FEBS Letters 583 (2009) 363–368

journal homepage: www.FEBSLetters .org

Proof of function of a putative 3-hydroxyacyl-acyl carrier protein dehydratasefrom higher plants by mass spectrometry of product formation

Adrian Brown *, Val Affleck, Johan Kroon, Antoni SlabasSchool of Biological and Biomedical Sciences, Science Laboratories, University of Durham, South Road, Durham DH1 3LE, UK

a r t i c l e i n f o a b s t r a c t

Article history:Received 3 November 2008Revised 9 December 2008Accepted 10 December 2008Available online 25 December 2008

Edited by Ulf-Ingo Flügge

Keywords:DehydrataseFatty acid synthaseMass spectrometryAcyl carrier protein

0014-5793/$34.00 � 2008 Federation of European Biodoi:10.1016/j.febslet.2008.12.022

Abbreviations: ACP, acyl carrier protein; FAS, fatspectrometry; CoA, coenzyme A; DH, 3-hydroxyacyl-Amass unit.

* Corresponding author. Fax: +44 191 3341201.E-mail address: a.p.brown@durham.ac.uk (A. Brow

The predicted mature portion of a putative 3-hydroxyacyl-ACP dehydratase (DH) from Arabidopsiswas linked to an N-terminal poly-histidine-tag and the fusion protein expressed in Escherichia coli.Soluble dehydratase was present on induction at 25 �C and pure dehydratase eluted from a nickel-affinity column in 0.2–0.5 M imidazole. High concentrations of imidazole were necessary to retainenzyme solubility. The dehydratase reaction is reversible and 3-hydroxybutyryl- and 2-butenoyl-ACP substrates were prepared from E. coli apo-ACP. Analysis of these suggested contamination ofapo-ACP with dehydratase and an additional reverse-phase chromatographic step was required dur-ing acyl carrier protein (ACP) preparation. Activity of purified dehydratase was demonstrated bymass spectrometry using 2-butenoyl-ACP, providing the first functional experimental evidence forplant DH gene sequences.� 2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction

Fatty acids are synthesised by a cyclical series of reactionsoccurring on acyl-substrates linked to acyl carrier protein (ACP).Each cycle consists of condensation, reduction, dehydration andfurther reduction steps, and two-carbon units are repetitivelyadded until the appropriate chain length is reached. The third stepof elongation, conversion of 3-hydroxyacyl-ACP to trans-2-enoyl-ACP by dehydration, is catalysed by 3-hydroxyacyl-ACP dehydra-tase (DH). Fatty acid synthase (FAS) enzymes are classified as typeI, composed of large (>200 kDa) multifunctional proteins, or type II,in which individual catalytic activities are on separate polypep-tides [1]. Type II FAS systems are found in most bacteria and plantplastids and the synthesis of a C18 unsaturated fatty acid from theinitial Coenzyme A (CoA) substrates, acetyl-CoA and malonyl-CoA,requires co-ordinated activity of eight enzymes [2,3]. We have aninterest in the organisation of FAS enzymes in plants and wantedto express and purify all Arabidopsis thaliana type II FAS compo-nents, including DH.

The model system for type II FAS enzyme expression and anal-ysis is Escherichia coli, for which the structure of all the enzymes

chemical Societies. Published by E

ty acid synthase; MS, massCP dehydratase; amu, atomic

n).

has been determined [4]. E. coli contains two DH enzymes, one ofwhich (FabA) is required for unsaturated fatty acid synthesis andis a dual dehydratase/isomerase enzyme, acting primarily on C10substrates [5]. The FabA gene was among the first FAS genes iso-lated [6], mutation of FabA causing an unsaturated fatty acid aux-otrophic phenotype. The other hydroxyacyl-ACP dehydratase inE. coli (FabZ) is the core enzyme required for fatty acid elongationfrom C4 to C18 and was identified later as the site of suppression ofa temperature-sensitive lipid A biosynthetic mutant [7]. The func-tion of FabA and FabZ enzymes from E. coli has been demonstratedby analysis of ACP-derivatives on native gels containing urea [5].

