Association of Diverse Staphylococcus aureus Populations with Pseudomonas aeruginosa Coinfection and Inflammation in Cystic Fibrosis Airway Infection

Abstract

Staphylococcus aureus is one of the most common pathogens isolated from the airways of cystic fibrosis (CF) patients and often persists for extended periods. There is limited knowledge about the diversity of S. aureus in CF. We hypothesized that increased diversity of S. aureus would impact CF lung disease. Therefore, we conducted a 1-year observational prospective study with 14 patients with long-term S. aureus infection. From every sputum, 40 S. aureus isolates were chosen and characterized in terms of phenotypic appearance (size, hemolysis, mucoidy, and pigmentation), important virulence traits such as nuclease activity, biofilm formation, and molecular typing by spa sequence typing. Data about coinfection with Pseudomonas aeruginosa and clinical parameters such as lung function, exacerbation, and inflammatory markers in blood (C-reactive protein [CRP], interleukin 6 [IL-6], and S100A8/9 [calprotectin]) were collected. From 58 visits of 14 patients, 2,319 S. aureus isolates were distinguished into 32 phenotypes (PTs) and 50 spa types. The Simpson diversity index (SDI) was used to calculate the phenotypic and genotypic diversity, revealing a high diversity of PTs ranging from 0.19 to 0.87 among patients, while the diversity of spa types of isolates was less pronounced. The SDI of PTs was positively associated with P. aeruginosa coinfection and inflammatory parameters, with IL-6 being the most sensitive parameter. Also, coinfection with P. aeruginosa was associated with mucoid S. aureus and S. aureus with high nuclease activity. Our analyses showed that in CF patients with long-term S. aureus airway infection, a highly diverse and dynamic S. aureus population was present and associated with P. aeruginosa coinfection and inflammation. IMPORTANCE Staphylococcus aureus can persist for extended periods in the airways of people with cystic fibrosis (CF) in spite of antibiotic therapy and high numbers of neutrophils, which fail to eradicate this pathogen. Therefore, S. aureus needs to adapt to this hostile niche. There is only limited knowledge about the diversity of S. aureus in respiratory specimens. We conducted a 1-year prospective study with 14 patients with long-term S. aureus infection and investigated 40 S. aureus isolates from every sputum in terms of phenotypic appearance, nuclease activity, biofilm formation, and molecular typing. Data about coinfection with Pseudomonas aeruginosa and clinical parameters such as lung function, exacerbation, and inflammatory markers in blood were collected. Thirty-two phenotypes (PTs) and 50 spa types were distinguished. Our analyses revealed that in CF patients with long-term S. aureus airway infection, a highly diverse and dynamic S. aureus population was associated with P. aeruginosa coinfection and inflammation.

Keywords: Pseudomonas aeruginosa; Staphylococcus aureus; cystic fibrosis; diversity; persistent infection.

FIG 1

Pictures of the seven most abundant phenotypes (PTs). S. aureus strains were cultured on Columbia blood agar and Congo red agar (CRA) plates and are characterized by the following parameters: size, mucoidy, hemolysis, β-toxin, and pigmentation. CRA was used to facilitate the discrimination of mucoid isolates. Special characteristics for mucoid isolates on CRA are a pyramidal-shaped morphology with rough wrinkled edges and a dry crystalline consistency (PT30 and -15), while nonmucoid isolates appear as flat and smooth colonies with a more moist consistency (PT4, -7, -1, -6, and -22). PT4 is normal (n), nonmucoid (nm), hemolytic (h), β-toxin negative (-), and gray (g). PT7 is normal, nonmucoid, hemolytic, β-toxin positive (+), and gray. PT1 is normal, nonmucoid, nonhemolytic (nh), β-toxin negative, and gray. PT6 is normal, nonmucoid, hemolytic, β-toxin negative, and yellow (y). PT30 is an SCV, mucoid (m), hemolytic, β-toxin negative, and gray. PT22 is an SCV, nonmucoid, hemolytic, β-toxin negative, and gray. PT15 is normal, mucoid, hemolytic, β-toxin negative, and yellow.

FIG 2

spa clonal complexes (spa-CCs) of S. aureus isolates in the prospective study. This figure demonstrates the clonal relatedness of all 50 spa types of the 2,319 S. aureus isolates from 14 CF patients during the study period (two CF centers). The analysis was performed using the BURP algorithm as implemented in the Ridom StaphType software (BURP, Ridom StaphType software; Ridom GmbH, Würzburg, Germany). The main founder spa type is shown in blue in the middle of each cluster. The size of each circle demonstrates the number of isolates with the respective spa type. Colored loops surrounding the circles indicate the patient with the respective spa type. For CC6 to CC9 isolates, the software could not determine a founder spa type, and therefore, the isolates with these respective spa types are presented as “no founder.” Eight spa types could not be related to other spa types and were therefore determined as “singletons.” t005, t084, t091, t267, t1204, t2202, t3127, and t13342 were singletons. spa types with less than five repeats were excluded, which resulted in the exclusion of one spa type (t463).

