Biosynthesis of Membrane Lipids
JOHN E. CRONAN, JR.,1* AND CHARLES O. ROCK2
[SECTION
EDITOR:
JOHN E. CRONAN, JR.]
Posted October 7, 2008
Departments of Microbiology and Biochemistry, University of Illinois, Urbana, IL 61801,1 and Protein Science Division, Department of Infectious Diseases, St. Jude Children's Research Hospital, Memphis, TN 381012
*Corresponding author. Mailing address: Department of Microbiology, University of Illinois, B103 CLSL, 601 S. Goodwin Avenue, Urbana, IL 61801. Phone: (217) 333-7919, Fax: (217) 244-6697, E-mail:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
There have been a number of major advances in our knowledge of lipid synthesis in Escherichia coli and (largely by analogy) Salmonella enterica since we reviewed the subject in the last print predecessor of EcoSal (55). First, that review was written just prior to the release of the genome sequences of E. coli K-12 and S. enterica serovar Typhimurium LT2. Hence, in the last edition we could not state with confidence that all of the enzymes of a protein family had been found by genetics or biochemistry. With the genome sequences in hand, the possible players are known. Second, at the time of the prior review, our knowledge of the structural biology aspect of lipid synthesis was virtually nil; the only structures available were those of various acyl carrier protein (ACP) species obtained by NMR. We now have crystal structures of each of enzymes of the fatty acid synthetic cycle plus those of the enzymes of unsaturated fatty acid synthesis (for a recent review, see reference 300). In several cases structures of the unliganded protein and of the protein liganded with a substrate and/or inhibitor are available. We also have significant, albeit incomplete, structures of the enzymes that catalyze reactions ancillary to saturated and unsaturated fatty acid synthesis such as holo-ACP synthase, the FadR regulatory protein, the four acetyl-CoA carboxylase subunits, and BirA biotin protein ligase, the enzyme responsible for essential attachment of biotin to the AccB subunit of acetyl-CoA carboxylase. Moreover, where we lack direct evidence, the structures of related enzymes from other organisms often provides useful hypotheses that are readily tested in E. coli and S. enterica. The third advance is in the genetics of lipid synthesis. This has profited from the major advances in E. coli and S. enterica genetic technology resulting from the exploitation of the phage λ homologous recombination proteins and of several site-specific recombination systems that allow manipulation of chromosomal sequences at an unprecedented scale. These methods have permitted the dependence of growth on function of the genes of lipid synthesis to be tested with much greater rigor than previously possible and have resulted in the isolation of mutants in genes (e.g., acpP) (62) where conventional approaches had been fruitless for decades. There are now physiologically useful mutants available in each of the genes of the fatty acid synthetic cycle (except fabZ) and unsaturated fatty acid synthesis and in most of genes encoding the ancillary enzymes. Fourth, advances in mass spectroscopy have allowed very accurate and detailed analyses of lipid compositions as well as detection of the interactions of lipid biosynthetic proteins with one another and with proteins outside the lipid pathway (22, 101, 102). Moreover, mass spectroscopy has recently provided values for protein molecules/cell for most of the fatty acid synthetic proteins (192). Finally, roles for bacterial fatty acids, other than as lipid structural components, have been uncovered. Some of these, such as lipoic acid synthesis (Chapter Biotin and Lipoic Acid: Synthesis, Attachment, and Regulation), are found in common laboratory strains of E. coli and S. enterica, whereas others, such as acylation of protein toxins, are found only in certain pathogenic strains of these bacteria. Still other roles for bacterial fatty acids, such as synthesis of acyl-homoserine quorum-sensing molecules and polyhydroxyalkanoates, are not found in E. coli and S. enterica. However, upon introduction of the relevant gene(s), synthesis of these foreign molecules readily proceeds in E. coli and S. enterica. The combination of these advances have resulted in use of E. coli and S. enterica for discovery of new antimicrobials targeted to lipid synthesis (144, 295, 316, 317) and in deciphering the molecular actions of known antimicrobials. Because of this knowledge explosion, the bacterial lipid field and, in particular, the fatty acid synthesis pathway have been extensively reviewed in recent years and the present authors have frequently been authors of these publications. The topics of our reviews include the structural biology and enzymology of fatty acid synthesis (300), antimicrobial agents that target fatty acid synthesis (23, 121, 122, 314), and the functions of membrane lipids (42, 313). Given space considerations and the fact that our recent reviews have appeared for the most part in widely available publications, we will often refer to these rather than repeat our prior analyses. However, we will focus on E. coli and S. enterica and will add any new information that has become available.
The compositions of the membrane lipids of E. coli and S. enterica are among the simplest found in biology. These bacteria contain three major phospholipids: phosphatidylethanolamine (PE), phosphatidylglycerol (PG), and cardiolipin (CL) (also called diphosphatidylglycerol). Traces of other phospholipids (e.g., monoacylphosphatidylglycerol, phosphatidylserine [PS]) are also found. Phospholipids contain about 90% of the total fatty acyl groups of E. coli, the remainder being found in the lipid A component of the lipopolysaccharide, in lipoproteins, and in traces of neutral lipids (diglyceride and free fatty acids).