Plant DH enzymes have been purified from a number of species[8,9] and several putative genes have been identified in sequencedatabases but there is, to our knowledge, no definitive proof offunction of the gene product. It is becoming increasingly apparentthat such confirmation is important, as gene identification byhomology alone can give rise to erroneous conclusions. For exam-ple, identification of the first diacylglycerol acyltransferase se-quence followed analysis of expressed proteins with homology toacyl-CoA:cholesterol acyltransferase [10]. Whilst plant DH hasbeen purified to apparent homogeneity from spinach [9], no aminoacid sequence data is available, which would greatly strengthenputative database assignments.

We report here the expression, purification and assay of activityof an A. thaliana type II 3-hydroxyacyl-ACP dehydratase usingcrotonoyl (2-butenoyl)-ACP substrate. This was made from anacyl-CoA analogue and recombinant E. coli apo-ACP using

lsevier B.V. All rights reserved.

364 A. Brown et al. / FEBS Letters 583 (2009) 363–368

holo-ACP synthase (EC 2.7.8.7). Of particular note is the require-ment to further purify ACP to remove contaminants, which other-wise apparently equilibrate enzyme substrates during synthesisand prevent reaction analysis. DH activity was monitored by a di-rect assay using mass spectrometry (MS) that identifies both sub-strate and product and to our knowledge this is the first proof offunction of a putative DH from plants. The confirmation of plantDH amino acid sequence allows us to make preliminary conclu-sions on potential molecular interactions of three FAS components.

2. Materials and methods

2.1. Materials

High purity formic acid and acetonitrile for MS analysis werefrom Merck and Rathburn. Crotonoyl-CoA and 3-hydroxybutyryl-CoA were from Sigma and chemicals were reagent-grade or higher.

2.2. Expression and purification of A. thaliana 3-hydroxyacyl-ACPdehydratase

DNA coding for a 161 amino acid DH fragment was inserted inpET28a (Novagen). IPTG-mediated protein expression in E. coli wasat 25 �C for 4 h (See Supplementary data for further details) and DHpurification used a 1 ml HisTrapTM HP nickel-affinity column (GEHealthcare). Bacteria were lysed with lysozyme/DNAase treatmentin nickel-binding buffer A (20 mM NaH2PO4, 20 mM imidazole0.5 M NaCl pH 7.4) and the lysate centrifuged and filtered beforecolumn loading. The column was washed with 5 ml buffer A beforestep elution with 10 ml buffer A containing 0.2 M imidazole, fol-lowed by 2 � 5 ml buffer A containing 0.5 M imidazole pH 7.4.DH was in the 0.5 M imidazole fraction, and 16% (v/v) glycerolwas added before snap-freezing and storage at �20 �C.

2.3. Purification of E. coli ACP

Expressed E. coli apo-ACP was purified by anion-exchange chro-matography as described [11] and freeze-dried after dialysis intowater. ACP was re-suspended in buffer A (50 mM sodium phos-phate pH 7.2) plus 2 M (NH4)2SO4 and applied onto a butyl-sephar-

Fig. 1. Alignment of putative plant DH and E. coli FabZ sequences. Predicted full-length(NP_001155645) and E. coli FabZ (NP_414722) were aligned with Clustalw2 (http://wwidentical in all sequences and full stops indicate conserved substitutions. The black triang

ose column previously equilibrated with 1.7 M (NH4)2SO4 in bufferA. Protein was eluted with a gradient of 1.7–0 M (NH4)2SO4 in buf-fer A. ACP-fractions were dialysed into water and freeze-dried andACP re-suspensed in 20 mM Tris/HCl pH 6.8 for use.