FIG 3

Biofilm formation and nuclease activity for all isolates and for all patients. This figure shows the amount of biofilm formation (A) and nuclease activity (B) for all S. aureus isolates from all patients at all visits. (A) We arbitrarily determined the categories low and high biofilm formation if the isolates formed less or more than 10% of the biofilm of the positive control (RP62A, a coagulase-negative Staphylococcus that produces large amounts of biofilm compared to S. aureus). In patients 2 to 6, no high biofilm-forming isolates have been observed, while an increasing biofilm formation of isolates was visible for patients 1, 12, 13, and 14. In patients 8 and 11, a decrease of biofilm formation was determined, while from patients 7, 9, and 10, a large percentage of high biofilm-forming isolates over all visits were detected. (B) For the nuclease activity, we distinguished groups with low and high nuclease activity if less or more than 100% of the nuclease activity of the positive control (AH1263) were measured. Half of the patients (patients 6 to 10, 12, and 13) carried more than 92% of isolates with high nuclease activity, while in the other patients, only small numbers of isolates with high nuclease activity were detected. In patient 5 only, no isolates with high nuclease activity were recovered.

FIG 4

Heatmaps of S. aureus isolates of patients 1 and 9 with inflammatory and lung function data of every visit. The heatmaps illustrate all data for 40 S. aureus isolates collected during every visit in terms of spa type (dominant spa type/related to the dominant spa type/not related), phenotype (PT) (mucoidy, hemolysis, β-toxin, SCV phenotype [all positive/negative], pigment [gray/white/yellow] as well as biofilm formation [low < 10% / high > 10% of the positive control] and nuclease activity [low < 100% / high > 100% of the positive control]). Below the heatmap, data of all visits are reported for months from first visit, exacerbation, P. aeruginosa coinfection, SDI for spa type and PTs. Data for S100A8/9, CRP, IL-6, and FEV₁%pred are given for every visit in a table and illustrated as a graph with visits of exacerbation marked with a red arrow.

FIG 4

Heatmaps of S. aureus isolates of patients 1 and 9 with inflammatory and lung function data of every visit. The heatmaps illustrate all data for 40 S. aureus isolates collected during every visit in terms of spa type (dominant spa type/related to the dominant spa type/not related), phenotype (PT) (mucoidy, hemolysis, β-toxin, SCV phenotype [all positive/negative], pigment [gray/white/yellow] as well as biofilm formation [low < 10% / high > 10% of the positive control] and nuclease activity [low < 100% / high > 100% of the positive control]). Below the heatmap, data of all visits are reported for months from first visit, exacerbation, P. aeruginosa coinfection, SDI for spa type and PTs. Data for S100A8/9, CRP, IL-6, and FEV₁%pred are given for every visit in a table and illustrated as a graph with visits of exacerbation marked with a red arrow.

FIG 4

Heatmaps of S. aureus isolates of patients 1 and 9 with inflammatory and lung function data of every visit. The heatmaps illustrate all data for 40 S. aureus isolates collected during every visit in terms of spa type (dominant spa type/related to the dominant spa type/not related), phenotype (PT) (mucoidy, hemolysis, β-toxin, SCV phenotype [all positive/negative], pigment [gray/white/yellow] as well as biofilm formation [low < 10% / high > 10% of the positive control] and nuclease activity [low < 100% / high > 100% of the positive control]). Below the heatmap, data of all visits are reported for months from first visit, exacerbation, P. aeruginosa coinfection, SDI for spa type and PTs. Data for S100A8/9, CRP, IL-6, and FEV₁%pred are given for every visit in a table and illustrated as a graph with visits of exacerbation marked with a red arrow.

FIG 5

Simpson diversity index (SDI) of phenotypes (PTs). The SDIs of PTs at every patients’ visit are shown to present the dynamics of diversity.

FIG 6

Lung function of patients correlated with inflammatory markers S100 A8/A9, CRP, and IL-6. Data of the patients’ lung function (FEV₁% predicted) were collected at every visit and correlated with the following inflammatory markers from patients’ serum S100A8/A9 (A), CRP (B), and IL-6 (C). Applying GEE, the linear fit line shows a negative correlation for all inflammatory parameters with the FEV₁%pred revealing that decreased lung function is associated with higher inflammatory parameters (P < 0.0001).

FIG 7

Association of inflammatory markers S100 A8/A9, CRP, and IL-6 and SDI of phenotypes (PTs). Inflammatory markers from patients’ sera for CRP (A), IL-6 (B), and S100A8/A9 (C), which were determined at every visit, were correlated with the SDI of PTs by GEE. The most sensitive parameter was IL-6, for which a positive association with the SDI of PTs (P = 0.001) was observed, while there was no clear trend for S100A8/9 and CRP.

FIG 8

Increasing nuclease activity in β-toxin-positive S. aureus isolates. Nuclease activity was measured for all S. aureus isolates and correlated with β-toxin status of isolates as exemplified here for isolates of patients 8 and 12. The graph distinguishes β-toxin-negative and -positive isolates, showing an increase of nuclease activity in β-toxin-positive isolates over time (P < 0.001), as analyzed by GEE.

FIG 9

P. aeruginosa coinfection associated with mucoid growth and nuclease activity of S. aureus isolates. (A) The pie charts show that a new coinfection with P. aeruginosa was associated with a lower number of mucoid S. aureus isolates compared to patients with no coinfection, while in patients with chronic P. aeruginosa coinfection, the highest number of mucoid S. aureus isolates were recovered. (B) Also, the number of S. aureus isolates with high nuclease activity was highest in patients with chronic P. aeruginosa coinfection.