PE, the major phospholipid of E. coli and S. enterica (and of gram-negative bacteria in general), comprises about 75% of the total phospholipid. The relative amounts of the other lipids (PG and CL) depend on the growth phase of the cultures; PG is dominant in log-phase cells whereas CL is dominant in stationary-phase cells. The phospholipids, neutral lipids, and lipoproteins of E. coli have similar acyl chain compositions consisting of the saturated fatty acids, palmitic (hexadecanoic) acid, and myristic (tetradecanoic) acid plus the monounsaturated fatty acids, palmitoleic (cis-9-hexadecenoic) acid and cis-vaccenic (cis-11-octadecenoic) acid. Traces of other fatty acids (lauric [dodecanoic] acid, stearic [octadecanoic] acid, and cis-7-tetradecenoic acid) are also present in the phospholipids. The fatty acid compositions of E. coli cultures depend on both growth temperature and growth phase. At 37°C, the phospholipids of a typical E. coli strain contain about 45% palmitic acid, 2% myristic acid, 35% palmitoleic acid, and 18% cis-vaccenic acid. However, in cultures grown at 25°C, the percentages of palmitic acid and cis-vaccenic acid are reversed. This temperature-dependent compositional alteration (thermal regulation) is thought to be a mechanism to optimize membrane lipid fluidity. The mechanism of this compositional change is understood in some detail and will be discussed below. It should be noted that very old cultures of E. coli were reported to contain linoleic (cis-9-cis-12-octadecdienoic) acid (233) but (in agreement with the older literature) we were unable to detect this fatty acid in the same E. coli K-12 strain (52).
The variation of fatty acid composition with growth phase is due to the conversion of palmitoleic acid and (to a lesser extent) cis-vaccenic acid to their cyclopropane derivatives by addition of a methylene group (from S-adenosylmethionine) across the double bond of the phospholipid-bound fatty acyl moiety (see below). This membrane bilayer modification reaction has received some recent attention and will also be discussed below. It should be noted that most workers in the field have used E. coli and thus more compositional data are available for this bacterium. However, in cases in which S. enterica was examined, the two organisms seem identical. Therefore, information obtained with one organism can be assumed to be directly applicable to the other. The analytical data on the lipid compositions of these organisms were essentially complete over 25 years ago and thus we will defer to previous reviews for the original references (44, 57).
All of the phospholipids, neutral lipids, and lipoproteins of E. coli and S. enterica are membrane components; no cytoplasmic, extracellular, or periplasmic lipid-containing structures are known. Both the phospholipids and the neutral lipids are distributed between the inner (cytoplasmic) membrane and the outer membrane. The outer membrane contains roughly one-third of the total phospholipid and has been reported to consist largely or exclusively of PE. In contrast, the phospholipids of the inner membrane have been reported to contain roughly equal amounts of PE and of the two acidic lipids, PG and CL. It should be noted, however, that although the outer membrane has generally been reported to be enriched in PE relative to the cytoplasmic membrane, the magnitude of the enrichment varies greatly among reports. It seems reasonable to accept the more extreme values because such values are generally found in the more carefully documented reports and because the lack of enrichment can be attributed to transfer of lipids between the two membranes during cellular lysis. Phospholipid fatty acid compositions of the two membranes are similar, although the inner membrane seems somewhat enriched in unsaturated fatty acids. This enrichment can probably be attributed to the increased concentration of PG in the cytoplasmic membrane, which contains a greater fraction of unsaturated fatty acid than do the other phospholipids. There is some evidence for flow of lipid between the inner and outer membranes of E. coli, although the transfer mechanism remains a black box (130, 165, 222). Although phospholipid export to the outer leaflet of the inner membrane is partially blocked in MsbA-deficient cells, decreased MsbA function does not affect export to the outer membrane (70).
The enzymes of lipid biosynthesis are distributed between the cytosol and the inner membrane. The fatty acid synthetic enzymes plus PlsX and G3P synthase (GpsA) are cytosolic, whereas the enzymes of phospholipid synthesis are firmly bound to the inner membrane. Three exceptions to these generalizations have been reported. First, the ACP, which carries the growing fatty acid chain and donates the completed acyl chain to G3P, was reported to be localized to the inner face of the cytoplasmic membrane on the basis of autoradiography of intact cells (287). However, a more refined colloidal gold-antibody technique has demonstrated that ACP is distributed throughout the cytoplasm of E. coli (137). The second exception was the observation that the phospholipid biosynthetic enzyme, PS synthase, was found associated with the ribosomes rather than with the inner membranes of disrupted cells. Subsequent work has shown this association to be an artifact of cell disruption. In the presence of the lipid substrate of the enzyme, CDP-diglyceride, the enzyme is released from the ribosomes and binds to the inner membrane lipid bilayer. These findings are referenced in our first chapter in the print predecessor of EcoSal (56). The most recent exception is that of PlsX, a functional component of the PlsX-PlsY acyltransferase. PlsX is purified as a soluble protein, although in vivo imaging suggests that it is associated with the membrane in vivo (223) (see below).
Lipid synthesis in E. coli and S. enterica may be conveniently divided into three sections: (i) the synthesis of the fatty acids that are responsible for the characteristic hydrophobicity of lipids, (ii) the attachment of the completed fatty acids to water-soluble phosphorylated polyols in the first steps in the synthesis of phospholipids (and lipid A), and (iii) modification of the acylated polyols to give the major cellular lipids. The synthesis of lipid A will be addressed elsewhere in this volume and we will restrict our discussion to the synthesis of the membrane phospholipids. However, it should be noted that the fatty acid synthetic pathway is required for the synthesis of essential molecules other than membrane lipids such as the protein-bound coenzymes, lipoic acid, and (probably) biotin (Chapter Biotin and Lipoic Acid: Synthesis, Attachment, and Regulation). In other gram-negative bacteria fatty acid synthesis appears to be required for the synthesis of the acylated homoserine lactones involved in density-dependent signaling (87). Moreover, the synthesis of polyketide antibiotics and poly-β-hydroxybutyrate proceeds by modified fatty acid pathways.
