Abstract

Three structural features of lipid A (addition of palmitate [C16 fatty acid], addition of aminoarabinose [positively charged amino sugar residue], and retention of 3-hydroxydecanoate [3-OH C10 fatty acid]) were determined for Pseudomonas aeruginosa isolates from patientswith cystic fibrosis (CF; n = 86), from the environment (n = 13), and from patients with other conditions (n = 14). Among P. aeruginosa CF isolates, 100% had lipid A with palmitate, 24.6% with aminoarabinose, and 33.3% retained 3-hydroxydecanoate. None of the isolates from the environment or from patients with other conditions displayed these modifications. These results indicate that unique lipid A modifications occur in clinical P. aeruginosa CF isolates.

Pseudomonas aeruginosa is a ubiquitous environmental gramnegative bacterium found in soil and water. This opportunistic pathogen can cause infections in individuals with host-defense defects, including those with cystic fibrosis (CF), tissue injury due to burns, or immunodeficiency due to cancer chemotherapy [1]. CF, the most common lethal genetic disease in Caucasians, is associated with chronic respiratory tract infections that are the principal cause of morbidity and mortality in this disorder [2]. P. aeruginosa is the predominant respiratory pathogen among individuals with CF and is associated with progressive pulmonary disease that often leads to death during the first 4 decades of life [2]. Successful adaptation of P. aeruginosa to the CF airway—essential for bacterial persistence, replication, and resistance to antibiotics—is poorly understood.

Our previous studies, which focused on 7 isolates from 4 patients with CF, 2 laboratory adapted isolates, and 5 isolates from patients with conditions other than CF, showed that an early adaptation of P. aeruginosa to the CF airway was synthesis of a modified lipid A, the bioactive component of lipopolysaccharide (LPS) [3]. CF-specific changes in P. aeruginosa lipid A structure included the addition of aminoarabinose and palmitate, leading to resistance to host innate immune defenses (such as cationic antimicrobial peptides) [3] and increased proinflammatory signaling [4]. Interestingly, these changes were observed in P. aeruginosa isolates from patients with CF under growth conditions (high magnesium concentrations) that normally repress lipid A modifications in vitro. Additionally, these modifications were stable, as they were not lost after continual passage in rich medium. These results suggest that lipid A modifications are constitutively and stably expressed in P. aeruginosa isolates from CF patients after growth in vitro. In contrast, in isolates from persons with acute infection (blood) and in environmental isolates, lipid A modifications were induced in vitro under magnesium-limited conditions but were repressed under magnesium-replete conditions. P. aeruginosa clinical isolates from patients with CF who had severe lung disease displayed an additional lipid A modification—namely, retention of 3-hydroxydecanoic acid at the lipid A 3-O position, presumably due to lack of expression of PagL, a recently identified 3-O position lipid A deacylase [5]. This lipid A modification was associated with enhanced resistance to β-lactam antibiotics but not aminoglycosides [6]. These findings suggest that, in a limited number of clinical isolates, the synthesis of specific lipid A structures is important to P. aeruginosa infection of CF airways

The present study was designed to determine the prevalence of CF-associated P. aeruginosa lipid A modifications (addition of palmitate, addition of aminoarabinose, and retention of 3- hydroxydecanoate) among P. aeruginosa strains from patients with chronic CF airway infection, from patients with other acute or chronic infections, or from the environment.

Materials and methods.P. aeruginosa strains were obtained from the following sources: (1) patients <3 years old with a clinical diagnosis of CF who were participating in a multicenter study (in Washington, Texas, and Ohio) of early airway infection and inflammation [7]; and (2) patients >6 years old with a clinical diagnosis of CF and either mild or severe lung disease who were participating in a 2-center study (in Washington and Ohio) of lung disease severity. The appropriate institutional review boards reviewed and approved the use of medical records and specimens for both studies, and patients or their legal guardians gave informed consent. Non-CF clinical isolates (bronchiectasis) were from J.L.B., acute infection isolates (blood, ear, eye, urinary tract infection) were from S. Lory (Harvard University), and environmental isolates were from the American Type Culture Collection and D. Speert (University of British Columbia). A complete strain list is available on request.

Bacterial cells (pelleted from 10 mL) for LPS analysis were obtained after overnight growth with aeration in Luria-Bertani broth supplemented with 1 mmol/L MgCl2, a condition known to repress palmitoylation and aminoarabinosylation of lipid A in laboratory-adapted P. aeruginosa strains.

