Ecology and Evolution of Interspecies Interactions between Pseudomonas aeruginosa and Staphylococcus aureus

Under natural conditions, bacteria rarely live in isolation but frequently interact with members of other species. Such ‘interspecies interactions’ do not only occur in the environment, but also during bacterial infections. In fact, many bacterial infections are ‘polymicrobial’, which means that they are caused by multiple different species. Polymicrobial infections have frequently been observed to be more virulent and more difficult to treat, leading to the hypothesis that interspecies interactions may lead to adverse outcomes for patients. The bacterial pathogens Pseudomonas aeruginosa and Staphylococcus aureus have emerged as a particularly relevant model system to study interspecies interactions. The reason for this is that they are among the most important pathogens in polymicrobial infections, such as cystic fibrosis lung or chronic wound infections. At the molecular level, their interactions are well characterized, with most studies concluding that P. aeruginosa inhibits and often outcompetes S. aureus by producing a variety of inhibitory compounds. However, as both species can co-exist in infections, our current understanding is likely incomplete. This has led to the idea that not molecular mechanisms alone, but also ecological and evolutionary factors should be considered to fully understand interactions between P. aeruginosa and S. aureus. The aim of my PhD thesis is to exactly do this, and to fill some of the existing knowledge gaps by combining molecular, ecological, and evolutionary principles in order toimprove our understanding of bacterial interspecies interactions.In Project 1, I performed pairwise competition experiments and assessed the relevance of three different ecological factors in interactions between P. aeruginosaand S. aureus: the S. aureus strain genetic background, the relative species frequency, and the culturing condition (affecting species mixing). Consistent with existing literature, I found that under many conditions, P. aeruginosa was the dominant species, but all three ecological factors affected the interspecies competitive balance. Competitiveness of S. aureus correlated with the virulence level of the respective strain and followed positive frequency-dependent fitness patterns, depending on the competition pair used. The latter means that the species that is initially more common in the population wins the competition, a finding that offers an explanation as to why P. aeruginosa might be constrained in invading established S. aureus populations. Overall, P. aeruginosa was most dominant under static incubation, emphasizing the importance of physical properties prevailing in the environment. Together, Project 1 showed that different S. aureus strains vary in their ability to compete with P. aeruginosa, that frequency effects play an important role, and that the culturing condition can further shift the competitive balance between species. Notably, we propose that the same ecological factors (strain background, species frequency and environmental conditions) also vary in infections, and thus predict that they may influence species interaction patterns within a host as well. Although inhibition of S. aureus by P. aeruginosa is well-described, little is known about whether and how S. aureus responds towards this attack. I addressed this aspect in Project 2. First, I performed supernatant assays and confirmed that growth of all three S. aureus strains was inhibited by the cell-free supernatant of P. aeruginosa. To determine how S. aureus responds towards growth inhibition over time, I then performed an experimental evolution experiment, during which I passaged the three S. aureus strains for one month in the presence or absence of cell-free P. aeruginosa supernatant. Replicate populations from all three S. aureus strains rapidly adapted to the P. aeruginosa supernatant and no longer showed growth inhibition after evolution, indicating resistance. This pattern was accompanied by several, S. aureus strain-specific, phenotypic changes, indicating that each S. aureus strain adapted to P. aeruginosa-induced growth inhibition in an independent way. I corroborated this observation by whole genome sequencing of the 150phenotypically characterizedclones, where I found that most mutations that had appeared during evolution were unique to the specific treatment condition and to the S. aureus strain genetic background. Particularly interesting was the finding that many mutated genes belonged to the category ‘membrane transporter’. S. aureus membrane transporters are known to control virulence and antibiotic resistance and have been suggested as novel drug targets. The fact that the presence of P. aeruginosa inhibitory compounds selects for mutations in S. aureus membrane transporters, is therefore of clinical relevance and could affect treatment efficacy. On the eco-evolutionary level, the S. aureus adaptation mechanisms that I found in Project 2 could also explain why co-infections with P. aeruginosa and S. aureus and co-existence of the two species seem frequent despite strong antagonism in vitro. Most studies that examined interactions between P. aeruginosa and S. aureushave been performed with planktonic batch cultures. However, in infections, the two species frequently colonize surfaces and form microcolony aggregates that develop into biofilms. Interactions should therefore mainly take place at the boundaries of such microcolonies. In Project 3, I performed time-lapse fluorescence microscopy and assessed interactions between single P. aeruginosa and S. aureus cells in growing microcolonies. The solid substrate dramatically altered the competitive balance between the two species, and all three S. aureus strains were able to inhibit the growth of P. aeruginosa. Conversely, P. aeruginosa itself only had a minor effect on S. aureusgrowth, but upregulated quorum sensing gene expression in mixed microcolonies, consistent with the concept of ‘competition sensing’. The fact that there were, once more, strain-specific differences in interspecies behavior, indicates that S. aureus strains may vary in their extent to which they produce inhibitory compounds targeting P. aeruginosa on surfaces. Overall, Project 3 revealed that P. aeruginosa and S. aureus influence each other on surfaces in ways that were not anticipated from planktonic batch culture studies, and it reinforces the importance to apply an ecologically relevant context to study interspecies interactions. In a broader context, my three PhD thesis projects show that eco-evolutionary parameters significantly impact interaction patterns between P. aeruginosa and S. aureus, and that the relationship between the two pathogens is more complicated than ‘P. aeruginosa inhibits and displaces S. aureus’. My results therefore contribute to a better understanding of interactions between P. aeruginosa and S. aureus, two of the most important pathogens occurring in human polymicrobial infections