2.4. Synthesis of acyl-ACP substrates

This was as described [11] but with 50 lM apo-ACP, 200 lMacyl-CoA and the synthesis of acylated ACP-derivatives was as-sessed by mobility shift on native gels containing urea [12]. Fol-lowing dialysis into water, the identity of ACP-derivatives wasanalysed by addition of 10% formic acid/50% acetonitrile and staticinfusion into a Q-star Pulsar I tandem mass spectrometer (AppliedBiosystems). Substrates were freeze-dried and re-suspended in20 mM Tris/HCl pH 6.8 before storage at �20 �C.

2.5. Assay of DH activity by electrospray-ionisation-MS

Assays contained 10 mM NaH2PO4, 0.25 M NaCl, 0.25 M imidaz-ole pH 7.4, dehydratase and 20 lM crotonoyl-ACP. Addition of DHcontaining imidazole and NaCl meant samples had to be de-saltedbefore MS analysis. For this, 8 ll samples were acidified with 2.7 ll10% formate and applied to a C4 Zip-tip� (Millipore) that had beentreated with 2 � 10 ll ACN and 2 � 10 ll 1% formate. After wash-ing in 3 � 10 ll 1% formate, ACPs were eluted in 6 ll 50% ACN, 1%formate for infusion into the Q-star mass spectrometer. The dataacquisition window was routinely 800–1600 m/z and spectra werede-convoluted using the Bayesian Protein Reconstruct Tool in Bio-Analyst� (Applied Biosytems/MDS-Sciex). Typically, mass graphsbetween 8 and 10000 atomic mass unit (amu) were calculated.

3. Results and discussion

3.1. Expression and purification of acyl-ACP dehydratase

A recent survey of the A. thaliana genome for genes associatedwith acyl-lipid metabolism [http://www.plantbiology.msu.edu/lipids/genesurvey/index.htm] found two sequences homologousto DHs – At2g22230 and At5g10160. The proteins contain 220and 218 amino acids and the predicted mature polypeptides are

open reading frames from A. thaliana (At2g22230), B. napus (AAK60545), O. sativaw.ebi.ac.uk/Tools/clustalw/) using default parameters. Open circles denote residuesle shows the N-terminal amino acid of A. thaliana dehydratase chosen for expression.

45

29

20.2

14

1 2 3 4 5

kDa

Fig. 2. Purification of histidine-tagged A. thaliana DH by nickel-affinity chroma-tography. Samples from purification were: (1) soluble cell lysate, (2) flow duringsample loading, (3) column wash (4) 0.2 M imidazole elution, and (5) 0.5 Mimidazole elution. Expressed DH in the cell lysate is indicated by dots and mobilityof molecular weight markers shown.

A. Brown et al. / FEBS Letters 583 (2009) 363–368 365

86% identical. At2g22230 was selected for expression because of abetter reliability score (0.925 vs. 0.872) for plastid-location usingTargetP [See http://lipids.plantbiology.msu.edu/Localization.htmfor description]. Alignment of putative plant DH sequences andE. coli FabZ (Fig. 1) revealed a conserved sequence likely to be re-quired for enzyme activity and a further 16 amino acid conservedregion towards the N-terminus of the three plant proteins. The var-iable N-terminal sequences in the plant proteins are probable plas-tid-import peptides and accordingly primers were designed forexpression of a 161 amino acid protein from glutamate 60 to theC-terminal serine 220 of the protein encoded by At2g22230.Amplified DNA was cloned into pET28a to enable expression inE. coli with an N-terminal histidine-tag.