LPS was isolated using the rapid small-scale isolation method for mass spectrometry (MS) analysis, as described elsewhere [8]. Briefly, 1.0 mL of TRI Reagent (Molecular Research Center) was added to the cell culture pellet (10 mL of an overnight culture), resuspended, and incubated at room temperature for 15 min. Two hundred microliters of chloroform was added, vortexed, and incubated at room temperature for 15 min. Samples were centrifuged for 10 min at 13,400 g, and the aqueous layer was removed. Five hundred microliters of water was added to the lower layer and vortexed well. After 15–30 min, the sample was spun down, and the aqueous layers were combined. The process was repeated 2 more times. The combined aqueous layers were lyophilized overnight.

Lipid A was isolated after hydrolysis in 1% SDS at pH 4.5, as described elsewhere [9]. Briefly, 500 mL of 1% SDS in 10 mmol/L Na-acetate (pH 4.5) was added to a lyophilized LPS sample. Samples were incubated at 100°C for 1 h and lyophilized. The dried pellets were washed in 100 µL of water and 1 mL of acidified ethanol (100 mL of 4N HCl in 20 mL of 95% ethanol [EtOH]). Samples were centrifuged at 2300 g for 5 min. The lipid A pellet was further washed (3 times) in 1 mL of 95% EtOH. The entire series of washes was repeated twice. Samples were resuspended in 500 µL of water, frozen on dry ice, and lyophilized.

Negative-ion matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) experiments were performed as described elsewhere [3, 10], with the following modifications. Lyophilized lipid A was dissolved with 10 µL of 5-chloro-2- mercaptobenzothiazole MALDI matrix in chloroform/methanol at 1:1 (vol/vol) and then applied (1 µL) onto the sample plate. All MALDI-TOF experiments were performed using a Bruker Autoflex II MALDI-TOF mass spectrometer (Bruker Daltonics). Each spectrum was an average of 300 shots. ES Tuning Mix (Agilent) was used as an external calibrant for the MALDI-TOF analysis. Lipid A structural modifications identified from individual mass spectra included the addition of palmitate (m/z [mass to charge ratio], 1685), addition of aminoarabinose (m/z, 1578 or 1748), and loss of deacylation of the 3 position fatty acid (m/z, 1855) (see figure 1 for individual lipid A structures).

The prevalence of each structural modification was determined among patients with CF and among patients with conditions other than CF. Prevalence was determined as the proportion of affected patients in each group, and a corresponding binomial exact 95% confidence interval was calculated. For each structural modification, prevalence was compared between the CF and non-CF groups by performing a binomial exact unconditional test based on the χ2 statistic to obtain a 2-sided P value.

Results. A total of 113 nonduplicate P. aeruginosa isolates were obtained from patients with CF, patients with acute (blood, ear, eye, or urinary tract) or chronic (bronchiectasis) infection, and the environment (soil, vegetables, water, or aquatic life). Negative-ion MALDI-TOF MS was used to analyze lipid A structures isolated from these strains. Individual isolates were analyzed to determine the presence of 3 specific lipid A modifications: addition of palmitate (C16 fatty acid), addition of aminoarabinose (positively charged amino sugar residue), or retention of 3-hydroxydecanoate (3-OH C10 fatty acid). Examples of individual lipid A structures are shown in figure 1.

The prevalence of the individual lipid A modifications is shown in table 1. Lipid A purified from all 86 isolates from 58 patients with CF was positive for palmitate addition. In the CF infant cohort, this modification was observed in clinical isolates from very young children with CF (all initial isolates from patients <1 year of age, with the earliest isolate from a 4-monthold patient), suggesting that this adaptation to the CF airway occurs very quickly after colonization. Aminoarabinose modification was observed in 40% (44% of the individual isolates), 38% (35% of the individual isolates), and 25% (17% of the individual isolates) of the patients in the infant CF, mild CF, and severe CF cohorts, respectively.

Discussion. This survey of lipid A structure among CF and non-CF strains of P. aeruginosa has defined specific structural modifications that are absolutely, relatively, or temporally associated with CF lung infection. The addition of palmitate to lipid A was always found in CF strains of this organism (P < .0001), whereas aminoarabinose modification was present in less than half of these strains (P = .018) after growth in vitro under conditions that repress these lipid A modifications. Interestingly, loss of lipid A deacylase activity was seen only in strains from patients with CF who had more severe lung disease.

