HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 18, 15558-15565, May 3, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/18/15558    most recent M201399200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, Y.-M. Articles by Rock, C. O. Search for Related Content PubMed PubMed Citation Articles by Zhang, Y.-M. Articles by Rock, C. O. Social Bookmarking                      What's this?

The FabR (YijC) Transcription Factor Regulates Unsaturated Fatty Acid Biosynthesis in Escherichia coli^*

Yong-Mei Zhang^§, Hedia Marrakchi^§, and Charles O. Rock^¶

From the  Department of Infectious Diseases, Protein Science Division, St. Jude Children's Research Hospital, Memphis, Tennessee 38105 and the ^¶ Department of Molecular Biosciences, University of Tennessee Health Science Center, Memphis, Tennessee 38163

Received for publication, February 11, 2002, and in revised form, February 20, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Unsaturated fatty acid biosynthesis is a vital facet of Escherichia coli physiology and requires the expression of two genes, fabA and fabB, in the type II fatty acid synthase system. This study links the FabR (YijC) transcription factor to the regulation of unsaturated fatty acid content through the regulation of fabB gene expression. The yijC (fabR) gene was deleted by replacement with a selectable cassette, and the resulting strains (fabR::kan) possessed significantly elevated levels of unsaturated fatty acids, particularly cis-vaccenate, in their membrane phospholipids. The altered fatty acid composition was observed in the fabR::kan fabF1 double mutant pinpointing fabB as the condensing enzyme responsible for the increased cis-vaccenate production. The fabR::kan strains had 4- to 8-fold higher levels of fabB and a 2- to 3-fold increase in fabA transcripts as judged by Northern blotting, Affymetrix array analysis, and real-time PCR. FabR did not regulate the enzymes of fatty acid -oxidation. The elevated level of fabB mRNA was reflected by higher condensing enzyme activity in fabR::kan fabF1 double mutants. Thus, FabR functions as a repressor that potently controls the expression of the fabB gene, which in turn, modulates the physical properties of the membrane by altering the level of unsaturated fatty acid production.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Unsaturated fatty acid biosynthesis is required to maintain membrane structure and function in Escherichia coli and many other organisms. E. coli possesses a type II fatty acid synthase system, and the double bond is introduced into the growing acyl chain at the ten-carbon -hydroxydecanoyl-ACP¹ intermediate (for reviews, see Refs. 1, 2). Early genetic studies identified two genes, fabA and fabB, that were essential for olefin formation. Inactivating mutations in either one of these genes leads to an absolute requirement for unsaturated fatty acids for growth (1). FabA introduces the double bond into the acyl chain through its dual activities as a -hydroxydecanoyl-ACP dehydratase and a trans-2,cis-3-decanoyl-ACP isomerase (3). Because there is a second -hydroxyacyl-ACP dehydratase (FabZ) that only produces the trans-2 isomer (4, 5), unsaturated fatty acid formation also depends on the ability of the type II system to efficiently divert the cis-3 intermediate to the unsaturated branch of the elongation pathway. The FabB-condensing enzyme fulfills this function presumably by efficiently catalyzing the elongation of cis-3-decenoyl-ACP (1, 2). The fabA and fabB genes are always found together in bacterial genomes (6), supporting the idea that these two gene products work in tandem to generate unsaturated fatty acids.

FadR is a transcriptional regulator that modulates the expression of the fabB and fabA genes. FadR was discovered through the analysis of a mutation that results in the constitutive induction of the -oxidation enzymes (7), and led to its characterization as a repressor of fatty acid -oxidation genes (8). FadR is released from its DNA binding sites by long-chain acyl-CoAs (9-12), which bind to the carboxyl terminus of the protein and release the amino-terminal winged helix domain from the DNA (13-15). An interesting twist in the FadR story began with the observation that fabA(Ts) fadR double mutants were unable to grow at the permissive temperature without an unsaturated fatty acid supplement (16). This suggested a positive effect of FadR on fabA expression, and it was soon demonstrated that FadR is a transcriptional activator that binds to the 40 region of the fabA gene, a site common for activators of ⁷⁰-responsive promoters (17, 18). FadR is also a positive regulator of the fabB gene, although the changes in fabB expression in fadR mutants are not as great as with fabA (19). Thus, FadR acts as a repressor of -oxidation genes and an activator of the two genes required for unsaturated fatty acid synthesis (20).

