofilms remains the primary cause of morbidity and mortality in CF patients [84].

H. influenzae and S. aureus are initial colonizers of the lungs of young children with CF [129],[130]; by the age of twenty, 60–70% of CF patients present intermittent and chronic colonization by P. aeruginosa [131]; and, eventually, the lung is terminally colonized by B. cepacia (Table 1) [132]. P. aeruginosa and B. cepacia rarely infect healthy human lungs [133], making them later colonizers. The consensus is that P. aeruginosa surpasses S. aureus in the resident microbial community at a later age and becomes the prevalent pathogen in the CF lung [123],[134],[135]. In fact, it is the dominant microorganism in at least 50% of adult CF patients [136]. However, recent studies demonstrated that S. aureus is often co-isolated with P. aeruginosa in at least 30% of adult CF patients [137]–[139].

CF infection with P. aeruginosa has been associated with more rapid lung failure, the worst clinical results, and premature death [135],[139],[140]. Moreover, CF patients infected with P. aeruginosa are more vulnerable to developing secondary infections, for example with species from the Burkholderia cepacia complex, which are associated with cepacia syndrome. This syndrome predisposes patients to usually fatal necrotizing pneumonia [51],[141].

The B. cepacia complex consists of several phenotypically indistinguishable genomovars (i.e., different species that are phylogenetically closely related) [142]. Its members form mucoid biofilms, engaging in a close network of interactions with P. aeruginosa, which promote the exchange of genetic material [50], prompting a more rapid decline in pulmonary function [78].

Additionally, fungi and yeasts inhabit the individuals' airways, in which Aspergillus fumigatus and C. albicans are the most prevalent fungi and yeast, respectively [82], being identified in up to 50% of CF patients (Table 1) [143],[144]. These fungi are able to form biofilms with P. aeruginosa [31],[135],[145],[146]. Moreover, the metagenomic characterization of the lung microbiome allowed for the description of new pathogens in the CF lung, such as Ralstonia mannitolilytica, which is associated with accelerated disease progression and increased mortality [147]. More than 450 viral genotypes were also described [148], some of them being related to the onset of CF pulmonary exacerbations [149]. Interestingly, CF biofilms consisting of P. aeruginosa and either the anaerobic emerging species Inquilnus limosus or Dolosigranulum pigrum, rarely observed elsewhere, are found to present an increased resistance to most antibiotics [150].

The role of multispecies interactions is of paramount importance in predicting patients' health. In the next chapter, the role of biofilm-species interactions in shaping virulence, antimicrobial resistance, and disease progression will be discussed.

5.   Interaction between pairs of pathogens in biofilms during co-infection

P. aeruginosa is extremely difficult to eliminate mainly due to the low permeability of its cells, attributed to the gram-negative cell wall, added to the presence of efficient efflux systems from five superfamilies (major facilitator superfamily, ATP-binding cassette superfamily, resistance nodulation division family, small multidrug resistance family, and multidrug and toxic compound extrusion family), the expression of intrinsic and acquired antibiotic resistance mechanisms, the ability to express a considerable number of extracellular virulence factors, and the ability to adapt to several environmental conditions [151].

The ESKAPE group is the acronym for Enterococcus faecium, S. aureus, K. pneumoniae, A. baumannii, P. aeruginosa, and Enterobacter spp., which are highly virulent and antibiotic resistant pathogens [152]. When P. aeruginosa forms biofilm complexes with microorganisms belonging to the ESKAPE group, such as either S. aureus or A. baumannii, its elimination is almost unachievable, leading to treatment failure in diseases like CF, urinary tract infections, or chronic wound infections (Table 1) [153]. Therefore, the mechanisms behind the interaction between this bacterial species and other microbes' have been the subject of many studies. Next, the interaction between P. aeruginosa and other microbes in biofilms during co-infection will be described.

5.1.   P. aeruginosa and S. aureus

P. aeruginosa and S. aureus are highly prevalent pathogens found in chronic wound infections [26], including in diabetic foot infections [25],[122], surgical site infections, necrotizing fasciitis [154],[155], burn wounds [156], and CF [157].

