fatty acid biosynthesis in pseudomonas aeruginosa is ... · fatty acid biosynthesis in pseudomonas...

Fatty Acid Biosynthesis in Pseudomonas aeruginosa Is Initiated by theFabY Class of �-Ketoacyl Acyl Carrier Protein Synthases

Yanqiu Yuan, Meena Sachdeva, Jennifer A. Leeds, and Timothy C. Meredith

Infectious Diseases Area, Novartis Institutes for BioMedical Research, Cambridge, Massachusetts, USA

The prototypical type II fatty acid synthesis (FAS) pathway in bacteria utilizes two distinct classes of �-ketoacyl synthase (KAS)domains to assemble long-chain fatty acids, the KASIII domain for initiation and the KASI/II domain for elongation. The centralrole of FAS in bacterial viability and virulence has stimulated significant effort toward developing KAS inhibitors, particularlyagainst the KASIII domain of the �-acetoacetyl-acyl carrier protein (ACP) synthase FabH. Herein, we show that the opportunis-tic pathogen Pseudomonas aeruginosa does not utilize a FabH ortholog but rather a new class of divergent KAS I/II enzymes toinitiate the FAS pathway. When a P. aeruginosa cosmid library was used to rescue growth in a fabH downregulated strain ofEscherichia coli, a single unannotated open reading frame, PA5174, complemented fabH depletion. While deletion of all fourKASIII domain-encoding genes in the same P. aeruginosa strain resulted in a wild-type growth phenotype, deletion of PA5174alone specifically attenuated growth due to a defect in de novo FAS. Siderophore secretion and quorum-sensing signaling, partic-ularly in the rhl and Pseudomonas quinolone signal (PQS) systems, was significantly muted in the absence of PA5174. The defectcould be repaired by intergeneric complementation with E. coli fabH. Characterization of recombinant PA5174 confirmed apreference for short-chain acyl coenzyme A (acyl-CoA) substrates, supporting the identification of PA5174 as the predominantenzyme catalyzing the condensation of acetyl coenzyme A with malonyl-ACP in P. aeruginosa. The identification of the func-tional role for PA5174 in FAS defines the new FabY class of �-ketoacyl synthase KASI/II domain condensation enzymes.

Fatty acid synthesis (FAS) is a primary metabolic pathway cen-tral to both eukaryotes and bacteria. A wide range of cellular

processes are dependent on fatty acids, from the biosyntheses ofessential cellular structural components (membrane phospholip-ids, lipoproteins, and lipoglycans), cofactors (lipoate and biotin),and energy storage reserves to diffusible secondary metabolites(quorum-sensing [QS] signal molecules, siderophores, and bio-surfactants). Although common to bacteria, plants, and animals,fatty acids are assembled using two distinct types of biosyntheticmachinery. In the mammalian FAS I type, fatty acids are builtusing a single multisubunit polypeptide that serves as the acylcarrier protein (ACP) tether and source of enzymatic activities foracyl chain elongation (77, 78). Plants and bacteria, with the excep-tion of some actinomycetes (27, 71), utilize the disassociated FASII pathway for de novo fatty acid production (17, 49, 86). In thissystem, multiple discrete enzymes assemble the nascent fatty acidchain on a freely diffusible phosphopantetheinylated ACP. Thedisassociated and modular organization of bacterial FAS II en-ables the synthesis of diverse fatty acid products utilized by mul-tiple pathways, a versatility that is unmatched in FAS I systems,where the scope is more limited to long-chain saturated fatty acids(86).

While the architecture of the FAS enzymatic machinery differsconsiderably, the overall catalytic strategy is similar among bothFAS I and FAS II model organisms (78, 86). Fatty acids are assem-bled two carbons at a time through iterative rounds of Claisen-type condensations between malonyl-ACP donor and nascent acylacceptor. In the bacterial FAS II prototype pathway of Escherichiacoli, the initial condensation between malonyl-ACP and acetylcoenzyme A (acetyl-CoA) to form �-acetoacetyl-ACP is catalyzedby the initiating enzyme �-ketoacyl ACP synthase III (FabH) (84).The �-keto carbonyl is next reduced to a hydroxyl group by theNADPH-dependent �-ketoacyl-ACP reductase (FabG) (2), dehy-drated by �-hydroxyacyl-ACP dehydratase (FabZ/FabA) to form

trans-2-enoyl-ACP (42, 54), and then reduced to a saturatedacyl-ACP by NADH-dependent enoyl-ACP reductase (FabI) (4).The acyl-ACP intermediate is condensed with malonyl-ACPby the elongation �-ketoacyl ACP synthases I/II (FabB/FabF)(31), the product of which reenters the reduction cycle for anotherround of elongation. In E. coli, unsaturated fatty acids are pro-duced from trans-2-decenoyl-ACP by the cis-/trans-enoyl-ACPisomerase activity of FabA (42). The cis-3-decenoyl-ACP is thenspecifically extended by the FabB elongase to make unsaturatedfatty acids (69). Intermediates from the FAS pathway also providethe �-hydroxyl myristoyl-ACP and lauroyl-/myristoyl-ACPneeded for lipopolysaccharide biosynthesis (64).

The divergence between mammalian and bacterial FAS path-ways, coupled with the multifaceted role in both the viability andvirulence of many pathogenic bacteria, makes fatty acid inhibitionan attractive strategy for the development of new antimicrobialagents (8, 32, 37, 92). Much effort has been devoted toward un-derstanding bacterial FAS enzymes and cognate inhibitors, partic-ularly for the initiating condensing enzyme FabH (9, 44, 56, 91).Unlike bacterial FAS II which has discrete initiating and elongat-ing �-ketoacyl synthases (KAS), both the priming and the subse-quent elongation condensations are catalyzed by the same KASdomain in FAS I systems using first acetyl-ACP and then subse-quently �-ketoacyl-ACP acceptor. Thus, there is no equivalentKAS enzymatic activity in mammalian cells that directly utilizes

Received 7 May 2012 Accepted 21 June 2012

Published ahead of print 29 June 2012

Address correspondence to Timothy C. Meredith, [email protected].

Supplemental material for this article may be found at http://jb.asm.org/.

Copyright © 2012, American Society for Microbiology. All Rights Reserved.

doi:10.1128/JB.00792-12

October 2012 Volume 194 Number 19 Journal of Bacteriology p. 5171–5184 jb.asm.org 5171

on July 20, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

on July 20, 2020 by guest

http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from


.org/D

ownloaded from


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

http://dx.doi.org/10.1128/JB.00792-12

http://jb.asm.org

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

http://jb.asm.org/

acetyl-CoA, and the KASIII domain of FabH is almost exclusivelya bacterium-specific domain (12). A dedicated protein for fattyacid initiation allows FabH to serve as a posttranscriptional regu-latory checkpoint for de novo fatty acid flux. In E. coli, FabH issubject to stringent feedback regulation by long-chain acyl-ACPso as to coordinate the rate of initiation with that of elongation(36). FabH is highly conserved, possessing an invariant triad ofactive site residues (28), as well as being essential across a diversespectrum of both Gram-positive and Gram-negative bacteria thatincludes E. coli (47), Streptomyces coelicolor (67), and Lactococcuslactis (47). The uniqueness of the KASIII domain, conservationamong FabH orthologs, and regulatory role in controlling FASflux all suggest potential for broad-spectrum and bacterial selec-tive inhibition.

It has recently been reported that de novo FAS can be bypassedby exogenous fatty acid uptake from serum in certain Gram-pos-itive pathogens, including Streptococcus (6). However, otherGram-positive bacteria such as Staphylococcus are not subject tobypass of FAS by exogenous fatty acids due to differences in FASregulatory mechanisms (58). The essentiality of the FAS pathwayhas not been questioned in Gram-negative bacteria, in large partdue to the dependence of lipopolysaccharide biogenesis on FASprecursors (59). To this end, we initiated a program to study FASinitiation in the opportunistic nosocomial human pathogen Pseu-domonas aeruginosa. P. aeruginosa is a versatile Gram-negativepathogen, associated with infections in burn patients, in the respi-ratory tracts of cystic fibrosis patients, and more generally amongthe immunocompromised (43). We hypothesized that inhibitionof FabH-mediated FAS initiation in P. aeruginosa would not onlyprevent growth through depleting fatty acid pools for membranephospholipid and lipopolysaccharide biogenesis but also suppressquorum sensing, since all three major QS signal molecules containfatty acid moieties (87). We herein provide evidence that P.aeruginosa does not utilize a FabH KASIII ortholog as the primarymeans for FAS initiation like other bacteria studied thus far butinstead uses the unannotated open reading frame (ORF) PA5174.Depletion of the PA5174 gene product elicited a pleiotropicgrowth and QS-defective phenotype. The levels of siderophoresand numerous virulence factors decreased, which is consistentwith a general role of PA5174 in FAS initiation. The identificationof the functional role for PA5174, the inaugural member of ahighly divergent class of KASI/II domain enzymes, defines the newFabY class of �-ketoacyl synthase condensation enzymes presentin P. aeruginosa.

MATERIALS AND METHODSBacterial strains and growth conditions. E. coli strains were both K-12descendants (EPI100-TiR and the reference strain BW25113), while all P.aeruginosa strains were derived from the reference strain PAO1 (80). E.coli and P. aeruginosa were grown in LB-Miller medium at 37°C unlessotherwise noted. Antibiotic markers were selected with gentamicin (Gent)(100 �g/ml in P. aeruginosa or Chromobacterium violaceum and 10 �g/mlin E. coli), carbenicillin (Carb) (150 �g/ml in P. aeruginosa and 100 �g/mlin E. coli), chloramphenicol (Cam) (20 �g/ml in E. coli), and kanamycin(Kan) (25 �g/ml in E. coli). The bacterial strains and plasmids used in thisstudy are listed in Table 1.

Construction of P. aeruginosa deletion strains. The conjugation vec-tor pEX18ApGW encoding carbenicillin resistance (Carbr), oriT transferorigin, and sucrose counterselection (sacB) was utilized for all P. aerugi-nosa gene deletion constructs (15). Unmarked, in-frame internal genedeletion cassettes were made using �1-kb PCR amplicons (P1/P2 and

P3/P4) of homologous flanking DNA to each KASIII domain-containingopen reading frame (PA0998, PA0999, PA3286, and PA3333). Approxi-mately 90 bp was retained at both the N and C termini to mitigate polareffects. Deletion cassettes (�2 kb) were assembled by overlap PCR andinserted into pEX18ApGW by either standard DNA restriction/ligation(T4 DNA ligase) or by using the Gateway system cloning protocol (Invit-rogen). The PA5174-E. coli fabH exchange vector pTMT123 was con-structed by assembly of 5 DNA fragments (linearized pEX18ApGW [Hin-dIII/KpnI] with PCR products using primer pairs for upstream flankingDNA [PA5174 P1/P2], Gentr cassette [Gent for PA5174/rev], FabH of E.coli [fabH Ec for/rev], and downstream flanking DNA [PA5174 P3/P4])using the In-fusion system (Clontech). An internal DNA fragment of E.coli fabH in pTMT123 was removed by digestion (MfeI/SphI), end re-paired, and self ligated to generate the PA5174 knockout vectorpTMT124. The pEX18ApGW cviI::gentR vector was likewise assembledfrom 4 DNA fragments (linearized pEX18ApGW [HindIII/KpnI] withPCR products of primer pairs cviI P1/P2, Gent for cviI/rev, and cviI P3/P4) using the In-fusion system. Primers are listed in Table S1 in the sup-plemental material.

The vectors were mobilized in trans using either E. coli SM10 (76) or E.coli Stellar cells (Clontech) transformed with the helper plasmid pRK2013(24) as the conjugation-proficient donor strains. The E. coli donor and P.aeruginosa recipient (106 CFU in 10 �l for each strain) were cospotted onLB agar and incubated at 37°C for 8 h to overnight. The mating mixtureswere then scraped off the plate, resuspended, and plated on Pseudomonasisolation agar (PIA) to counterselect E. coli and with antibiotic to select forintegrated vector. Merodiploid colonies were confirmed by PCR, out-grown for 4 h in LB at 37°C, and then spread on LB agar containing 7%sucrose (LB–7% sucrose agar) (without NaCl) to counterselect coloniesharboring unresolved plasmid. Mutant alleles were checked in P. aerugi-nosa by colony PCR using primers annealing outside the targeted region aswell as confirming anticipated antibiotic sensitivity/resistance. The qua-druple knockout strain TMT16 was constructed stepwise by 4 successiverounds of gene deletion. The PA5174::gentR deletion strain TMT39 wasback complemented by electroporation (14) of the shuttle vector pZEN-PA5174, which was made by ligation of the PA5174 gene (amplified withprimers pZEN for NdeI/PA5174 pZEN rev XbaI) into a similarly re-stricted pZEN vector. The C. violaceum cviI::gentR (TMT45) reporterstrain was directly isolated by streaking the pTMT127 mating mixture tosingle colonies at 28°C on sucrose LB agar containing Gent and Carb. ThecviI deletion was confirmed in unpigmented candidate colonies (due toloss of homoserine lactone signaling [51]) by PCR analysis.

Construction and cosmid complementation of the fabH inducible E.coli strain TMY19. The L-arabinose (L-Ara)-inducible FabH strainTMY19 was constructed in E. coli EPI100 in two steps (Fig. 1). First, anintegration cassette targeted to the D-glucitol transporter operon(gutAEB) was constructed by cloning the Salmonella enterica subsp. en-terica (ATCC 19430) fabHSe (the Se superscript indicates that the fabHgene was from S. enterica) ortholog (amplified with primers fabH Se forEcoRI/fabH Se rev KpnI) in front of the PBAD promoter in pKD4-PBAD.The resulting plasmid (containing araC-PBADfabHSe) was used as the tem-plate for PCR amplification of the integration cassette (primers gutAEBpKD4 P1/gutAEB pKD4 P2). The 40 bp of homology to the D-glucitoloperon on the termini of the integration cassette directed site-specificrecombination using the Red recombinase system (18). Insertion offabHSe at the gut locus was confirmed by PCR analysis as well as by loss ofD-glucitol catabolism using MacConkey-sorbitol indicator plates (52).The resulting strain, merodiploid in fabH orthologs, was then subjected toa second round of Red recombinase-mediated gene disruption using afabHEc::camR (the Ec superscript indicates that the fabH gene was from E.coli) cassette (amplified using primers fabH Ec KO P1/fabH Ec KO P2with pKD3 as the template). The L-Ara conditional strain TMY19 wasmaintained on LB agar plus L-arabinose (0.2%) to enable initiation of fattyacid biosynthesis by heterologous expression of the integrated PBAD-fab-HSe cassette.

Yuan et al.

5172 jb.asm.org Journal of Bacteriology


.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

The pWEB cosmid cloning kit (Epicentre) was used to introduce alibrary of pWEB cosmids harboring P. aeruginosa PAO1 genomic DNAinto E. coli TMY19. Cosmids were directly introduced into strain TMY19that had been pregrown in medium containing 0.2% L-Ara by lambdaphage transduction as instructed by the manufacturer. Transductantswere isolated on unsupplemented LB agar (100 �g/ml Carb) for pWEBselection. Aliquots were plated in parallel on LB agar with L-Ara as thepositive control, along with aliquots of bacteria that were not infected tomonitor the spontaneous induction background. The plates were incu-bated at 37°C for 32 h before comparison. A total of 17 larger colonies thatstood out from the background growth due to leakiness of the PBAD pro-moter were selected for further analysis. Cosmids (pTMYcos1 to pTMY-cos17) were isolated, and junction sites were sequenced using pWebupseq/pWebdown seq (seq stands for sequencing) primers (cosmid sum-mary listed in Table S2 in the supplemental material). Purified cosmids

were then retransformed by electroporation into pTMY19 to confirmphenotypic rescue. The unidentified open reading frame PA5174, com-mon to all E. coli TMY19 growth-rescuing cosmids and immediately prox-imal to the vector junction site, was amplified and cloned into pWEBusing the In-fusion system (Clontech). The plasmid pWEB-PA5174 wasthen electroporated into strain TMY19 to evaluate complementation.

FabH complementation in E. coli BW25113. The PA5174 open read-ing frame was cloned into the isopropyl-�-D-thiogalactopyranoside(IPTG)-inducible expression vector pET24b(�) (Novagen) using thePCR product of primers pET24-PA5174 for/rev. The expression plasmidpET-PA5174 was transformed into E. coli BW25113. The Red recombi-nase system was used to delete the fabH gene using the fabHEc::camRcassette (see above) in order to generate strain TMY32.

Growth analysis. For E. coli and P. aeruginosa strains, starter cul-tures were made by scraping cells off LB agar plates that had been

TABLE 1 Bacterial strains and plasmids used in this study

Bacterial strain or plasmid Relevant genotype or phenotypea Source or reference

E. coli strainsEPI100-T1R Phage T1 resistant [F� mcrA �(mrr-hsdRMS-mcrBC) �80dlacZ�M15 �lacX74 recA1

endA1 araD139 �(ara leu)7697 galU galK �� rpsL nupG tonA]Epicentre

TMY19 EPI100-T1R fabH::camR gutA::kanR araC PBAD-fabHSe; Kanr Camr; D-glucitolnegative; L-arabinose conditional growth

This study

BW25113 E. coli -12 wild type [�(araD-araB)567 �lacZ4787(::rrnB-3) �� rph-1 �(rhaD-rhaB)568 hsdR514]

CGSC7636b

TMY32 BW25113 fabH::camR [pET-PA5174]; Kanr Camr This study

P. aeruginosa strainsNB52019 P. aeruginosa PAO1prototroph K767 K. PooleTMT01 NB52019 �PA3333 (fabH2) using pJLM34 This studyTMT02 NB52019 �PA0999 (pqsD) using pTMT107 This studyTMT12 NB52019 �PA3286 using pTMT111 This studyTMT15 NB52019 �PA0998 (pqsC) using pTMT114 This studyTMT16 NB52019 �PA0998 �PA0999 �PA3333 �PA3286 This studyTMT38 NB52019 PA5174::gentR fabHEc using pTMT123; Gentr This studyTMT39 NB52019 PA5174::gentR fabH= using pTMT124; Gentr This studyTMT41 TMT39 (pZEN-PA5174); Gentr Carbr This studyNB56503 Chromobacterium violaceum ATCC 31532 ATCCTMT45 NB56503 cviI::gentR using pTMT127; Gentr; HSL negative This study

PlasmidspWEB ColE1 cos ori; Carbr Neor host vector for cosmid library EpicentrepTMYcos1 to pTMYcos17 pWEB-NB52019 genome library selected in E. coli TMY19c This studypWEB-PA5174 pWEB with the PA5174 gene; Carbr This studypKD4 oriR� kanR; Kanr 18pKD4-PBAD pKD4 with the L-arabinose-inducible araC PBAD cassette This studypKD3 oriR� camR; Camr 18pUCGM pUC19 vector with the Gentr cassette for the PCR template This studypET24b(�) IPTG-inducible T7 promoter for protein expression; Kanr NovagenpET-PA5174 pET24b(�) with PA5174 and C-terminal His tag; Kanr This studypEX18ApGW Gene replacement conjugation vector; oriT�sacB�; Carbr 15pJLM34 pEX18ApGW-�PA3333 (in-frame deletion) This studypTMT107 pEX18ApGW-�PA0999 (in-frame deletion) This studypTMT111 pEX18ApGW-�PA3286 (in-frame deletion) This studypTMT114 pEX18ApGW-�PA0998 (in-frame deletion) This studypTMT123 pEX18ApGW-PA5174::gentR fabHEc; Gentr This studypTMT124 pEX18ApGW-PA5174::gentR fabH= fragment; Gentr This studypZEN E. coli (pMB1 ori)-P. aeruginosa (pAK1900 ori) shuttle vector with lacIq and

IPTG-inducible lacO PTRC; Carbr

This study

pZEN-PA5174 pZEN with the PA5174 gene This studypTMT127 pEX18ApGW- cviI::gentR This study

a Carbr, carbenicillin resistant; Camr, chloramphenicol resistant; Kanr, kanamycin resistant; fabHEc, fabH from E. coli; fabHSe, fabH from S. enterica.b Strain CGSC7636 at the Coli Genetic Stock Center (CGSC).c Insert coordinates listed in Table S2 in the supplemental material.

Initiation of Fatty Acid Biosynthesis by FabY

October 2012 Volume 194 Number 19 jb.asm.org 5173


.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

supplemented with appropriate selection antibiotics and/or inducers.Cultures were typically resuspended to �2 105 CFU/ml in LB anddistributed into clear 96-well flat-bottom untreated microplates (Co-star 3370). The plates were incubated at 37°C, and the optical densityat 600 nm (OD600) was recorded every 10 min on a Spectramax platereader with intermittent shaking.

Fatty acid composition analysis. Bacteria were streaked on LB agarplates and incubated overnight at 37°C. Biomass was scraped from thesurface and suspended in phosphate-buffered saline (PBS). The cell pelletwas thoroughly washed three times with PBS. Lipids were saponified,methylated, extracted, and washed according to the Sherlock microbialidentification system (Microbial ID, Inc., Newark, DE). Fatty acid methylester (FAME) composition was determined by gas chromatography withflame ionization detection. Structure assignments were made by compar-ison of retention times to those of authentic FA standards.

