Microorganisms often dwell and interact in complex environmental
communities. These interactions may benefit multiple parties (e.g., protection in
biofilms) or just one member (e.g., metabolite exchange). However, interspecies
interactions can be antagonistic interactions in which one organism inhibits or kills
its neighbor, especially involving nutrient and niche competition. In either case,
polymicrobial interactions can alter the behavior, physiology, and persistence of
microbes impacting human infections (1).
Scientists have gained a greater appreciation for the constituents and causes of human
polymicrobial infections, including respiratory tract infections (1). In addition
to identifying the microbiota, studies have begun
to define the interactions between microbes in these infections. The cystic fibrosis
(CF) lung environment represents an ideal location for diverse microbes to interact.
Although diverse bacterial species colonize the airways of individuals with CF,
Pseudomonas aeruginosa and Staphylococcus aureus
are two key pathogens in terms of prevalence, sputum abundance, and associations with
lung disease (2–4). P. aeruginosa has long been considered the
dominant CF pathogen on the basis of circumstantial evidence. First, the U.S. CF
Foundation Patient Registry reports an inverse relationship between the prevalence
of
both species; specifically, S. aureus declines during late teenage
years, whereas P. aeruginosa increases (2) (Figure 1). Given the
relatively high S. aureus prevalence in children and young adults
(60–80%), P. aeruginosa is often believed to actively supplant
S. aureus in secretions (5,
6). Second, P. aeruginosa
outcompetes and suppresses S. aureus growth during in
vitro coculture by producing antistaphylococcal compounds (7), which are induced by
S.
aureus (8) and are detected in CF
sputum (9, 10). P. aeruginosa benefits from iron released after
S. aureus lysis (11),
potentially explaining why it targets S. aureus.
Figure 1.
Data from the U.S. Cystic Fibrosis Foundation Patient Registry showing the
prevalence of individuals with cystic fibrosis who had positive culture results
for a bacterial species according to age group during 2018. The age cohort is
presented on the x-axis, and the percentages of positive
culture results are displayed on the y-axis for the following
organisms: Pseudomonas aeruginosa, Staphylococcus
aureus (methicillin-susceptible), MRSA, Haemophilus
influenzae, Achromobacter species, Burkholderia
cepacia complex, and Stenotrophomonas maltophilia.
Although coinfections are not displayed, individuals may have been colonized
with multiple organisms. Reprinted by permission from Reference 2.
MRSA = methicillin-resistant S. aureus.
Based on this evidence, an assumption is that P. aeruginosa suppresses
S. aureus in the CF lung, shifting prevalence with age. However,
the Patient Registry figure depicts cross-sectional rather than longitudinal data
of
individuals and lacks quantitative culture results to measure changes in bacterial
abundance (Figure 1). Rather than succumbing to
P. aeruginosa antagonism, S. aureus may simply
disappear with age, either in response to host physiological changes or with treatment,
allowing other pathogens to occupy the space. S. aureus prevalence and
coinfection rates with P. aeruginosa have risen over the last decade
(2), which also challenges the assumed
in vivo dominance of P. aeruginosa over S.
aureus. Finally, in vitro coculture models involve cells
in a metabolically active state and in physical contact or immediate proximity to
each
other, both of which may not exist in vivo.
In this issue of the Journal, Fischer and colleagues (pp. 328–338) thoughtfully scrutinize
the dynamic relationship between
P. aeruginosa and S. aureus. The authors examined
retrospective, longitudinal, and quantitative culture data from people with CF who
regularly expectorated sputum (12). This large
collection of culture data provided a unique resource to assess the presence and
densities of these pathogens over an extended timeframe. Among patients with CF who
provided ≥10 sputum or BAL quantitative cultures over 13 years, a majority of
patients had cultures positive for each organism, high rates of simultaneous
coinfection, and high bacterial densities of each (median log10
colony-forming units/ml of 6.52 for P. aeruginosa and 6.42 for
S. aureus). This permitted longitudinal analyses of changes in
culture abundance after acquisition of the competing species or during simultaneous
coinfection. Contrary to the longstanding assumptions of P. aeruginosa
dominance, the authors found that S. aureus had stable, long-term
coexistence with P. aeruginosa in CF samples. Regardless of whether
S. aureus preceded the introduction of P.
aeruginosa or whether both organisms were cocultured early in the study,
S. aureus bacterial densities did not decline with time in the
presence of P. aeruginosa. Interestingly, coinfections actually
increased rather than decreasing over time, and replacement of S.
aureus by P. aeruginosa rarely occurred. In comparison,
Haemophilus influenzae, another early CF pathogen, did not compete
well against either organism. S. aureus is typically categorized
according to methicillin susceptibility (methicillin-susceptible S.
aureus [MSSA] and methicillin-resistant S. aureus [MRSA]).
National surveillance has shown an increase in MRSA culture positivity in U.S. patients
with CF from 2000 to 2010 (2). This increase was
even more pronounced at the authors’ center, raising doubts whether persistence
was related to changes in S. aureus susceptibility rather than
resilience to P. aeruginosa antagonism. To address this question, the
authors repeated their analyses for MSSA and MRSA individually and found both subtypes
had similar durations of infection and maintained high culture abundances when
coinfecting with P. aeruginosa, although MRSA sputum densities were
generally higher and persisted longer than MSSA.
Because these findings are provocative, a number of questions emerge. How generalizable
are these single site findings to the broader CF population and other centers? The
authors analyzed quantitative culture results from subjects who expectorated. These
subjects represented ∼40% of the clinic population, were relatively older, had
worse lung disease, and had higher P. aeruginosa culture rates than the
other patients with CF in the center. The interactions between P.
aeruginosa and S. aureus may differ in younger, healthier
patients with newly acquired bacteria. For example, persistent (late) CF P.
aeruginosa isolates lose their competitive advantage in
vitro over S. aureus compared with recently acquired
(early) P. aeruginosa isolates (13, 14). In addition, the
relatively high MRSA rates at this center raise doubts about whether treatment and
antibiotic susceptibility alter outcomes between P. aeruginosa and
S. aureus. A prospective, multicenter study encompassing a larger,
younger, and healthier patient population with lower MRSA prevalence and including
a
longitudinal linkage between microbiology and antibiotic usage would alleviate these
limitations.
Regardless of limitations, these results suggest that S. aureus is more
resilient than previously believed, but how does S. aureus coexist with
P. aeruginosa in vivo, given P. aeruginosa’s in
vitro dominance? These two species may be compartmentalized within the
airways and only mix on expectoration. The concept of microbial compartmentalization
in
the CF airway has previously been demonstrated (15). Alternatively, S. aureus may
adapt to the presence of
P. aeruginosa, facilitating its in vivo
coexistence. For example, P. aeruginosa and certain antibiotics select
for S. aureus small colony variants in vitro that are
tolerant to P. aeruginosa antagonism (9). The detection of small colony variants was
not evaluated in this
report.
Importantly, what does this information mean for the health and care of people with
CF?
Patients with CF coinfected with P. aeruginosa and S.
aureus reportedly have worse respiratory outcomes than those with single
infections (16). Characteristics associated
with persistence, pathogenesis, and response to therapy of each species were affected
by
interactions between these bacteria in vitro (7). Fischer and colleagues (12) conclude
that P. aeruginosa and S.
aureus can coinfect for much longer than previously anticipated. These
results suggest that concurrent treatments directed at both organisms may improve
CF
clinical outcomes.