The promoter regions of the fabA and fabB genes are highly related, suggesting that their expression is similarly regulated. The FadR binding site is not the only region of sequence similarity in the fabB and fabA promoters, and their alignment is shown in Fig. 1 with the sites of transcription initiation indicated by the arrows for both fabA (18) and fabB (21). Recently, McCue et al. (22) used a bioinformatic "phylogenetic footprinting" method to identify putative transcription factor binding sites in bacterial genomes and located a strongly predicted site in the promoter regions of the fabA and fabB genes. They went on to show that a protein, YijC, specifically bound to a DNA affinity column carrying the predicted transcription factor palindrome (Fig. 1). The oligonucleotide 5'-GGCGTACAAGTGTACGCT was used to isolate YijC protein from E. coli cell-free extracts (22). The YijC binding sequence on fabB consists of a central CGTACAXXTGTACG palindrome, whereas the predicted site in fabA has a 2-bp mismatch (Fig. 1). Based on the location of these binding sites, McCue et al. (22) proposed renaming the yijC gene fabR for Fatty Acid Biosynthesis Regulator. However, there are no direct data demonstrating that FabR actually influences the levels of fabA or fabB mRNA or alters the production of unsaturated fatty acids by the pathway. The goal of this study was to determine whether FabR (YijC) regulates unsaturated fatty acid formation by examining the effects of deleting this gene on fatty acid metabolism.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Location of the FadR and FabR binding sites in the fabB and fabA promoters. The sequence of the promoter regions of the fabA and fabB gene are very similar. The transcriptional start sites are indicated by the arrows for both fabA (18) and fabB (21). The FadR (17, 18) and FabR (22) binding sites are indicated by the brackets.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials and Bacterial Strains-- [-³²P]dCTP (3000 Ci/mmol) and [2-¹⁴C]malonyl-CoA (55 mCi/mmol) were purchased from Amersham Biosciences, Inc.. Restriction enzymes and other molecular biology reagents were from Promega Life Science, Invitrogen, and New England BioLabs. ACP was purchased from Sigma Chemical Co., and myristoyl-ACP was prepared using the acyl-ACP synthetase method (23). The E. coli strains that we use in this study and their relevant genotypes are listed in Table I. Plasmid pDM4 (24), expressing the fabB gene, was constructed by cloning fabB with its promoter into pBR322, and plasmid pSJ21, expressing the fabA gene, was constructed by cloning the fabA into pBluescript. Transductions with P1 phage were performed as described previously by Miller (25). DNA sequencing, Affymetrix microarray analysis, and oligonucleotide synthesis were performed by the Hartwell Center at St. Jude.

Generation of an fabR Deletion Strain-- The gene targeting strategy shown in Fig. 2A was used to generate a fabR deletion strain of E. coli. The 813-bp fragment upstream of fabR was amplified by PCR with primers P_S (5'-CGCGAGCTCACATCTGTTGGATGATATGG, a SacI site is underlined) and P_Br (5'-CGCGGATCCCATCACGATGTCTGAATCC, a BamHI site is underlined). The SacI-BamHI-digested 813-bp fragment was cloned into SacI-BamHI site of pUC19. The 788-bp fragment downstream of fabR was amplified similarly by PCR with primers P_Bf (5'-CGCGGATCCATGTGAAGGACGAGTAATG, a BamHI site is underlined) and P_H (5'- CCCAAGCTTATACAACATCGCAGCTAAC, a HindIII site is underlined). Then the BamHI-HindIII-digested 788-bp fragment was inserted into the BamHI-HindIII site of the plasmid harboring the 813-bp upstream fragment. The BamHI-digested kanamycin-resistant gene fragment was cloned into the BamHI site of the plasmid harboring both the upstream (813-bp) and downstream (788-bp) PCR fragments, yielding plasmid pUCFabRK. This plasmid was digested with SacI and AflIII. The 3.3-kb fragment, containing both PCR fragments, kanamycin gene-replacing fabR and a 350-bp fragment of pUC19, was purified from 1% agarose gel. The 3.3-kb linear DNA (25 ng) was transformed into strain PDJ1 (recD) by electroporation, and kanamycin-resistant transformants were isolated. The fabR::kan strain grew with the same doubling time as the wild type on both rich and minimal media.

Genomic Analyses of fabR Deletion Strain-- PCR and Southern blot analyses were used to confirm the genotype of the fabR deletion strain. PCR experiments were performed with a primer pair, P1 (5'-GATCACTCCAGCACCATAG) and P2 (5'-TTACTGGAGCTGTACTGCG), to amplify a fragment encompassing 1 kb both upstream and downstream of fabR. The PCR products from both wild type and fabR::kan strains were digested with enzymes NcoI, BamHI, and HindIII individually. Two other primer pairs, P1 and P_K1 (5'-ATCTTGTGCAATGTAACATCAGAG), P_K2 (5'-AGTCAGCAACACCTTCTTCACG) and P2, were used to amplify two 1-kb fragments in fabR::kan strain. The PCR products were purified from 1% agarose gel, and DNA sequencing was performed in the Hartwell Center with ABI Prism 3700 DNA analyzer. Southern blots were performed on genomic DNA isolated from both wild type and fabR::kan strains by the phenol/chloroform/isoamyl alcohol extraction method (26). Three different restriction enzymes, NdeI, EcoRI, and ScaI, were used to digest 5 µg of genomic DNA. The digested genomic DNA was separated by electrophoresis on a 0.8% agarose gel. The probe was the 230-bp ClaI-ScaI fragment downstream of fabR (Fig. 2A), and was labeled with [-³²P]dCTP. Southern transfer, hybridization, and washing were carried out by standard procedures (27). Bands were detected with a Molecular Dynamics Storm 860 imager, and their intensity was determined using the ImageQuaNT 5.1 program.

