The Pseudomonas aeruginosa Lipid A Deacylase: Selection for Expression and Loss within the Cystic Fibrosis Airway
http://www.100md.com
细菌学杂志 2006年第1期
Departments of Medicine,Pediatrics,Microbiology,Genome Sciences, University of Washington, Seattle, Washington 98195,Department of Biochemistry and Cell Biology, National Institute of Infectious Diseases, Tokyo 162-8640, Japan,Department of Microbiology, J. H. Quillen College of Medicine, Johnson City, Tennessee 37614
ABSTRACT
Lipopolysaccharide (LPS) is the major surface component of gram-negative bacteria, and a component of LPS, lipid A, is recognized by the innate immune system through the Toll-like receptor 4/MD-2 complex. Pseudomonas aeruginosa, an environmental gram-negative bacterium that opportunistically infects the respiratory tracts of patients with cystic fibrosis (CF), can synthesize various structures of lipid A. Lipid A from P. aeruginosa strains isolated from infants with CF has a specific structure that includes the removal of the 3 position 3-OH C10 fatty acid. Here we demonstrate increased expression of the P. aeruginosa lipid A 3-O-deacylase (PagL) in isolates from CF infants compared to that in environmental isolates. PagL activity was increased in environmental isolates by growth in medium limited for magnesium and decreased by growth at low temperature in laboratory-adapted strains of P. aeruginosa. P. aeruginosa PagL was shown to be an outer membrane protein by isopycnic density gradient centrifugation. Heterologous expression of P. aeruginosa pagL in Salmonella enterica serovar Typhimurium and Escherichia coli resulted in removal of the 3-OH C14 fatty acid from lipid A, indicating that P. aeruginosa PagL recognizes either 3-OH C10 or 3-OH C14. Finally, deacylated lipid A species were not observed in some clinical P. aeruginosa isolates from patients with severe pulmonary disease, suggesting that loss of PagL function can occur during long-term adaptation to the CF airway.
INTRODUCTION
The interaction between the host airway environment and specific bacterial pathogens is critical to the development of airway disease in cystic fibrosis (CF), the most common lethal autosomal recessive genetic disease of Caucasians (30, 35). Shortly after birth, the respiratory tracts of most patients with CF are infected with the opportunistic gram-negative bacterium Pseudomonas aeruginosa (6, 11, 34). P. aeruginosa adapts to the environment of the CF airway with the accumulation of a variety of mutations that lead to mucoidy, the formation of biofilms, and the alteration of lipopolysaccharide (LPS) (13, 16, 17, 45). These adaptations are associated with chronic airway inflammation that leads to severe and progressive pulmonary dysfunction, resulting in premature death. This inflammatory process results, at least in part, from stimulation of the innate immune system by CF-specific modifications of P. aeruginosa lipid A. Lipid A is the bioactive component of LPS, a pathogenic factor of gram-negative bacteria that consists of three distinct regions: O antigen, core, and lipid A (1, 17, 25). Both O antigen and core consist of polysaccharide chains, whereas lipid A consists of fatty acid and phosphate groups bonded to a glucosamine disaccharide (37, 42).
Like lipid A of other gram-negative bacteria, P. aeruginosa lipid A contains penta- and/or hexa-acylated structures that consist of a (1,6)-linked disaccharide of glucosamine with phosphate groups at the 1 and 4' positions, amide-linked fatty acids at the 2 and 2' positions, and ester-linked fatty acids at the 3 and 3' positions (Fig. 1A) (2, 17, 27). The fatty acids attached to the P. aeruginosa glucosamine residues are shorter (C10 and C12 versus C12 and C14) than those in enterobacterial lipid A. P. aeruginosa further modifies its lipid A structure by the addition of secondary or "piggyback" fatty acids (C12 and 2-OH C12) to generate a hexa-acylated structure (a mass to charge ratio [m/z] equal to 1,616; Fig. 1C). This hexa-acylated lipid A structure can be modified by the removal or deacylation of the fatty acid (3-OH C10) at the 3 position to generate a penta-acylated structure (m/z = 1,447; Fig. 1D). Finally, this second penta-acylated structure can be modified by the acyloxyacyl addition of a C16 fatty acid at the 3' position to generate a hexa-acylated form (m/z = 1,685; Fig. 1E) of lipid A (16). This palmitate-containing hexa-acylated structure was observed in lipid A isolated from all P. aeruginosa clinical isolates from individuals with CF after growth in magnesium-replete medium, while it was absent from laboratory-adapted strains (PAO-1, PAK, and PA14) grown under the same conditions (data not shown).
P. aeruginosa isolated from either the upper or lower airway of young children with CF synthesized a more highly acylated lipid A (hexa-acylated, m/z = 1,685; Fig. 1E), while P. aeruginosa isolated from non-CF patients with acute infections synthesized a penta-acylated form (m/z = 1,419; Fig. 1B) (17). Lipid A structural modifications observed specifically in LPS from P. aeruginosa clinical isolates from patients with CF include the addition of 2-hydroxy laurate (2-OH C12), deacylation of the 3-position fatty acid, and addition of palmitate (C16) leading to mass spectrometric detection of ions with m/z ratios of 1,616 (Fig. 1C), 1,447 (Fig. 1D), and 1,685 (Fig. 1E), respectively. These changes in acylation are associated with resistance to host antimicrobial peptides and an alteration in the innate inflammatory response by way of Toll-like receptor 4 signaling (16, 17, 25). The fact that these modifications were not observed in acute clinical isolates suggests that environmental organisms, the sources for acute infections, did not express the enzymes required for these changes.
Environmentally regulated LPS modification enzymes have been observed in a wide variety of pathogenic and nonpathogenic gram-negative bacterial species. Interestingly, each bacterial species has unique regulatory and biosynthetic mechanisms that nevertheless result in similar lipid A modifications (15, 37, 42). Lipid A modifications in P. aeruginosa are regulated by two-component regulatory systems PhoP/PhoQ and PmrA/PmrB. The P. aeruginosa PhoP/PhoQ regulatory system is required for addition of palmitate and aminoarabinose in response to magnesium-limiting growth conditions (17, 28, 29, 32). The addition of aminoarabinose can also be regulated by a second two-component regulatory system, PmrA/PmrB. Based on work with other organisms, it is predicted that three enzymes are necessary for the synthesis of CF-specific lipid A (16): PagP, an outer membrane palmitoyl transferase (3-5); LpxO, an oxygenase required for the synthesis of 2-OH fatty acids (20); and PagL, a lipase required for the deacylation of the 3-O-position fatty acid (19, 38).
In this paper, we examine the function and regulation of the P. aeruginosa PagL homolog encoded by PA4661 and its role in the generation of CF-specific lipid A modifications. We demonstrate that P. aeruginosa PagL is required for the generation of the CF-specific penta-acylated lipid A structure. In addition, we show that P. aeruginosa PagL enzymatic activity is regulated both by magnesium concentration and temperature, and its activity can be lost in individuals with severe CF pulmonary disease.
MATERIALS AND METHODS
Bacterial strains and growth conditions. Bacterial strains used are shown in Table 1. Bacterial cells for LPS analysis were obtained after overnight growth with aeration in N-minimal medium (38 mM glycerol, 0.1% [wt/vol] Casamino Acids) supplemented with either 1 mM or 8 μM MgCl2 (18) or in LB medium supplemented with 1 mM MgCl2.
Clinical P. aeruginosa isolates. P. aeruginosa strains were obtained from CF patients of Children's Hospital and Regional Medical Center and the University of Washington Medical Center (both in Seattle, Washington) who met the following criteria: clinical diagnosis of CF; CF transmembrane conductance regulator (CFTR) genotype homozygous for the F508 allele; age of 7 years or older; and severe obstructive lung disease, defined by median percent forced expiratory volume in 1 second predicted during the 24-month period prior to culture. These strains had previously been isolated by quantitative sputum culture as part of a clinical-care study of CF lung severity, and were retrieved from storage at –80°C. The most abundant P. aeruginosa morphotype from each patient was analyzed. The Children's Hospital and Regional Medical Center Institutional Review Board reviewed and approved the use of medical records and specimens for this study, and patients gave informed consent.
Recombinant DNA techniques. Plasmids used in these studies are shown in Table 1. The pagL coding region was amplified from genomic DNA prepared from the P. aeruginosa laboratory-adapted isolate PAO-1 by PCR with Pfu Turbo DNA polymerase (Stratagene, La Jolla, CA). For expression in P. aeruginosa, the primers used for PCR were P. aeruginosa PagL2_attB2 (5'-TGTACAAAAAAGCAGGCTTAATGCGCAGCTTGAATG) and P. aeruginosa PagL2_attB1 (5'-TGTACAAGAAAGCTGGGTATCGCAGCCATGAAAAAG). PCR-amplified pagL DNA was cloned into the Gateway-compatible vector pDONR201 (Invitrogen, Carlsbad, CA) by site-specific recombination. The resulting product was then introduced into Escherichia coli DH5 cells by heat shock, and the bacterial transformants were selected on plates containing kanamycin (50 μg/ml). The P. aeruginosa pagL gene was then introduced into the broad-host-range plasmid pDEST JN105 (generated by the addition of Gateway-specific sequences, pDEST to JN105) under control of the arabinose PBAD promoter, and the resulting expression construct was named pPAPagL. For expression in enteric bacteria, primers used for the PCR were KK9 (5'-ATCATCCCATGGTAATGAAGAAACTACTTCCGCTG) and KK10 (5'-GAACTGCAGCTAGATCGGGATCTTGTAGAA). The amplified DNA fragment was cloned into the Nco1/Pst1 sites of pBAD24 (24) under control of the arabinose PBAD promoter, and the resulting expression construct was named pBAD24-PAPagL. The inserts of the expression constructs were verified by sequencing.
Removal of transposon insertion sequences in P. aeruginosa. Transposon insertion sequences from a reference strain library are flanked by loxP sites in the P. aeruginosa chromosome (26). To remove the transposon sequences, a donor strain, pCreI (E. coli SM10-pir containing the Cre recombinase gene), was grown in 5 ml LB supplemented with 100 μg/ml ampicillin at 37°C. The recipient P. aeruginosa strain 32751 was grown at 42°C in 5 ml LB without aeration. The next day, the donor strain was subcultured 50:50 into 5 ml LB containing ampicillin (100 μg/ml) and incubated for 40 minutes at 37°C. Thirty microliters of recipient strain was added to 900 μl of subcultured donor strain. The cells were pelleted and resuspended in 50 μl LB growth medium. The entire mixture was spotted onto prewarmed LB agar plates and incubated for 1 hour at 37°C, followed by streaking for single colonies from the spot onto LB agar with 10 μg/ml chloramphenicol. Colonies were apparent after 24 h at 37°C. Candidates were then tested on LB agar and LB agar supplemented with 60 μg/ml tetracycline to verify the loss of transposon sequences.