When A. thaliana DH was expressed in E. coli at 30 �C or 37 �Cusing standard IPTG-mediated induction, induced protein was inthe insoluble (28000 � g pellet) fraction and to maintain solubility,expression was performed at 25 �C. Purification of DH was by nick-el-affinity chromatography, with DH eluting in buffer containing0.5 M imidazole (Fig. 2). The high concentration of imidazole re-quired for elution might reflect the fact that expressed DH is poly-meric and gel-filtration experiments have previously suggestedthat plant DH is a tetramer [9]. The identity of purified DH was

8936

8917

holo

Na+

K+

5.0e5

4.0e5

3.0e5

2.0e5

1.0e5

Inte

nsity

/ cp

s

8800 8850 8900 8950 9000 9050 9100 9150 9200

Mass / amu

0.0

[Crotonoyl-ACP + H]+

[3-OH-butyryl-ACP + H]+

Fig. 3. MS analysis of crotonoyl-ACP and 3-hydroxbutyryl-ACP substrates made with a staACP substrates synthesised from: (A) crotonoyl-CoA, and (B) 3-hydroxybutyryl-CoA.

confirmed by peptide mass fingerprinting of a tryptic digest froman SDS–PAGE gel slice. Supplementary Fig. S1 shows the expressedprotein sequence and tryptic peptides used for protein identifica-tion with two different ionisation methods.

Purified DH was in 20 mM NaH2PO4, 0.5 M NaCl and 0.5 M imid-azole pH 7.4 and attempts were made to reduce the ionic strengthby dialysis. It was possible to remove NaCl by dialysis and retainsoluble DH in 20 mM NaH2PO4, 0.5 M imidazole pH 7.4, but dialysisinto 0.5 M NaCl, 20 mM NaH2PO4 buffer caused DH precipitation.Dialysis into 0.5 M NaCl, 20 mM NaH2PO4 containing 250, 125and 50 mM imidazole resulted in, respectively 80%, 39% and 9% sol-uble DH remaining after centrifugation. Purified DH was stored at�20 �C in the 0.5 M imidazole elution buffer plus 16% glycerol,freezing without glycerol caused precipitation. It is unclear whyimidazole is required for DH solubility.

3.2. Synthesis of acyl-ACP substrates following additional apo-ACPpurification

Acyl-ACP substrates were required for proof of DH activity, asthe plant enzyme does not utilise acyl-CoA analogues [9], whichhave simplified recent analysis of DH enzymes from protozoanspecies [13,14]. Alternative methods for DH ACP substrate synthe-sis have been reported, including in situ preparation from satu-rated acyl-CoAs and appropriate purified FAS enzymes [5] andchemical synthesis of crotonoyl-ACP (for assay of the reverse DHreaction) with crotonic anhydride and reduced ACP [15]. MS anal-ysis showed crotonoyl-ACP made this way to be multiply-deriva-tised at unknown sites in addition to the active thiol [16]. Usingholo-ACP synthase (AcpS) to transfer acyl-phosphopantetheinefrom crotonoyl-CoA to apo-ACP resulted in purer substrate [16]and this method was chosen for the present study. The dehydra-tase reaction is freely reversible and can be assayed with 3-hydroxyacyl-ACP or with a 2-enoyl-ACP. Holo-ACP synthase isactive with a variety of acyl-CoA substrates but availability of thefour-carbon compounds 3-hydroxybutyryl- and crotonoyl-CoAmeant these were the acyl chain lengths chosen.

3-Hydroxybutyryl- and crotonoyl-ACP were synthesised usingE. coli apo-ACP purified by mono-Q chromatography as describedpreviously [11] and reaction products analysed by electrospray-ionisation MS. The duration of sample infusion alters acquiredspectra slightly but the reproducibility of ACP-derivative masseswas ±1 kDa. MS analysis of the initial ACP substrates led to the