Macromolecular labeling of fatty acids. Colonies were scraped fromLB agar plates that had been grown overnight and resuspended to anOD600 of 0.1 (�107 CFU/ml) in M9 minimal salts medium supplementedwith glycerol (0.4% [vol/vol]), uridine (80 �g/ml), and tryptic soy broth(7% [vol/vol]). Cultures were incubated with shaking at 37°C for 3 h toreturn the cells back to the exponential growth phase before being backdiluted to an OD600 of 0.1 with prewarmed supplemented minimal me-dium. Aliquots of 50 �l were then added to 50 �l of prewarmed mediumin 96-well round-bottom microplates (catalog no. 351177; BD Falcon)that contained either [3H]acetate (3.47 Ci/mmol) (catalog no. NET003H;Perkin Elmer) for fatty acid or [3H]uridine (25.5 Ci/mmol) (catalog no.NET174; Perkin Elmer) for RNA precursor radiolabeling (final concen-tration of 10 �Ci/ml for each). After being incubated for 30 min at 37°C,wells were quenched with 100 �l of ice-cold 20% trichloroacetic acid. Theplates were held at 4°C overnight to precipitate macromolecules, and theprecipitates were collected by filtration (Perkin Elmer UniFilter-96 GF/C). The filters were washed extensively with water, dried, and sealed, andthen 40 �l of scintillant (Perkin Elmer Betaplate Scint) was added prior tocounting (Perkin Elmer MicroBeta Trilux).

Extraction and assay of quorum-sensing signal molecules. Cultureswere grown in LB medium for 24 h at 37°C with shaking (250 rpm) inbaffled flasks. Under these conditions, all cultures reached stationaryphase at least 8 h before harvesting and had similar final cell densities.Bacteria were pelleted by centrifugation (3,000 g, 5 min, room temper-ature [RT]), and the culture supernatants were decanted and passedthrough a sterile filter (0.45 �m). Quorum-sensing signal molecules weretwice extracted from 10 ml of supernatant for each sample using either an

equal volume of ethyl acetate for homoserine lactones (HSL) (74) or anequal volume of acidified ethyl acetate for 2-heptyl-3-hydroxy-4(1H)-quinolone (PQS) (25). The organic extracts were concentrated to drynessusing a nitrogen bubbler, and the residue was resuspended in 100 �l ofethyl acetate (HSL) or methanol (PQS). The extracts were analyzed bythin-layer chromatography (TLC). For the rhl-dependent acylated HSL,15 �l of each sample was spotted on Partisil LKC18 (Whatman) reverse-phase TLC plates, while 3 �l of each sample was spotted on reverse-phaseoctadecyl Si-C18 (Baker) TLC plates for the las-dependent 3-oxo HSL.The plates were developed with 60:40 (vol/vol) methanol (MeOH)-H2Oas described previously (74). The 3-oxo acyl HSL-inducible �-galactosi-dase bioreporter strain Agrobacterium tumefaciens NL4(pZLR4) ATCCBAA-240 was used in an agar overlay to visualize 3-oxo HSL spots (10),while the acyl HSL spots were detected with the reporter C. violaceumstrain TMT45 as has been described for similar C. violaceum reporterstrains (51). PQS content was assessed by normal-phase TLC on activatedsilica 60 F254 plates (Merck) that had been pretreated with a 5% KH2PO4

soak before development with a 95:5 (vol/vol) dichloromethane-metha-nol mobile phase. The plates were visualized by transillumination withUV light (25). Authentic acyl- and 3-oxo-acyl-HSL (Sigma, CaymanChemical, or Santa Cruz Biotechnology) and PQS (Sigma) standards wereused for compound identification by comparing migration distances (Rf

values).Rhamnolipid production on SW agar. Rhamnolipid secretion was

assessed using SW agar indicator plates as described previously (61).Briefly, cultures were pregrown in SW minimal medium containing glyc-erol {20 g/liter glycerol, 0.7 g/liter KH2PO4, 0.9 g/liter Na2HPO4, 2 g/literNaNO3, 0.4 g/liter MgSO4 · H2O, 0.1 g/liter CaCl2 · 2H2O, and 2 ml/liter ofa trace element solution [2 g/liter FeSO4 · 7H2O, 1.5/liter g MnSO4 · H2O,and 0.6 g/liter (NH4)6Mo7O24 · 4H2O]}. Wells were cut into the surface ofSW agar plates (0.2 g cetyltrimethylammonium bromide [Sigma], 0.005 gmethylene blue [MP Biomedical], and 12 g agar to 1 liter of the abovemedium) and filled with 10 �l of stationary-phase culture. The plates wereincubated at 37°C for 48 h and then placed at 4°C overnight to intensifythe development of the blue halo complexation zone before photograph-ing.

Pseudomonas exoproduct assays. Rhamnolipids were extracted fromsupernatants produced by 24-h cultures grown in LB at 37°C with shaking(250 rpm). Aliquots (1 ml) of supernatant were acidified by adding 100 �lof 20% citric acid, and the aqueous phase was then twice extracted with anequal volume of chloroform-methanol (2:1 [vol/vol]). Pooled organicfractions were concentrated using a nitrogen bubbler, and the residue wasresuspended in 1 ml of methanol. Rhamnose content was measured usingthe anthrone colorimetric assay (93) and quantitated using L-rhamnose(Sigma) as the standard. The elastase proteolytic activity in culture super-natants, grown as described for the rhamnose assay, was determined usingthe elastin Congo red assay (20). After a 3-h digestion period at 37°C (100mM Tris, 1 mM CaCl2 [pH 7.5] with 20 mg of elastin Congo red substrate[Sigma] plus 100 �l of spent supernatant), suspensions were clarified bycentrifugation (16,000 g, 5 min, RT), and the absorbance at 495 nm wasmeasured to determine the amount of liberated dye. To measure pyocya-nin production, Pseudomonas cultures were grown in glycerol alanineminimal medium [1% (vol/vol) glycerol, 6 g/liter L-alanine, 2 g/literMgSO4, 0.1 g K2HPO4, 18 mg/liter FeSO4, 1 mg/liter biotin] at 37°C withshaking for 42 h (70). Pyocyanin-containing aliquots (5 ml) of clarifiedsupernatants were extracted with chloroform (3 ml), separated, and thenback extracted into the aqueous phase by the addition of 1 ml of 0.2 N HCl(23). The absorbance of the pink aqueous phase was measured at 520 nmand quantified using a pyocyanin (Cayman Chemical) standard curve.

Pseudomonas swarming assay. Swarming assays were performed onmodified M8 minimal medium agar supplemented with MgSO4 (1 mM),glucose (0.2%), Casamino Acids (0.5%), and solidified with 0.4% agar (7).The plates were dried for 2 h at room temperature, inoculated, and incu-bated overnight at 37°C before imaging.

FIG 1 Cosmid complementation strategy for identification of FabH-type ac-tivity in P. aeruginosa PAO1. The E. coli strain TMY19 has a chloramphenicolcassette (camR) inserted within fabH (b1091) and needs L-Ara supplementa-tion to induce expression of the PBAD-regulated S. enterica fabH ortholog (in-serted within the D-glucitol phosphotransferase operon) in order to sustainwild-type growth rates. The operon is transcriptionally silent under standardculture conditions in the absence of D-glucitol (52). A sheared library of wild-type P. aeruginosa PAO1 genomic DNA was ligated into the pWEB cosmidvector, packaged, and transduced by phage into E. coli TMY19. Transductantswere directly selected on unsupplemented LB agar to identify cosmids (pTMY-cos1 to pTMYcos17) that complement depletion of FabH.

Yuan et al.



.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

Siderophore secretion assay. The CAS-LB plates were prepared by amodified version of the published protocol (72). A 10 blue dye solutionconsisting of 1 mM chrome azurol S (CAS) (Acros), 2 mM cetyltrimeth-ylammonium bromide, and 500 �M FeCl3 · 6H2O was sterilized by auto-claving. A 10-ml aliquot was added to 100 ml of molten LB-Miller agar(1.5%) immediately before being poured. Plates were dried for 1 h at RTprior to inoculation.

Recombinant PA5174 protein expression and purification. Re-combinant PA5174 protein was obtained by transforming E. coliBL21(DE3)(pLysS) (Novagen) with the His tag expression vector pET-PA5174. Colonies were inoculated into 1 liter of LB medium and grown at37°C with shaking until mid-exponential growth (OD600 of 0.6). Proteinproduction was induced with 1 mM IPTG, and expression was allowed tocontinue for 3 h. The biomass was then harvested by centrifugation (5,000 g, 10 min, RT) and stored at �80°C. The pellet was subjected to 2 roundsof freeze-thaw cycles and then incubated in lysis solution (1 Bugbuster[Novagen], 5 kU/ml recombinant lysozyme, 25 U/ml benzonuclease)with gentle shaking at room temperature for 20 min. The solution wasclarified by centrifugation (15,000 g, 30 min, 4°C), and the supernatantwas gently shaken with nickel-nitrilotriacetic acid (Ni-NTA) His bindresin (1 h on ice). The slurry was loaded into an empty column, washedwith 50 ml of binding buffer (50 mM Tris, 500 mM NaCl, 10 mM imida-zole [pH 7.5]), 20 ml of wash buffer (binding buffer with 50 mM imida-zole total), and eluted (binding buffer plus 200 mM imidazole total) in1-ml fractions. Fractions containing protein of the expected size (�70kDa) by SDS-polyacrylamide gel electrophoresis (PAGE) were pooled andconcentrated (Amicon Ultra, 30-kDa molecular size cutoff; Millipore).The sample was further purified by size exclusion chromatography usinga Superdex200 16/60 column with an isocratic elution buffer (20 mMTris-HCl, 150 mM NaCl, 1 mM D/L-dithiothreitol, and 10% glycerol).Protein fractions with PA5174 were aliquoted and flash frozen at �80°C.

Recombinant PA5174 and FabH enzyme assays. A continuous, cou-pled spectrophotometric assay was used to measure the �-acetoacetyl syn-thase activity of P. aeruginosa PA5174 and E. coli FabH. Acetyl-CoA andmalonyl-ACP were added as substrates, and activity was monitored byNADPH consumption using the coupled E. coli �-ketoacyl-ACP reduc-tase (FabG) as has been described previously (56). A final 100-�l reactionmixture consisting of 0.1 M K2HPO4/KH2PO4 buffer (pH 7.0), 50 �Mholo-ACP, and 5 mM dithiothreitol (DTT) was incubated in a 96-wellclear Costar plate at 30°C for 5 min to obtain fully reduced holo-ACP. TheE. coli FabD malonyl-CoA:ACP transacylase (80 ng) was added along with100 �M malonyl-CoA and incubated for another 15 min to generatemalonyl-ACP in situ. Next, 100 �M NADPH, 2 �g of E. coli FabG, and 100�M acetyl-CoA was mixed in before adding PA5174 or E. coli FabH toinitiate the reaction. The UV absorption at 340 nm was monitored on aPHERAstar microplate reader.

Conformation-sensitive urea-PAGE was used to separate and analyzeacyl-ACP products (68). Radiolabeled malonyl-ACP was produced in situusing the same procedure as described above, except a total of 40 �Mmalonyl-CoA (10 �M [2-14C]malonyl-CoA) was added to the 20-�l re-action mixture. For each saturated straight-chain acyl-CoA acceptortested, 200 �M acyl-CoA (C2 to C16 in length; Sigma), along with 0.1 �g ofPA5174 (final concentration of 70 nM) was added, and the reaction mix-tures were incubated at room temperature for 1 h. The products wereseparated by 0.5 M urea–16% PAGE and visualized by film autoradiogra-phy. Where indicated, �-ketoacyl reaction products were further reducedto acyl-ACP using 1 �g each of E. coli FabG/FabI/FabA plus 200 �M eachof the NADH and NADPH cofactors.

Bioinformatics analysis. The Basic Local Alignment Search Tool(BLAST) (http://blast.ncbi.nlm.nih.gov/) search algorithm was used toidentify E. coli FabH (b1091) KASIII domain orthologs in the sequencedP. aeruginosa PAO1 genome (80). Likewise, the open reading frames withstand-alone KASI/II domains in other whole-genome sequenced Pseu-domonas species were identified by BLAST analysis using the KASI/II do-main proteins of the reference strain P. aeruginosa PAO1 (PA5174/

PA1373/PA1609/PA2965). All KASI/II domain-containing proteinsequences (48 total sequences from 4 P. aeruginosa strains, 7 other Pseu-domonas species, Azotobacter vinelandii DJ, and E. coli K-12 MG1655)were aligned using Clustal Omega v1.0.3 with mBed-like clustering guidetree and mBed-like iteration with default settings for iteration number(75). The phylogenetic tree was inferred using the neighbor-joining op-tion with partial deletion for gap treatment with the MEGA5 program(82). The bootstrap consensus tree was inferred from 1,000 replicates withdistances computed using the Poisson correction method.

RESULTSP. aeruginosa does not use a KASIII domain for fatty acid initi-ation. Multiple P. aeruginosa FAS genes (fabAB [38], fabDG-acp-fabF [46], and fabI and fabZ [40]) have been identified based onsequence similarity to orthologs in E. coli, suggesting a conservedFAS pathway. Unlike the situation in E. coli, however, the P.aeruginosa fabH gene is noticeably absent from any FAS-associ-ated cluster and must reside in a unique chromosomal location.Analysis of the genome of P. aeruginosa PAO1 for ORFs encodingproteins similar to the experimentally verified E. coli FabH(b1091) by BLAST search yielded four candidate KASIII-contain-ing orthologs (Fig. 2). All four proteins (PA3333 [FabH2],PA3286, PA0998 [PqsC], and PA0999 [PqsD]) possess 25 to 30%amino acid identity to E. coli FabH and contain at least two of thethree signature Cys-His-Asn active site residues that form the cat-alytic triad (19). While PA0998 and PA0999 belong to the Pseu-domonas quinolone signal (PQS) operon (22), a dual function orside activity capable of masking the �FabH phenotype could notbe ruled out, so all candidates were subjected to deletion analysis.In-frame deletions were constructed singly or sequentially in thesame genetic background to generate the KASIII-less P. aeruginosastrain TMT16. The fatty acid compositions of lipid extracts fromall of the strains were similar regardless of whether the KASIIIproteins were deleted singly or in combination (see Table S3 in thesupplemental material) and were similar to published values (11).The growth curves of the entire KASIII domain deletion strain setwere nearly superimposable (Fig. 2C). Collectively, the data indi-cate that P. aeruginosa does not primarily utilize a KASIII proteinfor fatty acid initiation but rather has a unique FAS initiationprotein.

PA5714 complements FabH depletion in E. coli. In order toidentify P. aeruginosa candidate proteins with FabH-type enzy-matic activity, we constructed a regulated FabH strain in E. coli(Fig. 1). We first inserted an araC-PBAD-fabHSe cassette within theD-glucitol (gutAEB) catabolism operon, utilizing the S. entericaFabH ortholog to minimize integration at the native fabH locus(47) and loss of D-glucitol metabolism as a convenient phenotypicmarker to screen recombinant clones for site-specific integration.Using L-arabinose (L-Ara)-supplemented media, the E. coli fabHgene was next disrupted to create the TMY19 strain, which is de-pendent on L-Ara to realize growth rates comparable to that of thewild type. A random cosmid library of P. aeruginosa PAO1genomic DNA was introduced into strain TMY19 and then platedon unsupplemented media in order to find cosmids capable ofcomplementing the fabH depleted growth defect. Cosmids from17 large colonies with enhanced growth were sequenced and backtransformed into TMY19 to confirm the rescue phenotype (re-sults summarized in Table S2 in the supplemental material). Ofthe 17 cosmids, 14 at least partially rescued growth in the absenceof L-Ara supplementation. All of the complementing cosmidsharbored P. aeruginosa genomic inserts that started immediately




.org/D

ownloaded from

http://blast.ncbi.nlm.nih.gov/

http://jb.asm.org

http://jb.asm.org/

upstream of the unannotated ORF PA5174 and extended 20 to 40kb downstream (Fig. 3A). PA5174 is predicted to encode a 634-amino-acid protein belonging to the KASI/II domain family. Thestrong bias for selection of inserts with PA5174 juxtaposed to thecosmid backbone immediately suggested that expression ofPA5174 by plasmid-based promoters was necessary for comple-mentation. To confirm that PA5174 alone was responsible forcomplementation, we cloned the PA5174 gene without its nativepromoter region into the pWEB cosmid so as to be under thecontrol of the T7 phage promoter. The growth characteristics ofthe resulting strain TMY19(pWEB-PA5174) were similar to thoseseen after L-Ara supplementation of the parent strain (Fig. 3B);PA5174 is thus by itself sufficient to complement FabH and re-store initiation of FAS.

Since FabH expression is not fully repressed in uninduced E.coli TMY19, it was necessary to verify that PA5174 could com-pletely replace fabH in E. coli and alone support the cellular de-mand for de novo FAS. The PA5174 gene was cloned into an IPTG-inducible plasmid (pET-PA5174) and transformed into E. coliBW25113. Deletion of the fabH gene was attempted using the Redrecombinase system (18) with a fabH::camR integration cassette.No fabH disrupted mutants were obtained in wild-type BW25113,consistent with the essential function of FabH in E. coli (47), but

viable colonies were obtained in the presence of plasmid-encodedPA5174. Growth analysis confirmed complete fabH-PA5174 crosscomplementation (Fig. 3C), nominating PA5174 as the P. aerugi-nosa FabH functional ortholog.

PA5174 deletion in P. aeruginosa attenuates growth by de-creasing de novo fatty acid synthesis. The genomic context indi-cates that PA5174 is likely a monocistronic gene (Fig. 3A), withminimal anticipated polarity concerns upon deletion. Since noloss-of-function transposon disruption mutants have been re-ported (no insertions in P. aeruginosa PAO1 [88] and only anextreme C-terminal transposon mutant in P. aeruginosa PA14[48]), PA5174 likely has an essential cellular role. To test this, allelicexchange vectors that either replaced PA5174 with a gentamicin re-sistance-E. coli fabH (pTMT123) cassette or with a gentamicin resis-tance-E. coli fabH nonfunctional fragment (pTMT124) were con-structed. Cointegrants were passively resolved, and the sacB�

integration vector was counterselected with sucrose. Allelic exchangewith fabH gave rise to hyperpigmented blue-green gentamicin-resis-tant mutants (TMT38 [PA5174::fabH]), while the nonfunctionalfabH fragment (TMT39 [�PA5174]) resulted in small, translucentcolonies that were gentamicin resistant and took 2 days to appear.Analysis of the growth characteristics in liquid LB confirmed a mark-edly attenuated growth phenotype for the �PA5174 strain (doubling

FIG 2 Fatty acid �-ketoacyl acyl carrier protein (ACP) synthase III (KASIII) domain-containing proteins in P. aeruginosa PAO1. (A) The E. coli K-12 MG1655FabH sequence (b1091) was used to query the P. aeruginosa PAO1 genome for similar protein sequences using the BLAST search tool. Four candidate P.aeruginosa FabH-encoding open reading frames at 3 distinct genomic loci all containing the KASIII conserved domain were identified: PA3333 (white) (FabH2,E-value � 8e�66), PA3286 (white) (E-value � 3e�25), PA0998 (white) (PqsC, E-value � 3e�17), and PA0999 (gray) (PqsD, E-value � 2e�68). (B) Alignmentof putative active site regions with the Cys112-His244-Asn274 catalytic triad of E. coli FabH (FabH Ec) (indicated by asterisks). Identical amino acid residues areshaded with a black background, and partially conserved substitutions are shaded in gray. (C) Growth curves of P. aeruginosa strains with all KASIII candidategenes (wild type [Œ]) or with KASIII candidate genes deleted singly (PA0999 [�], PA3286 [�], PA3333 [Œ], PA0998 [o]) or in combination (TMT16 [�]) inLB broth at 37°C.

Yuan et al.



.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

time increased from 19 to 62 min in the absence of PA5174). Com-plementation with either E. coli fabH (TMT38) or PA5174 plasmid(TMT41) retained growth rates comparable to that of the wild type(Fig. 4A).

Since a defect in FAS upon PA5174 deletion was suspected, theFAS de novo flux was next measured by macromolecular labelingusing radiolabeled precursors to monitor FAS ([3H]acetate sub-strate) and RNA ([3H]uridine substrate) pathway activity. BothRNA and FAS activity were lower in the �PA5174 mutant (Fig.4B), in part reflecting slower overall metabolic activity. However,the larger decrease in the rate of label incorporation for FAS incomparison to RNA indicates particular attenuation specific toFAS. Both FAS and RNA pathway activity returned to near wild-type levels in the strain complemented with PA5174 (TMT41).Collectively, the data support a central role for the PA5174 geneproduct in FAS initiation in P. aeruginosa, analogous to the func-tion of FabH in E. coli.