Expression of P. aeruginosa PagL in E. coli and Salmonella enterica serovar Typhimurium. For P. aeruginosa, the wild-type laboratory-adapted isolate PAO-1 and transposon mutant strain 32751 were transformed with pPAPagL by electroporation and selected on plates containing gentamicin (100 μg/ml). The transformant was cultivated overnight at 37°C in LB medium containing 100 μg/ml gentamicin. E. coli strain XL1-Blue (Stratagene, La Jolla, CA) and Salmonella enterica serovar Typhimurium strain CS015 (31) were transformed by electroporation with pBAD24-PAPagL or pBAD24. The transformant was cultivated at 37°C in LB medium containing 100 μg/ml ampicillin until A600 was approximately 0.5, and then expressions of recombinant pagL were induced by the addition of L-(+)-arabinose (2 mg/ml). After cultivation for 4 h in the presence of arabinose, cells were collected for lipid A analysis.
Separation of inner and outer membranes. Washed membranes from a laboratory-adapted wild-type P. aeruginosa PAK strain were prepared as previously described (38) from cells grown in N-minimal media (pH 7.4) containing 10 μM MgCl2 at either 15°C or 37°C. Membranes were resuspended in 10 mM HEPES, pH 7.0, containing 0.05 mM EDTA at a protein concentration of 5 mg/ml and applied to a sucrose gradient as described previously (23, 33). The seven-step gradient was then subjected to ultracentrifugation in a Beckman SW40.1 rotor for 19 h at 3°C for the separation of inner and outer membranes. Fractions were collected from the gradient and assayed for NADH oxidase activity (46) and level of protein using the bicinchonic acid assay (Pierce, Rockford, IL). Two distinct protein peaks representing the inner and outer membrane fractions were present (data not shown). Fractions containing <2% of the total NADH oxidase activity were deemed outer membrane fractions. Finally, each fraction was assayed for PagL 3-O-deacylase activity as described by Trent et al. (38), using the tetra-acylated lipid A precursor, lipid IVA.
LPS purification and lipid A isolation. LPS was isolated using the rapid small-scale isolation method for mass spectrometry analysis as described previously (44). Briefly, 1.0 ml of Tri-Reagent (Molecular Research Center, Cincinnati, OH) was added to cell culture pellets (10 ml LB or 50 ml N-minimal medium of an overnight culture), resuspended, and incubated at room temperature for 15 minutes. Two hundred microliters of chloroform was added, vortexed, and incubated at room temperature for 15 minutes. Samples were centrifuged for 10 minutes at 12,000 rpm, and the aqueous layer was removed. Five hundred microliters of water was added to the lower layer and vortexed well. After 15 to 30 min, the sample was spun down and the aqueous layers were combined. The process was repeated two more times. The combined aqueous layers were lyophilized overnight.
Lipid A was isolated after hydrolysis in 1% (wt/vol) sodium dodecyl sulfate at pH 4.5 as described previously (7). Briefly, 500 μl of 1% (wt/vol) sodium dodecyl sulfate in 10 mM Na-acetate, pH 4.5, was added to a lyophilized sample. Samples were incubated at 100°C for 1 hour and lyophilized. The dried pellets were washed in 100 μl of water and 1 ml of acidified ethanol (100 μl 4N HCl in 20 ml 95% ethyl alcohol [EtOH]). Samples were centrifuged at 5,000 rpm for 5 minutes. The lipid A pellet was further washed (three 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.
Fatty acid analysis. LPS fatty acids were derivatized to fatty acid methyl esters and analyzed by gas chromatography as described previously (12, 36). Briefly, LPS fatty acids were derivatized to fatty methyl esters with 2 M methanolic HCl at 90°C for 18 h (Alltech, Lexington, KY) and identified and quantified by gas chromatography (GC) using an HP 5890 series II with a 7673 autoinjector. Pentadecanoic acid (10 μg; Sigma, St. Louis, MO) was added as an internal standard.
Mass spectrometry procedures. Negative-ion matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS) experiments were performed as described previously, with the following modifications (17, 22). Lyophilized lipid A was dissolved with 10 μl 5-chloro-2-mercaptobenzothiazole (CMBT) MALDI matrix in chloroform/methanol, 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 Inc., Billerica, MA). Each spectrum was an average of 300 shots. ES tuning mix (Agilent, Palo Alto, CA) was used to calibrate the MALDI-TOF.
RESULTS
Characterization of a P. aeruginosa transposon mutant strain that lacks deacylase activity. As shown by Trent et al., PagL was initially identified in Salmonella enterica serovar Typhimurium (38); it was later shown by Geurtsen et al. to exist in a variety of other gram-negative bacteria, including P. aeruginosa (19). To demonstrate that the predicted open reading frame (PA4661) functioned in such a capacity in P. aeruginosa, a strain with a transposon insertion in the P. aeruginosa pagL homolog (32751) was used. Lipid A was isolated after growth in N-minimal medium supplemented with 1 mM MgCl2 and analyzed for absence of the deacylated, penta-acylated ion species (m/z = 1,447; Fig. 1D) using MALDI-TOF MS. MALDI-TOF MS analysis of lipid A from mutant 32751 gave a predominant hexa-acylated ion species with an m/z of 1,616 (Fig. 2B and 1C) that was consistent with the retention of the 3-OH C10 fatty acid residue at the 3 position. In contrast, the laboratory-adapted strain, PAO-1, gave a predominant penta-acylated ion species with an m/z of 1,447 (Fig. 2A and 1D). Loss of deacylase enzymatic activity was confirmed by capillary gas chromatography using flame ionization detection analysis of LPS isolated after growth in N-minimal medium supplemented with 1 mM MgCl2 (Fig. 3A). Similar MS and GC results were observed for lipid A isolated from 32751 and strain 32751Tn, in which the transposon was removed (data not shown).
The location of the transposon in mutant strain 32751 was shown by sequence analysis to be inserted after nucleotide 196 in the open reading frame PA4661 (www.pseudomonas.com). PA4661 is predicted to encode a hypothetical protein of 173 amino acids in length. Analysis of the amino acid sequence using the SignalP 3.0 program indicated the presence of a 23-amino-acid type I signal peptide. Separation by isopycnic density gradient centrifugation revealed that PagL deacylase activity was localized to the outer membrane protein peak, with no deacylase activity found in fractions containing the inner membrane protein NADH oxidase (data not shown), indicating that P. aeruginosa PagL is an outer membrane protein. The specific enzymatic activity of the deacylase in the P. aeruginosa membranes was less than the Salmonella enterica serovar Typhimurium PagL enzyme activity (0.01 to 0.015 nmol/min/mg and 0.5 nmol/min/mg, respectively) (38). Therefore, PA4661 likely encodes a functional P. aeruginosa outer membrane lipid A deacylase.
Expression of pagL in trans indicated that P. aeruginosa PA4661 is responsible for deacylase activity. To further demonstrate that PA4661 encoded the deacylase enzyme in P. aeruginosa, the pagL gene was amplified by PCR from genomic DNA prepared from the wild-type laboratory-adapted isolate, PAO-1. Deacylase activity was recovered from the P. aeruginosa pagL mutant as shown qualitatively by MS (Fig. 2B and D; presence of m/z = 1,447) and quantitatively by GC (Fig. 3B; decreased levels of 3-OH C10 fatty acid) analysis after transformation with pPAPagL. The ion species at m/z values of 1,367 and 1,536 (Fig. 2) correspond to the loss of a phosphate residue (m/z = 80) and are probably artifacts of the MS analysis. In addition, adjacent peaks to the penta- and hexa-acylated ion species that differ by 16 m/z units represent the addition of an additional hydroxyl group (OH) (m/z = 1,463 and 1,632).
In addition, heterologous expression of P. aeruginosa pagL in the enteric bacteria Salmonella enterica serovar Typhimurium and E. coli resulted in the loss of a 3-OH C14 from the 3 position of enteric lipid A, confirming the heterologous expression results of Geurtsen et al. in the E. coli strain, BL21 STAR (19) (data not shown). Based on these data, the P. aeruginosa PagL functions as a deacylase in Salmonella enterica serovar Typhimurium and E. coli and indicates a lack of specificity for fatty acid length.
P. aeruginosa deacylase activity in acute clinical and environmental isolates was activated by growth in media limited for magnesium. Previously, we have demonstrated that P. aeruginosa laboratory-adapted strains (PAO-1 and PAK) and isolates from patients with CF constitutively generate a penta-acylated lipid A structure (m/z = 1,447; Fig. 1D) as a result of 3-O deacylation, i.e., the removal of the 3-OH C10 at the 3-position of lipid A (2, 27). Interestingly, deacylated lipid A structures were not observed in P. aeruginosa isolates from patients with acute infections (blood, urinary tract infection [UTI], and eye) or from the environment (16, 17). This observation suggested that the regulation of deacylase activity in CF clinical isolates and laboratory-adapted strains was altered relative to acute infection isolates.
We examined whether deacylase activity in P. aeruginosa isolates from acute clinical infection isolates (blood, UTI, and eye) was regulated by growth conditions. Five acute infection isolates were analyzed for deacylated lipid A species after growth under magnesium-limited conditions, a growth condition known to induce lipid A modifications (17, 21, 22). Lipid A was isolated from individual P. aeruginosa clinical isolates after growth in minimal medium supplemented with either low (8 μM) or high (1 mM) magnesium concentrations. MALDI-TOF MS analyses of the individual lipid A preparations isolated from all acute infection isolates were similar when grown under the same conditions. A representative P. aeruginosa blood isolate grown in high-magnesium medium gave a dominant [M-H]– ion at an m/z of 1,419 (Fig. 4A) representative of a penta-acylated lipid A structure that retained the 3-OH C10 fatty acid at the 3 position (Fig. 1B). In contrast, lipid A isolated from cells grown in low-magnesium medium had a more complex MS spectrum containing ion species that represent both penta- (m/z = 1,447) and hexa-acylated species (m/z = 1,616, 1,685, and 1,815) (Fig. 4B). The additional ion species at m/z = 1,815 corresponded to the addition of an aminoarabinose residue (m/z = 131) to one of the terminal phosphates in the hexa-acylated species, m/z = 1,685 (Fig. 1E). Peaks adjacent to the penta- and hexa-acylated ion species that differ by 16 m/z units represent the addition of another hydroxyl group (OH) (m/z = 1,463, 1,632, 1,701, and 1,831). These ion species represent lipid A structures constitutively produced in clinical isolates from patients with CF and demonstrate that acute infection isolates of P. aeruginosa have inducible enzyme systems as compared to laboratory-adapted or CF isolates. These results are in contrast to the work of Geurtsen et al. that suggested that P. aeruginosa PagL enzymatic activity was not regulated by growth under different magnesium concentrations for the laboratory-adapted P. aeruginosa strain PAO-1 (19). However, their immunoblotting results were from laboratory-adapted P. aeruginosa strains grown in LB, a rich magnesium-replete growth condition. Similar results were observed for lipid A isolated from 11 environmental isolates (data not shown). Therefore, an alteration in PagL activity occurs in isolates from the CF airway and in laboratory-adapted bacteria.