Studies regarding wound infections have shown that, S. aureus is usually located surrounding the surface of the wound and P. aeruginosa in the deeper sections of the wound [48],[158]. However, the location of P. aeruginosa in deeper regions of the wound, which may seem contradictory since it is a strict aerobic species, requires further explanation. Interestingly, according to Trizna et al. [159], after vancomycin treatment, S. aureus can be found in the middle-lower layers, i.e., biofilms' spatial organization shifts from segregated to mixed after antibiotic-induced stress. Therefore, the localization of both P. aeruginosa and S. aureus may not be segregated, being influenced by stress, which may benefit both species [159].

Although not being fully elucidated, during co-existence in polymicrobial infections, S. aureus and P. aeruginosa can potentially have both antagonistic and cooperative roles, mediated by several virulence factors (Table 2).

There is evidence that P. aeruginosa competes with S. aureus during early co-infections; however, in chronic long-term infections, P. aeruginosa may co-exist with S. aureus, with the presence of both bacteria reaching an equilibrium. This finding is in accordance with a previous study [178], that provides indications that chronicity increases the virulence of P. aeruginosa, by promoting its co-existence with S. aureus [25]. Importantly, S. aureus does not have any effect on P. aeruginosa's transcriptome, which suggests that this bacterial species is the main pathogen controlling infection progression [179].

During early co-infections, P. aeruginosa has an antagonistic behavior towards S. aureus, associated with the production of toxins such as staphylolysin (LasA), rhamnolipids, cis-2-decenoic acid, and long-chain N-acyl homoserine lactones [160] (Table 2). LasA cleaves the glycyl-alanine and glycyl-glycine bonds in the peptidoglycan layer of S. aureus, inducing bacterial cell lysis [160]–[162].

Rhamnolipids and cis-2-decenoic acid have anti-adhesive properties and promote biofilm dispersion, and together, they disrupt established biofilms and inhibit their formation by different bacteria and fungi in a dose-dependent manner [162],[164]. Moreover, cis-2-decenoic acid has been found to enhance bacterial metabolic activity and reverse the persistence state of S. aureus. Persister cells are phenotypic variants of regular cells representing a subpopulation of non-mutants, described as tolerant cells that can be present in both planktonic cultures and biofilms [180],[181]. Persisters are dormant metabolically quiescent cells with reduced metabolism, which enables them to tolerate high concentrations of antimicrobials [182]. Therefore, the presence of persister cells in chronic infections hampers their treatment [182],[183]. The reversal of a persister state to an antimicrobial-susceptible state can be due to a higher membrane permeability induced by cis-2-decenoic acid and the consequent increase in antimicrobial uptake [162],[164]. Regarding N-acyl homoserine lactones, they inhibit the growth of S. aureus in a dose-dependent manner [165]. At sub-inhibitory concentrations, they do the opposite, leading to robust biofilm formation and host cell invasion by S. aureus [165],[184].

P. aeruginosa produces several respiratory toxins, including hydrogen cyanide, pyocyanin, and mainly quinoline N-oxides, which interfere with the electron transport chain (cytochrome system), blocking the anaerobic respiration of S. aureus [160]. Pyocyanin is a potent phenazine, a prominent virulence factor, and a pseudomonal iron-scavenger, that induces the generation of reactive oxygen species, such as hydrogen peroxide and superoxide radicals, that can lead to cell death [167]. Quinoline N-oxides are quorum sensing-regulated virulence factors that block the oxidation of cytochrome b1 and the reduction of cytochrome a2 in S. aureus [168]. At higher concentrations, quinoline N-oxides inhibit bacterial growth; however, at lower concentrations, they may induce the selection of S. aureus SCV [160],[169].