The rhl QS system is muted by deletion of PA5714. The rhl QSsignal molecules N-butanoyl-L-homoserine lactone (C4-HSL)and N-hexanoyl-L-homoserine lactone (C6-HSL) are synthesizedfrom S-adenosylmethionine and �-butanoyl-ACP/�-hexanoyl-ACP, respectively (65). Since both molecules are formed from�-acetoacetyl-ACP, the putative product of PA5174, muted sig-naling in the cognate C4-HSL/C6-HSL-responsive rhl QS regula-tory circuit would be expected in the �PA5174 strain if PA5174encodes the major FAS initiation enzyme. Acyl-HSL was extracted

from culture supernatants and separated by TLC. The acyl-HSLspots were detected using an agar overlay of a C. violaceum re-porter strain which produces a purple chromophore in responseto acyl-HSL (51). The amount of acyl-HSL in the �PA5174 strainwas nearly undetectable in comparison to that in the wild type(Fig. 5A). Production of acyl-HSL actually increased beyond wild-type levels in the PA5174 complemented strain (TMT41), con-firming a tight association between PA5174 expression and acyl-HSL synthesis.

We next looked at levels of rhamnolipids among the PA5174strain panel. Rhamnolipids are secreted surfactant glycolipidsformed by acylation of L-rhamnose using the FAS pathway inter-mediate �-hydroxydecanoyl-ACP (93). Since rhamnolipids de-pend on the rhl QS system for expression (66) and are themselvesassembled from FAS pathway intermediates, the levels of rham-nolipids should reflect defects in FAS initiation. To detect rham-nolipids, cultures were spotted into wells cut into SW agar indicatorplates (61). Blue halos in the agar indicate zones of ion-pairing com-plexation between the methylene blue dye, cetyltrimethylammoniumbromide, and secreted rhamnolipids. While the wild-type strain andthe PA5174-complemented strain produced prominent halos, nohalo was observed with the �PA5174 mutant strain (Fig. 5B).Rhamnolipid quantification by colorimetric detection of rham-nose showed a 5-fold decrease in the �PA5174 mutant while therewas a 3-fold increase in the complemented strain relative to thewild type (Fig. 5C). Swarming motility on semisolid agar was also

FIG 3 Complementation of FabH depletion in E. coli K-12 by a P. aeruginosa cosmid library. (A) The plasmid-genomic DNA insert junction sites of cosmids thatrescued growth of the FabH-depleted E. coli strain TMY19. All 14 sites were located in the immediate 5= intergenic region of the unknown open reading framePA5174 and contained 24 to 42 kb of downstream DNA. All inserts except for pTMYcos13 (asterisk) placed PA5174 proximal to the T7 promoter on the pWEBvector backbone. The number of base pairs intervening between the respective cosmid-junction site and the start codon of PA5174 is indicated in parentheses.Cosmids selected for further growth analysis are underlined. (B) Growth curves of E. coli strain TMY19 grown in various conditions at 37°C. E. coli TMY19 wasgrown in LB only (Œ) or in LB plus 0.2% L-arabinose inducer (o). E. coli TMY19 was grown in LB plus 0.2% L-arabinose inducer with cosmid pTMY2 (�),cosmid pTMY10 (�), cosmid pTMY17 (Œ) or plasmid pWEB-PA5174 (�). (C) Growth curves of the IPTG-inducible PA5174 E. coli strain TMY32 with fabHdeleted in LB (in the absence of IPTG [�] or in the presence of 1 mM IPTG [o]) were measured at 37°C in LB and compared to that of the parent wild-type strainBW25113 (�).




.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

evaluated in order to estimate rhamnolipid secretion, as the sur-face wetting properties of the amphiphilic glycolipid are thoughtto be critical for breaking the surface tension that otherwise hin-ders flagellar propulsion (45). Strains were spotted in the middleof swarming plates, and the degree of spreading was comparedafter overnight incubation (Fig. 5D). Consistent with the rham-nolipid assays, the �PA5174 strain was nonmotile, while both thewild-type and complemented strains formed short, flowering ten-drils characteristic of some P. aeruginosa PAO1 isolates (83).

Both las and PQS QS-dependent exoproducts are downregu-lated by deletion of PA5714. Together with the rhl system, the lasand Pseudomonas quinolone signal molecules form the QS net-work of P. aeruginosa (33, 41, 87). The las and PQS signal mole-cules both utilize �-ketoacyl medium-chain fatty acid metabolitesto acylate HSL to make N-(3-oxododecanoyl)-L-HSL (3-oxo-C12-HSL) (40) and to synthesize 2-heptyl-3-hydroxy-4-quinolone(PQS) (62), respectively. The 3-oxo-acyl-HSL fraction was ex-tracted from culture supernatants, separated by TLC, and visual-ized using the inducible �-galactosidase reporter strain Agrobac-terium tumefaciens NL4(pZLR4) (10). While spot intensities werenearly equivalent between the wild type and the �PA5174 strain,

3-oxo-acyl-HSL levels were increased in the plasmid-comple-mented strain (Fig. 6A). The �PA5174 mutant apparently cansynthesize 3-oxo-acyl-HSL to wild-type levels in LB liquid cultureafter 24 h of incubation despite the FAS defect. The secreted viru-lence factor elastase (lasB) is in large part induced by the las QSsystem (30). Unlike the levels of the 3-oxo-acyl-HSL inducer, elas-tase activity was almost undetectable in the �PA5174 strain, whileplasmid complementation with PA5174 restored elastase to 2-foldabove the wild type (Fig. 6B).

The uncoupling of 3-oxo-HSL inducer concentration and elas-tase activity in the �PA5174 strain suggested that another partic-ipant involved in lasB induction was absent. PQS-mediated sig-naling, which itself is subject to hierarchal control by the las QSsystem (87), also contributes to lasB induction (60). We thus an-alyzed the levels of PQS in supernatant extracts (Fig. 6C). The PQSlevel was reduced to a trace in the supernatant extract whenPA5174 was deleted, similar in quantity to the level in the PQSnegative-control �pqsC strain. In contrast, PQS production was

FIG 4 Growth characteristics and complementation of the P. aeruginosa�PA5174 strain. (A) Growth of the P. aeruginosa PAO1 parent strain (�),�PA5174 strain (�), PA5174 back-complemented with plasmid (TMT41[o]), and the E. coli FabH-complemented strain (TMT38 [�]) in LB at 37°C.Basal expression of PA5174 on the plasmid is sufficient for full complementa-tion in the absence of IPTG induction. (B) The rate of radiolabeled precursorincorporation by FAS ([3H]acetate [black bars]) and RNA ([3H]uridine [whitebars]) biosynthetic pathways was measured after a 30-min incubation periodin supplemented M9 minimal medium. The percent incorporated label wasnormalized to the value of the wild-type (Wt) P. aeruginosa PAO1 control.Error bars represent the standard deviations from the means calculated fromthree independent experiments.

FIG 5 Characterization of rhl quorum sensing and dependent exoproductsproduced in P. aeruginosa �PA5174 strain. (A) The acylated HSL fraction wasextracted from the supernatants of stationary-phase cultures grown in LBbroth at 37°C for 24 h and separated by reverse-phase TLC. The levels ofacylated HSL were estimated using a soft agar overlay of the C. violaceumreporter strain TMT45 and compared to standards (C4-HSL [20 nmol] andC6-HSL [0.2 nmol]). The cvi QS system of C. violaceum detects C6-HSL withthe highest sensitivity (51), limiting quantitative comparisons to within thesame HSL acyl chain length. (B) Cultures of P. aeruginosa were spotted intowells cut into the agar of SW plates. Blue halos indicate zones of ion-pairingcomplexation between the methylene blue dye, cetyltrimethylammoniumbromide, and secreted rhamnolipids. (C) Rhamnose levels in stationary-phaseculture supernatants were measured using the colorimetric anthrone assay.(D) Swarming motility was evaluated on M8 agar plates supplemented withCasamino Acids and glucose after overnight incubation at 37°C.

Yuan et al.



.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

markedly enhanced in the complemented strain. PQS signaling isabsolutely necessary for production of the phenazine secondarymetabolite pyocyanin (29). Assay of pyocyanin levels in cultureextracts mirrored the respective PQS levels in each of the strains,ranging from a trace in �PA5174 to a 5-fold increase in the plas-mid-complemented strain (Fig. 6D). The pyocyanin levels mea-sured from cultures grown in liquid broth are consistent with thecharacteristic blue-green hyper- and hypopigmentation initiallyobserved on agar.

PA5174 deletion in P. aeruginosa reduces production of sid-erophores. The dominant siderophore produced by fluorescentpseudomonads, including P. aeruginosa, is the iron chelator py-overdine (85). Since pyoverdine is coregulated by QS systems (79)and is assembled on a myristate fatty acid residue (21, 34), weinvestigated whether siderophore secretion was attenuated in the

�PA5174 strain by streaking on LB-chrome azurol S (CAS) indi-cator plates (72). Zones that change from blue to yellow-orangeindicate where the siderophores have sequestered Fe3� away fromthe blue CAS-Fe3� complex. While the wild-type and plasmid-complemented P. aeruginosa strains generated prominent zonesof clearing, the �PA5174 strain did not produce a zone, indicatinga lack of relative siderophore production (Fig. 7). In addition topyoverdine, a second siderophore (pyochelin) and PQS can alsochelate Fe3� (5). The Fe3� sequestration zone produced by thePQS� control strain TMT02 (�pqsD) was similar to that of thewild type (Fig. 7). Thus, the absence of dedicated siderophores inthe �PA5174 strain is responsible for the difference observed inthe Fe3� sequestration zone.

Characterization of recombinant PA5174 �-ketoacyl ACPsynthase activity. Recombinant PA5174 was expressed in E. coliand purified by His tag affinity chromatography in order tocharacterize in vitro enzymatic activity. Using a FabG coupledcontinuous spectrophotometric assay to monitor NADPHconsumption via reduction of �-ketoacyl-ACP product (56),the specific activities of PA5174 and E. coli FabH with malonyl-ACP and acetyl-CoA as substrates were 13.4 and 7.4 pmol/min/ng under the same assay conditions (Fig. 8A). The FASinitiation activity of PA5174 is on par with E. coli FabH andwith previously reported data (13, 63).

We next examined the reaction product of PA5174 using con-formation-sensitive urea-PAGE with [2-14C]malonyl-ACP (68).The putative �-acetoacetyl-ACP product of PA5714 migratesfaster than the starting substrate [2-14C]malonyl-ACP and to thesame extent as the E. coli FabH standard reaction (Fig. 8B). Tofurther identify the product, the FAS reduction cycle enzymesfrom E. coli (FabG/FabA/FabI) were included in the reaction mix-ture. The radioactive acyl-ACP band migrated faster on the gel,indicating that the product of PA5174 is a substrate for FabG/FabA/FabI. The electrophoretic mobility of the reduced product isidentical to that of the E. coli FabH butyryl-ACP standard. Thelack of additional bands indicates that PA5174 cannot efficientlycarry out the next round of condensation for chain elongationusing acyl-ACP as a substrate, as do canonical KASI/II domainFAS condensing enzymes.

The substrate specificity of PA5174 enzyme was probed usingvarious straight-chain saturated acyl-CoA (C2 to C16) acceptors.Reaction products were analyzed by conformation-sensitive urea-PAGE as described above using [2-14C]malonyl-ACP. PA5174

FIG 6 Characterization of las and PQS-dependent quorum sensing in the P.aeruginosa �PA5174 strain. (A) The 3-oxo-HSL produced by the las quorum-sensing system was extracted from the supernatants of stationary-phase cul-tures grown in LB broth at 37°C for 24 h and separated by reverse-phase TLC.The levels of 3-oxo-HSL were estimated using a soft agar overlay of the A.tumefaciens NTL4(pZLR4) �-galactosidase reporter strain and compared tostandards ([3-oxo-C6-HSL [1 pmol], 3-oxo-C8-HSL [0.2 pmol], 3-oxo-C10-HSL [20 pmol], and 3-oxo-C12-HSL [100 pmol]). The tra QS system of A.tumefaciens detects 3-oxo-C8-HSL with the highest sensitivity, limiting quan-titative comparisons to within the same HSL acyl chain length (74). (B) Theelastase activity (LasB) in culture supernatants was measured using a Congored-elastin dye release assay. (C) The amount of PQS was compared by directUV illumination of culture supernatant extracts separated by normal-phaseTLC on silica 60 F254 plates. Synthetic PQS standard (20 nmol) and P. aerugi-nosa TMT15 (�pqsC) were used as the positive and negative controls, respec-tively. (D) Pyocyanin was extracted from supernatants obtained from culturesgrown in glycerol alanine minimal media for 42 h at 37°C, acidified, and quan-tified by absorbance at 520 nm. Values (in �g/ml of supernatant) were deter-mined using a standard curve generated with purified pyocyanin.

FIG 7 Siderophore secretion in P. aeruginosa �PA5174. The wild-type P.aeruginosa PAO1 (Wt), TMT39 (�PA5174), �PA5174(pZEN-PA5174)(TMT41), and �pqsD PQS� (TMT02) strains were streaked onto LB agarplates containing CAS and incubated at 37°C for 24 h. Zones of yellow-orangeclearing indicate sequestration of Fe3� away from CAS- Fe3� complexes bysecreted siderophores.




.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

utilized short-chain acyl-CoA as substrates, including acetyl-CoA,butyryl-CoA, and hexanoyl-CoA, but did not use any of the lon-ger-chain acyl-CoA substrates (Fig. 8C). A second radioactiveband migrating slightly ahead of the malonyl-ACP substrate wasapparent for the long-chain substrates. However, this band islikely [2-14C]acetyl-ACP, which is produced through decarboxyl-ation of [2-14C]malonyl-ACP. Decarboxylation upon extendedreaction times has previously been noted for all three types of KASenzymes (1, 13) and becomes apparent in the absence of suitableacyl-CoA acceptors. All of the in vitro characteristics of PA5174are entirely consistent with its assignment as the FAS initiation

enzyme in P. aeruginosa, catalyzing the formation of �-aceto-acetyl-ACP from acetyl-CoA and malonyl-ACP.

DISCUSSION

The enzymatic pathway of FAS type II biosynthesis in bacteria hasbeen well defined in large part using E. coli and Salmonella asmodel organisms (17, 86). While exceptions in the overarchingcatalytic strategy to assemble fatty acids remain rare, novel FASenzymes with little sequence similarity to their functional coun-terparts in the canonical FAS type II pathway are becoming in-creasingly common, as organisms beyond members of the familyEnterobacteriaceae are subjected to detailed study (59). For in-stance, P. aeruginosa encodes a second enoyl-ACP reductase(FabV) which bears no sequence similarity to FabI (94). In thisreport, we provide evidence that FAS initiation in P. aeruginosa isalso unique in using the KASI/II domain-containing proteinPA5174. The viability of a P. aeruginosa mutant that lacks allKASIII domain-containing proteins (Fig. 2C), the attenuation ofgrowth due to the FAS defect in the �P�5174 mutant (Fig. 4A),intergeneric cross complementation between P. aeruginosaPA5174 and E. coli fabH (Fig. 3 and 4), the pleiotropic phenotypestemming from depletion of fatty acid-dependent metabolites(Fig. 5 to 7), and the in vitro enzymatic characterization (Fig. 8) allsupport the conclusion that PA5174 encodes the predominant�-acetoacetyl-ACP synthase of P. aeruginosa. We propose distin-guishing �-ketoacyl-ACP synthases with KASI/II domains and apreference for short-chain acyl-CoA acceptor substrates as thenew FabY class of condensing enzymes.

The viability of the �fabY mutant indicates that an alternatesecondary mechanism for FAS initiation must exist. In a fabHmutant of Lactococcus lactis, growth can be restored by overex-pression of FabF (55). A suppression mechanism whereby acetyl-ACP is generated either through FabF-mediated decarboxylationof malonyl-ACP or acetyl-CoA:ACP transacylation was proposed.Acetyl-ACP, in turn, could then be used as the acceptor substrate(albeit a poor one) in a condensation reaction with malonyl-ACPto generate essential �-acetoacetyl-ACP pools. There are threeKASI/II candidates in P. aeruginosa PAO1 (FabB, FabF, and FabF2[Fig. 9]), as well as four KASIII enzymes (Fig. 2A), to supply eitherof the putative decarboxylation/transacylation activities requiredto bypass FabY. Alternatively, any of the KASIII proteins couldhave an intrinsic secondary FabH-type activity that becomes bio-logically relevant only in the absence of FabY, as deletion of allKASIII proteins in a wild-type background was aphenotypic withrespect to growth (Fig. 2C).

Other KASI/II proteins could theoretically utilize an acetyl-CoA acceptor in place of acetyl-ACP to make �-acetoacetyl-ACP,particularly since FabY now demonstrates that substrate specific-ity of a KASI/II domain is not necessarily confined to acyl-ACPacceptor substrates. Phylogenetic analysis of KASI/II domain en-zymes in pseudomonads, however, would seem to make this theleast likely of all the possible FabY bypass mechanisms (Fig. 9).There are three KASI/II domain protein clusters conserved amongmultiple Pseudomonas species (FabB, FabF, and FabY). The otherKASI/II proteins are more narrowly distributed, suggesting strain-specific auxiliary cellular roles outside primary FAS metabolism.The FabB and FabF clusters (as defined by their respective E. coliorthologs; green labels) tightly cluster and have representativesfrom all 12 Pseudomonas strains analyzed, consistent with theirrole in FAS elongation. The FabY cluster is highly divergent from

FIG 8 Enzymatic activity and acyl-CoA acceptor substrate specificity of re-combinant PA5174. (A) The continuous FabG coupled assay was initiated with15 nM FabH (�) or PA5174 (Œ), and the reaction mixture was incubated at30°C. Reaction progress was monitored by UV absorbance at 340 nm to assayNADPH consumption. Background absorbance (no enzyme addition) wassubtracted from each curve before plotting. (B) Discontinuous conformation-sensitive urea-PAGE was employed to separate acyl-ACP products with incor-porated [2-3H]malonyl-ACP substrate. Gels were sequentially imaged usingCoomassie blue staining and autoradiography. All lanes contained [2-3H]ma-lonyl-ACP and acetyl-CoA substrates. Lanes: 1, substrates only; 2, E. coli FabH;3, PA5174; 4, E. coli FabH plus FabG/FabA/FabI; 5, PA5174 plus FabG/FabA/FabI. The positions of malonyl-ACP (malonyl), �-acetoacetyl-ACP or acetyl-ACP (�-acac/ac), and �-butyryl-ACP (�-butyryl) are labeled. (C) The acyl-CoA substrate specificity of PA5174 was assayed using [2-3H]malonyl-ACPsubstrate alone (control [Con]) or with various acyl-CoA acceptors (C2 toC16). Reaction products were separated by urea-PAGE, stained, and sequen-tially imaged. Under these assay conditions, the [2-3H]�-acetoacetyl-ACP re-action product using C2-CoA as the acceptor is not adequately separated fromthe [2-3H]acetyl-ACP side product produced by decarboxylation of [2-3H]malonyl-ACP unless further reduced by FabG/FabA/FabI as in lanes 4 and5 of panel B.

Yuan et al.



.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

its FabF/FabB KASI/II counterparts, having the least similar se-quence and being �200 amino acids longer. The sequence diver-gence likely reflects acquisition of short-chain acyl-CoA substrateselectivity, a catalytic activity not shared with distantly relatedKASI/II proteins. Interestingly, FabY is also more narrowly dis-tributed at the species level, being confined to all sequenced P.aeruginosa strains (orange labels), Pseudomonas fulva (red labels),Pseudomonas mendocina (blue labels), and Azotobacter vinelandii(purple labels). The A. vinelandii genome is included with Pseu-domonas, as its taxonomic distinction within a separate genus isdebatable based on 16S rRNA analysis (57, 73). Indeed, the con-servation of FabY supports the notion that P. aeruginosa is moreclosely related to A. vinelandii than to other Pseudomonas species.The A. vinelandii genome is also noteworthy in that there are noKASIII orthologs in the sequenced genome (unpublished obser-vation), and as with the quadruple knockout P. aeruginosa strainTMT16, FAS initiation must be entirely KASIII independent.Conversely, there are no FabY orthologs in the genomes of Pseu-domonas stutzeri, Pseudomonas putida, Pseudomonas fluorescens,and Pseudomonas syringae (all black labels), indicating that FabYwas either relatively recently acquired by, or only retained in, theP. aeruginosa-specific branch after speciation of the pseudomonadlineage. How the other Pseudomonas species initiate FAS is un-

clear, but it is likely that a KASIII domain protein may be involved.In Pseudomonas syringae pv. tabaci 6605, a FabH ortholog (orf3)located within the flagellin glycosylation island has been identified(81). Deletion of the monocistronic orf3 had no effect on proteinglycosylation, but a role in FAS initiation was suggested based ona defect in HSL QS that is reminiscent of the �FabY phenotypeobserved here in P. aeruginosa. Once more, a KASIII orthologembedded in a flagellin glycosylation island is absent in P. aerugi-nosa PAO1 (b-type flagellin) (3) but is present in P. putida, P.fluorescens, and P. syringae (unpublished observation). The pres-ence of a putative FabH ortholog within an element subject toextensive rearrangement and horizontal gene exchange suggests apossible genetic means for establishing species-dependent FASinitiation pathways among pseudomonads.