8917

8935

Na+

holo

K+

2.0e5

1.6e5

1.2e5

8.0e4

4.0e4

8800 8850 8900 8950 9000 9050 9100 9150 9200

Mass / amu

0.0

Inte

nsity

/ cp

s

[Crotonoyl-ACP + H]+

[3-OH-butyryl-ACP + H]+

ndard apo-ACP preparation. Mass-graphs from 8800 to 9200 m/z of freshly prepared

366 A. Brown et al. / FEBS Letters 583 (2009) 363–368

conclusion that the synthesis reactions were probably contami-nated with DH. In a stored crotonoyl-ACP sample, the ratio ofcrotonoyl- to 3-hydroxybutyryl-ACP was found to be 1:3.1, closeto the measured equilibrium value of 1:2.85 for E. coli DH [17],and after further parallel syntheses the ratios were 1:3.2 and1:5.8 respectively in substrates made with crotonoyl- and3-hydroxybutyryl-CoA. The mass graphs of ACPs from these tworeactions were essentially identical (Fig. 3), with peakscorresponding to [crotonoyl-ACP+H]+ (8917 amu), [3-hydroxy-butyryl-ACP+H]+ (8936 or 8935 mass), alkali metal adducts ofhydroxybutyryl-ACP and holo-ACP identifiable in both samples.

As the acyl-CoA compounds used for synthesis were not appar-ently cross-contaminated when analysed by MS, an explanation forthis observation was contamination of ACP-synthesis mixtureswith DH. A likely source of this was purified E. coli apo-ACP, whichwas therefore further purified by reverse-phase chromatography

A 21

5nm

/ mA

U

600

500

400

300

200

100

0

100ACP

Load Wash Time / mi0 40 80

Fig. 4. Purification of E. coli ACP on butyl-sepharose following anion-exchange chromato(NH4)2SO4 and eluted with a gradient of 1.7–0 M (NH4)2SO4. Percentage buffer B containintogether with duration of sample loading and washing (dotted line) are shown. Fractionsconfirmed they contained ACP.

8917

Na+ (8939)

2.4e5

1.8e5

1.2e5

6.0e4

Inte

nsity

/ cp

s

8800 8850 8900 8950 9000 9050 9100 9150 9200

Mass / amu

0.0

[crotonoyl-ACP + H]+

Fig. 5. MS analysis of acyl-ACP substrates prepared with additionally purified E. coli ACP.either: (A) crotonoyl-CoA or (B) 3-hydroxybutyryl-CoA and apo-ACP purified with rever8965 m/z units.

after the initial anion-exchange step. ACP purification by decreas-ing ammonium sulphate concentration on a reverse-phase columnhas been reported [18] and high salt concentrations might be ex-pected to disrupt binding of any ACP contaminants. Samples ofE. coli apo-ACP were applied to a butyl-sepharose column (Fig. 4)and a substantial proportion of the ACP applied to the columnbound and was eluted in the gradient. Importantly a small A280

peak was observed during column loading and since ACP has alow absorbance at 280 nm [19] due to a lack of aromatic aminoacids, this peak presumably corresponded to unbound contami-nants. Highlighted ACP-fractions from this additional purificationstep were selected for subsequent substrate syntheses and reactionproducts analysed by MS.

In contrast to the analysis described above, the 3-hydroxybuty-ryl- and crotonoyl-ACP mass graphs showed clear differences(Fig. 5). In particular there was no peak corresponding to

% B

uffe

r B

0

20

40

60

80

100

nutes120 160

graphy. Chromatogram from a 7 ml butyl-sepharose column loaded with ACP in 2 Mg 0 M (NH4)2SO4 (dotted line), column eluant A280 (dashed line) and A215 (solid line)from the marked area were pooled for substrate synthesis after gel electrophoresis

8935

Na+ (8957)

8917

8917

8935 Na+

8800 8850 8900 8950 9000 9050 9100 9150 9200

Mass / amu

0.0

2.0e5

1.6e5

1.2e5

8.0e4

4.0e4

Inte

nsity

/ cp

s

8920 8940 8960

[3-OH-butyryl-ACP + H]+

Mass- graphs from 8800 to 9200 m/z obtained from ACP substrates synthesised withse-phase chromatography. The panel insert in (B) is an enlargement from 8911 to

8918

plus Na+

(8940)