RNA Analyses of fabR::kan Deletion Strain-- Total RNA was isolated from exponentially growing cells, both wild type and fabR::kan, using a MasterPure RNA purification kit (Epicentre Technologies). The same RNA sample was used for Northern blot, Affymetrix array, and TaqMan RT-PCR analyses. Briefly, cell pellets from 40-ml cultures were resuspended in lysis buffer (containing 0.17 mg/ml proteinase K) and incubated at 65 °C for 15 min. After removing the debris and proteins by centrifugation, the RNA was precipitated with isopropanol. The remaining DNA was removed by treating RNA preparations with DNase I (0.025 unit/µl) at 37 °C for 20 min. RNA samples were isopropanol-precipitated, washed twice with 75% ethanol, and redissolved in TE.

Northern blots were performed with 10 µg of total RNA separated by electrophoresis on a 1% agarose formaldehyde gel. The probes were the 320-bp HincII-MunI fragment of plasmid pSJ21 (for fabA) and the 280-bp EcoRI-ClaI fragment of pDM4 (for fabB). Northern transfer, hybridization, and washing were performed using standard procedures (28). Detection and quantitation of the blot was performed using a Molecular Dynamics Storm 860 imager. Staining the membrane with methylene blue assessed equal loading of the samples, and the intensity of the bands was consistent in all lanes.

Affymetrix array analyses were performed using messenger RNA species enriched by the specific degradation of the 16 S and 23 S ribosomal RNAs, which constitute 90% of the total prokaryotic RNA population. Moloney murine leukemia virus reverse transcriptase (Epicentre Technologies) and primers specific to 16 S and 23 S rRNA were used to synthesize the complementary cDNAs. Then rRNAs were removed enzymatically by treatment with RNase H. The cDNA molecules were degraded by DNase I digestion, and the enriched mRNAs were purified using Qiagen RNeasy columns. The RNA was fragmented by heat and ion-mediated hydrolysis, and the 5'-end RNA termini were enzymatically modified by T4 polynucleotide kinase and ATPS. A biotin group was then conjugated to the thiolated RNA. After purification of the product (using the RNA/DNA mini kit, Qiagen), the efficiency of the labeling was assessed using a gel-shift assay based on the retardation of the biotinylated RNA upon addition of NeutrAvidin molecules. 1.5-4.0 µg of fragmented RNA (biotin-labeled target) was hybridized for 16 h at 45 °C to E. coli oligonucleotide arrays (Affymetrix) containing all known E. coli genes together with more than 2700 probe sets designed to intergenic sequences. Arrays were washed at 25 °C with 6 × SSPE (0.9 M NaCl, 60 mM NaH₂PO₄, 6 mM EDTA, and 0.01% Tween 20) followed by a stringent wash at 50 °C with 100 mM MES, 0.1 M NaCl, 0.01% Tween 20. The arrays were then stained with phycoerythrin-conjugated streptavidin (Molecular Probes), and the fluorescence intensities were determined using a laser confocal scanner (Hewlett-Packard). The scanned images were analyzed using Microarray software (Affymetrix). Sample loading and variations in staining were standardized by scaling the average of the fluorescence intensities of all genes on an array to a constant target intensity for all arrays used. The expression data were analyzed as previously described (29). The signal intensity for each gene was calculated as the average intensity difference, represented by the formula, (PM  MM)/number of probe pairs, where PM and MM denote perfect-match and mismatch probes.

Oligonucleotide primers and probes for real-time RT-PCR were designed with Primer Express 1.0 software (ABI Prism, PerkinElmer Life Sciences/Applied Biosystems), and the probes were purchased from Applied Biosystems. The probes consisted of an oligonucleotide labeled at the 5'-ends with the reporter dyes, 3FAM or VIC, and at the 3'-ends with the quencher dye TAMRA (PE Applied Biosystems) and are listed in Table II.