Temperature regulation of deacylase activity in laboratory-adapted isolates of P. aeruginosa. At low growth temperatures (<21°C), bacteria alter lipid A acylation patterns, presumably to maintain and/or adjust membrane fluidity of the outer membrane as an adaptive response. Such adaptation includes induction of acyltransferase activity that leads to an increase in the proportion of cis-unsaturated fatty acids, such as C16:1, in the lipid A structure (8, 10, 40). To determine the effect of temperature on the regulation of P. aeruginosa deacylase activity, lipid A was isolated from two laboratory-adapted wild-type P. aeruginosa strains, PAO-1 and PAK, grown over a wide range of temperatures (15°C to 42°C). This temperature range corresponds to temperatures encountered in diverse environments, from soil to mammals. Lipid A was isolated at various temperature points (15, 21, 30, 37, and 42°C) after overnight growth in LB medium, and the fatty acids were analyzed by capillary gas chromatography using flame ionization detection (Fig. 5A). These experiments showed that when grown at low temperatures (15 or 21°C), P. aeruginosa deacylase enzymatic activity was altered in both laboratory-adapted isolates tested, as indicated by the increase in the levels of 3-OH C10 fatty acid attached to lipid A (3 and 3' positions) (Fig. 5A).
Mass-spectrometric analysis of lipid A isolated from PAO-1 and PAK grown over the same range of temperatures revealed a single major peak at an m/z of 1,447 in each strain upon growth at 42°C, consistent with the presence of a penta-acylated species with the 3 position 3-OH C10 fatty acid having been deacylated (Fig. 5C and E and 1D). Interestingly, the small difference in levels of 3-OH C10 fatty acid observed between PAK and PAO-1 after growth at 15°C (34.6% versus 30.5%, respectively) was associated with slightly disparate lipid A profiles between these two strains (Fig. 5A). The predominant peaks observed from lipid A isolated from PAK after growth at 15°C are at m/z values of 1,616 and 1,632 (Fig. 5B) and are consistent with a hexa-acylated lipid A species with a 3-OH C10 fatty acid at the 3 position and one (m/z = 1,616) or two (m/z = 1,632) acyloxyacyl 2-OH C12 fatty acid residues (Fig. 1C). In contrast, a second set of peaks (m/z = 1,447 and 1,463) was observed for lipid A isolated from PAO-1 after growth at 15°C (Fig. 5D), consistent with the presence of a penta-acylated lipid A structure also observed after growth at elevated temperatures (Fig. 1D). These results suggest that P. aeruginosa deacylase activity is reduced at lower temperatures and the level of inhibition is strain dependent.
Interestingly, the level of 3-OH C10 in lipid A isolated from the transposon insertion mutant 32751 was similar to levels observed for lipid A isolated from PAK after growth in low temperature conditions. Membranes isolated from PAK grown at 15°C show an approximate 10- to 15-fold reduction in PagL 3-O-deacylase activity (0.0015 to 0.0022 nmol/min/mg) compared to membranes isolated from PAK grown at 37°C (0.022 nmol/min/mg). These results further suggested that in PAK, less PagL enzymatic activity was present in the membranes at low temperature, indicating that part of the difference between environmental isolates and CF and laboratory-adapted isolates involves less production or stability of the enzyme.
Some P. aeruginosa clinical isolates from CF patients with severe pulmonary disease are deficient in lipid A deacylase activity. To explore the possibility that changes in lipid A structure might contribute to lung disease progression in CF patients chronically infected with P. aeruginosa, we performed an analysis of lipid A isolated from longitudinal clinical isolates of three patients: a child with mild CF lung disease, a child with severe lung disease, and an adult with mild CF lung disease, as defined by serial measurements of lung function. Lipid A was isolated from the individual P. aeruginosa clinical isolates after growth in minimal medium supplemented with high (1 mM) magnesium concentrations. MALDI-TOF MS analyses of the individual lipid A preparations isolated from three clinical isolates are shown. A P. aeruginosa isolate from the child with mild CF lung disease (CF565) revealed dominant ions that represent both penta- (m/z = 1,447; Fig. 1D) and hexa-acylated species (m/z = 1,616; Fig. 1C) (m/z = 1,685, Fig. 1E) (Fig. 6A). Adjacent peaks to the penta- and hexa-acylated ion species that differ by 16 m/z units represent the addition of an additional hydroxyl group. Similar MS spectra were obtained for P. aeruginosa isolates from the adult with mild CF lung disease (data not shown).
In contrast, lipid A from clinical isolates from the child with severe CF lung disease and from a second CF patient with severe lung disease (10128 and SE22) had ion species that represent both hexa- (m/z = 1,616; Fig. 1C) and a novel hepta-acylated species (m/z = 1,854; Fig. 1F) (Fig. 6B and C, respectively). The hepta-acylated structure was consistent with four fatty acids (two 3-OH-C10 and two 3-OH C12) attached to the glucosamine backbone with three secondary "piggyback" fatty acids (C12, 2-OH C12, and C16). These results are consistent with a loss of deacylase enzymatic activity in these P. aeruginosa isolates. Adjacent peaks to the hexa- and hepta-acylated ion species that differ by 16 m/z units represent the addition of an additional hydroxyl group. Consistent with this hypothesis, deacylated lipid A species (m/z = 1,447; Fig. 1D) (m/z = 1,685; Fig. 1E) were recovered upon transformation of these isolates with pPAPagL, as determined by MS and GC analyses (data not shown), indicating that the deficiency of deacylated lipid A species in isolates from CF patients with severe lung disease is due to loss of deacylase enzymatic activity.
This analysis suggested a potential association between severe obstructive CF lung disease and loss of P. aeruginosa lipid A deacylation. In order to evaluate this potential association, lipid A was purified from P. aeruginosa clinical isolates of 21 patients with severe obstructive lung disease and analyzed by MALDI-TOF MS. This analysis showed that lipid A purified from P. aeruginosa isolates from seven patients (33%) had hexa-acylated (m/z = 1,616; Fig. 1C) or novel hepta-acylated structures (m/z = 1,854; Fig. 1F) but lacked the corresponding deacylated lipid A structures (m/z = 1,447 or 1,685; Fig. 1D and E, respectively) observed in isolates from other CF patients with severe lung disease, patients with milder CF lung disease, acute infections, and environmental isolates (data not shown). The presence of a full-length copy of the pagL gene was confirmed for all isolates used in these studies by colony PCR using pagL-specific oligonucleotides (data not shown). These results suggest that loss of lipid A deacylase activity can occur during progression of CF lung disease. This suggests that as disease progresses, selection may occur for loss of activity that results in the formation of a hepta-acylated lipid A.
DISCUSSION
Adaptation of P. aeruginosa to life in the airways of individuals with CF results in the expression of specific lipid A structures not observed in P. aeruginosa isolates from acute infections (blood, UTI, and eye). The modifications that distinguish these isolates' lipid A structures include the addition of palmitate and aminoarabinose and the deacylation of the 3-position fatty acid (3-OH C10) by PagL enzymatic activity. Our results show that the P. aeruginosa PagL enzymatic activity, though constitutively active in P. aeruginosa clinical isolates from patients with CF and laboratory-adapted wild-type isolates, was regulated in clinical isolates from both acute infections and environmental isolates. Acute P. aeruginosa isolates produce a precursor penta-acylated lipid A (Fig. 1B) structure not normally observed in isolates from patients with CF or laboratory-adapted isolates (Fig. 1C to E). Interestingly, acute P. aeruginosa isolates have the ability to produce CF-like lipid A modification when grown under magnesium-limited conditions (m/z = 1,447 and 1,685) (Fig. 1). These results indicate that the biosynthetic pathways necessary for the synthesis of CF-specific lipid A were intact and able to be regulated.
We further demonstrated that P. aeruginosa regulated PagL enzymatic activity in laboratory-adapted isolates upon growth under low temperature, a condition that mimics environmental growth conditions. Interestingly, addition of an unsaturated C16:1 fatty acid residue, a cold-shock regulated response typical among enteric bacteria, was not observed in P. aeruginosa (8). Instead, GC and MS analysis of P. aeruginosa lipid A isolated after growth at low temperatures (15°C) indicated that PagL enzymatic activity was completely inhibited in the laboratory-adapted wild-type isolate, PAK, suggesting a role for deacylase regulation of membrane structure and fluidity in response to temperature. The loss of this activity on growth in nutrient broth at 37°C or in the CF airway may promote PagL expression.
To determine the role of PagL in the generation of CF-specific lipid A structures, a P. aeruginosa transposon mutant, 35721 (19), was analyzed by MS and GC (Fig. 2 and 3). The insertion of the transposon in mutant strain 32751 mapped to open reading frame PA4661, which was predicted to encode a hypothetical protein of 173 amino acids in length. PA4661 is part of a group of six genes (hemA, prfA, hemK, moeB, murI, and PA4661), of which the first five genes form a putative operon. Genes in the five-gene operon include: hemA and hemK, required for heme biosynthesis (41); moeB, required for molybdopterin cofactor biosynthesis (43); prfA, encoding a homolog of the E. coli peptide chain release factor that directs the termination of translation in response to specific peptide chain termination codons (9); and murI, required for biosynthesis of D-glutamate and peptidoglycan (14). A 69-nucleotide intragenic region is present between murI and PA4661, suggesting that PA4661 is not part of the previous operon.
Deacylase activity was recovered in the P. aeruginosa transposon mutant background upon transformation and expression with either the P. aeruginosa PA4661 (Fig. 2 and 3) or Salmonella enterica serovar Typhimurium pagL gene (data not shown). Interestingly, two additional transposon insertion mutants located in the conserved C terminus of the protein retained deacylase activity, suggesting that this region of the proteins is not required for this activity. Heterologous expression of P. aeruginosa PA4661 in either a Salmonella enterica serovar Typhimurium PhoP-null strain or E. coli (both of which lack deacylase activity) resulted in the presence of deacylated lipid A species by MS and GC analysis. Taken together, these results suggest a relaxed chain length specificity for the P. aeruginosa and Salmonella enterica serovar Typhimurium pagL genes, since they have the ability to remove either longer (3-OH C14) or shorter (3-OH C10) fatty acid residues from the 3 position of lipid A, respectively. Finally, in contrast to the recently published results from Geurtsen et al., secondary addition of a C16 fatty acid after heterologous expression of P. aeruginosa pagL in either E. coli or Salmonella enterica serovar Typhimurium was not observed (19). Since different E. coli strain backgrounds were used for the heterologous-expression studies (XL1-Blue versus BL21 Star), strain-specific differences, such as constitutive expression of the genes necessary for the addition of aminoarabinose or palmitate observed in E. coli BL21 Star (39), may play a role in secondary lipid A modifications.
Lipid A isolated from P. aeruginosa clinical isolates from patients with CF (1 to 3 years of age) demonstrates constitutive expression of deacylase enzyme activity. A penta-acylated lipid A structure that contains a 3-OH C10 fatty acid (Fig. 1B) in the 3 position was not observed in CF isolates but was observed in isolates from acute infections or environmental isolates. Using clinical isolates from patients with severe pulmonary disease (as defined by lung function values), we showed that lipid A isolated from one-third of the patients' clinical isolates lacked deacylase activity compared to those from patients with mild pulmonary disease. Severe patients' isolates that lacked deacylase activity resulted in synthesis of hexa- or hepta-acylated structures that retained the 3-OH C10 fatty acid at the 3 position. The hepta-acylated lipid A structure (m/z = 1,854; Fig. 1F) was modified by an acyloxyacyl addition of C16:0 at the 3'-position fatty acid. Deacylated lipid A structures were recovered in these P. aeruginosa clinical isolates upon complementation with the P. aeruginosa pagL gene, indicating that these strains have lost deacylase activity. The characterization of these clinical isolates may result in the identification of novel mutations in the P. aeruginosa pagL gene that affect activity of the enzyme, or the identification of regulatory pathway alterations that result in the synthesis of CF-specific lipid A modifications. Therefore, deciphering the role of this late adaptation of P. aeruginosa to the CF airway may be important in understanding and preventing the progression of chronic CF lung disease.