3.0e6

2.0e6

1.0e6

0.0

Inte

nsity

/ cp

s

8800 8900 9000 9100 9200Mass / amu

8918

plus Na+

(8940)

3.0e6

2.0e6

1.0e6

0.0In

tens

ity /

cps

8900 8920 8940 8960

Mass / amu

A1 Crotonoyl-ACP plus buffer A2 Crotonoyl-ACP plus buffer B Crotonoyl-ACP plus DH

Inte

nsity

/ cp

s

8918

8936

3.0e6

2.0e6

1.0e6

0.0

plus Na+

(8958)

8900 8920 8940 8960

Mass / amu

Fig. 6. MS analysis of crotonoyl-ACP in an assay of purified DH. Regions of mass-graphs derived from assay samples containing crotonoyl-ACP with either: (A) DH elutionbuffer or (B) purified DH enzyme. A2 is an enlargement of part of the graph in A1 and reactions were incubated at 25 �C for 16 min before sampling for MS infusion.

A. Brown et al. / FEBS Letters 583 (2009) 363–368 367

3-hydroxybutyryl-ACP (mass 8935) in the crotonoyl-ACP prepara-tion (Fig. 5A). The [M+H]+ form of crotonoyl-ACP (mass 8917) is themost abundant ion in this sample and only the [M+Na]+ adduct ofthis (mass 8939) is present in the mass region 8930–8940 amu.Some crotonoyl-ACP was however present in the 3-hydroxybuty-ryl-ACP substrate (Fig. 5B), perhaps arising from contaminationin the acyl-CoA sample used for synthesis. The area of the massgraph peak for crotonoyl-ACP in this sample was 3–4% the sumof the [3-hydroxybutyryl-ACP+H]+ and [3-hydroxybutyryl-ACP+Na]+

peaks at 8935 and 8957 amu, respectively. These substrates didnot contain detectable holo-ACP (8848 m/z), probably because re-verse chromatography removed it from the apo-ACP used forsynthesis.

Application of apo-ACP to reverse-phase chromatography wasrequired for synthesis of the two substrates and the observedmasses are consistent with acidified ACP-derivatives. Holo-ACPhas a theoretical mass of 8847 kDa [19], which leads to derivedmasses of 8933 kDa and 8915 kDa for hydroxybutyryl- and croto-noyl-ACP. Measured values are +1 or +2 kDa from these and a dif-ference of 18 kDa is consistently detected between the two ACP-derivatives. It has been reported that AcpS can utilise saturatedacyl-CoAs up to C10 [20] and crotonoyl-CoA [16] but we believethis is the first report of using the method to make a 3-hydroxy-acyl-ACP-derivative.

3.3. Assay of 3-hydroxyacyl-ACP dehydratase activity

We observed no differences in UV absorbance between thehydroxybutyryl- and crotonoyl-ACP compounds, which surprisedus as they have been reported in the literature [8,9]. Spectropho-tometric assays were not possible with these ACP-derivatives butMS analysis enabled a direct assay of purified A. thaliana DH bydetection of a mass change of 18 kDa. The presence of functionalDH should change the proportions of ACP-derivatives present andthis can be seen using crotonoyl-ACP, monitoring the reversereaction (Fig. 6). The most abundant ions present in the substrateare the H+ and Na+ adducts of crotonoyl-ACP (mass 8918 and8940) and addition of purified DH resulted in the appearanceof a new mass peak (8936 kDa) corresponding to the [3-hydrox-yacyl-ACP+H]+ ion, which was not present when buffer alonewas added. After incubation with DH (Fig. 6B), the ratio of pro-tonated crotonoyl-substrate (8918 kDa) to 3-hydroxyacyl-product(8936 kDa) was 1:3.3, close to the value observed for E. coli DHat equilibrium, and this value does not change on addition ofmore DH.