The mechanism of fatty acid synthesis is strongly conserved in procaryotes and eukaryotes (the archaea synthesize isoprenoid-derived lipids) and proceeds in two stages: initiation and cyclic elongation. Intermediates of the pathway are diverted to introduce the double bond of the unsaturated fatty acid species, to provide the 3-hydroxy and medium-chain fatty acids of lipid A (236), the octanoyl-ACP used in the lipoic acid synthetic pathway, and the unknown early intermediates of the biotin synthetic pathway (Chapter Biotin and Lipoic Acid: Synthesis, Attachment, and Regulation. A key feature of the fatty acid synthetic pathway is that all of the intermediates are covalently bound to ACP, a small, very acidic, and extremely soluble protein (231). The carboxyl groups of the fatty acyl intermediates are in thioester linkage to the thiol of the 4'-phosphopanthetheine (4'-PP) prosthetic group which in turn linked to Ser-36 of ACP through a phosphodiester bond. These ACP thioesters are the substrates for the enzymes of the pathway. ACP is one of the most abundant proteins in E. coli and constitutes about 0.25% of the total soluble protein (6 × 104 to 8 × 104 molecules per cell) (192, 244). Indeed, ACP is reported to be the third most abundant protein in E. coli (being slightly exceeded by only two translation apparatus proteins, RplL and TufB) (192). The secondary structure predicted from the amino acid sequence (245) has been largely confirmed by high-resolution nuclear magnetic resonance (NMR) spectroscopy and X-ray crystallography. ACP (molecular weight [MW] 8,860) is composed of a preponderance of acidic residues largely grouped into three α-helices (helices I, II, and IV) oriented in an up-down-down topological arrangement to form a helical bundle plus a short fourth helix (helix III) that seems of lower stability and is found both almost parallel and almost perpendicular to the three-helix bundle in the various structures now available. The loop connecting helices I and II is quite long (17 residues) relative to the loops connecting the other helices. There is good agreement between the NMR and crystal structure data and both approaches indicate that the protein can attain a variety of related conformations. A difficulty in comparing the structures in detail is that we lack a crystal structure of holo-ACP (the species with the prosthetic group, but lacking an acyl chain), although the crystal structure of apo-ACP (the form lacking the prosthetic group) in the presence of high concentrations of zinc has been solved (232). However, we can extend our analyses by considering the structures of ACPs from other organisms that are known to functionally replace that of E. coli in vivo (63). A crystal structure of Bacillus subtilis holo-ACP in a complex with its cognate AcpS 4'-PP transferase and a high-resolution NMR structure of B. subtilis holo-ACP are useful in this regard (224). The two holo-ACP structures are quite similar, although in the complex with AcpS the long loop region connecting the first two ACP helices becomes partially structured into a short helix. However, modeling of the E. coli ACP NMR data indicates that the data are best fit by the assumption that ACP exists in at least two conformations that differ in the structures of the loop regions and by some distortions of the helices (157, 158). More recent NMR structures of the Plasmodium falciparum apicoplast ACP confirm that ACP exists as two structural isomers that interconvert slowly on the NMR timescale (~microseconds) (258). Probably because of technical advances, these are the first NMR data in which interactions between the prosthetic group and protein residues (other than the modified serine) have been observed. Indeed, the two structures differ primarily in the orientation of the 4'-PP prosthetic group. In the minor form, reorientation of the prosthetic group results in loss of the short third helix. The major (about 65%) form of P. falciparum ACP, called the open form, has a structure similar to the X-ray crystal structure of butyryl thioester of E. coli ACP (250), whereas the minor form has a more closed conformation. Conversion of the closed (minor) form to the open (major) form is thought necessary to give the extra flexibility needed to insert the acyl chain into the hydrophobic pocket predicted from prior studies of E. coli ACP. Finally, two high-resolution crystal structures of the Aquifex aeolicus ACP have recently been released (PDB codes 2EHS and 2EHT) that show different orientations of helices II and IV and the loop regions. These seem to be apo-ACP structures formed in the presence of Zn ions.
The structural plasticity of ACP seen in the absence of acylation of the prosthetic group thiol is also seen, albeit to a lesser extent, in the acylated forms. The crystal structures of a series of acyl-ACPs having saturated acyl chains from C4 to C10 have been solved at high resolution (251). As in the case of P. falciparum ACP two different binding modes for the 4'-PP prosthetic group are seen. In one form, the structure is stabilized mainly through hydrophobic interactions between the hydrocarbon portion of the acyl chain and residues of the hydrophobic cavity primarily formed by helices II, III, and IV plus neighboring loop residues. In this form, the pantoic acid and β-alanine moieties of the prosthetic group are exposed to the solution and are highly mobile such that they are not observed in the crystal structures. In contrast, the prosthetic group of the other acyl-ACP form has multiple contacts with the protein throughout its length and the acyl chain is thrust much deeper into the protein resulting in sequestration of the thioester in the hydrophobic core of the protein. The two structures are not dependent on crystallization conditions since both may be found in the same crystal (251). Thus, these structures probably reflect two of the perhaps many acyl-ACP conformations that exist in solution and seem likely to be related to the open and closed forms of P. falciparum ACP discussed above. In should be noted that although these authors ascribe a role to the hydroxyl of Thr39 of helix II in formation of the buried prosthetic group form, this seems unlikely since several of the foreign ACPs that functionally replace E. coli ACP in vivo (63) have an isoleucine at that position. The longer chain lengths seem to favor the second more structured form consistent with the finding that the longer the acyl chain the greater the stability of the acyl-ACPs to high pH-induced unfolding and thioester cleavage (45, 244). These structures also show that the volume of the hydrophobic cavity increases with acyl chain length by outward movement of helices I and III and the N-terminal end of helix II. Hence, the protein swells to accommodate longer acyl chains (251).