ACKNOWLEDGMENTS
We thank Lucas Hoffman for critical review of the manuscript.
This work was supported by grants from the National Institutes of Health (AI47938 and DK064954 to S.I.M.) and the Cystic Fibrosis Foundation (R.K.E.).
REFERENCES
Berger, M. 2002. Inflammatory mediators in cystic fibrosis lung disease. Allergy Asthma Proc. 23:19-25.
Bhat, R., A. Marx, C. Galanos, and R. S. Conrad. 1990. Structural studies of lipid A from Pseudomonas aeruginosa PAO1: occurrence of 4-amino-4-deoxyarabinose. J. Bacteriol. 172:6631-6636.
Bishop, R. E. 2005. Fundamentals of endotoxin structure and function. Contrib. Microbiol. 12:1-27.
Bishop, R. E., H. S. Gibbons, T. Guina, M. S. Trent, S. I. Miller, and C. R. Raetz. 2000. Transfer of palmitate from phospholipids to lipid A in outer membranes of gram-negative bacteria. EMBO J. 19:5071-5080.
Bishop, R. E., S. H. Kim, and A. El Zoeiby. 2005. Role of lipid A palmitoylation in bacterial pathogenesis. J. Endotoxin Res. 11:174-180.
Burns, J. L., R. L. Gibson, S. McNamara, D. Yim, J. Emerson, M. Rosenfeld, P. Hiatt, K. McCoy, R. Castile, A. L. Smith, and B. W. Ramsey. 2001. Longitudinal assessment of Pseudomonas aeruginosa in young children with cystic fibrosis. J. Infect. Dis. 183:444-452.
Caroff, M., A. Tacken, and L. Szabo. 1988. 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. 175:273-282.
Carty, S. M., K. R. Sreekumar, and C. R. Raetz. 1999. Effect of cold shock on lipid A biosynthesis in Escherichia coli. Induction at 12°C of an acyltransferase specific for palmitoleoyl-acyl carrier protein. J. Biol. Chem. 274:9677-9685.
Craigen, W. J., and C. T. Caskey. 1987. The function, structure and regulation of E. coli peptide chain release factors. Biochimie 69:1031-1041.
Cronan, J. E., Jr., and E. P. Gelmann. 1975. Physical properties of membrane lipids: biological relevance and regulation. Bacteriol. Rev. 39:232-256.
Dakin, C. J., A. H. Numa, H. Wang, J. R. Morton, C. C. Vertzyas, and R. L. Henry. 2002. Inflammation, infection, and pulmonary function in infants and young children with cystic fibrosis. Am. J. Respir. Crit. Care Med. 165:904-910.
Darveau, R. P., M. D. Cunningham, T. Bailey, C. Seachord, K. Ratcliffe, B. Bainbridge, M. Dietsch, R. C. Page, and A. Aruffo. 1995. Ability of bacteria associated with chronic inflammatory disease to stimulate E-selectin expression and promote neutrophil adhesion. Infect. Immun. 63:1311-1317.
Davies, J. C. 2002. Pseudomonas aeruginosa in cystic fibrosis: pathogenesis and persistence. Paediatr. Respir. Rev. 3:128-134.
Doublet, P., J. van Heijenoort, and D. Mengin-Lecreulx. 1992. Identification of the Escherichia coli murI gene, which is required for the biosynthesis of D-glutamic acid, a specific component of bacterial peptidoglycan. J. Bacteriol. 174:5772-5779.
Ernst, R. K., T. Guina, and S. I. Miller. 2001. Salmonella typhimurium outer membrane remodeling: role in resistance to host innate immunity. Microbes Infect. 3:1327-1334.
Ernst, R. K., A. M. Hajjar, J. H. Tsai, S. M. Moskowitz, C. B. Wilson, and S. I. Miller. 2003. Pseudomonas aeruginosa lipid A diversity and its recognition by Toll-like receptor 4. J. Endotoxin Res. 9:395-400.
Ernst, R. K., E. C. Yi, L. Guo, K. B. Lim, J. L. Burns, M. Hackett, and S. I. Miller. 1999. Specific lipopolysaccharide found in cystic fibrosis airway Pseudomonas aeruginosa. Science 286:1561-1565.
Garcia Vescovi, E., F. C. Soncini, and E. A. Groisman. 1996. Mg2+ as an extracellular signal: environmental regulation of Salmonella virulence. Cell 84:165-174.
Geurtsen, J., L. Steeghs, J. T. Hove, P. van der Ley, and J. Tommassen. 2005. Dissemination of lipid A deacylases (pagL) among gram-negative bacteria: identification of active-site histidine and serine residues. J. Biol. Chem. 280:8248-8259.
Gibbons, H. S., S. Lin, R. J. Cotter, and C. R. Raetz. 2000. Oxygen requirement for the biosynthesis of the S-2-hydroxymyristate moiety in Salmonella typhimurium lipid A. Function of LpxO, a new Fe2+/-ketoglutarate-dependent dioxygenase homologue. J. Biol. Chem. 275:32940-32949.
Guo, L., K. B. Lim, J. S. Gunn, B. Bainbridge, R. P. Darveau, M. Hackett, and S. I. Miller. 1997. Regulation of lipid A modifications by Salmonella typhimurium virulence genes phoP-phoQ. Science 276:250-253.
Guo, L., K. B. Lim, C. M. Poduje, M. Daniel, J. S. Gunn, M. Hackett, and S. I. Miller. 1998. Lipid A acylation and bacterial resistance against vertebrate antimicrobial peptides. Cell 95:189-198.
Guy-Caffey, J. K., M. P. Rapoza, K. A. Jolley, and R. E. Webster. 1992. Membrane localization and topology of a viral assembly protein. J. Bacteriol. 174:2460-2465.
Guzman, L.-M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.
Hajjar, A. M., R. K. Ernst, J. H. Tsai, C. B. Wilson, and S. I. Miller. 2002. Human Toll-like receptor 4 recognizes host-specific LPS modifications. Nat. Immunol. 3:354-359.
Jacobs, M. A., A. Alwood, I. Thaipisuttikul, D. Spencer, E. Haugen, S. Ernst, O. Will, R. Kaul, C. Raymond, R. Levy, L. Chun-Rong, D. Guenthner, D. Bovee, M. V. Olson, and C. Manoil. 2003. Comprehensive transposon mutant library of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 100:14339-14344.
Kulshin, V. A., U. Zahringer, B. Lindner, K. E. Jager, B. A. Dmitriev, and E. T. Rietschel. 1991. Structural characterization of the lipid A component of Pseudomonas aeruginosa wild-type and rough mutant lipopolysaccharides. Eur. J. Biochem. 198:697-704.
Macfarlane, E. L., A. Kwasnicka, and R. E. Hancock. 2000. Role of Pseudomonas aeruginosa PhoP-PhoQ in resistance to antimicrobial cationic peptides and aminoglycosides. Microbiology 146:2543-2554.
Macfarlane, E. L., A. Kwasnicka, M. M. Ochs, and R. E. Hancock. 1999. PhoP-PhoQ homologues in Pseudomonas aeruginosa regulate expression of the outer-membrane protein OprH and polymyxin B resistance. Mol. Microbiol. 34:305-316.
Miller, S. I., R. K. Ernst, and M. W. Bader. 2005. LPS, TLR4 and infectious disease diversity. Nat. Rev. Microbiol. 3:36-46.
Miller, S. I., A. M. Kukral, and J. J. Mekalanos. 1989. A two-component regulatory system (PhoP/PhoQ) controls Salmonella typhimurium virulence. Proc. Natl. Acad. Sci. USA 86:5054-5058.
Moskowitz, S. M., R. K. Ernst, and S. I. Miller. 2004. 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. 186:575-579.
Osborn, M. J., J. E. Gander, E. Parisi, and J. Carson. 1972. Mechanism of assembly of the outer membrane of Salmonella typhimurium. Isolation and characterization of cytoplasmic and outer membrane. J. Biol. Chem. 247:3962-3972.
Rosenfeld, M., R. L. Gibson, S. McNamara, J. Emerson, J. L. Burns, R. Castile, P. Hiatt, K. McCoy, C. B. Wilson, A. Inglis, A. Smith, T. R. Martin, and B. W. Ramsey. 2001. Early pulmonary infection, inflammation, and clinical outcomes in infants with cystic fibrosis. Pediatr. Pulmonol. 32:356-366.
Rowntree, R. K., and A. Harris. 2003. The phenotypic consequences of CFTR mutations. Ann. Hum. Genet. 67:471-485.
Somerville, J. E., Jr., L. Cassiano, B. Bainbridge, M. D. Cunningham, and R. P. Darveau. 1996. A novel Escherichia coli lipid A mutant that produces an antiinflammatory lipopolysaccharide. J. Clin. Investig. 97:359-365.
Trent, M. S. 2004. Biosynthesis, transport, and modification of lipid A. Biochem. Cell Biol. 82:71-86.
Trent, M. S., W. Pabich, C. R. Raetz, and S. I. Miller. 2001. A PhoP/PhoQ-induced Lipase (PagL) that catalyzes 3-O-deacylation of lipid A precursors in membranes of Salmonella typhimurium. J. Biol. Chem. 276:9083-9092.
Trent, M. S., A. A. Ribeiro, W. T. Doerrler, S. Lin, R. J. Cotter, and C. R. Raetz. 2001. Accumulation of a polyisoprene-linked amino sugar in polymyxin-resistant Salmonella typhimurium and Escherichia coli: structural characterization and transfer to lipid A in the periplasm. J. Biol. Chem. 276:43132-43144.
Vigh, L., B. Maresca, and J. L. Harwood. 1998. Does the membrane's physical state control the expression of heat shock and other genes Trends Biochem. Sci. 23:369-374.
Wandersman, C., and P. Delepelaire. 2004. Bacterial iron sources: from siderophores to hemophores. Annu. Rev. Microbiol. 58:611-647.
Whitfield, C., N. Kaniuk, and E. Frirdich. 2003. Molecular insights into the assembly and diversity of the outer core oligosaccharide in lipopolysaccharides from Escherichia coli and Salmonella. J. Endotoxin Res. 9:244-249.
Wuebbens, M. M., and K. V. Rajagopalan. 1995. Investigation of the early steps of molybdopterin biosynthesis in Escherichia coli through the use of in vivo labeling studies. J. Biol. Chem. 270:1082-1087.
Yi, E. C., and M. Hackett. 2000. Rapid isolation method for lipopolysaccharide and lipid A from gram-negative bacteria. Analyst 125:651-656.
Yu, H., and N. E. Head. 2002. Persistent infections and immunity in cystic fibrosis. Front. Biosci. 7:d442-d457.