Despite the antagonistic behavior of P. aeruginosa, S. aureus can counter-adapt to resist killing by this bacterial species by forming SCV and L-form colonies (Table 2). S. aureus evades P. aeruginosa quinoline N-oxides respiratory attack by forming SCV, which is a respiratory-defective phenotype that arises from mutations in metabolic genes [185], which is able to shift to fermentative metabolism due to insufficient oxygen availability in the biofilm [157],[170],[186]. S. aureus SCV are a slow-growing subpopulation of small and non-pigmented colonies, non-hemolytic, with reduced expression of virulence factors, and usually auxotrophic, i.e., they may lost the ability to synthesize certain substances, including thymidine, hemin, or menadione [160],[187],[188]. The enhanced survival of SCV is associated with a reduced activity of the accessory gene regulator (agr), which is a global regulator, related to the regulation of quorum-sensing systems and several virulence genes [187]. In the absence of environmental pressure, the temporary SCV phenotype can return to the wild-type phenotype [188]. To resist reactive oxygen species produced by P. aeruginosa, S. aureus SCV produce important antioxidant components, such as superoxide dismutases, catalase, and the pigment staphyloxanthin [189]. An SCV phenotype is associated with disease progression [190], as it exhibits reduced membrane potential, which is related to an increased tolerance to cationic antibiotics [135],[190], efficient evasion of the host immune system, and enhanced cell persistence [170],[187]. Therefore, the antagonistic interactions between P. aeruginosa and S. aureus mediated by quinoline N-oxides have the potential to directly influence disease progression [135],[160],[170].

To evade LasA-mediated lysis by P. aeruginosa, S. aureus may produce cell wall deficient L-form-like colonies that have lower surface charge and stronger hemolytic activity than the walled forms [171],[191]. Additionally, the aggregated characteristic of L-forms is identical to the one observed in biofilms, which could lead to enhanced persistence in the presence of stress and antibiotics [191], and promote an increased virulent infection [192].

During long-term chronic infections, S. aureus cooperates with P. aeruginosa by secreting exoproducts such as staphylococcal protein A (SpA) [160] (Table 2). Some exoproducts secreted by S. aureus enable P. aeruginosa to restore and enhance its swimming and swarming motility [172]. S. aureus SpA interacts with the polysaccharide locus (Psl) and the protein component of type IV pili (PilA) on the P. aeruginosa cell surface. When SpA attaches to Psl, this contact protects P. aeruginosa from phagocytosis by the host neutrophils [173]. Moreover, S. aureus SpA binds to the immunoglobulin Fcγ domain, preventing its own opsonophagocytosis and death. However, in the absence of Psl on the P. aeruginosa cell surface, SpA binds to PilA, inhibiting biofilm formation [173]. This is interesting since the genetic background of each P. aeruginosa strain can influence the progression of the disease, depending on either the presence or absence of Psl.

Regarding CF, it is recognized that a hallmark of chronic pulmonary infections is the presence of P. aeruginosa strains that are able to shift to a mucoid phenotype, due to the overproduction of the exopolysaccharide alginate, mediated by a mutation within the mucA anti-sigma factor [193]. Even though it is unexpected, the overproduction of alginate by mucoid strains inhibits P. aeruginosa's anti-staphylococcal activity, such as the production of rhamnolipids, and quinolone signals [174],[175]. Therefore, in the presence of mucoid variants, S. aureus strains are spared from P. aeruginosa killing [174], and both planktonic and sessile S. aureus forms are protected from antibiotic action [176],[177]. This indicates that P. aeruginosa shifts from a competitive to a cooperative approach during chronic CF, which benefits both species [177].