View this table: [in this window] [in a new window]   Table II TaqMan primers and probes (5'  3')

The reverse transcription was performed on total RNA prepared as above after a second step of DNase I treatment (DNA-free, AMBION). The reverse transcription mixture (20 µl) contained 500 ng of total RNA, 10 ng/µl random hexamers (Invitrogen), 33 units of RNAguard Ribonuclease inhibitor (Amersham Biosciences, Inc.), and 20 units/µl Superscript II reverse transcriptase (Invitrogen). Aliquots (1 µl) of the reverse transcription reaction were added to the real-time PCR reaction (30 µl) containing 600 nM of each forward and reverse primer and 166 nM of probe. Amplification and detection of specific products was performed with the ABI Prism 7700 sequence detection system (PE Applied Biosystems) with the following profile: 1 cycle at 50 °C, 1 cycle at 95 °C for 10 min, 40 cycles at 95 °C for 15 s, and 60 °C for 1 min. The critical threshold cycle (C_T) is defined as the cycle at which the fluorescence becomes detectable above background and is inversely proportional to the logarithm of the initial template molecules. The C_T values were used to calculate the relative number of fabA and fabB cDNA molecules in wild type and fabR::kan mutant. The quantity of cDNA for each experimental gene was normalized to the quantity of acpP cDNA in each sample.

Fatty Acid Composition Analysis-- Cultures (10 ml) of E. coli strains were grown to mid-log phase in M9 minimal medium (25) supplemented with 0.4% glucose, 0.01% methionine, 0.0005% thiamin, and harvested by centrifugation. The cell pellet was suspended in 1 ml of water, and the lipids were extracted as described by Bligh and Dyer (30) and fatty acid methyl esters were prepared by the addition of 2 ml of HCl/methanol to the dry extract. The fatty acid methyl esters were fractionated using a Hewlett-Packard Model 5890 gas chromatograph equipped with a flame ionization detector and a glass column (2 m by 4 mm, internal diameter) containing 3% SP2100 coated on Supelcoport (100/120 mesh) operated at 190 °C. Fatty acid methyl esters were identified by comparing their retention times with standards (Matreya).

Condensing Enzyme Assay-- The condensing enzyme activity assay was essentially the same as described previously (31). Cultures (20 ml) of E. coli strains (fabF1 and fabR::kan fabF1) were grown to mid-log phase in M9 minimal medium (25) supplemented with 0.4% glucose, 0.01% methionine, 0.0005% thiamin, and harvested by centrifugation. The cell pellet was suspended in 5 ml of 20 mM Tris-HCl, pH 7.6, containing 1 mM EDTA and 1 mM dithiothreitol. Cells were lysed by two passages through a French press cell at 20,000 p.s.i. The total cell lysate was centrifuged at 50,000 rpm at 4 °C for 1 h. The cell-free extract, a supernatant from ultracentrifugation, was saved for the assay. Protein content in the cell-free extract was determined with the Bradford assay using -globulin as the standard (32). The condensation assay contained 45 µM myristoyl-ACP, 50 µM [2-¹⁴C]malonyl-CoA, 100 µM ACP, 25 ng of FabD (malonyl-CoA:ACP transacylase) and the indicated concentrations of cell-free extract (0.1-1 µg) in a final volume of 40 µl. The mixtures were incubated at 37 °C for 15 min and then reduced with borohydride, extracted into toluene, and quantitated by scintillation counting. One unit of condensing enzyme activity corresponds to the formation of 1 µmol of -[¹⁴C]ketohexadecanoyl-ACP per minute.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Characterization of the fabR Deletion Strain-- A targeting construct was prepared that replaced the fabR (yijC) gene with a kanamycin-resistance cassette as described under "Experimental Procedures," and the linearized fragment was transformed into strain PDJ1 (recD). Kanamycin-resistant transformants were selected, and a combination of PCR and Southern blot analysis were used to verify that the kanamycin gene was inserted in place of fabR in the correct position in the genome. Primers P1 and P2 were designed outside the sequence that was used in the targeting construct for linear transformation (Fig. 2A). Substitution of the kanamycin gene for fabR increased the size of the PCR product from these two primers by 600 bp. Thus, the PCR product from the fabR::kan strain with P1 and P2 was 3 kb in length (Fig. 2B, lane 2) compared with the 2.4-kb product from wild type cells (Fig. 2B, lane 1). When these two PCR products were digested with three restriction enzymes (Fig. 2B, lanes 3 and 6, NcoI; lanes 4 and 7, BamHI; and lanes 5 and 8, HindIII), the products were of the predicted sizes (Fig. 2B). Sequence results of the PCR products read through the junction regions where the kanamycin gene was inserted (data not shown). P_K1 and P_K2 were designed from sequence of the kanamycin gene. PCR reactions with the two primer pairs, P1 and P_K1, P_K2 and P2, produced products of the correct size (1 kb) in the fabR::kan strain indicating the presence of the kanamycin gene in the correct locations, whereas no products were detected with these two primer sets in wild type cells (data not shown).