The present results indicate that P. aeruginosa isolates from patients with CF synthesize lipid A that is distinct from strains that have been isolated from other acute or chronic human disease sites or from the environment. These results indicate that adaptation to the CF airway environment, specifically the addition of palmitate and (for some strains) aminoarabinose, occurs soon after colonization of the airway, in contrast to other CF-associated P. aeruginosa phenotypes, which are typically noted later during the course of infection. These classic P. aeruginosa phenotypic characteristics include loss of flagellar-dependent motility, LPS changes (including loss of O-antigen), auxotrophy, decreased secretion of virulence factors, inability to produce pyocyanin and phage, antibiotic resistance, mucoidy, and formation of biofilms [11, 12]. Infection with mucoid P. aeruginosa is associated with accelerated progression of CF lung disease [13]. Despite this association, mucoid P. aeruginosa strains are also observed in other chronic diseases, including bronchiectasis (J.L.B., unpublished data) [14], and in P. aeruginosa isolated from urinary catheters [15]; these infections typically resolve or otherwise follow a more benign course. Thus, mucoidy may merely be a marker for coordinately regulated virulence determinants or alternatively may be a surrogate for the chronicity and sputum density of P. aeruginosa. Finally, loss of deacylase activity, associated with P. aeruginosa strains from patients with CF who have severe lung disease, may be an additional late phenotypic change in the bacterial surface structure.

Aminoarabinose modification of P. aeruginosa lipid A is of particular relevance to resistance to colistin, a specific isoform of the cationic antimicrobial peptide polymyxin commonly used to treat CF lung infection [16]. Although aminoarabinose modification alone—at least at the levels present in the early P. aeruginosa isolates from patients with CF—does not appear to be sufficient to confer this resistance, lipid A from colistin-resistant CF strains of P. aeruginosa is consistently found to be modified with aminoarabinose (S.M.M. and R.K.E., unpublished data). The results presented here suggest that some P. aeruginosa strains present in the CF population may be predisposed to develop colistin resistance before drug exposure, by virtue of preexisting aminoarabinose modification of lipid A.

In conclusion, the results of this study of the prevalence of specific lipid A modifications in P. aeruginosa clinical and environmental isolates provide a more complete picture of early P. aeruginosa adaptation to the CF airway. These modifications alter the bacterial cell surface and, thus, recognition by the host innate immune system, leading to chronic colonization of the airways of patients with CF. Further study of the regulation of enzymes required for the modifications of P. aeruginosa lipid A will be required to determine the mechanism by which CF strains of P. aeruginosa modify lipid A as part of the adaptation to the CF airway environment.

Acknowledgement

We thank Anahita Kivand and Ilana Cohen for critical review of the manuscript.