Inhibition of FAS initiation in P. aeruginosa is a compellingstrategy for developing new antimicrobial agents. A multitude ofcellular products, including essential as well as virulence-associ-ated factors, all depend on FAS initiation to directly provide bio-synthetic precursors or are themselves subject to transcriptionalregulation by fatty acid-containing signal molecules (Fig. 10). De-letion of FabY attenuated not only growth but also QS, sidero-phore secretion, and a host of virulence factors, many of which areby themselves attractive antibiotic development targets. Inhibi-

FIG 9 Phylogenetic tree and distribution of KASI/II domain-containing proteins among pseudomonads. All open reading frames within each genome havingsignificant similarity to P. aeruginosa PAO1 KASI/II proteins (E-value 1e�10) and that are not a subdomain of a polyketide synthase multidomain protein wereincluded in the analysis. Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The tree is drawn to scale, withbranch lengths in the units of number of amino acid substitutions per site. Analyses were conducted by using MEGA5 (82). The three main KASI/II groups (FabB,FabF, and FabY) are indicated, and all members within each cluster share at least partial genomic synteny. Locus tag abbreviations: PA, P. aeruginosa PAO1(orange); PSPA7, P. aeruginosa PA7 (orange); PA14, P. aeruginosa UCBPP-PA14 (orange); PLES, P. aeruginosa LESB58 (orange); Pmen, P. mendocina ymp(blue); MDS, P. mendocina NK-01 (blue); Psefu, P. fulva 12-X (red), PST, P. stutzeri A1501 (black), PputGB1, P. putida GB-1 (black); Psyr, P. syringae pv. syringaeB728a (black); PFLU, P. fluorescens SBW25 (black); Avin, Azotobacter vinelandii DJ (purple); b, E. coli (green).




.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

tion of iron acquisition by blocking siderophore production (53),and QS inhibition in particular (50), have garnered substantialinterest as strategies for treating P. aeruginosa infections. The threedominant QS signal molecules from the las, rhl, and PQS systems,all of which contain fatty acid moieties, together form a complex,cell density-dependent regulatory circuit controlling the expres-sion of a suite of virulence factors (87, 89). QS-regulated exoprod-ucts include the toxin pyocyanin which induces oxidative stress ininfected tissue, elastase which degrades elastin and other host ma-trix proteins, and rhamnolipids which have intrinsic cytolytic ac-tivity. QS signaling is also critical in orchestrating the transitionfrom planktonic to multicellular biofilm growth associated withinchronic infections (35). The levels of rhl acyl-HSL and PQS, as wellas all associated exoproducts, were markedly depressed in the ab-sence of FabY. The progression of QS in the �FabY mutant is likelydelayed by the lack of fatty acid precursors, as has also been ob-served with decreased FAS enoyl-ACP reductase activity (39). Thenear wild-type amounts of the las 3-oxo-HSL, which sits atop theQS hierarchal signal cascade, suggests that pools become availableas FAS demands for phospholipid/lipopolysaccharide biosynthe-sis wane during extended stationary-phase incubation. Con-versely, all QS exoproducts were upregulated well beyond wild-type levels in the multicopy plasmid FabY-complemented strain.

This was especially apparent for the PQS signal. Hypervirulence isassociated with early and robust quorum signaling in transmissi-ble epidemic clinical isolates (26). FabY and QS signal expressionappears to be correlated in P. aeruginosa PAO1 and warrants fu-ture study into the regulatory mechanisms connecting fatty acidflux and QS-mediated virulence.

The identification of FabY as the predominant FAS initiationenzyme in P. aeruginosa adds to the growing diversity in bacterialFAS pathways (59). Being narrowly confined among pathogens toP. aeruginosa, extending the spectrum of small-molecule inhibi-tors optimized against KASIII-type FabH targets to P. aeruginosamay prove to be challenging. However, simultaneous inhibition ofthe initiating and elongating KAS I/II enzyme family (FabB, FabF,and FabY) is potentially more feasible, given that a single domaincarries out all KAS steps in the FAS pathway of P. aeruginosa. TheKASI/II active site Cys-His-His triad is conserved in all 3 isozymes,including FabY, and the additional �200 amino acids of FabY arelocated outside the central KASI/II domain at the N and C termini(see Fig. S1 in the supplemental material). Insights into the struc-ture and biochemistry of FabY, as well as the mechanism of FabYFAS bypass, will be necessary to fully evaluate the ultimate poten-tial of FabY as a target for P. aeruginosa-specific FAS inhibitordevelopment. In the accompanying manuscript (90), we report

FIG 10 Cellular products dependent on FabY (i.e., FabH-type) activity in P. aeruginosa PAO1. The �-acetoacetyl-ACP product of FabY is used as a primer forelongation by the FAS elongation machinery, whose products are directly incorporated into a multitude of essential cell components, cofactors, and quorumsensing (QS) signaling molecules. Some exoproducts, including rhamnolipids, are dependent on fatty acid precursors as well as being subject to transcriptionalQS regulation. Molecular fragments originating from FabY enzymatic activity are shown in bold type (blue, red, or green). The depicted biotin pathway in P.aeruginosa PAO1 is based on the proposed bioC and/or bioH pathway of E. coli (16). E2, (�)-ketoacid dehydrogenase subunit E2-containing lipoylation domain.

Yuan et al.



.org/D

ownloaded from

http://dx.doi.org/10.1128/JB.00860-12

http://jb.asm.org

http://jb.asm.org/

that the PA3286 KASIII domain-containing condensing enzymeinitiates FAS in the absence of FabY through shunting of exoge-nous fatty acids from the �-oxidation degradation pathway intode novo FAS (90).

ACKNOWLEDGMENTS

We thank Herbert P. Schweizer (Colorado State University) for thepEX18ApGW plasmid and Carl Balibar (Novartis) for the pKD4-PBAD

plasmid.

REFERENCES1. Alberts AW, Bell RM, Vagelos PR. 1972. Acyl carrier protein. XV. Studies

of �-ketoacyl-acyl carrier protein synthetase. J. Biol. Chem. 247:3190 –3198.

2. Alberts AW, Majerus PW, Talamo B, Vagelos PR. 1964. Acyl-carrierprotein. II. Intermediary reactions of fatty acid synthesis. Biochemistry3:1563–1571.

3. Arora SK, Bangera M, Lory S, Ramphal R. 2001. A genomic island inPseudomonas aeruginosa carries the determinants of flagellin glycosyla-tion. Proc. Natl. Acad. Sci. U. S. A. 98:9342–9347.

4. Bergler H, et al. 1994. Protein EnvM is the NADH-dependent enoyl-ACPreductase (FabI) of Escherichia coli. J. Biol. Chem. 269:5493–5496.

5. Bredenbruch F, Geffers R, Nimtz M, Buer J, Haussler S. 2006. ThePseudomonas aeruginosa quinolone signal (PQS) has an iron-chelatingactivity. Environ. Microbiol. 8:1318 –1329.

6. Brinster S, et al. 2009. Type II fatty acid synthesis is not a suitable anti-biotic target for Gram-positive pathogens. Nature 458:83– 86.

7. Caiazza NC, Shanks RM, O’Toole GA. 2005. Rhamnolipids modulateswarming motility patterns of Pseudomonas aeruginosa. J. Bacteriol. 187:7351–7361.

8. Campbell JW, Cronan JE, Jr. 2001. Bacterial fatty acid biosynthesis:targets for antibacterial drug discovery. Annu. Rev. Microbiol. 55:305–332.

9. Castillo YP, Perez MA. 2008. Bacterial beta-ketoacyl-acyl carrier proteinsynthase III (FabH): an attractive target for the design of new broad-spectrum antimicrobial agents. Mini. Rev. Med. Chem. 8:36 – 45.

10. Cha C, Gao P, Chen YC, Shaw PD, Farrand SK. 1998. Production ofacyl-homoserine lactone quorum-sensing signals by gram-negative plant-associated bacteria. Mol. Plant Microbe Interact. 11:1119 –1129.

11. Chao J, Wolfaardt GM, Arts MT. 2010. Characterization of Pseudomonasaeruginosa fatty acid profiles in biofilms and batch planktonic cultures.Can. J. Microbiol. 56:1028 –1039.

12. Chen Y, et al. 2011. Structural classification and properties of ketoacylsynthases. Protein Sci. 20:1659 –1667.

13. Choi KH, Heath RJ, Rock CO. 2000. Beta-ketoacyl-acyl carrier proteinsynthase III (FabH) is a determining factor in branched-chain fatty acidbiosynthesis. J. Bacteriol. 182:365–370.

14. Choi KH, Kumar A, Schweizer HP. 2006. A 10-min method for prepa-ration of highly electrocompetent Pseudomonas aeruginosa cells: applica-tion for DNA fragment transfer between chromosomes and plasmidtransformation. J. Microbiol. Methods 64:391–397.

15. Choi KH, Schweizer HP. 2005. An improved method for rapid genera-tion of unmarked Pseudomonas aeruginosa deletion mutants. BMC Micro-biol. 5:30. doi:10.1186/1471-2180-5-30.

16. Cronan JE, Lin S. 2011. Synthesis of the alpha,omega-dicarboxylic acidprecursor of biotin by the canonical fatty acid biosynthetic pathway. Curr.Opin. Chem. Biol. 15:407– 413.

17. Cronan JE, Rock CO. 1996. Biosynthesis of membrane lipids, p 612– 636.In Neidhardt FC, et al. (ed), Escherichia coli and Salmonella: cellular andmolecular biology, 2nd ed, vol 1. American Society for Microbiology,Washington, DC.

18. Datsenko KA, Wanner BL. 2000. One-step inactivation of chromosomalgenes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci.U. S. A. 97:6640 – 6645.

19. Davies C, Heath RJ, White SW, Rock CO. 2000. The 1.8 A crystalstructure and active-site architecture of beta-ketoacyl-acyl carrier proteinsynthase III (FabH) from Escherichia coli. Structure 8:185–195.

20. Diggle SP, et al. 2003. The Pseudomonas aeruginosa quinolone signalmolecule overcomes the cell density-dependency of the quorum sensinghierarchy, regulates rhl-dependent genes at the onset of stationary phaseand can be produced in the absence of LasR. Mol. Microbiol. 50:29 – 43.

21. Drake EJ, Gulick AM. 2011. Structural characterization and high-throughput screening of inhibitors of PvdQ, an NTN hydrolase involvedin pyoverdine synthesis. ACS Chem. Biol. 6:1277–1286.

22. Dubern JF, Diggle SP. 2008. Quorum sensing by 2-alkyl-4-quinolones inPseudomonas aeruginosa and other bacterial species. Mol. Biosyst. 4:882–888.

23. Essar DW, Eberly L, Hadero A, Crawford IP. 1990. Identification andcharacterization of genes for a second anthranilate synthase in Pseudomo-nas aeruginosa: interchangeability of the two anthranilate synthases andevolutionary implications. J. Bacteriol. 172:884 –900.

24. Figurski DH, Helinski DR. 1979. Replication of an origin-containingderivative of plasmid RK2 dependent on a plasmid function provided intrans. Proc. Natl. Acad. Sci. U. S. A. 76:1648 –1652.

25. Fletcher MP, Diggle SP, Camara M, Williams P. 2007. Biosensor-basedassays for PQS, HHQ and related 2-alkyl-4-quinolone quorum sensingsignal molecules. Nat. Protoc. 2:1254 –1262.

26. Fothergill JL, et al. 2007. Widespread pyocyanin over-production amongisolates of a cystic fibrosis epidemic strain. BMC Microbiol. 7:45. doi:10.1186/1471-2180-7-45.

27. Gago G, Diacovich L, Arabolaza A, Tsai SC, Gramajo H. 2011. Fatty acidbiosynthesis in actinomycetes. FEMS Microbiol. Rev. 35:475– 497.

28. Gajiwala KS, et al. 2009. Crystal structures of bacterial FabH suggest amolecular basis for the substrate specificity of the enzyme. FEBS Lett.583:2939 –2946.

29. Gallagher LA, McKnight SL, Kuznetsova MS, Pesci EC, Manoil C. 2002.Functions required for extracellular quinolone signaling by Pseudomonasaeruginosa. J. Bacteriol. 184:6472– 6480.

30. Gambello MJ, Iglewski BH. 1991. Cloning and characterization of thePseudomonas aeruginosa lasR gene, a transcriptional activator of elastaseexpression. J. Bacteriol. 173:3000 –3009.

31. Garwin JL, Klages AL, Cronan JE, Jr. 1980. Structural, enzymatic, andgenetic studies of beta-ketoacyl-acyl carrier protein synthases I and II ofEscherichia coli. J. Biol. Chem. 255:11949 –11956.

32. Gerusz V. 2010. Recent advances in the inhibition of bacterial fatty acidbiosynthesis. Annu. Rep. Med. Chem. 45:295–311.

33. Girard G, Bloemberg GV. 2008. Central role of quorum sensing in reg-ulating the production of pathogenicity factors in Pseudomonas aerugi-nosa. Future Microbiol. 3:97–106.

34. Hannauer M, et al. 2012. Biosynthesis of the pyoverdine siderophore ofPseudomonas aeruginosa involves precursors with a myristic or a myristo-leic acid chain. FEBS Lett. 586:96 –101.

35. Haussler S. 2010. Multicellular signalling and growth of Pseudomonasaeruginosa. Int. J. Med. Microbiol. 300:544 –548.

36. Heath RJ, Rock CO. 1996. Regulation of fatty acid elongation and initi-ation by acyl-acyl carrier protein in Escherichia coli. J. Biol. Chem. 271:1833–1836.

37. Heath RJ, White SW, Rock CO. 2002. Inhibitors of fatty acid synthesis asantimicrobial chemotherapeutics. Appl. Microbiol. Biotechnol. 58:695–703.

38. Hoang TT, Schweizer HP. 1997. Fatty acid biosynthesis in Pseudomonasaeruginosa: cloning and characterization of the fabAB operon encodingbeta-hydroxyacyl-acyl carrier protein dehydratase (FabA) and beta-ketoacyl-acyl carrier protein synthase I (FabB). J. Bacteriol. 179:5326 –5332.

39. Hoang TT, Schweizer HP. 1999. Characterization of Pseudomonas aerugi-nosa enoyl-acyl carrier protein reductase (FabI): a target for the antimi-crobial triclosan and its role in acylated homoserine lactone synthesis. J.Bacteriol. 181:5489 –5497.

40. Hoang TT, Sullivan SA, Cusick JK, Schweizer HP. 2002. Beta-ketoacylacyl carrier protein reductase (FabG) activity of the fatty acid biosyntheticpathway is a determining factor of 3-oxo-homoserine lactone acyl chainlengths. Microbiology 148:3849 –3856.

41. Juhas M, Eberl L, Tummler B. 2005. Quorum sensing: the power ofcooperation in the world of Pseudomonas. Environ. Microbiol. 7:459 –471.

42. Kass LR, Brock DJ, Bloch K. 1967. Beta-hydroxydecanoyl thioesterdehydrase. I. Purification and properties. J. Biol. Chem. 242:4418 – 4431.

43. Kerr KG, Snelling AM. 2009. Pseudomonas aeruginosa: a formidable andever-present adversary. J. Hosp. Infect. 73:338 –344.

44. Khandekar SS, Daines RA, Lonsdale JT. 2003. Bacterial beta-ketoacyl-acyl carrier protein synthases as targets for antibacterial agents. Curr. Pro-tein Pept. Sci. 4:21–29.

45. Kohler T, Curty LK, Barja F, van Delden C, Pechere JC. 2000. Swarming




.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

of Pseudomonas aeruginosa is dependent on cell-to-cell signaling and re-quires flagella and pili. J. Bacteriol. 182:5990 –5996.

46. Kutchma AJ, Hoang TT, Schweizer HP. 1999. Characterization of aPseudomonas aeruginosa fatty acid biosynthetic gene cluster: purificationof acyl carrier protein (ACP) and malonyl-coenzyme A:ACP transacylase(FabD). J. Bacteriol. 181:5498 –5504.

47. Lai CY, Cronan JE. 2003. Beta-ketoacyl-acyl carrier protein synthase III(FabH) is essential for bacterial fatty acid synthesis. J. Biol. Chem. 278:51494 –51503.

48. Liberati NT, et al. 2006. An ordered, nonredundant library of Pseudomo-nas aeruginosa strain PA14 transposon insertion mutants. Proc. Natl.Acad. Sci. U. S. A. 103:2833–2838.

49. Lu YJ, Zhang YM, Rock CO. 2004. Product diversity and regulation oftype II fatty acid synthases. Biochem. Cell Biol. 82:145–155.

50. Martin CA, Hoven AD, Cook AM. 2008. Therapeutic frontiers: prevent-ing and treating infectious diseases by inhibiting bacterial quorum sens-ing. Eur. J. Clin. Microbiol. Infect. Dis. 27:635– 642.

51. McClean KH, et al. 1997. Quorum sensing and Chromobacterium viola-ceum: exploitation of violacein production and inhibition for the detec-tion of N-acylhomoserine lactones. Microbiology 143(Part 12):3703–3711.

52. Meredith TC, Woodard RW. 2005. Identification of GutQ from Esche-richia coli as a D-arabinose 5-phosphate isomerase. J. Bacteriol. 187:6936 –6942.

53. Miethke M, Marahiel MA. 2007. Siderophore-based iron acquisition andpathogen control. Microbiol. Mol. Biol. Rev. 71:413– 451.

54. Mohan S, Kelly TM, Eveland SS, Raetz CR, Anderson MS. 1994. AnEscherichia coli gene (FabZ) encoding (3R)-hydroxymyristoyl acyl carrierprotein dehydrase. Relation to fabA and suppression of mutations in lipidA biosynthesis. J. Biol. Chem. 269:32896 –32903.

55. Morgan-Kiss RM, Cronan JE. 2008. The Lactococcus lactis FabF fatty acidsynthetic enzyme can functionally replace both the FabB and FabF pro-teins of Escherichia coli and the FabH protein of Lactococcus lactis. Arch.Microbiol. 190:427– 437.

56. Nie Z, et al. 2005. Structure-based design, synthesis, and study of potentinhibitors of beta-ketoacyl-acyl carrier protein synthase III as potentialantimicrobial agents. J. Med. Chem. 48:1596 –1609.

57. Ozen AI, Ussery DW. 2012. Defining the Pseudomonas genus: where dowe draw the line with Azotobacter? Microb. Ecol. 63:239 –248.

58. Parsons JB, Frank MW, Subramanian C, Saenkham P, Rock CO. 2011.Metabolic basis for the differential susceptibility of Gram-positive patho-gens to fatty acid synthesis inhibitors. Proc. Natl. Acad. Sci. U. S. A. 108:15378 –15383.

59. Parsons JB, Rock CO. 2011. Is bacterial fatty acid synthesis a valid targetfor antibacterial drug discovery? Curr. Opin. Microbiol. 14:544 –549.

60. Pesci EC, et al. 1999. Quinolone signaling in the cell-to-cell communi-cation system of Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. U. S. A.96:11229 –11234.

61. Pinzon NM, Ju LK. 2009. Improved detection of rhamnolipid productionusing agar plates containing methylene blue and cetyl trimethylammo-nium bromide. Biotechnol. Lett. 31:1583–1588.

62. Pistorius D, et al. 2011. Biosynthesis of 2-alkyl-4(1H)-quinolones inPseudomonas aeruginosa: potential for therapeutic interference withpathogenicity. ChemBioChem 12:850 – 853.

63. Qiu X, et al. 2005. Crystal structure and substrate specificity of the beta-ketoacyl-acyl carrier protein synthase III (FabH) from Staphylococcus au-reus. Protein Sci. 14:2087–2094.

64. Raetz CR, Whitfield C. 2002. Lipopolysaccharide endotoxins. Annu. Rev.Biochem. 71:635–700.

65. Raychaudhuri A, Jerga A, Tipton PA. 2005. Chemical mechanism andsubstrate specificity of RhlI, an acylhomoserine lactone synthase fromPseudomonas aeruginosa. Biochemistry 44:2974 –2981.

66. Reis RS, Pereira AG, Neves BC, Freire DM. 2011. Gene regulation ofrhamnolipid production in Pseudomonas aeruginosa–a review. Bioresour.Technol. 102:6377– 6384.

67. Revill WP, Bibb MJ, Scheu AK, Kieser HJ, Hopwood DA. 2001. Beta-ketoacyl acyl carrier protein synthase III (FabH) is essential for fatty acidbiosynthesis in Streptomyces coelicolor A3(2). J. Bacteriol. 183:3526 –3530.

68. Rock CO, Cronan JE, Jr. 1981. Acyl carrier protein from Escherichia coli.Methods Enzymol. 71(Part C):341–351.

69. Rosenfeld IS, D’Agnolo G, Vagelos PR. 1973. Synthesis of unsaturatedfatty acids and the lesion in fabB mutants. J. Biol. Chem. 248:2452–2460.

70. Schaber JA, et al. 2004. Analysis of quorum sensing-deficient clinicalisolates of Pseudomonas aeruginosa. J. Med. Microbiol. 53:841– 853.