View larger version (39K): [in this window] [in a new window]   Fig. 2.   The fabR gene targeting strategy and characteristics of the fabR::kan deletion strain. A, a diagram of the fabR replacement in the genome showing the locations of the kanamycin gene insert, the PCR primers, and the diagnostic restriction enzymes used in constructing and analyzing the gene replacement. Nd, NdeI; E, EcoRI; S, ScaI; Nc, NcoI; C, ClaI; B, BamHI; H, HindIII. Kanamycin-resistant strains were isolated following linear transformation of strain PDJ1 (recD::Tn10) as described under "Experimental Procedures," and homologous recombination of the targeting construct was verified by PCR (B) and Southern blotting (C). B, PCR analysis of wild type and fabR::kan strains with primers P1 and P2. The PCR product from a wild type cell was 2.4-kb (lane 1), compared with the longer 3.0-kb product obtained from the fabR::kan strain MWF1 (lane 2). The PCR products of strain PDJ1 (wild type; lanes 3-5), and strain MWF1 (fabR::kan; lanes 6-8) were digested with three different restriction enzymes, NcoI (lanes 3 and 6), BamHI (lanes 4 and 7) and HindIII (lanes 5 and 8). The digested products showed patterns expected from the location of the restriction sites shown in A. C, Southern blot analysis was performed using a probe prepared from a 230-bp ClaI-ScaI fragment downstream of fabR gene as shown in A. The 32P-labeled probe was hybridized with genomic DNA fragments of the expected sizes from either strain MWF1 (fabR::kan) (lanes 2, 4, and 6) or strain PDJ1 (wild type) (lanes 1, 3, and 5) digested with ScaI (lanes 1 and 2), EcoRI (lanes 3 and 4), or NdeI (lanes 5 and 6) restriction enzymes.

Results from Southern blot analysis with the 230-bp ClaI-ScaI fragment corroborated the PCR results confirming that the kanamycin gene-replaced fabR gene in strain MWF1 was in the correct location without compromising the neighboring genes. The probe hybridized with a 650-bp fragment of wild type genomic DNA digested with NdeI (Fig. 2C, lane 5). The NdeI restriction site in the fabR gene disappeared in the fabR deletion strain, resulting in the probe hybridizing with a 3.2-kb fragment (Fig. 2C, lane 6). The probe also recognized bands of the correct sizes in the EcoRI and ScaI digests (Fig. 2C). In the case of the fabR::kan strain, the fragments were larger by 600 bp due to the presence of the kanamycin gene in place of fabR.

Deletion of fabR Leads to the Overproduction of Unsaturated Fatty Acids-- The effect of FabR deletion on the production of the major membrane fatty acids of E. coli was determined. Strain PDJ1 synthesized slightly more saturated fatty acids (SFA) than unsaturated fatty acids (UFA). The UFA:SFA ratio was 0.87, and the amount of cis-vaccenate (C18:111) and palmitoleate (C16:19) was approximately the same (Table III).² Strain MWF1 (fabR::kan) produced significantly more UFA increasing the UFA/SFA ratio to 1.77. Unlike the wild type cells, C18:111 was the predominant UFA species in the fabR::kan strain. A similar alteration of fatty acid composition was observed in cells that expressed fabB from a multicopy plasmid (pDM4) (Table III). In both cases, the 18:1/16:1 ratio was significantly increased (Table III). The similarity between the fatty acid composition alterations due to FabB overexpression (Table III and Ref. 24) and the composition of the fabR::kan strain pointed to an increased level of fabB expression as the underlying cause for the increased UFA and altered 18:1/16:1 ratio in strain MWF1. In contrast, increased expression of the fabA gene actually slightly elevated the amounts of SFA in the membranes (Table III) consistent with previously reported results (33). The activity of FabF regulates the cellular content of 18:111 (24, 34, 35), therefore, we transduced the fabR::kan allele into strain SJ109 (fabF1) and examined the fatty acid composition to test if the regulation of FabB alone was responsible for the elevation in UFA. As expected, the fabF1 mutant had reduced amounts of C18:111. The 18:111 levels in the fabR::kan fabF1 double mutant were elevated compared with the fabF control and were similar to the amount of C18:111 in the wild type cells (Table III). These compositional data strongly support the conclusion that FabR acts as a repressor of fabB, which in turn, modifies the membrane fatty acid composition.

The relationship between FabR and FadR regulation was investigated by determining the fatty acid composition in strain PDJ14 (fadR) and ANS4 (fadR fabR) (Table III). As reported earlier (16), the fadR strain PDJ14 had a higher level of saturated fatty acids compared with the control strain. The fadR fabR double-mutant strain ANS4 had essentially the same composition as the fadR strain with a similar UFA/SFA and 18:1/16:1 ratios. These data illustrate that the regulation of fatty acid composition by fabR requires a functional fadR gene.

