Introduction Pseudomonas (P.) aeruginosa is one of the most important causes of nosocomial infections and inhabiting a large number of environmental and host species. This distribution is not possible without extensive phenotypic adaptation. Isolates A total of 318 P. aeruginosa isolates were examined. These included 211 multi-resistant 4MRGN isolates, including 2 endemic strains, 38 sensitive human isolates, 22 animal isolates from dog ears, 45 water isolates from samples of the community water supply and the references PAO1 and PA14. Methods Biofilm formation was determined in a biofilm assay for spontaneous biofilm formation without addition of antibiotics and stress induced biofilm formation with 1 mg/L, 4 mg/L and 125 mg/L meropenem (MPM) respectively. A stress-response specific to the used concentration of MPM was also calculated. The in-vitro-fitness was determined in a competitive growth assay. The serum-sensitivity and PMN-sensitivity were determined in a killing assay with 30 min exposition to 5 % human serum or a PMN concentration of 10,000 cells/mL respectively. Furthermore, the genetically determined virulence factors toxA, exoS, exoT, exoU, exoY, cif, nan1, lasA and lasB were determined by PCR. Results The 4MRGN isolates, the animal isolates and the water isolates all showed significantly higher spontaneous biofilm formation than the sensitive human isolates. In each group, isolates showing an induction as well as an inhibition of biofilm formation as stress-response for all MPM concentrations, were observed. Inhibition almost always predominated, although on a lower scale for the 4MRGN isolates. A significantly higher in-vitro-fitness of the 4MRGN group in comparison to all other groups was observed. 4MRGN and animal isolates showed significantly higher serum-sensitivity than the water isolates and sensitive human isolates. Isolates with high serum-sensitivity showed also a significantly higher spontaneous biofilm formation. In the case of PMN-sensitivity, only the sensible human isolates showed significantly lower values than other groups. The genetically determined virulence factors cif, toxA and exoT were present in almost all isolates and lasA and exoY in most isolates. The genes exoS and exoU were exclusive, with a higher proportion of exoU observed in 4MRGN and water isolates. The nan1 gene was more frequently present in the water and sensible human isolates. The gene lasB was present less frequently in the 4MRGN group and sensible human isolate than in the animal and water isolates. Discussion The observed patterns of biofilm formation indicate, that an increase of biofilm formation is associated with a higher degree of adaptation. The ability to form extremely high amounts of biofilm by the two endemic strains underscores the link between biofilm formation and persistence. The variability in stress response shows that inducible biofilm formation is probably of greater importance in vivo than shown in literature. The significantly higher in-vitro-fitness of the 4MRGN isolates is remarkable. This is probably an intrinsic property of the isolates, which developed a 4MRGN profile despite competitive pressure. Groups with high serum-sensitivity seem to compensate this disadvantage with the ability to form higher amounts of biofilm. The differences in the phenotypic properties as well as in nan1 and lasB indicate an alteration in the virulence properties and the route of infection for the 4MRGN isolates. In contrast to the sensible clinical human isolates, there seems to be a shift in the route of infection towards a passive spread of biofilms. The similarity of the animal isolates to the 4MRGN isolates is possibly the result of advanced adaptation to the mammalian host. Through this similarity a zoonotic potential through amplification in the dog’s ear cannot be ruled out.weiterlesen