Although S. aureus has mechanisms to escape the toxicity and metabolic stress associated with the presence of P. aeruginosa and even sometimes cooperates with this bacterial species, S. aureus eventually dies in the presence of P. aeruginosa. This happens when S. aureus alters metabolic gene expression, affecting key metabolic pathways and reducing the energy used to survive. In these cases, the genes involved in nucleotide metabolism (nrd operons) are downregulated, limiting DNA synthesis, DNA repair control, and affecting S. aureus cell concentration [137],[139]. The switch from aerobic respiration to lactic acid fermentation allows P. aeruginosa to consume the lactate produced by S. aureus as a carbon source [160]. Moreover, genes involved in glycolysis and the pentose phosphate pathways are downregulated in S. aureus, revealing that there is a competition for nutrients [137],[139], which S. aureus loses. In P. aeruginosa, nitrogen starvation leads to the expression of ntrC, a response regulator to nitrogen limitation, and to the upregulation of genes involved in nitrogen assimilation, including those encoding glutamate dehydrogenase and synthase. On the other end, in S. aureus, nitrogen starvation leads to a downregulation of the glutamate synthase genes. Therefore, the additional availability of carbon and nitrogen provided by S. aureus may be beneficial for P. aeruginosa, allowing it to gain an advantage over other bacteria during colonization and during long-term chronic infections [25],[157],[177].

Additionally, in the presence of P. aeruginosa, S. aureus releases the peptidoglycan component N-acetyl glucosamine (GlcNAc) upon cell lysis or cell wall turnover during growth, which in turn inactivates the P. aeruginosa caseinolytic protease ClpXP. This protease has a critical role during quorum sensing homeostasis, leading to an increased production of virulence factors by P. aeruginosa, such as pyocyanin and exotoxins, that can damage S. aureus [22],[162],[194]. Moreover, lysed S. aureus cells might be a source of iron for P. aeruginosa [195]. Furthermore, the non-motile phenotype of S. aureus may bring some disadvantages during infection and colonization. Samad et al. [196] observed that the S. aureus location changed together with the one of P. aeruginosa by moving upwards along with the swimming cells of P. aeruginosa , leading to a proximity that can be harmful to S. aureus.

P. aeruginosa can also indirectly prejudice S. aureus by manipulating the innate immunity of the host (e.g., by inducing the production of phospholipase sPLA2-IIA by bronchial epithelial cells), which leads to S. aureus's death [197]. This may be simply a response of the host to which P. aeruginosa resists; however, it suggests that P. aeruginosa's interactions with the host can disrupt bacterial communities more broadly than previously expected [135],[198].

5.1.1.   P. aeruginosa and S. aureus interactions lead to increased antimicrobial resistance

S. aureus and P. aeruginosa normally show high antimicrobial resistance toward multiple compounds [199]. According to the European Center for Disease Control, from 2020 to 2022, approximately 30% of the P. aeruginosa isolates were resistant to at least one of the antimicrobial groups under surveillance [200]. According to Mottola et al. [199], 36% of fifty-three staphylococci isolated from DFU cases were multidrug-resistant. When bacteria like P. aeruginosa and S. aureus are present together in biofilms, resistance to antibiotics and to the human immune system can be up to 1,000 times higher than the one showed by their planktonic counterparts [199],[201],[202]. In chronic infections, hypoxia and thick dehydrated mucus provide optimal conditions for biofilm formation and contribute to adaptive resistance [1],[203]. Moreover, the proximity between bacteria in biofilms contributes to horizontal gene transfer and further antimicrobial resistance.

It was found that, in the presence of P. aeruginosa, S. aureus gains resistance to tetracycline and ciprofloxacin [137]. According to Briaud et al. [137], P. aeruginosa induces the over-expression of the Nor family genes (norA, norC, and of tet38), which encode efflux pump transporters responsible for increased antibiotic resistance in S. aureus. The most over-expressed gene is tet38, implicated in resistance against tetracycline. In addition, this transporter interacts with the CD36 receptor on epithelial cells, increasing S. aureus internalization, helping it to hide in epithelial cells and escape the attack from either the immune system or from P. aeruginosa [204].

Recent studies also suggest that S. aureus can enhance P. aeruginosa's antibiotic resistance ability [13],[205]. Beaudoin et al. [13] have demonstrated that the interaction between S. aureus SpA and P. aeruginosa Psl increases tobramycin resistance in P. aeruginosa. In the presence of S. aureus, this bacterial species synthesizes truncated lipopolysaccharides that lack a O-specific antigen and contribute to increased resistance to β-lactam antibiotics, but not against ciprofloxacin or polymyxin [205]. These studies show that P. aeruginosa is able to acquire resistance mechanisms against antibiotics targeting the cell wall in polymicrobial biofilms, such as β-lactams, and in protein biosynthesis, such as tetracycline [13],[160].