Steady-state Levels of fabB and fabA mRNA in the fabR Deletion Strain-- Because fabA and fabB are the two essential genes required for UFA synthesis, Northern blot experiments with ³²P-labeled probes specific for fabB or fabA were used to evaluate the levels of their respective mRNAs (Fig. 3). The mRNA levels for both genes were elevated in the fabR::kan strain (Fig. 3), although the magnitude of induction of fabB was significantly greater than fabA. The Northern blots revealed that deletion of the fabR gene resulted in a 7-fold elevation of fabB compared with a 3-fold elevation of fabA mRNA (Fig. 3).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Increased expression of both fabB and fabA in strain MWF1 (fabR::kan). Northern blot analysis showed that the level of fabA mRNA in strain MWF1 (fabR::kan) was three times higher than control (strain PDJ1). The level of fabB mRNA was 7-fold higher in FabR negative cells. The bands were localized, and the intensities were quantitated using a PhosphorImager. The size of the fabA mRNA was 550 bp, and the fabB mRNA was 1280 bp.

The fabR Strain Has Elevated Condensing Enzyme Activity-- The RNA analysis data suggest that fabR::kan strains will have higher levels of FabB protein and condensing enzyme activity. We tested this idea by performing condensing enzyme assays on cell-free extracts from strain ANS7 (fabR::kan fabF1). These experiments showed that the condensing enzyme specific activity in the cell-free extracts of strain ANS7 (fabR::kan fabF1) was 2.5 times higher than that of the strain SJ109 (fabF1) extracts using myristoyl-ACP as the substrate (Fig. 4). Because strain ANS7 did not have a functional FabF, we concluded that the increased condensing enzyme activity was due to higher amounts of FabB, which is consistent with the increased fabB mRNA levels. We did not perform a similar experiment with FabA due to the lack of the appropriate specific substrates to measure activity in crude extracts; however, previous work demonstrated a direct relationship between fabA mRNA level and FabA activity (33).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Increased condensing enzyme activity in strain ANS7 (fabR::kan). Cell-free extracts of strains SJ109 (fabF1) and ANS7 (fabR::kan fabF1) were prepared, dialyzed, and assayed for condensing enzyme activity using myristoyl-ACP as the substrate as described under "Experimental Procedures." The condensing enzyme specific activity in strain ANS7 extracts was 25.2 µmol/min/µg, which was 2.5 times higher than in strain SJ109 (10.5 µmol/min/µg).

Gene Expression in fabR and fadR Strains-- A comparison of the global transcription effect of FabR was addressed using the Affymetrix array technology to examine the changes in all transcripts. Our analysis of four sets of global gene expression profiles comparing mRNA levels in strain ANS8 (fabR::kan) to the wild type revealed that fabB and fabA were elevated 4- and 2-fold, respectively (Fig. 5A). The analogous array experiment comparing wild type and a fadR mutant showed that fabA transcription was significantly reduced in the absence of FadR (Fig. 5A). We did not observe any reduction in fabB transcript levels in fadR strains using this technology (Fig. 5A). In the fadR fabR double mutant, fabA mRNA was reduced to the same extent as in the fadR mutant (Fig. 5A). In this experiment, fabB transcripts were also lower in the double mutant compared with the fabR strain indicating that, in the absence of FabR, FadR had a significantly greater activating effect on fabB expression.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Levels of fabA and fabB mRNA in fabR and fadR strains determined by microarray analysis and real-time PCR. RNA preparations from strains UB1005 (wild type), PDJ14 (fadR::Tn10), ANS8 (fabR::kan), and ANS4 (fabR::kan fadR::Tn10) were used for the Affymetrix microarray analysis and real-time PCR. A, two independent fabR::kan RNA microarrays were compared with two control RNA microarrays to obtain four analyses of the amount of fabB and fabA mRNA in the two strains. The fabB and fabA fluorescence was normalized to the acpP mRNA value, and the results were averaged from four independent chip comparisons. In the FabR-negative cells, the fabA level was increased by 2-fold and the fabB level in the fabR mutant was four times higher than in wild type. The analysis of the double mutant fabR fadR showed a decrease in the fabA expression level, as observed in the case of the fadR mutant. The fabB mRNA levels in the fabR fadR mutant were 2-fold higher than in the wild type. B, TaqMan real-time PCR was performed with primers and probe sets specific for acpP, fabB, or fabA as described under "Experimental Procedures." The amounts of fabB and fabA cDNA were determined and normalized to the amount of acpP cDNA in the same samples. The data are means of four determinations on two independent RNA/cDNA preparations. The expression level of fabA in FadR-negative cells was 8-fold lower than in wild type cells, and no significant change was observed in the fabB levels. In the FabR-negative cells, the fabA level was elevated 3-fold and the fabB level was elevated 5-fold. Analysis of the fadR fabR double mutant showed a decrease in fabA expression, whereas the fabB levels were 1.5-fold higher than in the fadR strain.