Addition of purified DH to crotonoyl-ACP changed the propor-tions of acyl-ACP-derivatives in a manner consistent with an activeenzyme. The MS analysis reported here however does not readilyallow kinetic experiments to be done. In experiments with croto-noyl-ACP for example, the proportionality of reaction extent to en-zyme amount and time are difficult to measure because of thesimilarity in mass changes due to [M+Na]+ adduct formation andaddition of water in the catalysed reaction. Thus, for intermediatepoints during a time-course assay, de-convolution of the charge-series spectra to produce the mass graph results in apparentlyintermediate-sized ACP-derivatives of 8938 and 8937 kDa beingidentified as the ratio of [crotonoyl-ACP+Na]+ (8939 kDa) and [3-hydroxybutyryl-ACP+H]+ (8935 kDa) changes. It is only when incu-bations reach apparent equilibrium (for example that shown inFig. 6B) that [crotonoyl-ACP+H]+ and [3-hydroxybutyryl-ACP+H]+,separated by 18 amu, are clearly resolved.

We have expressed, purified and demonstrated activity of aputative A. thaliana DH. The database-entry for this lists the pres-ence of InterPro Beta-hydroxyacyl-(acyl-carrier-protein) dehydra-tase FabZ and FabA/FabZ domains but to our knowledge this isthe first definitive proof of function of a plant dehydratase enzyme,which will be of benefit in further studies on A. thaliana FAS pro-teins. Whilst we have detailed knowledge of the structure of sev-eral plant FAS enzymes after X-ray crystallography, little isknown about the interactions of the FAS components with thoseof different catalytic function. It is of interest that a sequencedanomalous band that co-purified with both enoyl-ACP reductaseand acyl-ACP thioesterase from developing rapeseed [21] cannow be definitely assigned as 3-hydroxyacyl-ACP dehydratase,indicating that these three enzymes are possibly associated in vivo.

Acknowledgement

We thank BBSRC for financial support during this work.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.febslet.2008.12.022.

References

[1] Magnuson, K., Jackowski, S., Rock, C.O. and Cronan Jr., J.E. (1993) Regulation offatty acid biosynthesis in Escherichia coli. Microbiol. Rev. 57, 522–542.

[2] Harwood, J.L. (1996) Recent advances in the biosynthesis of plant fatty acids.Biochim. Biophys. Acta 1301, 7–56.

368 A. Brown et al. / FEBS Letters 583 (2009) 363–368

[3] Campbell, J.W. and Cronan Jr., J.E. (2001) Bacterial fatty acid biosynthesis:targets for antibacterial drug discovery. Annu. Rev. Microbiol. 55, 305–332.

[4] White, S.W., Zheng, J., Zhang, Y-M. and Rock, C.O. (2005) The structural biologyof type II fatty acid biosynthesis. Annu. Rev. Biochem. 74, 791–831.

[5] Heath, R.J. and Rock, C.O. (1996) Roles of the FabA and FabZ b-hydroxyacyl-acyl carrier protein dehydratases in Escherichia coli fatty acid biosynthesis. J.Biol. Chem. 271, 27795–27801.

[6] Cronan Jr., J.E., Li, W.B., Coleman, R., Narasimhan, M., de Mendoza, D. andSchwab, J.M. (1988) Derived amino acid sequence and identification of activesite residues of Escherichia coli beta-hydroxydecanoyl thioester dehydrase. J.Biol. Chem. 263, 4641–4646.

[7] Mohan, S., Kelly, T.M., Eveland, S.S., Raetz, C.R.Z. and Anderson, M.S. (1994) AnEscherichia coli gene (FabZ) encoding (3R)-hydroxymyristoyl acyl carrierprotein dehydrase. J. Biol. Chem. 269, 32896–32903.

[8] Shimakata, T. and Stumpf, P.K. (1982) The procaryotic nature of the fatty acidsynthetase of developing Carthamus tinctorius L (safflower) seeds. Arch.Biochem. Biophys. 217, 144–154.