Our current picture is that ACP and its acylated derivatives can adopt many differing structures by sliding the helices relative to one another and by rearranging the loops and prosthetic group. This may allow acyl groups to slide in and out of the hydrophobic cavity such that a compromise is achieved between shielding the acyl chain from solvent and allowing access to the thioester-proximal acyl carbon atoms such that fatty acid synthesis can proceed. If so, the dynamics of this process will depend on the polarity and length of the acyl chain and on interactions with the other fatty acid synthetic proteins.
The prosthetic group of ACP undergoes metabolic turnover (141, 229, 230, 272) and the apo-protein is not only inactive in fatty acid synthesis but at high levels can be growth inhibitory (155). The primary enzyme catalyzing attachment of the prosthetic group is AcpS (61, 83, 227) although a back-up enzyme, AcpT, is also present (61, 83). AcpS and AcpT are 4'-phosphopantetheine transferases (4'-PP transferases) that transfer the 4'-phosphopantetheine portion from CoA to apo-ACP to give holo-ACP plus 3',5'-ADP. Although the known 4'-PP transferases comprise a single protein superfamily, the secondary structures of these proteins vary from monomers to trimers with E. coli AcpS reported to be dimeric (170, 202). Unlike the enzymes that attach biotin and lipoic acid to their cognate proteins, the 4'-PP transferases are, in general, quite specific for their protein substrates (169). Although E. coli AcpS attaches the prosthetic group to many fatty acid ACPs, some other fatty acid ACPs are not substrates (63) and it fails to modify the structurally related carrier proteins of polyketide and nonribosomal polypeptide synthesis (170). However, B. subtilis Sfp, which is responsible for modification of the carrier protein of surfactin, a nonribosomal peptide, readily modifies most carrier proteins regardless of origin or metabolic role (170).
Surprisingly, all known 4'-PP transferases accept acylated-CoA substrates and transfer acyl-4'-PP moieties to apo-ACP. E. coli AcpS utilizes acetyl-ACP in place of ACP at about half the catalytic efficiency and is also active with butyryl-, acetoacetyl-, and malonyl-CoAs (202). Again, Sfp is much more promiscuous and transfers acyl-4'-PP moieties modified with molecules such as biotin and fluorescein (206). It is unclear whether or not transfer of acyl-4'-PP moieties is physiologically relevant. However, the low levels of AcpS in E. coli argue that acyl-4'-PP transfer should not be able to replace a fatty acid synthetic enzyme such as FabD.
The second E. coli 4'-PP transferase, now called AcpT, was discovered by bioinformatic analyses. Lambalot and coworkers (169) performed homology searches of the E. coli K-12 genome using two small motifs conserved among PPTases that were subsequently shown to be involved in CoA binding (224, 242). Purified AcpT gave only barely detectable modification of E. coli AcpP in vitro (although it was not clear whether the protein had regained full enzymatic activity following its refolding in vitro) (169). Upon overexpression from high-copy-number plasmids AcpT complemented an E. coli acpS mutant (61, 83). Recent work indicates that AcpT can be largely (but not entirely) considered as an evolutionary relic (61). AcpT modifies two carrier proteins encoded in O-island 138, a cluster of fatty acid biosynthesis-like genes located adjacent to acpT found in the genome of the pathogenic E. coli strain O157:H7 but not in E. coli K-12. The two carrier proteins of O-island 138 of strain O157:H7 are not modified (or very poorly modified) by AcpS. AcpT is not an essential gene and cannot functionally replace AcpS in E. coli K-12 either in its native chromosomal location or upon insertion of acpT into the chromosomal location of acpS. However, in the absence of AcpS activity AcpT does allow very slow growth of E. coli, thus providing a rationale for its retention in E. coli K-12 in the absence of its cognate O-island 138 substrates (61). These results, together with phylogenetic analyses and comparisons of the E. coli and Shigella strains of known genome sequence, strongly argue that AcpT has been orphaned from its cognate substrates by a deletion event that occurred in a common ancestor of these organisms. This is one of the few cases where a chromosomal rearrangement has been functionally demonstrated to be a deletion event rather than an insertion event in the reference organism (61).
