In a previous EMR post from December 30 of 2014 (see link below), I discussed a study (Science, 2014) that offered evidence for reciprocal selection of host and pathogen iron-binding proteins arising out the competition for their shared ligand, which is critical to the metabolisms of both parties to the conflict. A recent paper (J. Bacteriol., 2015) by Filkins et al. demonstrates another sort of competition focused on the acquisition of iron that can affect human health. This conflict occurs between two species of bacterial pathogen associated with lung disease in cystic fibrosis (CF) patients.

A number of pathogenic and non-pathogenic bacteria can colonize the lungs of CF patients. Nevertheless, two pathogens dominate these often polymicrobial populations in the setting of CF: Staphylococcus aureus and Pseudomonas aeruginosa. S. aureus (Sa) is the most common pathogen found in pediatric CF patients, but P. aeruginosa (Pa) is the most common pathogen in adult CF patients, with the transition often beginning in late adolescence.

The authors chose to investigate this ecological transition in the predominant bacterial pathogen in the pulmonary tract of CF patients by using two in vitro systems. In the first, the pathogens (Sa and Pa; one, the other, or both) were cultured on a bronchial epithelial cell line carrying the most common CF-related mutation, deltaF508, in the cystic fibrosis transmembrane conductance regulator (CFTR) in homozygous form. The second culture system used a plastic substrate for one, the other, or both pathogens. Results were in many cases checked with alternative Sa and Pa strains to assess the generality of the findings.

In the initial set of experiments, Filkins et al. showed that Sa and Pa can coexist for some hours but by 16 hours, the numbers of Sa are substantially reduced, with greater loss of Sa on the epithelial cell line than on plastic. Sa killing proceeded for both biofilm and planktonic forms of Sa and could be mediated by both mucoid and non-mucoid strains of Pa.

Filkins et al. then demonstrated that exposure of Sa to Pa was associated with increased transcription of Sa genes involved in fermentation-related metabolic pathways. The Sa in such co-cultures exhibited less aerobic respiration and more lactic acid fermentation. Under these conditions, the population of Sa grew more slowly or decreased in numbers more quickly than in the absence of Pa.

The authors also demonstrated that the presence of Pa selected for small colony variants (SCV) of Sa. These variants clearly can better survive the chemical attack from Pa but grow relatively slowly. SCV Sa bacteria have mutations that inactivate the electron transport chain.

In additional experiments, the Pa bacteria were found to produce two siderophores, small molecules (pyoverdine and pyochelin in this particular instance) that bind iron, and 2-heptyl-4-hydroxyquinoline N-oxide (HQNO). Together, the authors suggest that these products of Pa are responsible for the important changes in Sa metabolism and viability. Another factor accounting for the effects of Pa on Sa may be the direct consumptition of oxygen. Additional mechanisms, such as competition for unidentified micronutrients or direct production of antimicrobial compounds, remain to be fully assessed. Host factors, such as immune responses, could not be evaluated in the in vitro experimental systems employed in these studies.

Experiments, in which genes essential for the synthesis of pyoverdine, pyochelin, and HQNO were individually or collectively deleted from the Pa genome, were used to address the relative roles of these metabolites. The authors concluded that killing of Sa by Pa was more dependent on pyoverdine, pyochelin, and HQNO on plastic surfaces than on the epithelial cells. Additional research will be required to evaluate the roles of these mechanisms of Pa-mediated killing of Sa in the lungs of CF patients.

The unsurprising existence of biologically consequential interactions among microbial cells, in this case of two clearly pathogenic species, in human tissues, as suggested by Filkins et al., prompts recognition of the enormous complexity that might characterize the relationships between host and microbiome in such locations as the gut or the skin. A reasonable inference is that the role played by any given bacterial species in the microbiome, relative to the host, could depend on the presence and numbers of one or many other microbial species.


Greenspan, N. Cellular ‘gold’: competition for iron as the cause of reciprocal positive selection of host and pathogen iron-binding proteins.

Barber MF, Elde NC. Nutritional immunity. Escape from bacterial iron piracy through rapid evolution of transferrin. Science. 2014 Dec 12;346(6215):1362-6. doi: 10.1126/science.1259329. PubMed PMID: 25504720.

Filkins LM, Graber JA, Olson DG, Dolben EL, Lynd LR, Bhuju S, O’Toole GA. Coculture of Staphylococcus aureus with Pseudomonas aeruginosa drives S. aureus towards fermentative metabolism and reduced viability in a cystic fibrosis model. J Bacteriol. 2015 Jul;197(14):2252-64. doi: 10.1128/JB.00059-15. Epub 2015 Apr 27. PubMed PMID: 25917910.