Regarding S. aureus SCV, they exhibit an intrinsic tolerance to antibiotics, which is not necessarily associated with antimicrobial resistance genes [206]. In this process, the shift to fermentative metabolism results in decreased ATP production, reducing the active transport of antibiotics that inhibit protein synthesis, such as tetracycline, aminoglycosides, and macrolides [160],[166],[170],[176],[177]. When S. aureus's intracellular ATP is reduced, it slows down cell proliferation, decreasing S. aureus's susceptibility to cell wall-targeting antibiotics [176],[177]. The reduced membrane potential and higher persistence of SCV contribute to antibiotic tolerance [170],[187].

5.2.   P. aeruginosa and B. cepacia complex

B. cepacia complex, or simply B. cepacia, is formed by at least 20 different species, including B. cepacia and B. cenocepacia, which are Gram-negative, catalase-producing, and lactose-non-fermenting bacteria [207]. B. cepacia seldom causes problems in healthy persons, as it is an opportunistic pathogen that is associated with pneumonia mainly in immunocompromised individuals with an underlying disease, like CF or the immunodeficiency chronic granulomatous disease [208].

Eventually, if B. cepacia complex bacteria gain access to lungs previously infected with P. aeruginosa, the close interactions between both bacterial species may influence virulence and result in a more rapid decline of human pulmonary function and eventually in death [78],[135]. Competition between these two bacteria is intense, as populations of both species have been found to be able to invade each other [51],[209],[210]. In a study on CF, most clinical isolates of P. aeruginosa (81%) and B. cenocepacia (57%) were found to be able to secrete bacteriocin-like toxins, such as pyocyanin, inhibiting each other's growth [211],[212].

Cooperative mechanisms between both species can also be detected, mainly benefiting B. cepacia [135]. When in coculture, P. aeruginosa upregulates putative B. cepacia virulence factors and increases its adhesion to the host epithelium [213]. Additionally, N-acyl homoserine lactones produced by P. aeruginosa can stimulate the production of lipases, proteases, and siderophores by Burkholderia, promoting its growth [210],[213]–[215]. Siderophores are very important elements in bacteria, since they are high-affinity, iron-chelating compounds for iron uptake, an element essential for growth, as several metabolically important proteins in living cells depend on iron [214],[215]. Finally, alginate production by P. aeruginosa is known to inhibit the host immune system and indirectly aid B. cenocepacia survival [216].

5.3.   P. aeruginosa and SMG members

Members of the SMG, also known as the Streptococcus anginosus group which include S. constellatus, S. intermedius, and S. anginosus, are gram-positive facultative anaerobic, catalase-negative, and nonmotile cocci [217]. They are common members of the airway microbiota and the genitourinary and gastrointestinal tracts, but can be involved in purulent infections, often resulting in abscess formation, which distinguishes them from other pathogenic streptococci, such as S. pyogenes and S. agalactiae [218]. SMG has recently been associated with clinically relevant pulmonary exacerbations in CF patients [84],[219].

Interestingly, in vitro investigations showed that P. aeruginosa differentially expresses several genes related to virulence factors and drug efflux pumps when in co-culture with streptococci [80]. Furthermore, the expression of some of these genes could be mediated by the streptococci quorum-sensing molecule AI-2 (section 4.1), which was found in high levels in the sputum of some CF patients. This result suggests that communication between microbes of the commensal microbiota through quorum sensing may exacerbate CF disease caused by conventional pathogens like P. aeruginosa [11],[80].