71. Schweizer E, Hofmann J. 2004. Microbial type I fatty acid synthases(FAS): major players in a network of cellular FAS systems. Microbiol. Mol.Biol. Rev. 68:501–517.

72. Schwyn B, Neilands JB. 1987. Universal chemical assay for the detectionand determination of siderophores. Anal. Biochem. 160:47–56.

73. Setubal JC, et al. 2009. Genome sequence of Azotobacter vinelandii, anobligate aerobe specialized to support diverse anaerobic metabolic pro-cesses. J. Bacteriol. 191:4534 – 4545.

74. Shaw PD, et al. 1997. Detecting and characterizing N-acyl-homoserinelactone signal molecules by thin-layer chromatography. Proc. Natl. Acad.Sci. U. S. A. 94:6036 – 6041.

75. Sievers F, et al. 2011. Fast, scalable generation of high-quality proteinmultiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7:539.

76. Simon R, Priefer U, Puhler A. 1983. A broad host range mobilizationsystem for in vivo genetic engineering - transposon mutagenesis in Gram-negative bacteria. Biotechnology 1:784 –791.

77. Smith S. 1994. The animal fatty acid synthase: one gene, one polypeptide,seven enzymes. FASEB J. 8:1248 –1259.

78. Smith S, Witkowski A, Joshi AK. 2003. Structural and functional orga-nization of the animal fatty acid synthase. Prog. Lipid Res. 42:289 –317.

79. Stintzi A, Evans K, Meyer JM, Poole K. 1998. Quorum-sensing andsiderophore biosynthesis in Pseudomonas aeruginosa: lasR/lasI mutantsexhibit reduced pyoverdine biosynthesis. FEMS Microbiol. Lett. 166:341–345.

80. Stover CK, et al. 2000. Complete genome sequence of Pseudomonasaeruginosa PAO1, an opportunistic pathogen. Nature 406:959 –964.

81. Taguchi F, et al. 2006. A homologue of the 3-oxoacyl-(acyl carrier pro-tein) synthase III gene located in the glycosylation island of Pseudomonassyringae pv. tabaci regulates virulence factors via N-acyl homoserine lac-tone and fatty acid synthesis. J. Bacteriol. 188:8376 – 8384.

82. Tamura K, et al. 2011. MEGA5: molecular evolutionary genetics analysisusing maximum likelihood, evolutionary distance, and maximum parsi-mony methods. Mol. Biol. Evol. 28:2731–2739.

83. Tremblay J, Deziel E. 2008. Improving the reproducibility of Pseudomo-nas aeruginosa swarming motility assays. J. Basic Microbiol. 48:509 –515.

84. Tsay JT, Oh W, Larson TJ, Jackowski S, Rock CO. 1992. Isolation andcharacterization of the beta-ketoacyl-acyl carrier protein synthase III gene(fabH) from Escherichia coli K-12. J. Biol. Chem. 267:6807– 6814.

85. Visca P, Imperi F, Lamont IL. 2007. Pyoverdine siderophores: frombiogenesis to biosignificance. Trends Microbiol. 15:22–30.

86. White SW, Zheng J, Zhang YM, Rock CO. 2005. The structural biologyof type II fatty acid biosynthesis. Annu. Rev. Biochem. 74:791– 831.

87. Williams P, Camara M. 2009. Quorum sensing and environmental ad-aptation in Pseudomonas aeruginosa: a tale of regulatory networks andmultifunctional signal molecules. Curr. Opin. Microbiol. 12:182–191.

88. Winsor GL, et al. 2011. Pseudomonas Genome Database: improved com-parative analysis and population genomics capability for Pseudomonasgenomes. Nucleic Acids Res. 39:D596 –D600.

89. Winstanley C, Fothergill JL. 2009. The role of quorum sensing in chroniccystic fibrosis Pseudomonas aeruginosa infections. FEMS Microbiol. Lett.290:1–9.

90. Yuan Y, Leeds JA, Meredith TC. 2012. Pseudomonas aeruginosa directlyshunts �-oxidation degradation intermediates into de novo fatty acid bio-synthesis. J. Bacteriol. 194:5185–5196.

91. Zhang HJ, Li ZL, Zhu HL. 2012. Advances in the research of beta-ketoacyl-ACP synthase III (FabH) inhibitors. Curr. Med. Chem. 19:1225–1237.

92. Zhang YM, White SW, Rock CO. 2006. Inhibiting bacterial fatty acidsynthesis. J. Biol. Chem. 281:17541–17544.

93. Zhu K, Rock CO. 2008. RhlA converts beta-hydroxyacyl-acyl carrierprotein intermediates in fatty acid synthesis to the beta-hydroxydecanoyl-beta-hydroxydecanoate component of rhamnolipids in Pseudomonasaeruginosa. J. Bacteriol. 190:3147–3154.

94. Zhu L, Lin J, Ma J, Cronan JE, Wang H. 2010. Triclosan resistance ofPseudomonas aeruginosa PAO1 is due to FabV, a triclosan-resistant enoyl-acyl carrier protein reductase. Antimicrob. Agents Chemother. 54:689 –698.

Yuan et al.



.org/D

ownloaded from

http://jb.asm.org

http://jb.asm.org/

Will the Initiator of Fatty Acid Synthesis in Pseudomonas aeruginosaPlease Stand Up?

Yong-Mei Zhanga and Charles O. Rockb

Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, South Carolina, USA,a and Department of Infectious Diseases, St.Jude Children’s Research Hospital, Memphis, Tennessee, USAb

Type II fatty acid synthesis (FASII) is vital for bacterial mem-brane biogenesis. The first condensation step in the pathway is

carried out by �-ketoacyl-acyl carrier protein (ACP) synthase III,or FabH, and subsequent condensation reactions are executed bythe FabB/F family of enzymes. The prototypical FabH activity ofEscherichia coli catalyzes the condensation of acetyl coenzyme A(acetyl-CoA) with malonyl-ACP to form the first �-ketoacyl-ACPintermediate (7). Variations on this theme are Gram-positive bac-teria that use 5-carbon branched-chain acyl coenzymes A (acyl-CoAs) to initiate fatty acid synthesis (1) and mycobacteria wherelong-chain acyl-CoAs are used by FabH to initiate mycolic acidsynthesis (2). Most bacteria have a single, essential fabH gene (9).However, the situation with Pseudomonas aeruginosa is far morecomplex, because there are at least three candidates for the FabHinitiator of FASII and four isoforms of the FabB/F elongation con-densing enzymes (Fig. 1). In this issue, two accompanying papers,“Fatty acid biosynthesis in Pseudomonas aeruginosa is initiated bythe FabY class of �-ketoacyl acyl carrier protein synthases” and“Pseudomonas aeruginosa directly shunts �-oxidation degrada-tion intermediates into de novo fatty acid biosynthesis”, both byYuan and colleagues (15, 16), have unraveled the complexity infabH gene functions in Pseudomonas aeruginosa. They reach thesurprising conclusion that none of the fabH genes are responsiblefor the initiation of fatty acid synthesis in P. aeruginosa! Rather,initiation of FASII is carried out by a unique gene named FabYthat is more closely related to the elongation condensing enzymes(FabB/F). Furthermore, one of the fabH homologs encoded byPA3286 plays a unique role in channeling 8-carbon acyl-CoA in-termediates arising from fatty acid �-oxidation into the FASIIelongation cycle. This is the first example of an enzyme that allows�-oxidation intermediates to be utilized by the FASII pathway.

Two classes of condensing enzymes. The condensing enzymesof FASII belong to the �/�-hydrolase superfamily and are dividedinto two classes based on their active site triads (Fig. 1). The FabHclass possesses a Cys-His-Asn catalytic triad and condenses anacyl-CoA with malonyl-ACP to initiate FASII. The FabB/F class ofelongation condensing enzymes carries out all subsequent con-densation reactions between the elongating acyl-ACP and malo-nyl-ACP using a Cys-His-His configuration. All biochemicallycharacterized condensing enzymes have fallen so far into one ofthese two classes based on the active site triad leading to confidentfunctional predictions. Although the FabH enzymes (FabHs) useacyl-CoA and FabB/F use acyl-ACP, the condensation reactionoccurs after the acyl chain has been transferred to the active sitecysteine and the subsequent binding of malonyl-ACP. The activesite chemistry performed by both of these enzymes appears iden-tical, so it has never been clear why initiation condensing enzymeshave one type of active site and the elongation enzymes have an-other type. The discoveries of Yuan and colleagues (15, 16) mean

that we can no longer confidently assign roles to a condensingenzyme based on the presence of a Cys-His-Asn or Cys-His-Hisactive site triad.

FabY, a FabB/F homolog that functions as a FabH. Deletingeach of the 3 fabH candidates to determine which gene was in-volved in FASII eventually led to the generation of a triple knock-out strain that had no overt growth phenotype. Yuan et al. (16)then identified PA5174 as an open reading frame capable of initi-ating FASII, using a P. aeruginosa cosmid library to select for genesthat cured the growth defect in an E. coli strain with downregu-lated fabH expression. Biochemical characterization of purifiedFabY showed that the enzyme efficiently catalyzed the condensa-tion of acetyl-CoA with malonyl-ACP. It is not known whetherFabY possesses FabB/F-like activity using acyl-ACP substrates.The fabY knockout exhibited a distinct growth defect and was alsodeficient in the formation of numerous secondary metabolites,quorum-sensing signaling molecules, and siderophores. All ofthese phenotypes are restored to normal by the expression of E.coli fabH. Thus, FabY is a condensing enzyme that performs a taskusually attributed to FabH but uses a Cys-His-His active site ofFabB/F enzymes.

Connecting fatty acid oxidation and biosynthesis. The dis-covery of FabY begs the question: what are the roles of all thosefabH genes? Yuan et al. (15) have defined the role of PA3286, thethird FabH in P. aeruginosa (FabH3) (Fig. 1A). The authors ob-served that the growth defect of the �fabY strain was cured bysupplementing the medium with decanoic acid. Experiments us-ing deuterated decanoic acid feeding established that FabH3shunts the octanoyl-CoA arising from the �-oxidation of fattyacids into the FASII pathway. Thus, FabH3 is an initiation con-densing enzyme with the unique property of utilizing octanoyl-CoA derived from fatty acid degradation. Subtle differences in the3-dimensional structures of the hydrophobic pockets that accom-modate the acyl-enzyme intermediate account for the acyl chainspecificity of the FabHs (5). Thus, PA3286 (FabH3) is predicted tohave a hydrophobic tunnel that acts as a molecular ruler to selec-tively accept an octanoyl-CoA. This substrate specificity is the keyto its function in P. aeruginosa lipid metabolism. Most obvious isthe savings in energy to produce a fatty acid chain starting from an8-carbon precursor compared to the standard 2-carbon primerprovided by acetyl-CoA. However, the importance of introducing

Published ahead of print 20 July 2012

Address correspondence to Charles O. Rock, [email protected].


doi:10.1128/JB.01198-12

COMMENTARY


http://dx.doi.org/doi:10.1128/JB.00792-12




http://dx.doi.org/10.1128/JB.01198-12

http://jb.asm.org

the �-oxidation intermediates into FASII at the 8-carbon step ismore clearly appreciated by understanding the critical importanceof FASII intermediates in P. aeruginosa physiology. 3-Hydroxyde-canoyl-ACP is the key intermediate in unsaturated fatty acid syn-thesis (20) and is also used for the synthesis of lipopolysaccharides(4), secreted rhamnolipids (21), and the poly-�-hydroxy-alkano-ate storage lipids (13). Octanoyl-ACP is the precursor to lipoicacid (19). The 10- and 12-carbon �-ketoacyl-ACP intermediatesare used in the biosynthesis of acyl-homoserine lactones (10) andalkyl-quinolone (12) signaling molecules. Thus, the reintroduc-tion of acyl chains at the 8-carbon stage allows for the productionof a spectrum of products derived from FASII.

Two enigmatic condensing enzymes remain. Two of theseven condensing enzymes in P. aeruginosa still lack a functionalannotation (Fig. 1). PqsD (FabH1) catalyzes a condensation reac-tion in the synthesis of dihydroxy- and alkyl-quinolines (12, 18).The function of PA3333 (FabH2) is unknown, and it is located inan operon with an ACP gene that is induced in stationary phase(14). PA3286 is responsible for initiating FASII from octanoyl-CoA (15). PA1609 (FabB) is required for unsaturated fatty acidbiosynthesis (6). Strains lacking PA2965 (FabF1) are deficient incis-vaccenic acid (8) and are unable to swarm, twitch, or swim(11). There is no clue to the function of PA1373 (FabF2). The mostdivergent of the FabB/F enzymes is PA5174 (FabY), which is nowknown as the FASII initiator (16). Neither PA1373 nor PA3333are essential for P. aeruginosa survival, so perhaps these en-zymes support the formation of secondary metabolites in sta-tionary phase, but this is just a guess at where to begin search-ing for their functions. Unraveling the transcriptional andbiochemical regulation of the condensing enzymes of P. aerugi-nosa should prove interesting.

ACKNOWLEDGMENTS

This work was supported by Department of Defense grant DM090161(Y.M.Z.), the Cystic Fibrosis Foundation grant ZHANG12I0 (Y.M.Z.),National Institutes of Health grants P20RR017677 SC COBRE in Lipido-mics and Pathobiology (subproject 8691 to Y.M.Z.), GM034496 (C.O.R.)and Cancer Center (CORE) support grant CA21765 and the AmericanLebanese Syrian Associated Charities.

REFERENCES1. Choi K-H, Heath RJ, Rock CO. 2000. �-Ketoacyl-acyl carrier protein

synthase III (FabH) is a determining factor in branched-chain fatty acidbiosynthesis. J. Bacteriol. 182:365–370.

2. Choi K-H, Kremer L, Besra GS, Rock CO. 2000. Identification andsubstrate specificity of �-ketoacyl-[acyl carrier protein] synthase III(mtFabH) from Mycobacterium tuberculosis. J. Biol. Chem. 275:28201–28207.

3. Davies C, Heath RJ, White SW, Rock CO. 2000. The 1.8 Å crystalstructure and active site architecture of �-ketoacyl-[acyl carrier protein]synthase III (FabH) from Escherichia coli. Structure 8:185–195.

4. Ernst RK, et al. 1999. Specific lipopolysaccharide found in cystic fibrosisairway Pseudomonas aeruginosa. Science 286:1561–1565.


6. Hoang TT, Schweizer HP. 1997. Fatty acid biosynthesis in Pseudomonasaeruginosa: cloning and characterization of the fabAB operon encoding�-hydroxyacyl-acyl carrier protein dehydratase (FabA) and �-ketoacyl-acyl carrier protein synthase I (FabB). J. Bacteriol. 179:5326 –5332.

7. Jackowski S, Rock CO. 1987. Acetoacetyl-acyl carrier protein synthase, apotential regulator of fatty acid biosynthesis in bacteria. J. Biol. Chem.262:7927–7931.

8. Kutchma AJ, Hoang TT, Schweizer HP. 1999. Characterization of aPseudomonas aeruginosa fatty acid biosynthetic gene cluster: purificationof acyl carrier protein (ACP) and malonyl-coenzyme A:ACP transacylase(FabD). J. Bacteriol. 181:5498 –5504.

9. Lai CY, Cronan JE. 2003. �-Ketoacyl-acyl carrier protein synthase III(FabH) is essential for bacterial fatty acid synthesis. J. Biol. Chem. 278:51494 –51503.

10. Moré MI, et al. 1996. Enzymatic synthesis of a quorum-sensing autoin-ducer through use of defined substrates. Science 272:1655–1658.

11. Overhage J, Lewenza S, Marr AK, Hancock REW. 2007. Identification ofgenes involved in swarming motility using a Pseudomonas aeruginosaPAO1 mini-Tn5-lux mutant library. J. Bacteriol. 189:2164 –2169.

12. Pistorius D, et al. 2011. Biosynthesis of 2-alkyl-4(1H)-quinolones inPseudomonas aeruginosa: potential for therapeutic interference withpathogenicity. ChemBioChem 12:850 – 853.

13. Rehm BH, Kruger N, Steinbuchel A. 1998. A new metabolic link betweenfatty acid de novo synthesis and polyhydroxyalkanoic acid synthesis. ThePhaG gene from Pseudomonas putida KT2440 encodes a 3-hydroxyacyl-acylcarrier protein-coenzyme A transferase. J. Biol. Chem. 273:24044–24051.

14. Schuster M, Lostroh CP, Ogi T, Greenberg EP. 2003. Identification,timing, and signal specificity of Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis. J. Bacteriol. 185:2066 –2079.

15. Yuan Y, Leeds JA, Meredith TC. 2012. Pseudomonas aeruginosa directly

FIG 1 The condensing enzymes in P. aeruginosa. The active site structures, locus tags, gene symbols, blocks of homology surrounding the key catalytic residues,and the functional annotations of the genes are provided for the three FabH (A) and four FabB/F (B) homologs in P. aeruginosa. The sequence alignment blockssurrounding the Cys-His-Asn triad of the FabH homologs and Cys-His-His of the FabB/F homologs of P. aeruginosa are compared to the active site residues ofthe prototypical E. coli FabH (EcFabH) and FabB (EcFabB), respectively. PqsD is sometimes annotated as FabH1. Identical residues are shown on a blackbackground, and conservatively substituted residues are shown on a gray background. The depicted active site structures are derived from the crystal structuresof E. coli FabH (Protein Data Bank [PBD] accession no. 1EBL) (3) and E. coli FabB (PDB accession no. 1G5X) (17). UFA, unsaturated fatty acid; 18:1�11,cis-vaccenic acid; C8-CoA, octanoyl-CoA; Ac-CoA, acetyl-CoA.

Commentary


http://jb.asm.org

shunts �-oxidation degradation intermediates into de novo fatty acid bio-synthesis. J. Bacteriol. 194:5185–5196.

16. Yuan Y, Sachdeva M, Leeds JA, Meredith TC. 2012. Fatty acid biosyn-thesis in Pseudomonas aeruginosa is initiated by the FabY class of�-ketoacyl acyl carrier protein synthases. J. Bacteriol. 194:5171–5184.

17. Zhang Y-M, et al. 2001. Identification and analysis of the acyl carrierprotein (ACP) docking site on �-ketoacyl-ACP synthase III. J. Biol. Chem.276:8231– 8238.

18. Zhang YM, Frank MW, Zhu K, Mayasundari A, Rock CO. 2008. PqsD isresponsible for the synthesis of 2,4-dihydroxyquinoline, an extracellular metabo-lite produced by Pseudomonas aeruginosa. J. Biol. Chem. 283:28788–28794.

19. Zhao X, Miller JR, Cronan JE. 2005. The reaction of LipB, the octanoyl-[acyl carrier protein]:protein N-octanoyltransferase of lipoic acid synthe-sis, proceeds through an acyl-enzyme intermediate. Biochemistry 44:16737–16746.

20. Zhu K, Choi K-H, Schweizer HP, Rock CO, Zhang Y-M. 2006. Twoaerobic pathways for the formation of unsaturated fatty acids in Pseu-domonas aeruginosa. Mol. Microbiol. 60:260 –273.

21. Zhu K, Rock CO. 2008. RhlA converts �-hydroxyacyl-acyl carrier proteinintermediates in fatty acid synthesis to the �-hydroxydecanoyl-�-hydroxydecanoate component of rhamnolipids in Pseudomonas aerugi-nosa. J. Bacteriol. 190:3147–3154.

Commentary


http://jb.asm.org

Pseudomonas aeruginosa Directly Shunts �-Oxidation DegradationIntermediates into De Novo Fatty Acid Biosynthesis

Yanqiu Yuan, Jennifer A. Leeds, and Timothy C. Meredith

Infectious Diseases Area, Novartis Institutes for BioMedical Research, Cambridge, Massachusetts, USA

We identified the fatty acid synthesis (FAS) initiation enzyme in Pseudomonas aeruginosa as FabY, a �-ketoacyl synthaseKASI/II domain-containing enzyme that condenses acetyl coenzyme A (acetyl-CoA) with malonyl-acyl carrier protein (ACP) tomake the FAS primer �-acetoacetyl-ACP in the accompanying article (Y. Yuan, M. Sachdeva, J. A. Leeds, and T. C. Meredith, J.Bacteriol. 194:5171-5184, 2012). Herein, we show that growth defects stemming from deletion of fabY can be suppressed by sup-plementation of the growth media with exogenous decanoate fatty acid, suggesting a compensatory mechanism. Fatty acids eightcarbons or longer rescue growth by generating acyl coenzyme A (acyl-CoA) thioester �-oxidation degradation intermediates thatare shunted into FAS downstream of FabY. Using a set of perdeuterated fatty acid feeding experiments, we show that the openreading frame PA3286 in P. aeruginosa PAO1 intercepts C8-CoA by condensation with malonyl-ACP to make the FAS interme-diate �-keto decanoyl-ACP. This key intermediate can then be extended to supply all of the cellular fatty acid needs, includingboth unsaturated and saturated fatty acids, along with the 3-hydroxyl fatty acid acyl groups of lipopolysaccharide. HeterologousPA3286 expression in Escherichia coli likewise established the fatty acid shunt, and characterization of recombinant �-keto acylsynthase enzyme activity confirmed in vitro substrate specificity for medium-chain-length acyl CoA thioester acceptors. Thepotential for the PA3286 shunt in P. aeruginosa to curtail the efficacy of inhibitors targeting FabY, an enzyme required for FASinitiation in the absence of exogenous fatty acids, is discussed.