Zhou, Z., K. A. White, A. Polissi, C. Georgopoulos, and C. R. Raetz. 1998. Function of Escherichia coli MsbA, an essential ABC family transporter, in lipid A and phospholipid biosynthesis. J. Biol. Chem. 273:12466-12475.(Robert K. Ernst,1Kristin )
ABSTRACT
Lipopolysaccharide (LPS) is the major surface component of gram-negative bacteria, and a component of LPS, lipid A, is recognized by the innate immune system through the Toll-like receptor 4/MD-2 complex. Pseudomonas aeruginosa, an environmental gram-negative bacterium that opportunistically infects the respiratory tracts of patients with cystic fibrosis (CF), can synthesize various structures of lipid A. Lipid A from P. aeruginosa strains isolated from infants with CF has a specific structure that includes the removal of the 3 position 3-OH C10 fatty acid. Here we demonstrate increased expression of the P. aeruginosa lipid A 3-O-deacylase (PagL) in isolates from CF infants compared to that in environmental isolates. PagL activity was increased in environmental isolates by growth in medium limited for magnesium and decreased by growth at low temperature in laboratory-adapted strains of P. aeruginosa. P. aeruginosa PagL was shown to be an outer membrane protein by isopycnic density gradient centrifugation. Heterologous expression of P. aeruginosa pagL in Salmonella enterica serovar Typhimurium and Escherichia coli resulted in removal of the 3-OH C14 fatty acid from lipid A, indicating that P. aeruginosa PagL recognizes either 3-OH C10 or 3-OH C14. Finally, deacylated lipid A species were not observed in some clinical P. aeruginosa isolates from patients with severe pulmonary disease, suggesting that loss of PagL function can occur during long-term adaptation to the CF airway.
INTRODUCTION
The interaction between the host airway environment and specific bacterial pathogens is critical to the development of airway disease in cystic fibrosis (CF), the most common lethal autosomal recessive genetic disease of Caucasians (30, 35). Shortly after birth, the respiratory tracts of most patients with CF are infected with the opportunistic gram-negative bacterium Pseudomonas aeruginosa (6, 11, 34). P. aeruginosa adapts to the environment of the CF airway with the accumulation of a variety of mutations that lead to mucoidy, the formation of biofilms, and the alteration of lipopolysaccharide (LPS) (13, 16, 17, 45). These adaptations are associated with chronic airway inflammation that leads to severe and progressive pulmonary dysfunction, resulting in premature death. This inflammatory process results, at least in part, from stimulation of the innate immune system by CF-specific modifications of P. aeruginosa lipid A. Lipid A is the bioactive component of LPS, a pathogenic factor of gram-negative bacteria that consists of three distinct regions: O antigen, core, and lipid A (1, 17, 25). Both O antigen and core consist of polysaccharide chains, whereas lipid A consists of fatty acid and phosphate groups bonded to a glucosamine disaccharide (37, 42).
Like lipid A of other gram-negative bacteria, P. aeruginosa lipid A contains penta- and/or hexa-acylated structures that consist of a (1,6)-linked disaccharide of glucosamine with phosphate groups at the 1 and 4' positions, amide-linked fatty acids at the 2 and 2' positions, and ester-linked fatty acids at the 3 and 3' positions (Fig. 1A) (2, 17, 27). The fatty acids attached to the P. aeruginosa glucosamine residues are shorter (C10 and C12 versus C12 and C14) than those in enterobacterial lipid A. P. aeruginosa further modifies its lipid A structure by the addition of secondary or "piggyback" fatty acids (C12 and 2-OH C12) to generate a hexa-acylated structure (a mass to charge ratio [m/z] equal to 1,616; Fig. 1C). This hexa-acylated lipid A structure can be modified by the removal or deacylation of the fatty acid (3-OH C10) at the 3 position to generate a penta-acylated structure (m/z = 1,447; Fig. 1D). Finally, this second penta-acylated structure can be modified by the acyloxyacyl addition of a C16 fatty acid at the 3' position to generate a hexa-acylated form (m/z = 1,685; Fig. 1E) of lipid A (16). This palmitate-containing hexa-acylated structure was observed in lipid A isolated from all P. aeruginosa clinical isolates from individuals with CF after growth in magnesium-replete medium, while it was absent from laboratory-adapted strains (PAO-1, PAK, and PA14) grown under the same conditions (data not shown).
P. aeruginosa isolated from either the upper or lower airway of young children with CF synthesized a more highly acylated lipid A (hexa-acylated, m/z = 1,685; Fig. 1E), while P. aeruginosa isolated from non-CF patients with acute infections synthesized a penta-acylated form (m/z = 1,419; Fig. 1B) (17). Lipid A structural modifications observed specifically in LPS from P. aeruginosa clinical isolates from patients with CF include the addition of 2-hydroxy laurate (2-OH C12), deacylation of the 3-position fatty acid, and addition of palmitate (C16) leading to mass spectrometric detection of ions with m/z ratios of 1,616 (Fig. 1C), 1,447 (Fig. 1D), and 1,685 (Fig. 1E), respectively. These changes in acylation are associated with resistance to host antimicrobial peptides and an alteration in the innate inflammatory response by way of Toll-like receptor 4 signaling (16, 17, 25). The fact that these modifications were not observed in acute clinical isolates suggests that environmental organisms, the sources for acute infections, did not express the enzymes required for these changes.
Environmentally regulated LPS modification enzymes have been observed in a wide variety of pathogenic and nonpathogenic gram-negative bacterial species. Interestingly, each bacterial species has unique regulatory and biosynthetic mechanisms that nevertheless result in similar lipid A modifications (15, 37, 42). Lipid A modifications in P. aeruginosa are regulated by two-component regulatory systems PhoP/PhoQ and PmrA/PmrB. The P. aeruginosa PhoP/PhoQ regulatory system is required for addition of palmitate and aminoarabinose in response to magnesium-limiting growth conditions (17, 28, 29, 32). The addition of aminoarabinose can also be regulated by a second two-component regulatory system, PmrA/PmrB. Based on work with other organisms, it is predicted that three enzymes are necessary for the synthesis of CF-specific lipid A (16): PagP, an outer membrane palmitoyl transferase (3-5); LpxO, an oxygenase required for the synthesis of 2-OH fatty acids (20); and PagL, a lipase required for the deacylation of the 3-O-position fatty acid (19, 38).
In this paper, we examine the function and regulation of the P. aeruginosa PagL homolog encoded by PA4661 and its role in the generation of CF-specific lipid A modifications. We demonstrate that P. aeruginosa PagL is required for the generation of the CF-specific penta-acylated lipid A structure. In addition, we show that P. aeruginosa PagL enzymatic activity is regulated both by magnesium concentration and temperature, and its activity can be lost in individuals with severe CF pulmonary disease.
MATERIALS AND METHODS
Bacterial strains and growth conditions. Bacterial strains used are shown in Table 1. Bacterial cells for LPS analysis were obtained after overnight growth with aeration in N-minimal medium (38 mM glycerol, 0.1% [wt/vol] Casamino Acids) supplemented with either 1 mM or 8 μM MgCl2 (18) or in LB medium supplemented with 1 mM MgCl2.
Clinical P. aeruginosa isolates. P. aeruginosa strains were obtained from CF patients of Children's Hospital and Regional Medical Center and the University of Washington Medical Center (both in Seattle, Washington) who met the following criteria: clinical diagnosis of CF; CF transmembrane conductance regulator (CFTR) genotype homozygous for the F508 allele; age of 7 years or older; and severe obstructive lung disease, defined by median percent forced expiratory volume in 1 second predicted during the 24-month period prior to culture. These strains had previously been isolated by quantitative sputum culture as part of a clinical-care study of CF lung severity, and were retrieved from storage at –80°C. The most abundant P. aeruginosa morphotype from each patient was analyzed. The Children's Hospital and Regional Medical Center Institutional Review Board reviewed and approved the use of medical records and specimens for this study, and patients gave informed consent.
Recombinant DNA techniques. Plasmids used in these studies are shown in Table 1. The pagL coding region was amplified from genomic DNA prepared from the P. aeruginosa laboratory-adapted isolate PAO-1 by PCR with Pfu Turbo DNA polymerase (Stratagene, La Jolla, CA). For expression in P. aeruginosa, the primers used for PCR were P. aeruginosa PagL2_attB2 (5'-TGTACAAAAAAGCAGGCTTAATGCGCAGCTTGAATG) and P. aeruginosa PagL2_attB1 (5'-TGTACAAGAAAGCTGGGTATCGCAGCCATGAAAAAG). PCR-amplified pagL DNA was cloned into the Gateway-compatible vector pDONR201 (Invitrogen, Carlsbad, CA) by site-specific recombination. The resulting product was then introduced into Escherichia coli DH5 cells by heat shock, and the bacterial transformants were selected on plates containing kanamycin (50 μg/ml). The P. aeruginosa pagL gene was then introduced into the broad-host-range plasmid pDEST JN105 (generated by the addition of Gateway-specific sequences, pDEST to JN105) under control of the arabinose PBAD promoter, and the resulting expression construct was named pPAPagL. For expression in enteric bacteria, primers used for the PCR were KK9 (5'-ATCATCCCATGGTAATGAAGAAACTACTTCCGCTG) and KK10 (5'-GAACTGCAGCTAGATCGGGATCTTGTAGAA). The amplified DNA fragment was cloned into the Nco1/Pst1 sites of pBAD24 (24) under control of the arabinose PBAD promoter, and the resulting expression construct was named pBAD24-PAPagL. The inserts of the expression constructs were verified by sequencing.
Removal of transposon insertion sequences in P. aeruginosa. Transposon insertion sequences from a reference strain library are flanked by loxP sites in the P. aeruginosa chromosome (26). To remove the transposon sequences, a donor strain, pCreI (E. coli SM10-pir containing the Cre recombinase gene), was grown in 5 ml LB supplemented with 100 μg/ml ampicillin at 37°C. The recipient P. aeruginosa strain 32751 was grown at 42°C in 5 ml LB without aeration. The next day, the donor strain was subcultured 50:50 into 5 ml LB containing ampicillin (100 μg/ml) and incubated for 40 minutes at 37°C. Thirty microliters of recipient strain was added to 900 μl of subcultured donor strain. The cells were pelleted and resuspended in 50 μl LB growth medium. The entire mixture was spotted onto prewarmed LB agar plates and incubated for 1 hour at 37°C, followed by streaking for single colonies from the spot onto LB agar with 10 μg/ml chloramphenicol. Colonies were apparent after 24 h at 37°C. Candidates were then tested on LB agar and LB agar supplemented with 60 μg/ml tetracycline to verify the loss of transposon sequences.
Expression of P. aeruginosa PagL in E. coli and Salmonella enterica serovar Typhimurium. For P. aeruginosa, the wild-type laboratory-adapted isolate PAO-1 and transposon mutant strain 32751 were transformed with pPAPagL by electroporation and selected on plates containing gentamicin (100 μg/ml). The transformant was cultivated overnight at 37°C in LB medium containing 100 μg/ml gentamicin. E. coli strain XL1-Blue (Stratagene, La Jolla, CA) and Salmonella enterica serovar Typhimurium strain CS015 (31) were transformed by electroporation with pBAD24-PAPagL or pBAD24. The transformant was cultivated at 37°C in LB medium containing 100 μg/ml ampicillin until A600 was approximately 0.5, and then expressions of recombinant pagL were induced by the addition of L-(+)-arabinose (2 mg/ml). After cultivation for 4 h in the presence of arabinose, cells were collected for lipid A analysis.