5.4.   P. aeruginosa and fungi: A. fumigatus and C. albicans

The fungus A. fumigatus is one of the most common Aspergillus species causing disease in immunocompromised individuals. It is the most common fungus found in CF airways [144], and the clinical presentation of associated diseases may include no apparent respiratory failure, bronchitis, and bronchiectasis [82],[220]. Infection tends to occur after P. aeruginosa colonization, resulting in synergistic co-infections that cause more severe clinical outcomes compared with those promoted by each microorganism alone [135],[146],[221].

When they coexist in the CF lungs, the interactions between P. aeruginosa and A. fumigatus are extensive and mainly concerned to the production of P. aeruginosa phenazines, which are beneficial to the fungus. It is known that high concentrations of phenazines, including pyocyanin, phenazine-1-carboxylate, phenazine-1-carboxamide, and 1-hydroxyphenazine, inhibit A. fumigatus growth by inducing the production of reactive oxygen and nitrogen species [222],[223]. However, pyocyanin, phenazine-1-carboxylate, and phenazine-1-carboxamide are usually present in subinhibitory concentrations in vivo, promoting fungal growth by acting as iron-reducing agents. They sequester ferric iron (Fe³⁺), reducing it into the more soluble ferrous iron ions (Fe²⁺), enhancing iron uptake in the host where iron is tightly bound to proteins such as hemoglobin and with extremely limited availability [223]. It was also shown that the P. aeruginosa phenazine 1-hydroxyphenazine causes iron starvation of A. fumigatus by chelating iron, which is partially compensated by the fungal mechanisms adapted to iron starvation, such as the production of siderophores [135],[223].

During chronic infections, damaged host cells release iron in the form of haem groups and hemoglobin. In these infections, phenazine-1-carboxylic acid present in infected tissues may shift the redox equilibrium between Fe³⁺ and Fe²⁺, thereby helping to make iron become more bioavailable [224]. There is some evidence that P. aeruginosa siderophores, pyoverdine and pyochelin, are replaced by a haem assimilation system used by Pseudomonas to sequester iron, generating an iron-rich environment [225].

A. fumigatus infections generally occur after P. aeruginosa colonization because a) P. aeruginosa previously generates an iron-rich environment in which A. fumigatus can prosper; and/or because b) “damaged lungs” allow for better and faster colonization by pathogens [135].

The fungus C. albicans is a common, commensal microbe that colonizes the oropharyngeal cavity, gastrointestinal and vaginal tracts, and skin of healthy individuals without causing disease. The transition from an element of the commensal microbiota to an agent of opportunistic infections depends on different factors. Opportunistic C. albicans can be found in superficial mucocutaneous disorders, in parenteral nutrition feeding tubes [77], in denture stomatitis [73] and dental caries [72], and in CF [226]. C. albicans is a dimorphic yeast, and can switch from a round yeast, generally associated with commensal form, and germinate to an elongated hyphal form, generally associated with opportunistic infections. This transition is central to its pathogenesis [11],[226].

In CF patients, C. albicans is frequently found in the lower airways due to the high amounts of fungus colonizing the nasopharynx, mainly due to repeated rounds of antimicrobial therapy [143]. In its pathogenic hyphae form, the elongated structure is able to puncture the epithelial and endothelial layers to gain access to deeper tissues, which is crucial for biofilm formation [227],[228].

When coexisting, the interactions between P. aeruginosa and C. albicans are essentially antagonistic, generally towards the fungus [11]. These interactions are complex, and their influence on the human host is not fully understood. Studies revealed that when in co-culture, P. aeruginosa is able to inhibit C. albicans germination by secreting 3-oxo-C₁₂ homoserine lactone, decreasing Candida pathogenic potential [145]. However, the secretion of this molecule diminishes during chronic infections [229], and consequently prompts the development of the invasive filamentous form of the fungus [230]. Moreover, P. aeruginosa can attach to specific areas of the C. albicans hyphal surface, effectively eliminating the hyphal cell by inducing its lysis, though it is incapable of acting on the round yeast form of the fungus [11],[31].