[9] Shimakata, T. and Stumpf, P.K. (1982) Purification and characterizations ofb-ketoacyl-[acyl-carrier-protein] reductase, b-hydroxyacyl-[acyl-carrier-protein] dehydrase, and enoyl-[acyl-carrier-protein] reductase from Spinaciaoleracea leaves. Arch. Biochem. Biophys. 218, 77–91.

[10] Cases, S., Smith, S.J., Zheng, Y.-W., Myers, H.M., Lear, S.R., Sande, E., Novak, S.,Collins, C., Welch, C.B., Lusis, A.J., Erickson, S.K. and Farese Jr., R. (1998)Identification of a gene encoding an acyl CoA: diacylglycerol acyltransferase, akey enzyme in triacylglycerol synthesis. Proc. Natl. Acad. Sci. USA 95, 13018–13023.

[11] Roujeinikova, A., Baldock, C., Simon, W.J., Gilroy, J., Baker, P.J., Stuitje, A.R., Rice,D.W., Rafferty, J.B. and Slabas, A.R. (2002) Crystallization and preliminary X-ray crystallographic studies on acyl-(acyl carrier protein) from Escherichia coli.Acta Cryst. D58 (2), 330–352.

[12] Post-Beittenmiller, D., Jaworski, J.G. and Ohlrogge, J.B. (1991) In vivo pools offree and acylated acyl carrier proteins in spinach. Evidence for sites ofregulation of fatty acid biosynthesis. J. Biol. Chem. 266, 1858–1865.

[13] Sharma, S.K., Kapoor, M., Ramya, T.N.C., Kumar, S., Kimar, G., Modak, R.,Sharma, S., Surolia, N. and Surolia, A. (2003) Identification, characterization,and inhibition of Plasmodium falciparum b-hydroxyacyl-acyl carrier proteindehydratase (FabZ). J. Biol. Chem. 278, 45661–45671.

[14] Dautu, G., Ueno, A., Munyaka, B., Carmen, G., Makino, S., Kobayashi, Y. andIgarashi, M. (2008) Molecular and biochemical characterization of Toxoplasmagondii b-hydroxyacyl-acyl carrier protein dehydratase (FABZ). Parasitol. Res.112, 1301–1309.

[15] Weeks, G. and Wakil, S.J. (1968) Studies on the mechanism of fatty acidsynthesis. J. Biol. Chem. 243, 1180–1189.

[16] Fawcett, T., Copse, C., Simon, J.W. and Slabas, A.R. (2000) Kinetic mechanism ofNADH-enoyl-ACP reductase from Brassica napus. FEBS Lett. 484, 65–68.

[17] Mizugaki, M., Weeks, G., Toomey, R.E. and Wakil, S.J. (1968) Preparation andgeneral properties of b-hydroxybutyryl acyl carrier protein dehydrase. J. Biol.Chem. 243, 3661–3670.

[18] Crosby, J., Sherman, D.H., Bibb, M.J., Revill, W.P., Hopwood, D.A. and Simpson,T.J. (1995) Polyketide synthase acyl carrier proteins from Streptomyces:expression in Escherichia coli, purification and partial characterisation.Biochim. Biophys. Acta 1251, 32–42.

[19] Rock, C.O. and Cronan Jr., J.E. (1981) Acyl carrier protein from Escherichia coli.Meth. Enzymol. 71, 341–351.

[20] Roujeinikova, A., Simon, W.J., Gilroy, J., Rice, D.W., Rafferty, J.B. and Slabas, A.R.(2007) Structural studies of fatty acyl-(acyl carrier protein) thioesters reveal ahydrophobic binding cavity that can expand to fit longer substrates. J. Mol.Biol. 365, 135–145.

[21] Hellyer, A., Leadlay, P.F. and Slabas, A.R. (1992) Induction, purification andcharacterisation of acyl-ACP thioesterase from developing seeds of oil seedrape (Brassica napus). Plant Mol. Biol. 20, 763–780.