Separation of inner and outer membranes. Washed membranes from a laboratory-adapted wild-type P. aeruginosa PAK strain were prepared as previously described (38) from cells grown in N-minimal media (pH 7.4) containing 10 μM MgCl2 at either 15°C or 37°C. Membranes were resuspended in 10 mM HEPES, pH 7.0, containing 0.05 mM EDTA at a protein concentration of 5 mg/ml and applied to a sucrose gradient as described previously (23, 33). The seven-step gradient was then subjected to ultracentrifugation in a Beckman SW40.1 rotor for 19 h at 3°C for the separation of inner and outer membranes. Fractions were collected from the gradient and assayed for NADH oxidase activity (46) and level of protein using the bicinchonic acid assay (Pierce, Rockford, IL). Two distinct protein peaks representing the inner and outer membrane fractions were present (data not shown). Fractions containing <2% of the total NADH oxidase activity were deemed outer membrane fractions. Finally, each fraction was assayed for PagL 3-O-deacylase activity as described by Trent et al. (38), using the tetra-acylated lipid A precursor, lipid IVA.
LPS purification and lipid A isolation. LPS was isolated using the rapid small-scale isolation method for mass spectrometry analysis as described previously (44). Briefly, 1.0 ml of Tri-Reagent (Molecular Research Center, Cincinnati, OH) was added to cell culture pellets (10 ml LB or 50 ml N-minimal medium of an overnight culture), resuspended, and incubated at room temperature for 15 minutes. Two hundred microliters of chloroform was added, vortexed, and incubated at room temperature for 15 minutes. Samples were centrifuged for 10 minutes at 12,000 rpm, and the aqueous layer was removed. Five hundred microliters of water was added to the lower layer and vortexed well. After 15 to 30 min, the sample was spun down and the aqueous layers were combined. The process was repeated two more times. The combined aqueous layers were lyophilized overnight.
Lipid A was isolated after hydrolysis in 1% (wt/vol) sodium dodecyl sulfate at pH 4.5 as described previously (7). Briefly, 500 μl of 1% (wt/vol) sodium dodecyl sulfate in 10 mM Na-acetate, pH 4.5, was added to a lyophilized sample. Samples were incubated at 100°C for 1 hour and lyophilized. The dried pellets were washed in 100 μl of water and 1 ml of acidified ethanol (100 μl 4N HCl in 20 ml 95% ethyl alcohol [EtOH]). Samples were centrifuged at 5,000 rpm for 5 minutes. The lipid A pellet was further washed (three 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.
Fatty acid analysis. LPS fatty acids were derivatized to fatty acid methyl esters and analyzed by gas chromatography as described previously (12, 36). Briefly, LPS fatty acids were derivatized to fatty methyl esters with 2 M methanolic HCl at 90°C for 18 h (Alltech, Lexington, KY) and identified and quantified by gas chromatography (GC) using an HP 5890 series II with a 7673 autoinjector. Pentadecanoic acid (10 μg; Sigma, St. Louis, MO) was added as an internal standard.
Mass spectrometry procedures. Negative-ion matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS) experiments were performed as described previously, with the following modifications (17, 22). Lyophilized lipid A was dissolved with 10 μl 5-chloro-2-mercaptobenzothiazole (CMBT) MALDI matrix in chloroform/methanol, 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 Inc., Billerica, MA). Each spectrum was an average of 300 shots. ES tuning mix (Agilent, Palo Alto, CA) was used to calibrate the MALDI-TOF.
RESULTS
Characterization of a P. aeruginosa transposon mutant strain that lacks deacylase activity. As shown by Trent et al., PagL was initially identified in Salmonella enterica serovar Typhimurium (38); it was later shown by Geurtsen et al. to exist in a variety of other gram-negative bacteria, including P. aeruginosa (19). To demonstrate that the predicted open reading frame (PA4661) functioned in such a capacity in P. aeruginosa, a strain with a transposon insertion in the P. aeruginosa pagL homolog (32751) was used. Lipid A was isolated after growth in N-minimal medium supplemented with 1 mM MgCl2 and analyzed for absence of the deacylated, penta-acylated ion species (m/z = 1,447; Fig. 1D) using MALDI-TOF MS. MALDI-TOF MS analysis of lipid A from mutant 32751 gave a predominant hexa-acylated ion species with an m/z of 1,616 (Fig. 2B and 1C) that was consistent with the retention of the 3-OH C10 fatty acid residue at the 3 position. In contrast, the laboratory-adapted strain, PAO-1, gave a predominant penta-acylated ion species with an m/z of 1,447 (Fig. 2A and 1D). Loss of deacylase enzymatic activity was confirmed by capillary gas chromatography using flame ionization detection analysis of LPS isolated after growth in N-minimal medium supplemented with 1 mM MgCl2 (Fig. 3A). Similar MS and GC results were observed for lipid A isolated from 32751 and strain 32751Tn, in which the transposon was removed (data not shown).
The location of the transposon in mutant strain 32751 was shown by sequence analysis to be inserted after nucleotide 196 in the open reading frame PA4661 (www.pseudomonas.com). PA4661 is predicted to encode a hypothetical protein of 173 amino acids in length. Analysis of the amino acid sequence using the SignalP 3.0 program indicated the presence of a 23-amino-acid type I signal peptide. Separation by isopycnic density gradient centrifugation revealed that PagL deacylase activity was localized to the outer membrane protein peak, with no deacylase activity found in fractions containing the inner membrane protein NADH oxidase (data not shown), indicating that P. aeruginosa PagL is an outer membrane protein. The specific enzymatic activity of the deacylase in the P. aeruginosa membranes was less than the Salmonella enterica serovar Typhimurium PagL enzyme activity (0.01 to 0.015 nmol/min/mg and 0.5 nmol/min/mg, respectively) (38). Therefore, PA4661 likely encodes a functional P. aeruginosa outer membrane lipid A deacylase.
Expression of pagL in trans indicated that P. aeruginosa PA4661 is responsible for deacylase activity. To further demonstrate that PA4661 encoded the deacylase enzyme in P. aeruginosa, the pagL gene was amplified by PCR from genomic DNA prepared from the wild-type laboratory-adapted isolate, PAO-1. Deacylase activity was recovered from the P. aeruginosa pagL mutant as shown qualitatively by MS (Fig. 2B and D; presence of m/z = 1,447) and quantitatively by GC (Fig. 3B; decreased levels of 3-OH C10 fatty acid) analysis after transformation with pPAPagL. The ion species at m/z values of 1,367 and 1,536 (Fig. 2) correspond to the loss of a phosphate residue (m/z = 80) and are probably artifacts of the MS analysis. In addition, adjacent peaks to the penta- and hexa-acylated ion species that differ by 16 m/z units represent the addition of an additional hydroxyl group (OH) (m/z = 1,463 and 1,632).
In addition, heterologous expression of P. aeruginosa pagL in the enteric bacteria Salmonella enterica serovar Typhimurium and E. coli resulted in the loss of a 3-OH C14 from the 3 position of enteric lipid A, confirming the heterologous expression results of Geurtsen et al. in the E. coli strain, BL21 STAR (19) (data not shown). Based on these data, the P. aeruginosa PagL functions as a deacylase in Salmonella enterica serovar Typhimurium and E. coli and indicates a lack of specificity for fatty acid length.
P. aeruginosa deacylase activity in acute clinical and environmental isolates was activated by growth in media limited for magnesium. Previously, we have demonstrated that P. aeruginosa laboratory-adapted strains (PAO-1 and PAK) and isolates from patients with CF constitutively generate a penta-acylated lipid A structure (m/z = 1,447; Fig. 1D) as a result of 3-O deacylation, i.e., the removal of the 3-OH C10 at the 3-position of lipid A (2, 27). Interestingly, deacylated lipid A structures were not observed in P. aeruginosa isolates from patients with acute infections (blood, urinary tract infection [UTI], and eye) or from the environment (16, 17). This observation suggested that the regulation of deacylase activity in CF clinical isolates and laboratory-adapted strains was altered relative to acute infection isolates.
We examined whether deacylase activity in P. aeruginosa isolates from acute clinical infection isolates (blood, UTI, and eye) was regulated by growth conditions. Five acute infection isolates were analyzed for deacylated lipid A species after growth under magnesium-limited conditions, a growth condition known to induce lipid A modifications (17, 21, 22). Lipid A was isolated from individual P. aeruginosa clinical isolates after growth in minimal medium supplemented with either low (8 μM) or high (1 mM) magnesium concentrations. MALDI-TOF MS analyses of the individual lipid A preparations isolated from all acute infection isolates were similar when grown under the same conditions. A representative P. aeruginosa blood isolate grown in high-magnesium medium gave a dominant [M-H]– ion at an m/z of 1,419 (Fig. 4A) representative of a penta-acylated lipid A structure that retained the 3-OH C10 fatty acid at the 3 position (Fig. 1B). In contrast, lipid A isolated from cells grown in low-magnesium medium had a more complex MS spectrum containing ion species that represent both penta- (m/z = 1,447) and hexa-acylated species (m/z = 1,616, 1,685, and 1,815) (Fig. 4B). The additional ion species at m/z = 1,815 corresponded to the addition of an aminoarabinose residue (m/z = 131) to one of the terminal phosphates in the hexa-acylated species, m/z = 1,685 (Fig. 1E). Peaks adjacent to the penta- and hexa-acylated ion species that differ by 16 m/z units represent the addition of another hydroxyl group (OH) (m/z = 1,463, 1,632, 1,701, and 1,831). These ion species represent lipid A structures constitutively produced in clinical isolates from patients with CF and demonstrate that acute infection isolates of P. aeruginosa have inducible enzyme systems as compared to laboratory-adapted or CF isolates. These results are in contrast to the work of Geurtsen et al. that suggested that P. aeruginosa PagL enzymatic activity was not regulated by growth under different magnesium concentrations for the laboratory-adapted P. aeruginosa strain PAO-1 (19). However, their immunoblotting results were from laboratory-adapted P. aeruginosa strains grown in LB, a rich magnesium-replete growth condition. Similar results were observed for lipid A isolated from 11 environmental isolates (data not shown). Therefore, an alteration in PagL activity occurs in isolates from the CF airway and in laboratory-adapted bacteria.
Temperature regulation of deacylase activity in laboratory-adapted isolates of P. aeruginosa. At low growth temperatures (<21°C), bacteria alter lipid A acylation patterns, presumably to maintain and/or adjust membrane fluidity of the outer membrane as an adaptive response. Such adaptation includes induction of acyltransferase activity that leads to an increase in the proportion of cis-unsaturated fatty acids, such as C16:1, in the lipid A structure (8, 10, 40). To determine the effect of temperature on the regulation of P. aeruginosa deacylase activity, lipid A was isolated from two laboratory-adapted wild-type P. aeruginosa strains, PAO-1 and PAK, grown over a wide range of temperatures (15°C to 42°C). This temperature range corresponds to temperatures encountered in diverse environments, from soil to mammals. Lipid A was isolated at various temperature points (15, 21, 30, 37, and 42°C) after overnight growth in LB medium, and the fatty acids were analyzed by capillary gas chromatography using flame ionization detection (Fig. 5A). These experiments showed that when grown at low temperatures (15 or 21°C), P. aeruginosa deacylase enzymatic activity was altered in both laboratory-adapted isolates tested, as indicated by the increase in the levels of 3-OH C10 fatty acid attached to lipid A (3 and 3' positions) (Fig. 5A).