Pseudomonas aeruginosa is a versatile Gram-negative pathogen,being the causative agent of a wide range of both community-

associated (folliculitis and otitis externa) and health care-associ-ated (pneumonia, urinary tract, and bacteremia) bacterial infec-tions (15, 34, 39, 55). While generally not considered a normalmember of the human flora, the near ubiquitous environmentaldistribution of P. aeruginosa provides a ready reservoir for expo-sure and ensuing opportunistic infection. Infections due to P.aeruginosa are particularly prevalent among immunodeficient in-dividuals in whom cutaneous or mucosal barriers have beenbreached by ventilators, catheters, or through trauma, as seen inburn units. In �28,000 cases of all health care-associated infec-tions reported to the U.S. National Healthcare Safety Networkduring a 22-month period beginning in 2006, 7.9% were attrib-uted to P. aeruginosa (24). Chronic Pseudomonas infections areespecially problematic in the lungs of cystic fibrosis patients, inpart due to a genetic defect that facilitates bacterial colonizationthrough diminished mucociliary clearance (6). The breadth ofdifficult-to-treat P. aeruginosa-related infections, coupled with animpressive array of intrinsic and adaptive antibacterial resistancemechanisms (3, 38), makes developing new antipseudomonasdrugs a challenging priority.

We initiated a target evaluation program focused on fatty acidsynthesis (FAS) in P. aeruginosa. FAS plays a multifaceted role in bothmaintaining bacterial viability and virulence in P. aeruginosa, suggest-ing that inhibition of FAS in vivo may have added benefit beyondsimply blocking division through depleting fatty acid pools availablefor phospholipid biosynthesis. Aside from being compulsory compo-nents of membrane phospholipids (12), fatty acids are utilized bymultiple primary and secondary metabolic pathways in P. aeruginosa.Lipopolysaccharides (LPS), which are lipoglycans located in the ex-ternal leaflet of the outer membrane, are dependent on 3-hydroxy(3-OH) acyl-ACP (acyl carrier protein) FAS intermediates for com-plete acylation and in turn the establishment of the permeability bar-

rier (43). Lipoproteins, a large class of lipidated membrane proteinsthat includes components of the LPS transport machinery and of theresistance-nodulation cell division (RND)-type drug efflux pumps(44), depend on phospholipid donors for acylation (40). Even partialFAS inhibition could therefore induce LPS hypoacylation, decreaseLPS transport, and/or cripple efflux pumps, raising the potential forsynergistic combinations between FAS inhibitors and membrane-impermeable or efflux-susceptible antibiotics. Fatty acids are alsoused in the assembly of two important metabolic enzyme cofactors,lipoate and biotin. Lipoate is an essential cofactor of �-ketoacid de-hydrogenases (49), including pyruvate dehydrogenase. Pyruvate de-hydrogenase connects glycolytic flux to the tricarboxylic acid (TCA)cycle by forming acetyl coenzyme A (acetyl-CoA) from pyruvate. P.aeruginosa lacking pyruvate dehydrogenase does not express the typeIII secretion system, critical machinery involved in delivering cyto-toxic effectors, due to the metabolic defect (45). More recently, theessential cofactor biotin has also been proposed to hijack FAS biosyn-thetic enzymes for its own synthesis (11, 37). Biotin is required byenzymes involved in amino acid metabolism, gluconeogenesis, andFAS itself. The assembly of the high-affinity ferric iron siderophorepyoverdine is dependent on the fatty acid myristate carrier in P.aeruginosa (22). Finally, orchestrating the complex intercellular sig-naling and hierarchal gene regulation that is a hallmark of the P.aeruginosa social lifestyle are the three acylated quorum-sensing sig-nal molecules [Pseudomonas quinolone signal (PQS), N-(3-oxodo-

Received 16 May 2012 Accepted 21 June 2012

Published ahead of print 29 June 2012

Address correspondence to Timothy C. Meredith, [email protected].

Supplemental material for this article may be found at http://jb.asm.org/.


doi:10.1128/JB.00860-12



http://jb.asm.org/

http://dx.doi.org/10.1128/JB.00860-12

http://jb.asm.org

decanoyl)-L-homoserine lactone, and N-butanoyl-L-homoserine lac-tone] (30) and the cis-2-decenoic fatty acid diffusible signal factor(DSF) (14). Collectively, these systems coordinate the expression ofhundreds of genes (54), including pertinent virulence factors such asrhamnolipids, pyocyanin, and extracellular proteases, as well as thetransition from planktonic to antibiotic-recalcitrant growth withinbiofilms. In addition to being an essential target in and of itself, thecentral roles of fatty acids in intrinsic antimicrobial resistance, in sup-porting diverse intermediary metabolism, in sensing environmentalcues, and in mobilizing a suite of virulence factors all make FAS apromising antibacterial target in P. aeruginosa.

The initiating step of FAS in Escherichia coli is catalyzed by theenzyme �-acetoacetyl-ACP synthase encoded by fabH (52), whichcondenses malonyl-ACP with acetyl-CoA to form �-acetoacetyl-ACP. The enzyme is defined by the signature �-ketoacyl synthaseIII (KASIII) domain, which is present in multiple, highly similargenes within the genome of P. aeruginosa PAO1 (56, 57). Ourinitial attempts in the accompanying article (56) to assign theFabH-type activity of P. aeruginosa to a single KASIII orthologwere unsuccessful; rather, it was determined that the predominant�-acetoacetyl-ACP synthase is encoded by a new class of highlydivergent KASI/II-type synthases named fabY (formerly PA5174).Deletion of fabY confirmed a pleiotropic phenotypic consistentwith decreased FAS flux, as siderophore production, swarmingmotility, rhamnolipids, and fatty acid-dependent quorum-sens-ing signals (PQS and homoserine lactones) in tandem with theexpression of their cognate regulatory targets were attenuated(56). Although the doubling time of the �fabY deletion mutantwas three times longer in liquid media, the viability of the �fabYdeletion mutant indicated that another P. aeruginosa gene(s) iscapable of fatty acid initiation. In addition, we observed that thegrowth defect could be partially complemented by inclusion ofexogenous free fatty acids in the growth media. We thus set out toaddress the fabY-independent route of FAS in P. aeruginosa and tounderstand how fatty acids are incorporated into de novo phos-pholipids. We herein show that the KASIII domain-containingenzyme from the previously unannotated open reading framePA3286 shunts fatty acid degradation intermediates from the�-oxidation pathway by condensing octanoyl-CoA (C8-CoA)with malonyl-ACP to make �-keto-decanoyl-ACP, a key buildingblock common to saturated fatty acids (SFA), unsaturated fattyacids (UFA), and LPS.

MATERIALS AND METHODSBacterial strains and growth conditions. All P. aeruginosa strains werederived from the wild-type PAO1 strain (51), while Escherichia coli strainswere derived from the reference strain BW25113 (CGSC7636). Strainswere grown in LB-Miller medium at 37°C unless otherwise noted. Anti-biotic markers were selected with gentamicin (Gm) (100 �g/ml inP. aeruginosa and 10 �g/ml in E. coli), carbenicillin (Carb) (150 �g/ml inP. aeruginosa and 100 �g/ml in E. coli), tetracycline (Tet) (125 �g/ml in P.aeruginosa and 15 �g/ml in E. coli), chloramphenicol (Cam) (20 �g/ml inE. coli), and kanamycin (Kan) (50 �g/ml in E. coli). Bacterial strains andplasmids used in this study are listed in Table 1.

Growth analysis. For P. aeruginosa and E. coli strains, starter cultureswere made by scraping cells off LB agar plates with appropriate selectionantibiotics. Cultures were typically resuspended to �2 � 105 CFU/ml inLB with or without fatty acid supplement (1 to 100 �g/ml) and inducer (1mM isopropyl-�-D-1-thiogalactopyranoside [IPTG]). Cultures were in-oculated into clear 96-well flat-bottom untreated microplates (catalog no.3370; Costar) and incubated at 37°C. The optical density at 600 nm

(OD600) was recorded every 10 min on a Spectramax plate reader withintermittent shaking.

Fatty acid composition. For readily soluble fatty acids (C2:0 to C10:0),bacteria were streaked onto LB agar plates containing 100 �g/ml of thesodium salt of a given fatty acid (Sigma or CDN Isotopes for deuteratedfatty acids) and incubated overnight at 37°C. For longer-chain fatty acids(C14:0 and C16:0), liquid cultures were inoculated (2 � 105 CFU/ml) intoLB with 2 mg/ml of fatty acid-free bovine serum albumin (BSA; Sigma)carrier and grown overnight to stationary phase at 37°C. To induce thefatty acid transporters of E. coli, the medium was spiked with 10 �g/ml ofunlabeled palmitate (C16:0) along with 100 �g/ml of perdeuterated de-canoate. Biomass was scraped from the agar surface or collected by cen-trifugation, suspended in phosphate-buffered saline (PBS), and washedthree times with PBS. Lipids were saponified, methylated, extracted, andwashed according to the Sherlock microbial identification system (Micro-bial ID, Inc., Newark, DE). Fatty acid methyl ester (FAME) compositionwas determined by gas chromatography with flame ionization detection(GC-FID) or with mass spectrometry (GC-MS). An HP 6890 gas chro-matograph with an HP-5MS 30 m column was connected to an HP 5973mass selective detector. Results were analyzed on the HP MSD Chem-Station (version D.01.02). Structural assignments were made by compar-ison of retention times to authentic FAME standards, as well as from themass spectra.

Synthetic lethal scoring between fabY and KASIII domain or-thologs. The conjugation vector pEX18ApGW (9) carrying genes encod-ing carbenicillin resistance (Carbr), oriT transfer origin, and sucrosecounterselection (sacB) was used to deliver an aacC1 (Gm) resistancecassette with either a transcriptionally coupled and functional E. coli fabHgene (pTMT123) or an inactive fabH fragment (pTMT124) (56) (Table1). The cassettes were flanked by �1 kb of homologous DNA flanking thefabY gene in P. aeruginosa PAO1. Vectors were mobilized in trans using E.coli Stellar cells (Clontech) transformed with the helper plasmid pRK2013(16) as the conjugation-proficient donor strain along with different recip-ient P. aeruginosa strains as has been described previously (56). Merodip-loid colonies were confirmed by colony PCR and outgrown for 4 h in 1 mlof LB only at 37°C. Serially diluted aliquots were then spread on LB agarcontaining 7% sucrose (LB–7% sucrose agar) (without NaCl) to counter-select colonies harboring unresolved plasmid. In parallel, aliquots wereplated on LB-sucrose agar with Gm or Gm plus decanoate (100 �g/ml) toselect colonies from which fabY was replaced with fabH (pTMT123) ordeleted (pTMT124). Colonies were counted after incubation at 37°C for24 h (all pTMT123 constructs) or 48 h for slow-growing mutants(pTMT124 selected on Gm). In cases where less than 10 colonies arose andspontaneous sucrose tolerance was suspected, colonies were patched ontoLB agar with Carb to confirm loss of plasmid backbone. The entire exper-iment was repeated three separate times in each P. aeruginosa recipientstrain background.

Complementation of �PA3286. The pBAD-containing pRC9 plas-mid derivative of the site-specific integration vector pMini-CTX1 (26)was used to complement the �PA3286 deletion strain. The PA3286 gene,along with 483 bp upstream encompassing the putative native promoterregion, was amplified from P. aeruginosa PAO1 genomic DNA templateusing the primer pair CTX-PA3286 EcoRV/CTX-PA3286 XhoI (Table 1).The fragment was introduced into pRC9 using the In-fusion system(Clontech), removing the pBAD regulatory element in the process. Theresulting vector (pTMT131) was transferred into P. aeruginosa �PA3286by conjugation, after which the integrated vector backbone encoding thetetracycline resistance (Tetr) gene was removed by FLP-catalyzed recom-bination using pFLP2 (25) to generate the complemented strain TMT44(Table 1).

PA3286 complementation in E. coli. The PA3286 open reading framewas cloned by restriction and ligation into the IPTG-inducible expressionvector pET24b(�) (Novagen) using the PCR product of primers PA3286for NdeI/PA3286 rev HindIII (rev stands for reverse) (Table 1). The ex-pression plasmid pET-PA3286 was transformed into E. coli BW25113, in

Yuan et al.



http://jb.asm.org

which the fabH gene was deleted using the fabHEc::camR (fabHEc is thefabH gene from E. coli) cassette and the Red recombinase system as de-scribed previously (56). The TMT47 strain was verified by flanking PCR,and maintained on 1 mM IPTG for PA3286 induction.

Recombinant PA3286 protein expression and purification. Recom-binant PA3286 was obtained according to the protocol described forPA5174 (56). Briefly, E. coli BL21(DE3) Rosetta 2 (Novagen) cells weretransformed with the His tag expression vector pET-PA3286 (Table 1).Colonies were inoculated into 1 liter of LB medium and grown at 37°Cwith shaking until mid-exponential growth (OD600 of 0.6) before induc-tion with 1 mM IPTG. After 3 h of expression at 37°C, the biomass washarvested by centrifugation (5,000 � g, 10 min, room temperature [RT])and stored at 80°C. The pellet was subjected to 2 rounds of freeze-thawcycles, incubated in lysis solution (1� Bugbuster [Novagen], 5 kU/mlrecombinant lysozyme [Novagen], 25 U/ml benzonuclease [Novagen]),and then clarified by centrifugation. The supernatant was gently shakenwith nickel-nitrilotriacetic acid (Ni-NTA) His bind resin (1 h on ice). Theslurry was loaded into an empty column, washed with 50 ml of bindingbuffer (50 mM Tris, 500 mM NaCl, 10 mM imidazole [pH 7.5]), 20 ml ofwash buffer (binding buffer with 50 mM imidazole [total concentration]),and eluted (binding buffer plus 200 mM imidazole [total concentration]).Fractions containing protein of the expected size (43 kDa, 373 amino acids

without His tag) were pooled and concentrated (Amicon Ultra, 10-kDamolecular size cutoff; Millipore). The sample buffer was exchanged withstorage buffer (20 mM Tris-HCl, 150 mM NaCl, 1 mM D/L-dithiothreitol,and 10% glycerol) by three rounds of dilution (10 ml each) and concen-tration, after which the aliquots were flash frozen at 80°C.

Recombinant PA3286 �-ketoacyl synthase enzyme activity assay.Conformation-sensitive urea-PAGE was used to separate and analyzeacyl-ACP condensation products (46). Malonyl-ACP was produced insitu using a previously described procedure (56). For each saturatedstraight-chain acyl coenzyme A (acyl-CoA) acceptor substrate tested (C2

to C16 in length; Sigma), 200 �M acyl-CoA along with 0.1 �g of PA3286(final concentration of 70 nM) was added, and the reaction mixtures wereincubated at room temperature for 1 h. The products were separated by0.5 M urea–16% PAGE and stained with Coomassie blue dye.

RESULTSExogenous fatty acids C8 and longer rescue growth of P. aerugi-nosa �fabY. The P. aeruginosa �fabY strain has a pronouncedgrowth defect both in LB liquid medium and on agar plates. Theresults of initial experiments suggested that growth on agar couldat least partially be rescued by inclusion of the fatty acid decanoate

TABLE 1 Bacterial strains, plasmids, and primers used in this study

Bacterial strain,plasmid, or primer Relevant genotype or phenotypea or primer sequence Source or reference

E. coli strainsBW25113 E. coli -12 Wt [�(araD-araB)567 �lacZ4787(::rrnB-3) � rph-1 �(rhaD-rhaB)568 hsdR514] CGSCb

TMY32 BW25113 fabH::camR(pET-PA5174); Kanr Camr 56TMT47 BW25113 fabH::camR(pET-PA3286); Kanr Camr This study

P. aeruginosa strainsNB52019 P. aeruginosa PAO1 prototroph K767 K. PooleTMT01 NB52019 �PA3333 (fabH2) 56TMT02 NB52019 �PA0999 (pqsD) 56TMT12 NB52019 �PA3286 56TMT15 NB52019 �PA0998 (pqsC) 56TMT16 NB52019 �PA0998 �PA0999 �PA3333 �PA3286 56TMT39 NB52019 �fabY (PA5174)::aacC1; Gmr 56TMT44 TMT12 attB::PA3286�; Gmr This study

PlasmidspMini-CTX1 Chromosomal integration vector; FRT oriT int�ori (pMB1) tetR-FRT attP� MCS; Tetr 26pRC9 pMini-CTX-1 with araC-PBAD promoter cassette C. DeanpTMT131 pRC9 with the PA3286 gene and 483 bp upstream This studypFLP2 E. coli-P. aeruginosa shuttle vector with Flp recombinase; Carbr 25pET24b(�) IPTG-inducible T7 promoter for protein expression; Kanr NovagenpET-PA3286 pET24b(�) with PA3286 and C-terminal His tag; Kanr This studypKD3 oriR� camR; Camr 13pTMT123 pEX18ApGW-PA5174::aacC1 fabHEc; Gmr 56pTMT124 pEX18ApGW-PA5174::aacC1 fabH= fragment; Gmr 56

PrimersCTX-PA3286 EcoRVc TGCGACGCTGGCGATCTTCCTTGTGAATAC This studyCTX-PA3286 XhoIc CGGGCCCCCCCTCGAAACCAGCTCGCGGAAG This studyPA3286 for NdeId GTAGCTCATATGCATAAAGCCGTCATC This studyPA3286 rev HindIIId GTAGCTAAGCTTGTGTTTACGCAGGATC This studyfabH Ec KO P1 CGCCACATTGCCGCGCCAAACGAAACCGTTTCAACCATGGGCCGCCTACCTGTGACGGAA 56fabH Ec KO P2 CCGCCCCAGATTTCACGTATTGATCGGCTACGCTTAATGCATGAACTTCATTTAAATGGCGCG 56

a Carbr, carbenicillin resistant; Camr, chloramphenicol resistant; Kanr, kanamycin resistant; Gmr, gentamicin resistant; Tetr, tetracycline resistant; FRT, Flp recombinase target;MCS, multicloning site.b Strain CGSC7636 at the Coli Genetic Stock Center (CGSC).c Cloned using the In-fusion system (Clontech).d Cloned into vectors by DNA digestion and T4 DNA-catalyzed ligation.

Fatty Acid Shunting in P. aeruginosa


http://jb.asm.org

(C10) in the media. In liquid media, inclusion of as low as 1 �g/mlof C10 in liquid LB culture also restored growth, with rates ap-proaching those of the wild type at 100 �g/ml (Fig. 1A). To deter-mine whether a specific fatty acid chain length or range is requiredto suppress the growth-defective phenotype, the �fabY strain wasstreaked onto LB agar plates supplemented with 100 �g/ml ofstraight-chain saturated fatty acids from C2 to C16 in length (Fig.1B). The average colony size was noticeably larger for C8 and lon-ger fatty acids, whereas C6 down to C2 had minimal effect in com-parison to LB agar alone. P. aeruginosa PAO1 can thus utilizeexogenous medium- and long-chain fatty acids to compensate fordecreased de novo FAS flux in the absence of fabY.

P. aeruginosa shunts exogenous medium-chain-length fattyacids. Exogenous fatty acids are taken up and degraded 2 carbonsat a time by the fatty acid degradation (fad) �-oxidation cycle (2,10). The range of substrates that can enter and be efficiently uti-lized by the �-oxidation cycle is defined by a combination of fattyacid inducer specificity and acyl-CoA synthetase (FadD) substratepreference. In E. coli for instance, only fatty acids longer than C12

induce transcription of fad genes, while fatty acids must be at leastC8 to serve as FadD substrates (36, 53). The requirement for acylchain lengths of C8 or longer for growth rescue in P. aeruginosa�fabY could therefore reflect two scenarios; either a longer-chainfatty acid is required for shunting into de novo FAS downstream of

FabY or fatty acids must be efficiently utilized by the �-oxidationcycle in order to be degraded to acetyl-CoA. In the latter case, highconcentrations of intracellular acetyl-CoA pools would then ulti-mately be responsible for rescuing growth by improving turnoverrates of a putative low-affinity FabH/FabY-type enzyme. To dif-ferentiate between these two possibilities, we fed the wild-type and�fabY P. aeruginosa strains perdeuterated C10 and analyzed thefatty acid composition as fatty acid methyl ester (FAME) deriva-tives by gas chromatography with flame ionization detection (GC-FID) and mass spectrometry (Fig. 2 and 3). If the decanoate (C10)substrate is being completely degraded to acetyl-CoA, the labelshould become dispersed with random reincorporation into fattyacids, while a shunt would retain the label and generate uniquedeuterium-containing fatty acid peaks. Indeed, the FAME GCtrace of the wild type revealed additional peaks specific to samplesfed perdeuterated C10 (in comparison with unlabeled C10), elutingslightly ahead of every major constituent fatty acid peak (Fig. 2).Since the amount of deuterium incorporated into FAMEs is in-versely correlated with retention time (42), the shortened reten-tion time along with peak symmetry and relative abundance(�40% of unlabeled FAME peak area) suggested a shunt mecha-nism. We then fed perdeuterated C10 to the P. aeruginosa �fabYstrain, which has a rate of de novo FAS flux of less than 5% of thewild type as determined by phospholipid macromolecular label-ing with radiolabeled acetate (56). In this case, only the labeledpeak was observed, as would be expected in the �fabY geneticbackground, since the incorporation rate of the shunt pathwaygreatly exceeds de novo FAS initiation.