Soon after the identification of the prosthetic group of ACP as 4'-PP and of the AcpS 4'-PP transferase, Vagelos and Larrabee (284) reported an E. coli activity that removed the 4'-PP moiety of ACP. The enzyme (called ACP hydrolyase and ACP phosphodiesterase) was partially purified and shown to require a divalent ion and produce apo-ACP and 4'-PP. The enzyme was inactive on peptide fragments of ACP suggesting that it recognized the folded structure of ACP. Although crude preparations were occasionally used to convert holo-ACP to apo-ACP, this paper remained the only study of this enzyme for over 20 years until Fischl and Kennedy (82) reported purification of the enzyme to apparent homogeneity. These workers showed that ACP phosphodiesterase was an unusually stable enzyme of MW 25,000 and reported an N-terminal sequence attributed to the enzyme protein. Unfortunately, the N-terminal sequence was found to be that of a flavin-containing protein that lacked phosphodiesterase activity (D. Keating and J. Cronan, unpublished work; T. Stachelhaus and C. Walsh, personal communication) and which was subsequently shown to be an azoreductase (134, 215). It is clear that Fischl and Kennedy had determined the sequence of a major contaminating protein rather than that of the phosphodiesterase. Regrettably, based on this erroneous sequence attribution, many genes (often called acpD) have been annotated in bacterial genomes as encoding ACP phosphodiesterase rather than azoreductase (215). In order to prevent confusion with the mistaken AcpD annotations, the gene was named acpH on the basis of the original enzyme name (ACP hydrolyase) given by Vagelos and Larrabee (284). The acpH gene was recently isolated by expression cloning and was found to encode a protein of 23 kDa that readily aggregates upon overexpression (272, 273). Active enzyme was recovered by folding solubilized inclusion bodies and was found to be active on acyl-ACPs of fatty acyl chain lengths from C6 to C16. AcpH was active on the ACP of B. subtilis but inactive on that of Lactococcus lactis. Strains carrying deletions of acpH are fully viable, but the ACP prosthetic group is metabolically stable unlike that of wild-type strains (272). Upon AcpH overproduction, all of the cellular ACP is converted to the apo form (272). AcpH is not essential in the natural habitat because several strains of the shigellae (which are generally considered to be strains of E. coli) carry acpH genes that contain a frameshift mutation early in the coding sequence (272). AcpH has been shown to be a noncanonical member of the HD phosphatase/phosphodiesterase family (273).
The phosphopantetheine released by AcpH may be recycled into CoA biosynthesis via the phosphopantetheine adenylyltransferase or excreted into the medium (Chapter Biosynthesis of Pantothenic Acid and Coenzyme A). The CoA pool is approximately 8- to 10-fold larger than the ACP pool in normally growing cells) (Chapter Biosynthesis of Pantothenic Acid and Coenzyme A) and virtually all of the ACP is maintained in the active, holo-form in vivo (141, 155). Thus, the supply of prosthetic group does not limit fatty acid biosynthesis. The operation of the prosthetic group turnover cycle appears to be involved in governing the intracellular CoA concentration (Chapter Biosynthesis of Pantothenic Acid and Coenzyme A). During logarithmic growth, a significant pool of nonacylated holo-ACP can be found in vivo (83, 141, 145, 272). Inhibition of acyl-transfer, however, causes accumulation of acyl-ACPs and nonacylated ACP is no longer detectable (145, 249). Factors that restrain either chain elongation or fatty acid transfer to the membrane bilayer may cause the supply of ACP protein to be limiting, but the size of the ACP pool must be severely depleted before an effect on fatty acid and phospholipid synthesis can be detected (141, 227). The cellular concentration of ACP protein is controlled and massive overproduction of ACP is encoded by an inducible plasmid vector lethal to E. coli (155, 240) due to the close matching of AcpS and ACP levels in wild-type cells (83). The deleterious effects of ACP overexpression seem difficult to reconcile with the high levels of the protein normally present. However, most of the protein expressed in the inducible systems is apo-ACP (155), leading to the hypothesis that apo-ACP binds to and competitively inhibits the enzymes of lipid biosynthesis. Indeed, apo-ACP is a potent G3P acyltransferase inhibitor in vitro (155). ACP may play other roles in cell physiology. Proteomic studies argue that ACP has more interactions with other proteins than any other single E. coli protein (22). The only rivals in terms of the numbers of interacting partners are all large complexes containing multiple proteins: DNA polymerase III, RNA polymerase, and the ribosome. The list of proteins that interact with ACP included not only the expected lipid synthetic proteins, but also proteins involved in diverse cellular functions (22). Some of the diverse proteins reported were: MukB, a protein involved in partition of chromosome between daughter cells, SecA, a key protein translocation component, and SpoT, a protein required for adaptation of cell physiology to nutrient limitation (22). Moreover, in vitro ACP had previously been reported to be an essential cofactor for synthesis of a periplasmic oligosaccharide (270, 271), to stimulate the nicking reaction of transposon Tn3 (195), and to give improved binding of Tn7 transposase to its target (259).
Acyl-ACP has also been reported to be an acyl donor in protein acylation. The hemolysin toxin of E. coli is made as a protoxin, ProHlyA, which is matured in the cytosol to the active form (HlyA) by acylation at two internal lysine residues (133). This acylation is catalyzed by HlyC with the acyl donor being acyl-ACPs carrying saturated acyl chains of which lauroyl- and myristoyl-ACPs are the most effective substrates in vitro (303, 304, 305). HlyC lacks the usual acyltransferase active site motifs and its mechanism remains uncertain (304, 305). HlyC forms an acyl-enzyme intermediate, but the nature of the linkage between acyl group and protein has not been determined. Strikingly, the toxin isolated in vivo has been reported to contain myristate (68%), n-pentadecanoic (26%), and n-heptadecanoic (6%) amide-linked side chains (183). This implies that these cells synthesize saturated chains with an odd numbers of carbon atoms.
The fatty acid synthesis system of E. coli is the archetype of the Type II or dissociated fatty acid synthesis systems. The precursors for fatty acid biosynthesis are derived from the acetyl-CoA pool. Malonyl-CoA is required for all the elongation steps and is formed by the first committed step in fatty acid biosynthesis, acetyl-CoA carboxylase. E. coli acetyl-CoA carboxylase is composed of four individual proteins: biotin carboxylase, biotin carboxyl carrier protein, and two subunits required for the carboxyltransferase step (Fig. 1). Malonyl-CoA is utilized for fatty acid biosynthesis only following its conversion to malonyl-ACP by malonyl-CoA:ACP transacylase (Fig. 2).