Mass-spectrometric analysis of lipid A isolated from PAO-1 and PAK grown over the same range of temperatures revealed a single major peak at an m/z of 1,447 in each strain upon growth at 42°C, consistent with the presence of a penta-acylated species with the 3 position 3-OH C10 fatty acid having been deacylated (Fig. 5C and E and 1D). Interestingly, the small difference in levels of 3-OH C10 fatty acid observed between PAK and PAO-1 after growth at 15°C (34.6% versus 30.5%, respectively) was associated with slightly disparate lipid A profiles between these two strains (Fig. 5A). The predominant peaks observed from lipid A isolated from PAK after growth at 15°C are at m/z values of 1,616 and 1,632 (Fig. 5B) and are consistent with a hexa-acylated lipid A species with a 3-OH C10 fatty acid at the 3 position and one (m/z = 1,616) or two (m/z = 1,632) acyloxyacyl 2-OH C12 fatty acid residues (Fig. 1C). In contrast, a second set of peaks (m/z = 1,447 and 1,463) was observed for lipid A isolated from PAO-1 after growth at 15°C (Fig. 5D), consistent with the presence of a penta-acylated lipid A structure also observed after growth at elevated temperatures (Fig. 1D). These results suggest that P. aeruginosa deacylase activity is reduced at lower temperatures and the level of inhibition is strain dependent.
Interestingly, the level of 3-OH C10 in lipid A isolated from the transposon insertion mutant 32751 was similar to levels observed for lipid A isolated from PAK after growth in low temperature conditions. Membranes isolated from PAK grown at 15°C show an approximate 10- to 15-fold reduction in PagL 3-O-deacylase activity (0.0015 to 0.0022 nmol/min/mg) compared to membranes isolated from PAK grown at 37°C (0.022 nmol/min/mg). These results further suggested that in PAK, less PagL enzymatic activity was present in the membranes at low temperature, indicating that part of the difference between environmental isolates and CF and laboratory-adapted isolates involves less production or stability of the enzyme.
Some P. aeruginosa clinical isolates from CF patients with severe pulmonary disease are deficient in lipid A deacylase activity. To explore the possibility that changes in lipid A structure might contribute to lung disease progression in CF patients chronically infected with P. aeruginosa, we performed an analysis of lipid A isolated from longitudinal clinical isolates of three patients: a child with mild CF lung disease, a child with severe lung disease, and an adult with mild CF lung disease, as defined by serial measurements of lung function. Lipid A was isolated from the individual P. aeruginosa clinical isolates after growth in minimal medium supplemented with high (1 mM) magnesium concentrations. MALDI-TOF MS analyses of the individual lipid A preparations isolated from three clinical isolates are shown. A P. aeruginosa isolate from the child with mild CF lung disease (CF565) revealed dominant ions that represent both penta- (m/z = 1,447; Fig. 1D) and hexa-acylated species (m/z = 1,616; Fig. 1C) (m/z = 1,685, Fig. 1E) (Fig. 6A). Adjacent peaks to the penta- and hexa-acylated ion species that differ by 16 m/z units represent the addition of an additional hydroxyl group. Similar MS spectra were obtained for P. aeruginosa isolates from the adult with mild CF lung disease (data not shown).
In contrast, lipid A from clinical isolates from the child with severe CF lung disease and from a second CF patient with severe lung disease (10128 and SE22) had ion species that represent both hexa- (m/z = 1,616; Fig. 1C) and a novel hepta-acylated species (m/z = 1,854; Fig. 1F) (Fig. 6B and C, respectively). The hepta-acylated structure was consistent with four fatty acids (two 3-OH-C10 and two 3-OH C12) attached to the glucosamine backbone with three secondary "piggyback" fatty acids (C12, 2-OH C12, and C16). These results are consistent with a loss of deacylase enzymatic activity in these P. aeruginosa isolates. Adjacent peaks to the hexa- and hepta-acylated ion species that differ by 16 m/z units represent the addition of an additional hydroxyl group. Consistent with this hypothesis, deacylated lipid A species (m/z = 1,447; Fig. 1D) (m/z = 1,685; Fig. 1E) were recovered upon transformation of these isolates with pPAPagL, as determined by MS and GC analyses (data not shown), indicating that the deficiency of deacylated lipid A species in isolates from CF patients with severe lung disease is due to loss of deacylase enzymatic activity.
This analysis suggested a potential association between severe obstructive CF lung disease and loss of P. aeruginosa lipid A deacylation. In order to evaluate this potential association, lipid A was purified from P. aeruginosa clinical isolates of 21 patients with severe obstructive lung disease and analyzed by MALDI-TOF MS. This analysis showed that lipid A purified from P. aeruginosa isolates from seven patients (33%) had hexa-acylated (m/z = 1,616; Fig. 1C) or novel hepta-acylated structures (m/z = 1,854; Fig. 1F) but lacked the corresponding deacylated lipid A structures (m/z = 1,447 or 1,685; Fig. 1D and E, respectively) observed in isolates from other CF patients with severe lung disease, patients with milder CF lung disease, acute infections, and environmental isolates (data not shown). The presence of a full-length copy of the pagL gene was confirmed for all isolates used in these studies by colony PCR using pagL-specific oligonucleotides (data not shown). These results suggest that loss of lipid A deacylase activity can occur during progression of CF lung disease. This suggests that as disease progresses, selection may occur for loss of activity that results in the formation of a hepta-acylated lipid A.
DISCUSSION
Adaptation of P. aeruginosa to life in the airways of individuals with CF results in the expression of specific lipid A structures not observed in P. aeruginosa isolates from acute infections (blood, UTI, and eye). The modifications that distinguish these isolates' lipid A structures include the addition of palmitate and aminoarabinose and the deacylation of the 3-position fatty acid (3-OH C10) by PagL enzymatic activity. Our results show that the P. aeruginosa PagL enzymatic activity, though constitutively active in P. aeruginosa clinical isolates from patients with CF and laboratory-adapted wild-type isolates, was regulated in clinical isolates from both acute infections and environmental isolates. Acute P. aeruginosa isolates produce a precursor penta-acylated lipid A (Fig. 1B) structure not normally observed in isolates from patients with CF or laboratory-adapted isolates (Fig. 1C to E). Interestingly, acute P. aeruginosa isolates have the ability to produce CF-like lipid A modification when grown under magnesium-limited conditions (m/z = 1,447 and 1,685) (Fig. 1). These results indicate that the biosynthetic pathways necessary for the synthesis of CF-specific lipid A were intact and able to be regulated.
We further demonstrated that P. aeruginosa regulated PagL enzymatic activity in laboratory-adapted isolates upon growth under low temperature, a condition that mimics environmental growth conditions. Interestingly, addition of an unsaturated C16:1 fatty acid residue, a cold-shock regulated response typical among enteric bacteria, was not observed in P. aeruginosa (8). Instead, GC and MS analysis of P. aeruginosa lipid A isolated after growth at low temperatures (15°C) indicated that PagL enzymatic activity was completely inhibited in the laboratory-adapted wild-type isolate, PAK, suggesting a role for deacylase regulation of membrane structure and fluidity in response to temperature. The loss of this activity on growth in nutrient broth at 37°C or in the CF airway may promote PagL expression.
To determine the role of PagL in the generation of CF-specific lipid A structures, a P. aeruginosa transposon mutant, 35721 (19), was analyzed by MS and GC (Fig. 2 and 3). The insertion of the transposon in mutant strain 32751 mapped to open reading frame PA4661, which was predicted to encode a hypothetical protein of 173 amino acids in length. PA4661 is part of a group of six genes (hemA, prfA, hemK, moeB, murI, and PA4661), of which the first five genes form a putative operon. Genes in the five-gene operon include: hemA and hemK, required for heme biosynthesis (41); moeB, required for molybdopterin cofactor biosynthesis (43); prfA, encoding a homolog of the E. coli peptide chain release factor that directs the termination of translation in response to specific peptide chain termination codons (9); and murI, required for biosynthesis of D-glutamate and peptidoglycan (14). A 69-nucleotide intragenic region is present between murI and PA4661, suggesting that PA4661 is not part of the previous operon.
Deacylase activity was recovered in the P. aeruginosa transposon mutant background upon transformation and expression with either the P. aeruginosa PA4661 (Fig. 2 and 3) or Salmonella enterica serovar Typhimurium pagL gene (data not shown). Interestingly, two additional transposon insertion mutants located in the conserved C terminus of the protein retained deacylase activity, suggesting that this region of the proteins is not required for this activity. Heterologous expression of P. aeruginosa PA4661 in either a Salmonella enterica serovar Typhimurium PhoP-null strain or E. coli (both of which lack deacylase activity) resulted in the presence of deacylated lipid A species by MS and GC analysis. Taken together, these results suggest a relaxed chain length specificity for the P. aeruginosa and Salmonella enterica serovar Typhimurium pagL genes, since they have the ability to remove either longer (3-OH C14) or shorter (3-OH C10) fatty acid residues from the 3 position of lipid A, respectively. Finally, in contrast to the recently published results from Geurtsen et al., secondary addition of a C16 fatty acid after heterologous expression of P. aeruginosa pagL in either E. coli or Salmonella enterica serovar Typhimurium was not observed (19). Since different E. coli strain backgrounds were used for the heterologous-expression studies (XL1-Blue versus BL21 Star), strain-specific differences, such as constitutive expression of the genes necessary for the addition of aminoarabinose or palmitate observed in E. coli BL21 Star (39), may play a role in secondary lipid A modifications.
Lipid A isolated from P. aeruginosa clinical isolates from patients with CF (1 to 3 years of age) demonstrates constitutive expression of deacylase enzyme activity. A penta-acylated lipid A structure that contains a 3-OH C10 fatty acid (Fig. 1B) in the 3 position was not observed in CF isolates but was observed in isolates from acute infections or environmental isolates. Using clinical isolates from patients with severe pulmonary disease (as defined by lung function values), we showed that lipid A isolated from one-third of the patients' clinical isolates lacked deacylase activity compared to those from patients with mild pulmonary disease. Severe patients' isolates that lacked deacylase activity resulted in synthesis of hexa- or hepta-acylated structures that retained the 3-OH C10 fatty acid at the 3 position. The hepta-acylated lipid A structure (m/z = 1,854; Fig. 1F) was modified by an acyloxyacyl addition of C16:0 at the 3'-position fatty acid. Deacylated lipid A structures were recovered in these P. aeruginosa clinical isolates upon complementation with the P. aeruginosa pagL gene, indicating that these strains have lost deacylase activity. The characterization of these clinical isolates may result in the identification of novel mutations in the P. aeruginosa pagL gene that affect activity of the enzyme, or the identification of regulatory pathway alterations that result in the synthesis of CF-specific lipid A modifications. Therefore, deciphering the role of this late adaptation of P. aeruginosa to the CF airway may be important in understanding and preventing the progression of chronic CF lung disease.