Lower concentrations of P. aeruginosa phenazines in polymicrobial biofilms inhibit respiration and promote fermentation by C. albicans, increasing the production of products like ethanol, glycerol, and acetate. These metabolic effects impair the C. albicans biofilm formation and the fungal transition to the hyphal form [231]. In turn, ethanol leads to the enhanced development of P. aeruginosa biofilms and the formation of a more virulent mucoid phenotype [231],[232]. Furthermore, ethanol is an immunosuppressant, indirectly influencing the number and diversity of microbes within the polymicrobial biofilm [233],[234]. Ethanol has similar effects on other lung associated pathogens such as S. aureus [235] and A. baumanii [236], though the mechanisms behind these events have not yet been fully identified [135].

Nevertheless, interactions between P. aeruginosa and C. albicans are not unilaterally antagonistic. C. albicans produces the quorum-sensing alcohol farnesol, which inhibits the transition from yeast to hyphae, resulting in fewer hyphal forms. Moreover, the secretion of farnesol has been shown to restrain the swarming capacity of P. aeruginosa, which may hamper the preliminary deposition of P. aeruginosa onto the hyphae surface, resulting in lower lytic activity [11],[230]. According to Cugini et al. [237], farnesol leads to a substantial downregulation of the P. aeruginosa pqsA gene, which is a mediator of pyocyanin expression. Therefore, during the close coexistence of both microorganisms in different diseases, farnesol secretion may be protective for C. albicans by reducing P. aeruginosa pyocyanin levels [238]. Finally, C. albicans can reduce the expression of P. aeruginosa siderophores, namely of pyoverdine and pyochelin, limiting its growth and further virulence [239].

5.5.   P. aeruginosa and viruses

In a previous study, rhinovirus, influenza A virus, and influenza B virus were found to be the most frequent viruses colonizing the upper respiratory tract of children affected by temporary exacerbated CF, while the most common bacteria were P. aeruginosa and S. aureus [240]. In an in vitro work that studied the influence of RSV on the adherence of P. aeruginosa to epithelial cell monolayers, results showed that the presence of RSV increased adherence and further colonization of the host by this bacterial species during CF [83]. Another study aiming to determine if RSV could facilitate the development of acute infections by P. aeruginosa in vivo showed that a co-infection with both microorganisms lead to a paramount, nearly two-thousand-fold increase in bacteria colony counts in lung homogenates when compared to results from P. aeruginosa monomicrobial infection [241]. Therefore, viruses infecting the upper respiratory tract result in an exacerbation of CF symptoms [83],[240],[241], probably by serving as a bridge between the bacteria and host surfaces through the induction of a T-cell chemoattractant, named monokine, induced by gamma interferon [241]. They also exert immunomodulatory effects that inhibit bacterial clearance processes in the CF lung [241],[242].

6.   Conclusions

Bacteria, fungi, and viruses are often co-isolated together from complex polymicrobial biofilm communities in vivo and are responsible for clinical and pathological manifestations in humans. It is known that microbial communities within biofilms are spatially organized in either a mixed or segregated way and establish either physical or chemical interactions between them. These interactions can exhibit either cooperative or antagonistic effects, either benefiting or not benefiting the community, and may result in a synergistic outcome influencing bacterial clearance mechanisms. They can affect virulence directly or indirectly, modifying the clinical course of the disease, with biofilms being intimately associated with disease progression and worsening of patient's symptoms. P. aeruginosa is hard to eliminate as it can express several efflux systems and present the capacity to adapt to several environmental conditions, being the cause of many human diseases. In the presence of other microorganisms, it can establish interrelationships mostly beneficial to itself, forming strong biofilms that are even more resistant to the host immune system and to antimicrobial activity. Therefore, during the evaluation of a given disease, it is of utmost importance to determine if the infection is caused by polymicrobial biofilms and which microbes are involved, as their presence has significant implications for disease management. For example, the presence of antimicrobial resistant bacterial pathogens will influence the selection of antimicrobial therapy. As such, in future studies the challenge will be to identify potential targets for the inhibition of biofilm co-adhesion and development into large intraspecies and interspecies complex networks inside the human host.

Use of AI tools declaration

The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.