There are three known mechanisms for the initiation of fatty acid biosynthesis in E. coli (Fig. 2). First, FabH (3-ketoacyl-ACP synthase III) catalyzes the condensation of acetyl-CoA with malonyl-ACP to yield acetoacetyl-ACP. In the second pathway, the acetate moiety is first transferred from acetyl-CoA to acetyl-ACP by FabH. The acetyl-ACP is then condensed with malonyl-ACP by FabB (synthase I) or alternatively by FabF (synthase II). The third pathway involves the decarboxylation of malonyl-ACP by FabH, FabB, or FabF to form acetyl-ACP followed by subsequent condensation with malonyl-ACP. The evidence for the existence of these pathways and their relative contributions to the initiation of fatty acid biosynthesis is an area of current interest and is reviewed in more detail below.
The reactions of the fatty acid elongation cycle are outlined in Fig. 3. The first step is the condensation of malonyl-ACP with a growing acyl chain by a 3-ketoacyl-ACP synthase (either FabB or FabF). This is the only irreversible step in the elongation cycle, and thus it is not surprising that the 3-ketoacyl-ACP synthases play key roles in regulating the product distribution of the pathway. The resulting 3-keto-thioester is reduced by FabG, an NADPH-dependent 3-ketoacyl-ACP reductase, followed by removal of a water molecule by FabZ, a 3-hydroxyacyl-ACP dehydratase. The final reduction is catalyzed by FabI, an enoyl-ACP reductase, to form acyl-ACP, which in turn can serve as a substrate for another round of elongation (or if of sufficient chain length be transferred into complex lipids). The equilibrium of the FabI enoyl-ACP reductase reaction acts to pull the reversible 3-ketoacyl-ACP reductase and 3-hydroxyacyl-ACP dehydratase reactions to the right (114).
Two of the reactions of the cycle can be carried out by multiple discrete enzymes. As noted above there are three 3-ketoacyl-ACP synthases and two 3-hydroxyacyl-ACP dehydratases are also present. Due to their differing substrate specificities, each isozyme makes a unique contribution to the regulation of the distribution of products from the pathway (see below).
A specific dehydratase, 3-hydroxydecanoyl-ACP dehydratase (the fabA gene product), first described by Bloch and coworkers (19), catalyzes a key reaction at the point that unsaturated fatty acid biosynthesis branches from saturated fatty acid synthesis (Fig. 4). The FabA dehydratase catalyzes the dehydration reaction shown in Fig. 3, but it also isomerizes trans-2-decenoyl-ACP to cis-3-decenoyl-ACP (19). Thus, the FabA dehydratase is essential to the synthesis of unsaturated fatty acids. However, this protein is not the only gene product required for unsaturated fatty acid synthesis in E. coli. 3-Ketoacyl-ACP synthase I mutants (fabB) show that FabB is also required to produce unsaturated fatty acids.
The fatty acid biosynthetic pathway ends in the transfer of the acyl chains of the acyl-ACP end products into membrane phospholipids by a G3P acyltransferase system. The first acyltransferase reaction involves transfer of fatty acids to the 1-position of G3P, whereas the second transfer is catalyzed by a separate enzyme that esterifies the 2-position of the glycerol moiety. Like most phospholipids in nature, bacterial phospholipids have an asymmetric distribution of fatty acids between the 1- and 2-positions of the G3P backbone that is controlled in part by the acyl chain specificity of the acyltransferase steps. Although, the G3P acyltransferase system is a component of fatty acid biosynthesis, acyltransferase activity affects both the chain length distribution of the fatty acids found in membrane phospholipids and the rate of fatty acid biosynthesis (see below).
Enzymes catalyzing each of the reactions of the pathway have been demonstrated in E. coli, and each of the proteins has been purified to homogeneity and crystal structures are available for most of the proteins. The physiological roles of virtually all of the activities have now been established by genetic means. We will divide the enzymes into those of the initiation reaction, those of the elongation cycle, and those required to introduce the cis double bond, although there is considerable overlap. Both in E. coli and in other organisms, the protein names derived on the E. coli gene names have largely supplanted the classical names based on the reactions catalyzed. Therefore, we will generally use the gene-based protein names.
Acetyl-CoA carboxylase (ACC) catalyzes the first committed step of fatty acid (hence lipid) synthesis and its activity limits the rate of fatty acid synthesis (60).
A recent review of the multisubunit ACS that is focused on the E. coli enzyme is available (53). The overall ACC reaction is composed of two distinct half-reactions that require a biotin cofactor. ACC function requires that biotin be covalently coupled to a small (16.7 kDa) protein called AccB and (in the older literature) biotin carboxyl carrier protein (3, 53). The attachment reaction is catalyzed by a specific enzyme, biotin-apoprotein ligase (see below). AccB is the sole biotinylated protein in E. coli, whereas S. enterica has a second biotinylated protein, an inducible oxalacetate decarboxylase (301, 302).