The array experiments also revealed the up-regulation of the genes involved in -oxidation in fadR and fadR fabR double mutants (Table IV), as expected for fadR null strains (20). The exceptions were the yfcX and yfcY genes that were not known to be involved in fatty acid -oxidation, but whose predicted functions strongly point to a role for these gene products in -oxidation (Table IV and "Discussion"). We did not detect other fatty acid biosynthesis, fatty acid -oxidation, or additional known or unknown genes that were significantly regulated by FabR in the array experiments.

The TaqMan real-time PCR method is a sensitive and quantitative technology and was used to validate and extend the findings from the array analysis (Fig. 5B). The threshold cycle number (C_T) for each gene was determined and normalized to the C_T determined for the acpP gene in the same RNA sample. In these experiments, both the fabB and fabA mRNA levels were elevated in strain ANS8 (fabR::kan) 5- and 3-fold, respectively, compared with the wild type strain UB1005 confirming that FabR represses the mRNA levels of both genes. In the fadR mutant, the levels of fabA mRNA were significantly reduced, as anticipated. The level of fabB mRNA remained essentially the same in the fadR mutant strain. The levels of fabA mRNA in the fadR fabR double mutant were reduced to levels comparable to the fadR single mutant (Fig. 5B). Also, we observed a significant reduction in the fabB mRNA in the double mutant compared with the fabR::kan strain, supporting the idea that FadR has an activating effect on fabB transcription (19), especially in the absence of FabR.

Taken together these expression analyses point to FabR repression as a potent regulator of fabB expression and to a lesser extent a regulator of fabA expression. On the other hand, FadR activation is most important for the regulation of fabA expression, but also contributes to fabB regulation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Our study demonstrates the function of the FabR transcription factor in controlling UFA production through the regulation of fabB in the E. coli type II fatty acid synthase system. FabR controls the UFA production and thus directly influences the physical properties of the membrane bilayer (35, 36). Also, the increased FabB activity leads to the increased production of 18:111. The formation of 18:111 was originally hypothesized to be a unique property of the FabF (-ketoacyl-ACP synthase II) based on the reduced amount of this fatty acid characteristic of the fabF mutant phenotype (31, 34). Later, the enforced overexpression of FabB from a multicopy plasmid was demonstrated to drive 18:111 synthesis (24), illustrating that the substrate specificity of the condensing enzymes is not absolute. Our work validates this point by showing that the physiological regulation of FabB levels by FabR is also an important determinant of the cellular content of 18:111. The regulation of fabB is the most important function of FabR in determining fatty acid composition, and this discovery is a unique example of a transcription factor that exclusively regulates the expression of the enzymes of a type II fatty acid synthase.

Regulation of UFA content by the FabR repressor requires the presence of the FadR activator. FabR binds to a sequence found in the promoter regions of the fabA and fabB genes from E. coli (Fig. 1). The location of the predicted binding sequence and the data discussed above suggests that FabR regulates by repression and antagonizes the action of the FadR activator. Our experiments point to the level of FabB as a determinant of UFA synthesis, and the potent regulation of the fabB gene by FabR underscores a central role for this transcription factor in controlling membrane fluidity in bacteria that produce UFA using the FabA/FabB combination. However, FabR cannot accomplish this regulation without the participation of FadR. The fact that the UFA content is not increased in a fadR fabR strain compared with a fadR strain illustrates that the level of fabA expression, supported by the transcriptional activation by FadR, is required for UFA regulation by FabR. In the absence of FadR activation, we conclude that FabA activity is limiting for UFA synthesis, whereas in the presence of FadR activation, the level of FabB activity controls UFA production.

There are important questions that remain concerning the role of the FabR repressor in cell physiology. One important aim will be to determine the mechanism that controls its binding to DNA. FabR is a member of the TetR superfamily (Pfam00440 (37)) of bacterial regulatory proteins, and, by analogy to the known mechanisms used by these factors, it seems reasonable to postulate that a regulatory ligand binds to the carboxyl-terminal domain of the protein to release the amino-terminal DNA-binding domain from its target sequence. FadR has a carboxyl-terminal domain that is homologous to the regulatory domains of the TetR family members (14). The FadR factor controls fabA and fabB expression in response to long-chain acyl-CoA derived from exogenous fatty acids (9, 17), and it is tempting to speculate that FabR is controlled by a ligand that is an integral component of endogenous de novo fatty acid biosynthesis. A pathway intermediate such as a saturated long-chain acyl-ACP is an example of such a candidate ligand. However, we cannot rule out other mechanisms, such as phosphorylation. Another part of the FabR story that deserves attention is defining what mechanisms control the expression of fabR. Possibly, the fabR gene is controlled by another factor that ties UFA synthesis to other aspects of bacterial physiology in a fashion analogous to the control of acetate metabolism by FadR by virtue of its ability to control the expression of iclR, a transcriptional regulator of the glyoxylate bypass (38).