The structure of each deuterium-labeled FAME peak was as-signed using mass spectrometry (Fig. 3). In comparison to thecorresponding unlabeled FAME peak (see Fig. S1 in the supple-mental material), the deuterated FAME molecular ions were 15Da heavier and had odd mass numbers for all of the saturated andhydroxylated fatty acids. The odd masses suggested that the ter-minal �-carbon was deuterated (-CD3). The McLafferty rear-rangement ion, which is a characteristic fragmentation ion for a

FIG 1 Fatty acid rescue of the P. aeruginosa �fabY mutant. (A) Growth curvesin LB medium alone (P. aeruginosa PAO1 [�] and �fabY mutant [Œ]) or in LBmedium supplemented with decanoate (1 �g/ml [Œ], 10 �g/ml [�], or 100�g/ml [o]) for the �fabY mutant. Growth was measured at 37°C by recordingthe optical density at 600 nm. (B) The �fabY strain was streaked onto LB agarsupplemented with the indicated fatty acid (C2 to C12) at 100 �g/ml. The plateswere incubated at 37°C for 16 h before imaging.

FIG 2 Fatty acid composition analysis of P. aeruginosa. (A) Fatty acid methylesters (FAMEs) prepared from strains grown overnight on LB agar supple-mented with either 100 �g/ml of decanoate (C10:0) or perdeuterated decanoate(d19-C10:0) were analyzed by gas chromatography with flame ionization detec-tion. FAME peaks that contain deuterated fatty acids analyzed by mass spec-trometry in Fig. 3 are indicated with an asterisk. Wt, wild type.

Yuan et al.


http://jb.asm.org

FIG 3 Mass spectra of deuterated FAMEs. Unique FAME peaks appearing after supplementation with perdeuterated decanoate (indicated by asterisks in Fig. 2)were analyzed by mass spectrometry. The parent molecular ions ([M]�) and select fragment ions are indicated, including for the McLafferty ion (m/z 74). Thecorresponding unlabeled FAME mass spectra are included for comparison in the supplemental information (see Fig. S1 in the supplemental material).



http://jb.asm.org

methyl ester with an �-methylene group, remained at 74 Da. TheC-2 carbon was thus unsubstituted. The C10:0 3-OH (3-hydroxydecanoate) spectra contained the diagnostic carboxyl terminusion resulting from fragmentation between C-3 and C-4 (Fig. 3A).The resulting mass likewise indicated that C-3 was unsubstituted(103 Da). By inference, the terminal 7 carbons [CD3(CD2)6-; 15deuteriums total] were assigned as being fully deuterated. Theunsaturated C16:1�7c and C18:1�7c FAMEs were only 14 Da heavier(Fig. 3E and G), as would result from extraction of a single deute-rium at C-3 during FabA-catalyzed isomerization of trans-2-de-cenoyl-ACP to cis-3-decenoyl-ACP during anaerobic unsaturatedfatty acid (UFA) biosynthesis (33). Similar analysis of the otherspectra supports the conclusion that the 7 terminal carbon atomsof each fatty acid were fully substituted with deuterium. Since thecultures had been fed perdeuterated C10 fatty acid, deuterium la-bels were uniformly lost at C-2 and C-3. The data are consistentwith a fatty acid metabolism model whereby P. aeruginosa de-grades C10 via one round of the �-oxidation cycle to make C8-CoA. This intermediate is then shunted into FAS, at which pointthe C10 fatty acid is rebuilt by using unsubstituted malonyl-ACP asthe substrate. The terminally labeled �-keto-decanoyl-ACP estercan then be used to supply all the different types of fatty acidsneeded by the cell, including saturated (SFA), unsaturated (UFA),and 3-hydroxyl fatty acids.

PA3286 is required for de novo FAS initiation in the absenceof fabY and for fatty acid shunting. On the basis of the pattern oflabeled fatty acid incorporation (Fig. 3), we hypothesized a fattyacid shunt consisting of a single enzyme that condenses C8-CoAwith malonyl-ACP to make �-keto-decanoyl-ACP. This hypo-thetical catalytic activity is reminiscent of FabY/FabH, which con-dense acetyl-CoA with malonyl-ACP; the only difference beingthe acyl chain length of the CoA thioester acceptor. Since fattyacids were shunted in the absence of fabY and recombinant FabYdisplays an absolute substrate preference for short-chain acyl-CoA acceptors (56), we focused our search among the previouslyidentified four KASIII domain-containing proteins sharing simi-larity to FabH of E. coli (56). If one of the KASIII domain-contain-ing proteins were responsible for fatty acid shunting, then C10

supplementation should not rescue growth in a �fabY �KASIIIgenetic background. Allelic exchange plasmids were designed toreplace FabY with either a gentamicin resistance (Gmr) markeralone or in tandem with the functional ortholog fabHEc from E.coli to serve as a positive control (Fig. 4A). The fabY targetingplasmids were integrated into the chromosomes in a panel ofstrains with single KASIII domain-containing protein deletions(PA3333, PA0998, PA0999, and PA3286) as well as in a strainharboring all four KASIII deletions (TMT16). Merodiploid P.aeruginosa intermediates were verified by flanking PCR analysisand passively outgrown before plating on LB-sucrose agar tocounterselect unresolved clones. Aliquots were plated in parallelon plates with Gm to select for colonies in which the fabY allelehad been replaced, as well as on plates with Gm and C10 fatty acidsupplement. The fabY::Gmr deletion could be established withefficiencies comparable to the fabHEc exchange allele in the wild-type background (Fig. 4B), although as expected (56), the fabYdeletion (Gm selection) strain grew slowly in comparison to thefabHEc exchange control without C10 supplementation (Gm plusfatty acid [FA] selection). Conversely, the fabY deletion could notbe established in the TMT16 genetic background even with C10

supplementation, although the fabHEc allele could readily be es-

FIG 4 Synthetic lethal analysis for fabY with KASIII domain fabH or-thologs and fatty acid rescue. (A) Vectors conferring aacC1-mediated gen-tamicin (Gm) resistance that targeted the P. aeruginosa fabY gene weredesigned to either exchange fabH of E. coli (pTMT123 fabH�) or to deletefabY (pTMT124). Merodiploid intermediates (after step 1) were confirmedby PCR analysis and passively resolved by outgrowth in LB medium (step2). Aliquots were plated on LB-sucrose agar to counterselect unresolvedclones and to enumerate the sum of A- and B-type recombination events.In parallel, aliquots were plated on LB-sucrose agar plus Gm and on LB-sucrose agar plus Gm with 100 �g/ml of the fatty acid supplement decano-ate (FA) to select only the recombinants arising through A-type recombi-nation. (B) The pTMT123/124 vectors were individually introduced intothe wild-type (Wt) P. aeruginosa, TMT16 (�PA0998 �PA0999 �PA3333�PA3286), �PA3286, and TMT44 (�PA3286 attB::PA3286�). CFU result-ing from resolving either pTMT123 (black bars) or pTMT124 (hatchedbars) were determined by counting colonies that appeared after 24 h or 48h (#) of incubation at 37°C. The selection media and strain background areindicated below the x axis. Data are representative of 3 separate experi-ments. The results for the �PA0998, �PA0999, and �PA3333 mutantstrains are shown in Fig. S2 in the supplemental material. (C) Gas chro-matograms with FID detection of FAME extracts obtained from P. aerugi-nosa strains grown on LB agar with either 100 �g/ml of decanoate (C10:0) orperdeuterated decanoate (d19-C10:0). Asterisks indicate FAME peaksunique to d19-C10:0-fed samples with structures assigned as in Fig. 3.

Yuan et al.


http://jb.asm.org

tablished. The fabY deletion is thus synthetic lethal when intro-duced into the KASIII domain knockout strain TMT16. We thenrepeated this experiment in each of the single deletion strains todecipher whether a lone KASIII domain gene was the syntheticlethal partner of fabY or if the relationship was more complex. Themerodiploid �PA3286 deletion strain resolved with efficienciesthat closely mirrored that of TMT16 (Fig. 4B), whereas the otherthree single deletion strains (�PA3333, �PA0998, and �PA0999)behaved akin to the wild-type P. aeruginosa genetic background(see Fig. S2 in the supplemental material). Reintroduction of afunctional copy of the PA3286 gene into the chromosome of the�PA3286 mutant at the phage CTX attB site (TMT44) allowed thefabY deletion to now be established, indicating no confoundingpolarity issues and confirming a true synthetic lethal pairing be-tween fabY and PA3286.

The synthetic lethal relationship between PA3286 and fabY in-dicates shared acetyl-CoA:malonyl-ACP condensing activity. Un-like the P. aeruginosa �fabY strain, the �PA3286 deletion strain isaphenotypic with respect to growth (56). This suggests that thepresumptive FAS initiating activity of PA3286 that allows �fabY togrow, albeit slowly, is secondary to another main cellular role uti-lizing similar substrates. To determine whether PA3286 was in-deed the hypothesized C8-CoA:malonyl-ACP condensing enzymeresponsible for fatty acid shunting, we fed perdeuterated C10 to the�PA3286 mutant. Whereas perdeuterated C10 fatty acids notice-ably had a negative impact on the growth of the wild type and inparticular the �fabY mutant on LB agar, no inhibitory isotopeeffect was observed for the �PA3286 strain (data not shown).Since the primary isotope effect arising through breaking of C-Dbonds in shunted fatty acids to make UFA by the anaerobic FASpathway was likely responsible, this hinted at a lack of shunting inthe �PA3286 mutant. We next analyzed the FAME profile byGC-MS as had been done previously (Fig. 4C). The GC trace forthe �PA3286 mutant fed perdeuterated C10 showed no en blocincorporation of deuterated fatty acids and was nearly superim-posable with traces derived from samples that had been grownwith unlabeled C10. In the complemented PA3286 strain TMT44,deuterated FAME peaks having retention times and mass spectraidentical to those of deuterated FAMEs assigned for the wild typereappeared (Fig. 2). The data support the identification of thePA3286 enzyme as the enzyme responsible for shunting fatty acidsin vivo, as well as implicating PA3286 secondary enzymatic activityas the source of FAS initiation in the �fabY background.

PA3286 specifically shunts C8-CoA in vivo. While decanoateis likely intercepted as a C8-CoA �-oxidation cycle intermediateby PA3286 (Fig. 3 and 4C), we wanted to determine whether thiswas specific to C10 or for all fatty acids in general. Perdeuteratedlong-chain fatty acids (C14 and C16) were therefore fed to the wild-type and �PA3286 P. aeruginosa strains and analyzed by GC-MS(Fig. 5A). FAME peaks with retention times and mass spectraconsistent with the terminal 7 carbon atoms being fully substi-tuted with deuterium atoms were once again observed only in thewild-type strain upon either C14 or C16 supplementation. An ad-ditional FAME peak was present in the C16-fed samples for boththe wild-type and �PA3286 strains (marked by Œ). Analysis of themass spectrum identified the fatty acid as perdeuterated d31-C16

(Fig. 5B), based on the parent molecular ion mass (m/z 301) anda fully deuterated McLafferty ion [CD2 C(OD)�-OCH3; m/z 77]. Long-chain fatty acid-CoA esters can be recycled en bloc andincorporated into de novo phospholipids by the glycerol-phos-

phate and acylglycerol-phosphate acyltransferases PlsB/PlsC in P.aeruginosa (60). Under these growth conditions, the PA3286shunt clearly occurs in tandem with the direct en bloc PlsB/PlsC-catalyzed incorporation into phospholipids. Collectively, the datasuggest that PA3286 has high in vivo substrate specificity for C8-CoA intermediates and that the shunt is relevant during the catab-olism of both long- and medium-chain exogenous fatty acids.

PA3286 cross complements fabH and confers a fatty acidshunt in E. coli. Fatty acid metabolism in especially complex inpseudomonads, with an unusually large genetic allocation to boththe catabolism and anabolism of fatty acids (32, 50, 51). To furtherconfirm the two activities of PA3286 observed in P. aeruginosaPAO1, to probe the substrate specificity and to determine whetherPA3286 requires partner proteins for intercepting �-oxidation cy-cle intermediates, we introduced the PA3286 gene into the modelE. coli strain BW25113. In the presence of a high-copy-number

FIG 5 Fatty acid composition analysis of P. aeruginosa strains fed long-chainperdeuterated fatty acids. (A) Gas chromatograms with FID detection ofFAME extracts obtained from stationary-phase cultures of P. aeruginosagrown in liquid LB-BSA medium supplemented with either 100 �g/ml ofperdeuterated tetradecanoate (d27-C14:0) or hexadecanoate (d31-C16:0). Aster-isks indicate deuterated FAMEs of identical structure to those already assignedin Fig. 3 based on time of elution and mass spectra analysis. FAME peaksunique to d31-C16:0-fed cultures are indicated (Œ). (B) Mass spectra analysis ofthe d31-C16:0-specific FAME with the parent molecular ion ([M]�) and diag-nostic fragment ions labeled, including for the uniformly deuterium-labeledMcLafferty ion (m/z 77).



http://jb.asm.org

plasmid harboring PA3286, we were able to delete the otherwiseessential fabH gene that encodes the FAS initiating enzyme �-ace-toacetyl synthase III in enterobacteria. Unlike the fabY gene thatachieved full restoration of the growth rate (56), the PA3286 geneonly partially complemented fabH and indicates weak intrinsic�-acetoacetyl-ACP synthase catalytic activity (Fig. 6A). This isconsistent with our inability to identify PA3286 among a P. aerugi-nosa genomic cosmid library, which is introduced at low copynumber without any strong endogenous E. coli promoters (56).

We next fed the recombinant E. coli strain TMT47 [fabH::catR(pET-PA3286)] a mixture of perdeuterated C10 and unla-beled C16 fatty acid (10:1 [wt/wt]) in order to induce the fad genesinvolved in �-oxidation (Fig. 6B). The wild-type E. coli strain didnot incorporate the deuterium label for any of the peaks, indicat-ing that a medium-chain fatty acid shunt in order to extend exog-enous fatty acids is not native to E. coli, in agreement with previousreports (47). Likewise, replacement of fabH with fabY of P. aerugi-nosa did not result in fatty acid labeling. Only when fabH wasreplaced with PA3286 did deuterium-substituted FAME peaks be-come evident. Structure assignment based on the mass spectra of

all fatty acids common to P. aeruginosa and E. coli (labeled withasterisks; spectra similar to those in Fig. 3), as well as those specificto E. coli (see Fig. S3 in the supplemental material), were entirelyconsistent with complete labeling of the terminal 7 carbons.Hence, PA3286 is intrinsically specific for C8-CoA and can inter-cept �-oxidation intermediates without the need for a partnerprotein. The second aspect specific to the FAME profile of TMT47is the near absence of odd-numbered carbon fatty acids (indicatedby Œ in Fig. 6B) and of terminally labeled d7-C16:0 (indicated by �in Fig. 6B; mass spectrum is included in Fig. S3E in the supple-mental material). In E. coli, odd-numbered fatty acids arisethough utilization of propionyl primers as alternative substratesto acetyl-CoA for FAS by FabH (23, 27). The low-abundance d7-C16:0 peak that is present only in perdeuterated C10-fed E. coliwild-type or TMY32 [fabH::catR(pET-FabY)] may arise throughutilization of d7-butyryl-CoA produced by the �-oxidation cycleas a FAS primer by either FabH or FabY, respectively. In line withthe previously observed in vivo substrate specificity in P. aerugi-nosa, PA3286 did not efficiently utilize either short-chain acyl-CoA primer in E. coli. These short-chain acyl-CoA substrates are

FIG 6 Complementation of E. coli fabH by the P. aeruginosa PA3286 gene. (A) The E. coli strain TMT47 [fabH::camR(pET-PA3286)] was grown at 37°C in LBmedium alone (Œ) or with 1 mM IPTG (�) and compared to the parent Wt strain (�). Growth was monitored by measuring the optical density at 600 nm. (B)The gas chromatogram-FID trace for FAME samples prepared from E. coli strains grown on LB agar with perdeuterated decanoate (d19-C10:0; 100 �g/ml) andpalmitate (C16:0; 10 �g/ml) for induction of fatty acid degradation genes. The parent wild-type E. coli BW25113 and TMY32 [fabH::camR(pET-PA5174)] strainswere included as controls. Asterisks indicate deuterated FAMEs of identical structures to those already assigned in Fig. 3 based on time of elution and mass spectraanalysis. Mass spectra and assignment of deuterated FAME peaks unique to E. coli expressing PA3286 (C14:0, C14:0 3-OH, C17:0 cyclo �7c, and C19:0 cyclo �7c)along with the terminally labeled d7-C16:0 (�) peak present only in the absence of PA3286 are shown in Fig. S3 in the supplemental material. Odd-numbered acylchain FAMEs with reduced abundance in strain TMT47 are indicated (Œ).

Yuan et al.


http://jb.asm.org

present at much lower intracellular concentrations in comparisonto the glycolytic product/tricarboxylic acid cycle (TCA) interme-diate acetyl-CoA. Thus, while PA3286 can condense enough of theabundant acetyl-CoA substrate to maintain FAS initiation in theabsence of fabH, other low-affinity short-chain acyl-CoA sub-strates present in trace amounts are not appreciably utilized.

Characterization of recombinant PA3286 �-ketoacyl ACPsynthase activity. The PA3286 open reading frame is predicted toencode a 350-amino-acid protein (http://www.ncbi.nlm.nih.gov/). However, repeated attempts to express recombinantPA3286 failed to produce soluble protein. We thus chose an up-stream start codon and cloned the gene (encoding a 373-amino-acid protein) into a pET-24b(�) expression vector so as to appenda C-terminal His tag for purification. The longer protein was read-ily soluble and amenable to characterization. The substrate spec-ificity of the PA3286 enzyme was explored using various straight-chain saturated acyl-CoA (C2 to C16) as acceptors along withmalonyl-ACP, as has previously been described for FabY (56).Reaction products were separated by conformation-sensitiveurea-PAGE (46), and the gels were stained with Coomassie blue.PA3286 only weakly accepted short-chain acyl-CoA substrates,including acetyl-CoA, butyryl-CoA, and hexanoyl-CoA (Fig. 7).The substrate specificity suggests why shorter exogenous fatty ac-ids did not rescue growth in the P. aeruginosa �fabY mutant (Fig.1B), even though they can be utilized as the sole carbon source(31). The two best substrates were clearly C8-CoA and C10-CoA.When the experiment was repeated with C8-CoA and C10-CoAacceptors using radiolabeled [2-14C]malonyl-ACP, radioactivitywas incorporated into the faster-migrating bands (data notshown). This confirmed PA3286 is a malonyl-ACP condensingKAS as opposed to an acyl-CoA:ACP transacylase. CoA thioesteracceptors with longer acyl chains (C12, C14, and C16) were all ex-cluded as substrates. The in vitro substrate specificity is consistentwith the in vivo characteristics of PA3286, and together, theseresults support our assignment of PA3286 as a �-keto-decanoyl-ACP synthase that condenses malonyl-ACP with C8-CoA origi-nating from the fatty acid �-oxidation degradation cycle.

DISCUSSION

While it is clear that FAS in P. aeruginosa is a central metabolicpathway upon which the expression of multiple virulence andregulatory factors depend, exogenous C10 fatty acid noticeablymuted the growth-defective phenotype in the �fabY strain(Fig. 1). It might be expected, therefore, that the efficacy of asmall-molecule inhibitor of FabY would likewise be subject to thefatty acid content of the environment where it is administered.This is a particularly pressing concern with any intended anti-

pseudomonal agent, as an important segment of the intended pa-tient population has cystic fibrosis. P. aeruginosa can be found inthe respiratory tracts of 80% of all cystic fibrosis patients by thetime they are 18 years old (20), an environment where there is analready abundant lipid nutrient source in the form of pulmonarysurfactant. Pulmonary surfactant, being composed of 90% lipids(mostly dipalmitoyl phophatidylcholine) along with 10% surfac-tant proteins (1, 19), is an important in vivo nutrient source for P.aeruginosa (31, 48). Transcription of fad �-oxidation cycle genes isinduced in vivo, and mutants deficient in fatty acid degradationexhibited decreased fitness in a mouse lung infection model (32).Our goal in the present work was to address how the �fabY strainis rescued by exogenous fatty acids and whether the mechanism(s)might potentially compromise an inhibitor targeting FabY in P.aeruginosa.