ACKNOWLEDGMENTS
We thank Lucas Hoffman for critical review of the manuscript.
This work was supported by grants from the National Institutes of Health (AI47938 and DK064954 to S.I.M.) and the Cystic Fibrosis Foundation (R.K.E.).
REFERENCES
Berger, M. 2002. Inflammatory mediators in cystic fibrosis lung disease. Allergy Asthma Proc. 23:19-25.
Bhat, R., A. Marx, C. Galanos, and R. S. Conrad. 1990. Structural studies of lipid A from Pseudomonas aeruginosa PAO1: occurrence of 4-amino-4-deoxyarabinose. J. Bacteriol. 172:6631-6636.
Bishop, R. E. 2005. Fundamentals of endotoxin structure and function. Contrib. Microbiol. 12:1-27.
Bishop, R. E., H. S. Gibbons, T. Guina, M. S. Trent, S. I. Miller, and C. R. Raetz. 2000. Transfer of palmitate from phospholipids to lipid A in outer membranes of gram-negative bacteria. EMBO J. 19:5071-5080.
Bishop, R. E., S. H. Kim, and A. El Zoeiby. 2005. Role of lipid A palmitoylation in bacterial pathogenesis. J. Endotoxin Res. 11:174-180.
Burns, J. L., R. L. Gibson, S. McNamara, D. Yim, J. Emerson, M. Rosenfeld, P. Hiatt, K. McCoy, R. Castile, A. L. Smith, and B. W. Ramsey. 2001. Longitudinal assessment of Pseudomonas aeruginosa in young children with cystic fibrosis. J. Infect. Dis. 183:444-452.
Caroff, M., A. Tacken, and L. Szabo. 1988. 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. 175:273-282.
Carty, S. M., K. R. Sreekumar, and C. R. Raetz. 1999. Effect of cold shock on lipid A biosynthesis in Escherichia coli. Induction at 12°C of an acyltransferase specific for palmitoleoyl-acyl carrier protein. J. Biol. Chem. 274:9677-9685.
Craigen, W. J., and C. T. Caskey. 1987. The function, structure and regulation of E. coli peptide chain release factors. Biochimie 69:1031-1041.
Cronan, J. E., Jr., and E. P. Gelmann. 1975. Physical properties of membrane lipids: biological relevance and regulation. Bacteriol. Rev. 39:232-256.
Dakin, C. J., A. H. Numa, H. Wang, J. R. Morton, C. C. Vertzyas, and R. L. Henry. 2002. Inflammation, infection, and pulmonary function in infants and young children with cystic fibrosis. Am. J. Respir. Crit. Care Med. 165:904-910.
Darveau, R. P., M. D. Cunningham, T. Bailey, C. Seachord, K. Ratcliffe, B. Bainbridge, M. Dietsch, R. C. Page, and A. Aruffo. 1995. Ability of bacteria associated with chronic inflammatory disease to stimulate E-selectin expression and promote neutrophil adhesion. Infect. Immun. 63:1311-1317.
Davies, J. C. 2002. Pseudomonas aeruginosa in cystic fibrosis: pathogenesis and persistence. Paediatr. Respir. Rev. 3:128-134.
Doublet, P., J. van Heijenoort, and D. Mengin-Lecreulx. 1992. Identification of the Escherichia coli murI gene, which is required for the biosynthesis of D-glutamic acid, a specific component of bacterial peptidoglycan. J. Bacteriol. 174:5772-5779.
Ernst, R. K., T. Guina, and S. I. Miller. 2001. Salmonella typhimurium outer membrane remodeling: role in resistance to host innate immunity. Microbes Infect. 3:1327-1334.
Ernst, R. K., A. M. Hajjar, J. H. Tsai, S. M. Moskowitz, C. B. Wilson, and S. I. Miller. 2003. Pseudomonas aeruginosa lipid A diversity and its recognition by Toll-like receptor 4. J. Endotoxin Res. 9:395-400.
Ernst, R. K., E. C. Yi, L. Guo, K. B. Lim, J. L. Burns, M. Hackett, and S. I. Miller. 1999. Specific lipopolysaccharide found in cystic fibrosis airway Pseudomonas aeruginosa. Science 286:1561-1565.
Garcia Vescovi, E., F. C. Soncini, and E. A. Groisman. 1996. Mg2+ as an extracellular signal: environmental regulation of Salmonella virulence. Cell 84:165-174.
Geurtsen, J., L. Steeghs, J. T. Hove, P. van der Ley, and J. Tommassen. 2005. Dissemination of lipid A deacylases (pagL) among gram-negative bacteria: identification of active-site histidine and serine residues. J. Biol. Chem. 280:8248-8259.
Gibbons, H. S., S. Lin, R. J. Cotter, and C. R. Raetz. 2000. Oxygen requirement for the biosynthesis of the S-2-hydroxymyristate moiety in Salmonella typhimurium lipid A. Function of LpxO, a new Fe2+/-ketoglutarate-dependent dioxygenase homologue. J. Biol. Chem. 275:32940-32949.
Guo, L., K. B. Lim, J. S. Gunn, B. Bainbridge, R. P. Darveau, M. Hackett, and S. I. Miller. 1997. Regulation of lipid A modifications by Salmonella typhimurium virulence genes phoP-phoQ. Science 276:250-253.
Guo, L., K. B. Lim, C. M. Poduje, M. Daniel, J. S. Gunn, M. Hackett, and S. I. Miller. 1998. Lipid A acylation and bacterial resistance against vertebrate antimicrobial peptides. Cell 95:189-198.
Guy-Caffey, J. K., M. P. Rapoza, K. A. Jolley, and R. E. Webster. 1992. Membrane localization and topology of a viral assembly protein. J. Bacteriol. 174:2460-2465.
Guzman, L.-M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121-4130.
Hajjar, A. M., R. K. Ernst, J. H. Tsai, C. B. Wilson, and S. I. Miller. 2002. Human Toll-like receptor 4 recognizes host-specific LPS modifications. Nat. Immunol. 3:354-359.
Jacobs, M. A., A. Alwood, I. Thaipisuttikul, D. Spencer, E. Haugen, S. Ernst, O. Will, R. Kaul, C. Raymond, R. Levy, L. Chun-Rong, D. Guenthner, D. Bovee, M. V. Olson, and C. Manoil. 2003. Comprehensive transposon mutant library of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 100:14339-14344.
Kulshin, V. A., U. Zahringer, B. Lindner, K. E. Jager, B. A. Dmitriev, and E. T. Rietschel. 1991. Structural characterization of the lipid A component of Pseudomonas aeruginosa wild-type and rough mutant lipopolysaccharides. Eur. J. Biochem. 198:697-704.
Macfarlane, E. L., A. Kwasnicka, and R. E. Hancock. 2000. Role of Pseudomonas aeruginosa PhoP-PhoQ in resistance to antimicrobial cationic peptides and aminoglycosides. Microbiology 146:2543-2554.
Macfarlane, E. L., A. Kwasnicka, M. M. Ochs, and R. E. Hancock. 1999. PhoP-PhoQ homologues in Pseudomonas aeruginosa regulate expression of the outer-membrane protein OprH and polymyxin B resistance. Mol. Microbiol. 34:305-316.
Miller, S. I., R. K. Ernst, and M. W. Bader. 2005. LPS, TLR4 and infectious disease diversity. Nat. Rev. Microbiol. 3:36-46.
Miller, S. I., A. M. Kukral, and J. J. Mekalanos. 1989. A two-component regulatory system (PhoP/PhoQ) controls Salmonella typhimurium virulence. Proc. Natl. Acad. Sci. USA 86:5054-5058.
Moskowitz, S. M., R. K. Ernst, and S. I. Miller. 2004. 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. 186:575-579.
Osborn, M. J., J. E. Gander, E. Parisi, and J. Carson. 1972. Mechanism of assembly of the outer membrane of Salmonella typhimurium. Isolation and characterization of cytoplasmic and outer membrane. J. Biol. Chem. 247:3962-3972.
Rosenfeld, M., R. L. Gibson, S. McNamara, J. Emerson, J. L. Burns, R. Castile, P. Hiatt, K. McCoy, C. B. Wilson, A. Inglis, A. Smith, T. R. Martin, and B. W. Ramsey. 2001. Early pulmonary infection, inflammation, and clinical outcomes in infants with cystic fibrosis. Pediatr. Pulmonol. 32:356-366.
Rowntree, R. K., and A. Harris. 2003. The phenotypic consequences of CFTR mutations. Ann. Hum. Genet. 67:471-485.
Somerville, J. E., Jr., L. Cassiano, B. Bainbridge, M. D. Cunningham, and R. P. Darveau. 1996. A novel Escherichia coli lipid A mutant that produces an antiinflammatory lipopolysaccharide. J. Clin. Investig. 97:359-365.
Trent, M. S. 2004. Biosynthesis, transport, and modification of lipid A. Biochem. Cell Biol. 82:71-86.
Trent, M. S., W. Pabich, C. R. Raetz, and S. I. Miller. 2001. A PhoP/PhoQ-induced Lipase (PagL) that catalyzes 3-O-deacylation of lipid A precursors in membranes of Salmonella typhimurium. J. Biol. Chem. 276:9083-9092.
Trent, M. S., A. A. Ribeiro, W. T. Doerrler, S. Lin, R. J. Cotter, and C. R. Raetz. 2001. Accumulation of a polyisoprene-linked amino sugar in polymyxin-resistant Salmonella typhimurium and Escherichia coli: structural characterization and transfer to lipid A in the periplasm. J. Biol. Chem. 276:43132-43144.
Vigh, L., B. Maresca, and J. L. Harwood. 1998. Does the membrane's physical state control the expression of heat shock and other genes Trends Biochem. Sci. 23:369-374.
Wandersman, C., and P. Delepelaire. 2004. Bacterial iron sources: from siderophores to hemophores. Annu. Rev. Microbiol. 58:611-647.
Whitfield, C., N. Kaniuk, and E. Frirdich. 2003. Molecular insights into the assembly and diversity of the outer core oligosaccharide in lipopolysaccharides from Escherichia coli and Salmonella. J. Endotoxin Res. 9:244-249.
Wuebbens, M. M., and K. V. Rajagopalan. 1995. Investigation of the early steps of molybdopterin biosynthesis in Escherichia coli through the use of in vivo labeling studies. J. Biol. Chem. 270:1082-1087.
Yi, E. C., and M. Hackett. 2000. Rapid isolation method for lipopolysaccharide and lipid A from gram-negative bacteria. Analyst 125:651-656.
Yu, H., and N. E. Head. 2002. Persistent infections and immunity in cystic fibrosis. Front. Biosci. 7:d442-d457.
Zhou, Z., K. A. White, A. Polissi, C. Georgopoulos, and C. R. Raetz. 1998. Function of Escherichia coli MsbA, an essential ABC family transporter, in lipid A and phospholipid biosynthesis. J. Biol. Chem. 273:12466-12475.(Robert K. Ernst,1Kristin )