References

1.
Lyczak
JB
Cannon
CL
Pier
GB
Establishment of Pseudomonas aeruginosa infection: lessons from a versatile opportunist
Microbes Infect
 , 
2000
, vol. 
2
 (pg. 
1051
-
60
)
2.
Gibson
RL
Burns
JL
Ramsey
BW
Pathophysiology and management of pulmonary infections in cystic fibrosis
Am J Respir Crit Care Med
 , 
2003
, vol. 
168
 (pg. 
918
-
51
)
3.
Ernst
RK
Yi
EC
Guo
L
, et al.  . 
Specific lipopolysaccharide found in cystic fibrosis airway Pseudomonas aeruginosa
Science
 , 
1999
, vol. 
286
 (pg. 
1561
-
5
)
4.
Hajjar
AM
Ernst
RK
Tsai
JH
Wilson
CB
Miller
SI
Human Toll-like receptor 4 recognizes host-specific LPS modifications
Nat Immunol
 , 
2002
, vol. 
3
 (pg. 
354
-
9
)
5.
Geurtsen
J
Steeghs
L
Hove
JT
van der Ley
P
Tommassen
J
Dissemination of lipid A deacylases (PagL) among gram-negative bacteria: identification of active-site histidine and serine residues
J Biol Chem
 , 
2005
, vol. 
280
 (pg. 
8248
-
59
)
6.
Ernst
RK
Adams
KN
Moskowitz
SM
, et al.  . 
The Pseudomonas aeruginosa lipid A deacylase: selection for expression and loss within the cystic fibrosis airway
J Bacteriol
 , 
2006
, vol. 
188
 (pg. 
191
-
201
)
7.
Rosenfeld
M
Gibson
RL
McNamara
S
, et al.  . 
Early pulmonary infection inflammation and clinical outcomes in infants with cystic fibrosis
Pediatr Pulmonol
 , 
2001
, vol. 
32
 (pg. 
356
-
66
)
8.
Yi
EC
Hackett
M
Rapid isolation method for lipopolysaccharide and lipid A from gram-negative bacteria
Analyst
 , 
2000
, vol. 
125
 (pg. 
651
-
6
)
9.
Caroff
M
Tacken
A
Szabo
L
Detergent-accelerated hydrolysis of bacterial endotoxins and determination of the anomeric configuration of the glycosyl phosphate present in the “isolated lipid A” fragment of the Bordetella pertussis endotoxin
Carbohydr Res
 , 
1988
, vol. 
175
 (pg. 
273
-
82
)
10.
Guo
L
Lim
KB
Poduje
CM
, et al.  . 
Lipid A acylation and bacterial resistance against vertebrate antimicrobial peptides
Cell
 , 
1998
, vol. 
95
 (pg. 
189
-
98
)
11.
Nguyen
D
Singh
PK
Evolving stealth: genetic adaptation of Pseudomonas aeruginosa during cystic fibrosis infections
Proc Natl Acad Sci USA
 , 
2006
, vol. 
103
 (pg. 
8305
-
6
)
12.
Smith
EE
Buckley
DG
Wu
Z
, et al.  . 
Genetic adaptation by Pseudomonas aeruginosa to the airways of cystic fibrosis patients
Proc Natl Acad Sci USA
 , 
2006
, vol. 
103
 (pg. 
8487
-
92
)
13.
Li
Z
Kosorok
MR
Farrell
PM
, et al.  . 
Longitudinal development of mucoid Pseudomonas aeruginosa infection and lung disease progression in children with cystic fibrosis
JAMA
 , 
2005
, vol. 
293
 (pg. 
581
-
8
)
14.
Pujana
I
Gallego
L
Martin
G
Lopez
F
Canduela
J
Cisterna
R
Epidemiological analysis of sequential Pseudomonas aeruginosa isolates from chronic bronchiectasis patients without cystic fibrosis
J Clin Microbiol
 , 
1999
, vol. 
37
 (pg. 
2071
-
3
)
15.
Goto
T
Nakame
Y
Nishida
M
Ohi
Y
Bacterial biofilms and catheters in experimental urinary tract infection
Int J Antimicrob Agents
 , 
1999
, vol. 
11
 (pg. 
227
-
31
)
16.
Moskowitz
SM
Ernst
RK
Miller
SI
PmrAB a two-component regulatory system of Pseudomonas aeruginosa that modulates resistance to cationic antimicrobial peptides and addition of aminoarabinose to lipid A
J Bacteriol
 , 
2004
, vol. 
186
 (pg. 
575
-
9
)

Figures and Tables

Figure 1.

Proposed pathway for synthesis of cystic fibrosis-specific Pseudomonas aeruginosa lipid A. Addition of palmitate (Δm/z, +238; structures C and E), addition of aminoarabinose (Δm/z, +131; structure D; shown as either penta- [loss of 3-OH C10 in shaded oval] or hexa-acylated), and deacylation (or removal) of the 3 position fatty acid (Δm/z, -170; structures B and C) to the base hexa-acylated lipid A structure (structure A) are shown. For all structures, the molecular weight (m/z, or mass to charge ratio) of the singly charged lipid A species is indicated.

Figure 1.

Proposed pathway for synthesis of cystic fibrosis-specific Pseudomonas aeruginosa lipid A. Addition of palmitate (Δm/z, +238; structures C and E), addition of aminoarabinose (Δm/z, +131; structure D; shown as either penta- [loss of 3-OH C10 in shaded oval] or hexa-acylated), and deacylation (or removal) of the 3 position fatty acid (Δm/z, -170; structures B and C) to the base hexa-acylated lipid A structure (structure A) are shown. For all structures, the molecular weight (m/z, or mass to charge ratio) of the singly charged lipid A species is indicated.

Table 1.

Prevalence of lipid A structural modifications: comparison between patientswith cystic fibrosis (CF) and those with conditions other than CF (non-CF patients).

Table 1.

Prevalence of lipid A structural modifications: comparison between patientswith cystic fibrosis (CF) and those with conditions other than CF (non-CF patients).

Potential conflicts of interest: none reported.
Financial support: National Institutes of Health (grant AI047938 to R.K.E., grants K08HL67903 and M01RR-00037 to S.M.M., and grant K064954 to S.I.M.); Cystic Fibrosis Foundation (support to R.K.E. and grant MOSKOW01A1 to S.M.M.); Canadian Cystic Fibrosis Foundation (support to D.P.S.)