Based on the perdeuterated feeding experiments with P.aeruginosa and the defined isogenic mutants (Fig. 2 to 5), thesynthetic lethal relationship between PA3286 and fabY (Fig. 4B),the PA3286 cross complementation of fabH activity and fatty acidshunting in E. coli (Fig. 6), and in vitro characterization of recom-binant PA3286 (Fig. 7), we propose that fatty acids rescue the P.aeruginosa �fabY mutant through incorporation via the PA3286-mediated �-oxidation cycle to the FAS shunt (Fig. 8). In this path-way, C8-CoA intermediates originating from the oxidation of ex-ogenous fatty acids C8 and longer are intercepted by PA3286 andcondensed with malonyl-ACP to make the FAS intermediate�-keto-decanoyl-ACP. PA3286 belongs to the �-ketoacyl-acylcarrier protein synthase III family (KASIII), with E. coli FabHbeing the prototypical member (52). As in E. coli FabH and mostother characterized orthologs (18), PA3286 contains the class-defining Cys-His-Asn catalytic triad common to KASIII enzymes(56), and so the catalytic function of condensing malonyl-ACPwith an acyl-CoA is not altogether unexpected. However, KASIIIdomain-containing enzymes generally exhibit high preference forshort-chain acyl-CoA substrates (18). The use of longer-chainacyl-CoA primers by KASIII domain synthases does have prece-dent, as the Mycobacterium tuberculosis FabH prefers long-chainacyl-CoA substrates in vitro and in vivo during the synthesis ofmycolic acids (5, 8). With an overall sequence identity between E.coli FabH and PA3286 of 27%, there is significant difference be-tween the two proteins that may contribute to its substrate speci-ficity. Structural information of PA3286 is needed to shed morelight on how the KASIII fold has been co-opted by PA3286 intoaccommodating longer acyl chains and whether there are com-mon themes with FabH from M. tuberculosis.

The �-keto-decanoyl ACP thioester shunt product generatedby PA3286 is a versatile precursor that is uncommitted to thebiosynthesis of any specific essential fatty acid (Fig. 8). Pathwaysfor the biosynthesis of LPS (3-OH decanoyl-ACP/3-OH dode-canoyl-ACP), UFA (via trans-2-decenoyl-ACP), and SFA can allutilize nutrient-acquired fatty acids for extension without havingto degrade them down to C2-CoA before reincorporation via denovo FAS. This is critical, as FAS is bioenergetically the most costlysynthetic process for any membrane component (59); for in-stance, building an n-acyl-ACP thioester (where n � C10) fromC2-CoA consumes [(n/2) 1] ATP molecules along with [(n/2) 1] � 2 reducing equivalents. In comparison, degradation of n-acyl-CoA �-oxidation intermediates down to C2-CoA before re-assembly into n-acyl-ACP thioesters by FAS consumes a net[(n/2) 1] ATP molecules. However, the PA3286 C8 shunt uses

FIG 7 Acyl-CoA substrate specificity of recombinant PA3286. Reaction prod-ucts using malonyl-ACP (mACP) and saturated straight-chain acyl-CoAs (C2

to C16) as potential PA3286 substrates were separated with conformation-sensitive urea-PAGE. The gels were stained with Coomassie blue dye. Theposition of the malonyl-ACP-only control (malonyl-ACP) is shown to theright of the gel. � AGI, plus E. coli FabAGI coupling enzymes.



http://www.ncbi.nlm.nih.gov/

http://www.ncbi.nlm.nih.gov/

http://jb.asm.org

only [(n/2) 4] ATP molecules to accomplish the same feat,yielding a further net savings of 3 ATP molecules per metabolizedn-acyl fatty acid molecule (Fig. 8). The use of a C8 fatty acid chainlength for shunting by P. aeruginosa is a metabolically savvychoice, maximizing energy conservation while retaining fatty acidanabolic versatility. P. aeruginosa now joins certain Vibrio specieswhich are also capable of extending exogenous fatty acids en blocusing the FAS pathway (7, 21, 28). In the case of Vibrio, exogenousmedium-chain fatty acids are directly ligated to free ACP to formacyl-ACP thioesters. While mechanistically distinct and indepen-dent of the �-oxidation pathway, the end result is equivalent to thePA3286 shunt in that the acyl-ACP produced can be utilized forincorporation into LPS, SFA, and UFA (29).

It has been reported that the entire FAS can be bypassed byexogenous fatty acid uptake from serum in certain Gram-positivepathogens, including Streptococcus (4). The essentiality of the FASpathway has not been questioned in Gram-negative bacteria, inpart due to the dependence of LPS biogenesis on FAS supplied3-hydroxyacyl-ACP precursors (41). The LPS of P. aeruginosacontains both 3-hydroxydecanoyl and 3-hydroxydodecanoyl acyl

chains (35). Although the PA3286 shunt can supply 3-hydroxyde-canoyl-ACP (Fig. 8), it should be noted that FabG is still needed tointroduce the �-hydroxyl group. Further, even if a C16/C18 fattyacid source were available for direct incorporation into UFA/SFAmembrane phospholipids by the combined activities of the acyl-glycerol-phosphate acyltransferases PlsB/PlsC and the aerobic de-saturases DesB/DesC/DesA (60), at least one more turn from theFAS cycle enzymes would still be necessary to form 3-hydroxy-dodecanoyl-ACP. Hence, it seems unlikely that PA3286 wouldimpact the efficacy of FAS inhibitors beyond FabY. In the absenceof having access to a FabY-specific small-molecule inhibitor,whether FabY can be entirely bypassed by uptake of exogenousfatty acids still must remain an open question. However, if oneassumes that treatment with a FabY inhibitor will phenocopy the�fabY deletion strain, then it certainly seems likely that a FabYinhibitor could be compromised by exogenous fatty acids. Morethan 95% of the total extracted fatty acids were terminally labeledwith deuterium in the �fabY sample (Fig. 2). This suggests that theshunt can satisfy the majority of the cellular needs for acyl-ACPprimers, in effect making �-acetoacetyl synthase (FabY/FabH) ac-

FIG 8 The proposed PA3286-mediated fatty acid �-oxidation to synthesis shunt of P. aeruginosa PAO1. After facilitated diffusion across the outer membranethrough FadL (not shown), fatty acids are trapped in the cytoplasm by FadD1/FadD2-catalyzed vectorial esterification with CoA (32). Further metabolismdepends on acyl-CoA ester chain length. Long-chain fatty acid CoA esters (C16/C18-CoA) can either be recycled en bloc and incorporated into de novophospholipids by the glycerol-phosphate and acylglycerol-phosphate acyltransferases PlsB/PlsC (not shown) (60), or as with medium-chain acyl CoA esters (C14

to C10-CoA), be further degraded by the �-oxidation pathway (in blue). Once the preferred C8-CoA substrate chain length is reached, the CoA thioester isintercepted by PA3286 and condensed with malonyl-ACP to form the key uncommitted fatty acid intermediate �-keto-decanoyl-ACP (green). The �-keto-decanoyl-ACP metabolite can be utilized by the anaerobic unsaturated fatty acid (UFA) pathway, the saturated fatty acid (SFA) pathway, and in lipopolysac-charide (LPS) biosynthesis (C10:0/C12:0 3-OH). The terminal 7 carbons in �-keto-decanoyl-ACP that remain labeled with deuterium when fed perdeuterated fattyacids are underlined. PA3286 does not significantly contribute to de novo FAS biosynthesis (in red) (56) but becomes essential in the P. aeruginosa �fabYbackground (Fig. 4B) due to a cellular requirement for basal FabH-type acetyl-CoA:malonyl-ACP condensation activity. For clarity, not all putative FAS and�-oxidation isozymes are shown. TCA, tricarboxylic acid cycle.

Yuan et al.


http://jb.asm.org

tivity superfluous. Indeed, there may even be intrinsic cellularregulatory mechanisms to repress de novo FAS initiation by FabY,as synthesis and degradation in most bacteria are tightly regulatedin order to maintain homeostasis and to avoid a futile metaboliccycle (17, 58). We are currently studying these putative regulatorymechanisms in order to understand how the PA3286 shunt isintegrated into the global fatty acid metabolism network of P.aeruginosa.

ACKNOWLEDGMENTS

We thank Herbert P. Schweizer (Colorado State University) for thepMini-CTX1 plasmid and Charles Dean (Novartis) for the pRC9 plasmid.We also thank Karen Dohrman and Gary Jackoway of Microbial ID, Inc.(Newark, DE) for expert technical assistance in FAME analysis.

REFERENCES1. Bernhard W, et al. 1997. Lung surfactant in a cystic fibrosis animal model:

increased alveolar phospholipid pool size without altered compositionand surface tension function in cftrm1HGU/m1HGU mice. Thorax 52:723–730.

2. Black PN, DiRusso CC. 1994. Molecular and biochemical analyses offatty acid transport, metabolism, and gene regulation in Escherichia coli.Biochim. Biophys. Acta 1210:123–145.

3. Breidenstein EB, Fuente-Nunez C, Hancock RE. 2011. Pseudomonasaeruginosa: all roads lead to resistance. Trends Microbiol. 19:419 – 426.

4. Brinster S, et al. 2009. Type II fatty acid synthesis is not a suitable anti-biotic target for Gram-positive pathogens. Nature 458:83– 86.

5. Brown AK, et al. 2005. Probing the mechanism of the Mycobacteriumtuberculosis beta-ketoacyl-acyl carrier protein synthase III mtFabH: fac-tors influencing catalysis and substrate specificity. J. Biol. Chem. 280:32539 –32547.

6. Brugha RE, Davies JC. 2011. Pseudomonas aeruginosa in cystic fibrosis:pathogenesis and new treatments. Br. J. Hosp. Med. (Lond.) 72:614 – 619.

7. Byers DM. 1989. Elongation of exogenous fatty acids by the biolumines-cent bacterium Vibrio harveyi. J. Bacteriol. 171:59 – 64.

8. Choi KH, Kremer L, Besra GS, Rock CO. 2000. Identification andsubstrate specificity of beta-ketoacyl (acyl carrier protein) synthase III(mtFabH) from Mycobacterium tuberculosis. J. Biol. Chem. 275:28201–28207.

9. Choi KH, Schweizer HP. 2005. An improved method for rapid genera-tion of unmarked Pseudomonas aeruginosa deletion mutants. BMC Mi-crobiol. 5:30. doi:10.1186/1471-2180-5-30.

10. Clark DP, Cronan JE. 1996. Two-carbon compounds and fatty acids ascarbon sources, p 343–358. In Neidhardt FC, et al (ed), Escherichia coli andSalmonella: cellular and molecular biology, 2nd ed, vol 1. American Soci-ety for Microbiology, Washington, DC.

11. Cronan JE, Lin S. 2011. Synthesis of the alpha, omega-dicarboxylic acidprecursor of biotin by the canonical fatty acid biosynthetic pathway. Curr.Opin. Chem. Biol. 15:407– 413.

12. Cronan JE, Rock CO. 1996. Biosynthesis of membrane lipids, p 612– 636.In Neidhardt FC, et al (ed), Escherichia coli and Salmonella: cellular andmolecular biology, 2nd ed, vol 1. American Society for Microbiology,Washington, DC.

13. Datsenko KA, Wanner BL. 2000. One-step inactivation of chromosomalgenes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci.U. S. A. 97:6640 – 6645.

14. Davies DG, Marques CN. 2009. A fatty acid messenger is responsible forinducing dispersion in microbial biofilms. J. Bacteriol. 191:1393–1403.

15. Driscoll JA, Brody SL, Kollef MH. 2007. The epidemiology, pathogenesisand treatment of Pseudomonas aeruginosa infections. Drugs 67:351–368.

16. Figurski DH, Helinski DR. 1979. Replication of an origin-containingderivative of plasmid RK2 dependent on a plasmid function provided intrans. Proc. Natl. Acad. Sci. U. S. A. 76:1648 –1652.

17. Fujita Y, Matsuoka H, Hirooka K. 2007. Regulation of fatty acid metab-olism in bacteria. Mol. Microbiol. 66:829 – 839.


19. Galabert C, Jacquot J, Zahm JM, Puchelle E. 1987. Relationships be-

tween the lipid content and the rheological properties of airway secretionsin cystic fibrosis. Clin. Chim. Acta 164:139 –149.

20. Geller DE. 2009. Aerosol antibiotics in cystic fibrosis. Respir. Care 54:658 – 670.

21. Giles DK, Hankins JV, Guan Z, Trent MS. 2011. Remodelling of theVibrio cholerae membrane by incorporation of exogenous fatty acids fromhost and aquatic environments. Mol. Microbiol. 79:716 –728.

22. Hannauer M, et al. 2012. Biosynthesis of the pyoverdine siderophore ofPseudomonas aeruginosa involves precursors with a myristic or a myristo-leic acid chain. FEBS Lett. 586:96 –101.

23. Heath RJ, Rock CO. 1996. Inhibition of beta-ketoacyl-acyl carrier proteinsynthase III (FabH) by acyl-acyl carrier protein in Escherichia coli. J. Biol.Chem. 271:10996 –11000.

24. Hidron AI, et al. 2008. NHSN annual update: antimicrobial-resistantpathogens associated with healthcare-associated infections: annual sum-mary of data reported to the National Healthcare Safety Network at theCenters for Disease Control and Prevention, 2006 –2007. Infect. ControlHosp. Epidemiol. 29:996 –1011.

25. Hoang TT, Karkhoff-Schweizer RR, Kutchma AJ, Schweizer HP. 1998.A broad-host-range Flp-FRT recombination system for site-specific exci-sion of chromosomally-located DNA sequences: application for isolationof unmarked Pseudomonas aeruginosa mutants. Gene 212:77– 86.

26. Hoang TT, Kutchma AJ, Becher A, Schweizer HP. 2000. Integration-proficient plasmids for Pseudomonas aeruginosa: site-specific integra-tion and use for engineering of reporter and expression strains. Plas-mid 43:59 –72.

27. Ingram LO, Chevalier LS, Gabba EJ, Ley KD, Winters K. 1977. Propi-onate-induced synthesis of odd-chain-length fatty acids by Escherichiacoli. J. Bacteriol. 131:1023–1025.

28. Jiang Y, Chan CH, Cronan JE. 2006. The soluble acyl-acyl carrier proteinsynthetase of Vibrio harveyi B392 is a member of the medium chain acyl-CoA synthetase family. Biochemistry 45:10008 –10019.

29. Jiang Y, Morgan-Kiss RM, Campbell JW, Chan CH, Cronan JE. 2010.Expression of Vibrio harveyi acyl-ACP synthetase allows efficient entry ofexogenous fatty acids into the Escherichia coli fatty acid and lipid A syn-thetic pathways. Biochemistry 49:718 –726.

30. Jimenez PN, et al. 2012. The multiple signaling systems regulating viru-lence in Pseudomonas aeruginosa. Microbiol. Mol. Biol. Rev. 76:46 – 65.

31. Kang Y, Nguyen DT, Son MS, Hoang TT. 2008. The Pseudomonasaeruginosa PsrA responds to long-chain fatty acid signals to regulate thefadBA5 beta-oxidation operon. Microbiology 154:1584 –1598.

32. Kang Y, Zarzycki-Siek J, Walton CB, Norris MH, Hoang TT. 2010.Multiple FadD acyl-CoA synthetases contribute to differential fatty aciddegradation and virulence in Pseudomonas aeruginosa. PLoS One5:e13557. doi:10.1371/journal.pone.0013557.

33. Kass LR, Brock DJ, Bloch K. 1967. Beta-hydroxydecanoyl thioesterdehydrase. I. Purification and properties. J. Biol. Chem. 242:4418 – 4431.

34. Kerr KG, Snelling AM. 2009. Pseudomonas aeruginosa: a formidable andever-present adversary. J. Hosp. Infect. 73:338 –344.

35. King JD, Kocincova D, Westman EL, Lam JS. 2009. Lipopolysaccharidebiosynthesis in Pseudomonas aeruginosa. Innate Immun. 15:261–312.

36. Klein K, Steinberg R, Fiethen B, Overath P. 1971. Fatty acid degradationin Escherichia coli. An inducible system for the uptake of fatty acids andfurther characterization of old mutants. Eur. J. Biochem. 19:442– 450.

37. Lin S, Hanson RE, Cronan JE. 2010. Biotin synthesis begins by hijackingthe fatty acid synthetic pathway. Nat. Chem. Biol. 6:682– 688.

38. Lister PD, Wolter DJ, Hanson ND. 2009. Antibacterial-resistant Pseu-domonas aeruginosa: clinical impact and complex regulation of chromo-somally encoded resistance mechanisms. Clin. Microbiol. Rev. 22:582–610.

39. Moore NM, Flaws ML. 2011. Epidemiology and pathogenesis of Pseu-domonas aeruginosa infections. Clin. Lab Sci. 24:43– 46.

40. Okuda S, Tokuda H. 2011. Lipoprotein sorting in bacteria. Annu. Rev.Microbiol. 65:239 –259.

41. Parsons JB, Rock CO. 2011. Is bacterial fatty acid synthesis a valid targetfor antibacterial drug discovery? Curr. Opin. Microbiol. 14:544 –549.

42. Patton GM, Lowenstein JM. 1979. Measurements of fatty acid synthesisby incorporation of deuterium from deuterated water. Biochemistry 18:3186 –3188.

43. Raetz CR, Whitfield C. 2002. Lipopolysaccharide endotoxins. Annu. Rev.Biochem. 71:635–700.

44. Remans K, Vercammen K, Bodilis J, Cornelis P. 2010. Genome-wide



http://jb.asm.org

analysis and literature-based survey of lipoproteins in Pseudomonasaeruginosa. Microbiology 156:2597–2607.

45. Rietsch A, Mekalanos JJ. 2006. Metabolic regulation of type III secretiongene expression in Pseudomonas aeruginosa. Mol. Microbiol. 59:807– 820.

46. Rock CO, Cronan JE. 1981. Acyl carrier protein from Escherichia coli.Methods Enzymol. 7(Part C):341–351.

47. Silbert DF, Ruch F, Vagelos PR. 1968. Fatty acid replacements in a fattyacid auxotroph of Escherichia coli. J. Bacteriol. 95:1658 –1665.

48. Son MS, Matthews WJ, Jr, Kang Y, Nguyen DT, Hoang TT. 2007. Invivo evidence of Pseudomonas aeruginosa nutrient acquisition and patho-genesis in the lungs of cystic fibrosis patients. Infect. Immun. 75:5313–5324.

49. Spalding MD, Prigge ST. 2010. Lipoic acid metabolism in microbialpathogens. Microbiol. Mol. Biol. Rev. 74:200 –228.

50. Stanier RY, Palleroni NJ, Doudoroff M. 1966. The aerobic pseudomon-ads: a taxonomic study. J. Gen. Microbiol. 43:159 –271.

51. Stover CK, et al. 2000. Complete genome sequence of Pseudomonasaeruginosa PAO1, an opportunistic pathogen. Nature 406:959 –964.

52. Tsay JT, Oh W, Larson TJ, Jackowski S, Rock CO. 1992. Isolation andcharacterization of the beta-ketoacyl-acyl carrier protein synthase III gene(fabH) from Escherichia coli K-12. J. Biol. Chem. 267:6807– 6814.

53. Weeks G, Shapiro M, Burns RO, Wakil SJ. 1969. Control of fatty acid

metabolism. I. Induction of the enzymes of fatty acid oxidation in Esche-richia coli. J. Bacteriol. 97:827– 836.

54. Williams P, Camara M. 2009. Quorum sensing and environmental ad-aptation in Pseudomonas aeruginosa: a tale of regulatory networks andmultifunctional signal molecules. Curr. Opin. Microbiol. 12:182–191.

55. Wu DC, Chan WW, Metelitsa AI, Fiorillo L, Lin AN. 2011. Pseudomonasskin infection: clinical features, epidemiology, and management. Am. J.Clin. Dermatol. 12:157–169.

56. Yuan Y, Sachdeva M, Leeds JA, Meredith TC. 2012. Fatty acid biosyn-thesis in Pseudomonas aeruginosa is initiated by the FabY class of �-keto-acyl acyl carrier protein synthases. J. Bacteriol. 194:5171–5184.

57. Zhang YM, Frank MW, Zhu K, Mayasundari A, Rock CO. 2008. PqsDis responsible for the synthesis of 2,4-dihydroxyquinoline, an extracellularmetabolite produced by Pseudomonas aeruginosa. J. Biol. Chem. 283:28788 –28794.

58. Zhang YM, Rock CO. 2008. Membrane lipid homeostasis in bacteria.Nat. Rev. Microbiol. 6:222–233.

59. Zhang YM, Rock CO. 2009. Transcriptional regulation in bacterial mem-brane lipid synthesis. J. Lipid Res. 50(Suppl):S115–S119.

60. Zhu K, Choi KH, Schweizer HP, Rock CO, Zhang YM. 2006. Twoaerobic pathways for the formation of unsaturated fatty acids in Pseu-domonas aeruginosa. Mol. Microbiol. 60:260 –273.

Yuan et al.


http://jb.asm.org

fatty acid biosynthesis in pseudomonas aeruginosa is ... · fatty acid biosynthesis in pseudomonas...

Documents