The two ACC partial reactions are catalyzed by two different protein subcomplexes. Carboxylation of biotin is catalyzed by biotin carboxylase, a homodimeric enzyme composed of 55-kDa subunits encoded by accC. AccC copurifies in an unstable complex with AccB (31, 217). The complex has recently been shown to contain two molecules of AccB and four molecules of AccB, precisely the ratio of expression of the two proteins (31). The activity that transfers the carboxy group from the biotin moiety of AccB to acetyl-CoA is the carboxyltransferase component, a heterotetramer composed of two copies of each of two sequence-related (but separately encoded) subunits, called α and β. Subunit α is encoded by accA whereas accD encodes subunit β. The AccA and AccD proteins seem likely to have arisen by a gene duplication event. In cell extracts the overall ACC reaction (acetyl-CoA to malonyl-CoA) is lost and only the separate AccB-AccB and carboxyltransferase components are detected (3, 98). The overall ACC reaction can be reconstituted in vitro using high concentrations of the purified components (98) or by simultaneous overexpression of all four subunits (60). This result suggests that the AccB-AccB and carboxyltransferase components may not form a tight complex in vivo. We assume that the functional enzyme present in vivo is composed of one copy of each component (i.e., two molecules of each of the AccA, AccC, and AccD subunits and four molecules of AccB) and has a molecular mass of ca. 300 kDa. This assumption is based on results obtained with the acetyl-CoA carboxylase of Pseudomonas citronellolis (80). This enzyme can be readily isolated in the presence of high salt concentrations as a complex of about 280 kDa containing equal amounts of the four subunits, whereas low salt concentrations give two components analogous to E. coli. However, excepting the stable association in high salt, the P. citronellolis subunits are similar to those of E. coli and the AccB components of the two organisms are interchangeable in both in vitro half-reactions catalyzed by the carboxylase proteins of either organism (80). Moreover, the genes encoding the AccB protein of Pseudomonas aeruginosa complements an E. coli accB mutant and an open reading frame just downstream of the accB gene is almost certainly that of AccC (16).
All four acetyl-CoA carboxylase subunits have been purified to homogeneity (53). All four genes are essential (7, 94) and temperature-sensitive mutants in accB and accD are available (181, 182). AccB and the AccC biotin carboxylase are encoded in a small operon (180, 181). The active form of the AccB protein requires covalent attachment of biotin to lysine 122 of the protein, the reaction catalyzed by biotin-apoprotein ligase (Chapter Coenzymes, Cofactors, and Prosthetic Groups) (11). Biotin-apoprotein ligase is encoded by the birA gene and also is the repressor of the biotin biosynthetic operon (Chapter Coenzymes, Cofactors, and Prosthetic Groups) (11). The dual nature of the BirA protein results in the regulation of the rate of biotin biosynthesis by the rate of AccB synthesis (Chapter Coenzymes, Cofactors, and Prosthetic Groups) (11). The N-terminal half of AccB is not required for biotinylation of the protein and the first 20 residues are required for association with biotin carboxylase (31, 181). Mutational analysis of AccB and related proteins (53) has shown that essentially the entire C-terminal half of AccB is required for the modification of the protein by biotin ligase, although the biotinylated lysine is located in the midst of this minimal sequence (35 residues from the C terminus). Moreover, the C-terminal half of AccB has long been know to be very resistant to proteolysis consistent with a stable well-folded domain structure (81). These findings indicate that biotin ligase recognizes the structural domain rather than a linear sequence of amino acid residues (Chapter Coenzymes, Cofactors, and Prosthetic Groups). Biotinylation must be a posttranslational event because some C-terminal protein extensions inhibit biotinylation, a result incompatible with cotranslational modification. The biotinylation of the AccB biotin domain has been shown to stabilize the domain by interactions between the attached biotin moiety and a structured loop (called the thumb) that distinguishes the acetyl-CoA carboxylase biotin domains from those of other biotinylated enzyme proteins (47, 266).
The structure of the C-terminal segment of AccB has been determined by both X-ray and two-dimensional NMR analyses (53). The structures of both biotinylated and non-biotinylated forms of the 86 C-terminal residues were solved. The structures of both forms are essentially identical, although the apo structure is much more dynamic in solution (53). The protein is folded in two four-stranded antiparallel β-sheets. The biotin is attached to the ε-amino group of a conserved lysine residue located at the tip of a β-hairpin loop that protrudes from the compact globular domain. The BCCP domain structure is strikingly similar to the structures of domains modified with lipoic acid, although the amino acid sequences have <15% identity (Chapter Coenzymes, Cofactors, and Prosthetic Groups). Lipoyl domains are located at the N termini of the transferase subunits of the pyruvate and 2-oxoglutarate dehydrogenases, key enzymes in tricarboxylic acid cycle function (Chapter Tricarboxylic Acid Cycle and Glyoxylate Bypass). The lipoate moiety is attached to a specific lysine residue and shuttles acetate or succinate units from one active site to another. Both biotin and lipoic acid are classical examples of the "swinging arms" of metabolism first envisioned by Lynen as movement of the coenzyme plus the amino acid side chain between two active sites (225). Recent work indicates that Lynen’s idea was essentially correct, but overly conservative; the entire domain swings on the proline/alanine-rich hinges (located upstream of biotin domains and downstream of the lipoyl domains [225]). The finding of very similar structures in enzymes of differing amino acid sequence that catalyze very dissimilar reactions illustrates the unity of biology. Moreover, the linker region found downstream of the lipoyl domains can be substituted for the linker region found upstream of the AccB biotin domain with retention of AccB function (53).