The Affymetrix array analysis extended the number of genes directly controlled by FadR. Previous work has cataloged a substantial list of genes controlled by FadR that are either related to fatty acid metabolism or survival in stationary phase (20, 39, 40). Our experiments do not address stationary phase, but the new information in this report expands the repertoire of -oxidation genes subject to FadR regulation. The fadH and fadE genes are involved in -oxidation, and the prediction that they would be derepressed by FadR deletion is confirmed in this study (Table IV). We also found two new genes, yfcX and yfcY, that are also predicted to be involved in fatty acid -oxidation (Table IV). The yfcX gene encodes a protein with predicted multifunctional capabilities (enoyl-CoA hydratase, -hydroxyacyl-CoA dehydrogenase, and -hydroxybutyryl-CoA epimerase) and is thus similar in function to FadB and termed -oxidation complex II. The difference between the two complexes is that complex I contains 3-cis-2-trans-enoyl-CoA isomerase activity, which is predicted to be absent from complex II. The yfcY gene is predicted to be a -ketoacyl-CoA thiolase based on its similarity to fadA and is thus termed thiolase II. The reason for these two sets of similar -oxidation genes is unknown but may reflect differences in substrate specificity of the respective enzymes. Both of these genes have putative FadR binding sites located in regions downstream of the predicted transcriptional start site, consistent with a role for FadR in repressing their expression. Further genetic and biochemical characterization of these two genes will be required to establish their substrate specificity and roles in fatty acid -oxidation.

	ACKNOWLEDGEMENTS

We thank Matt Frank and Amy Sullivan for expert technical assistance and Divyen Patel and the Hartwell Center staff for assistance with the Affymetrix array analysis.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant GM34496, Cancer Center (CORE) Support Grant CA 21765, and the American Lebanese Syrian Associated Charities.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Both authors contributed equally to the work.

To whom correspondence should be addressed: Dept. of Infectious Diseases, Protein Science Division, St. Jude Children's Research Hospital, 332 N. Lauderdale, Memphis, TN 38105-2794. Tel.: 901-495-3491; Fax: 901-495-3099; E-mail: charles.rock@stjude.org.

Published, JBC Papers in Press, February 21, 2002, DOI 10.1074/jbc.M201399200

² The fatty acids are abbreviated as: 18:111, number of carbon atoms:number of double bonds; , the location of the double bond.

	ABBREVIATIONS

The abbreviations used are: ACP, acyl carrier protein; FabA, -hydroxydecanoyl-ACP dehydratase; FabB, -ketoacyl-ACP synthase I; FabF, -ketoacyl-ACP synthase II; SFA, saturated fatty acid; UFA, unsaturated fatty acid; FadR, fatty acid degradation regulator; and FabR, fatty acid biosynthesis regulator.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				L. A. Newberg, W. A. Thompson, S. Conlan, T. M. Smith, L. A. McCue, and C. E. Lawrence A phylogenetic Gibbs sampler that yields centroid solutions for cis-regulatory site prediction Bioinformatics, July 15, 2007; 23(14): 1718 - 1727. [Abstract] [Full Text] [PDF]

				Y. Tian, P. Jackson, C. Gunter, J. Wang, C. O. Rock, and S. Jackowski Placental Thrombosis and Spontaneous Fetal Death in Mice Deficient in Ethanolamine Kinase 2 J. Biol. Chem., September 22, 2006; 281(38): 28438 - 28449. [Abstract] [Full Text] [PDF]

				R. I. Clavijo, C. Loui, G. L. Andersen, L. W. Riley, and S. Lu Identification of Genes Associated with Survival of Salmonella enterica Serovar Enteritidis in Chicken Egg Albumen Appl. Envir. Microbiol., February 1, 2006; 72(2): 1055 - 1064. [Abstract] [Full Text] [PDF]

				S. Conlan, C. Lawrence, and L. A. McCue Rhodopseudomonas palustris Regulons Detected by Cross-Species Analysis of Alphaproteobacterial Genomes Appl. Envir. Microbiol., November 1, 2005; 71(11): 7442 - 7452. [Abstract] [Full Text] [PDF]

				M.-M. Sampaio, F. Chevance, R. Dippel, T. Eppler, A. Schlegel, W. Boos, Y.-J. Lu, and C. O. Rock Phosphotransferase-mediated Transport of the Osmolyte 2-O-{alpha}-Mannosyl-D-glycerate in Escherichia coli Occurs by the Product of the mngA (hrsA) Gene and Is Regulated by the mngR (farR) Gene Product Acting as Repressor J. Biol. Chem., February 13, 2004; 279(7): 5537 - 5548. [Abstract] [Full Text] [PDF]