The E. coli AccC biotin carboxylase is among the most intensively studied of the biotin-dependent enzymes because of its unusual ability to accept free biotin in place of the AccB-bound form (53). This property, although very inefficient and of no physiological consequence, greatly simplifies analysis of intermediates and allows the facile use of biotin analogs in enzymological investigations. Despite these advantages and recent structures obtained by X-ray crystallography (53), the mechanism of biotin carboxylation remains elusive (6, 53, 146). Carboxylation is thought to involve reaction of ATP and CO2 to form carboxyphosphate, an intermediate with an estimated half-life of 70 ms, which either reacts directly with biotin or collapses to CO2 and phosphate with subsequent attack of biotin on the now localized CO2 (6, 53, 146). The resolution of the current X-ray crystallographic structures does not permit definitive assignment of the active-site residues, but does show the enzyme to be a symmetrical dimer of a monomer composed of three distinct structural domains and has defined the ATP-binding residues and shows that, upon ATP binding, the central domain rotates relative to the other domains thereby closing off the active-site pocket (53). The ATP-AccC complex also shows that AccC has considerable structural homology with carbamyl phosphate synthase as predicted from sequence alignment (181) and shared mechanistic properties. Recent work has shown that monomeric derivatives of AccB constructed by mutagenesis of the dimer interface retain enzyme activity in vitro (260). However, these proteins are inactive in vivo (A. Smith and J. E. Cronan, unpublished data), presumably due to an inability to form the functional ACC complex. A crystal structure of the carboxytransferase component has recently been reported (17). Unfortunately, the structure is of fairly low resolution and thus not particularly informative. However, the structure does show that the carboxytransferase is a member of the crotonase superfamily and that the enzyme contains a zinc binding domain, a novel feature for a member of this superfamily. The role of the Zn binding domain is not clear (17).
The conversion of malonyl-CoA to malonyl-ACP is catalyzed by malonyl-CoA:ACP transacylase, the product of the fabD gene (197, 289). FabD is a monomeric protein that accepts the malonyl moiety from malonyl-CoA to form a stable malonyl enzyme intermediate in which the malonyl moiety is in ester linkage to a serine hydroxyl (221, 255). Nucleophilic attack of this intermediate by the sulfhydryl of ACP gives malonyl-ACP, the major building block of fatty acids. Crystal structures of FabD, FabD complexed with malonate, and the FabD-malonyl-CoA complex have been reported (221, 255). It should be noted that the original fabD89 temperature-sensitive allele is an amber mutation (290). The temperature sensitivity results from the amino acid inserted by a suppressor tRNA present in the original strain. It should be noted that several polyketide ACPs have been reported to self-malonate (4, 5, 127). That is, the malonyl moiety is transferred from CoA to ACP in the absence of malonyl-CoA:ACP transacylase. This has repeatedly been reported not to occur with E. coli ACP (127). However, self-malonylation was recently reported to occur with of P. falciparum ACP (208), one of the foreign ACPs that functionally replaces E. coli ACP in vivo (63). However, high-level expression of P. falciparum ACP in an E. coli FabD mutant failed to complement the mutation, indicating that the ability to self-malonate cannot functionally replace FabD (A. Smith and J. E. Cronan, unpublished data).
In contrast to the reactions that produce malonyl-ACP, the reactions whereby the methyl carbon atom and the immediately adjacent carbon atom (the last two carbons of the fatty acid chain by chemical nomenclature) are incorporated into fatty acid remain somewhat unclear. Isotopic labeling studies demonstrate that these "primer" carbons are largely derived directly from acetate (299). However, the possibility remains that some acetate enters by an indirect route such as via conversion to malonyl-ACP and subsequent decarboxylation of the malonyl-ACP to acetyl-ACP. The acetate source is acetyl-CoA produced mainly by the decarboxylation of pyruvate, although other sources are available depending on physiological conditions. Acetyl-CoA is a substrate for FabH (3-ketoacyl-ACP synthase III, which has also been called acetoacetyl-ACP synthase) and is incorporated directly to form the first four-carbon fatty acyl-ACP species (140, 278). Acetyl-CoA can also be converted into acetyl-ACP by a transacylase activity and the resulting acetyl-ACP can serve as the primer since alternative condensing enzymes such as FabB and FabF catalyze the initial condensation. For many years, the acetyl-CoA:ACP transacylase activity in E. coli was considered to be a discrete protein of rather low in vitro activity. However, the acetyl-CoA:ACP transacylase reaction was subsequently shown to be catalyzed by FabH (140, 278), and thus the acetyl transacylase activity measured previously may represent a side reaction of this 3-ketoacyl-ACP synthase (FabB and FabF have a similar activity [2]). An enzyme that was purified from E. coli based on its transacylase activity has properties quite similar to that of FabH (191). Another possibility is that the activity observed is that of a promiscuous thiolase as was reported in Streptomyces collinus (188).
In E. coli malonyl-ACP is utilized only in the elongation steps in fatty acid biosynthesis. However, FabH, FabB, and FabF are capable of initiating fatty acid synthesis in vitro in the absence of an added acetyl-ACP or acetyl-CoA as the primer. This synthesis occurs through a side reaction; malonyl-ACP is decarboxylated to produce acetyl-ACP. This reaction is readily demonstrated in vitro (2, 204), but its role in initiation in vivo awaits experimental verification. However, overproduction of synthase I renders E. coli resistant to the antibiotic thiolactomycin (278). A fabB point mutation has the same effect (142). Under these conditions it appears that FabB is the only 3-ketoacyl-ACP enzyme required for growth, which suggests that initiation of fatty acid synthesis by malonyl-ACP decarboxylation could be the major route utilized in the presence of this antibiotic and perhaps in other physiological conditions. Malonyl-CoA accumulates in the presence of cerulenin, but not when fatty acid synthesis is blocked later in the lipid synthetic pathway (at the G3P acyltransferase step) implicating a cerulenin-sensitive enzyme (FabB or FabF) in the degradation of malonyl-CoA (117). The degradation pathway presumably entails conversion of malonyl-CoA to malonyl-ACP followed by decarboxylation to acetyl-ACP. The question of whether one or several routes are used to initiate fatty acid synthesis remains an open question. Experiments in which cultures were labeled with deuterated acetate indicate that about 80% of the fatty acid methyl groups retain three deuterium atoms and thus cannot be the result of malonyl-CoA synthesis followed by malonyl-CoA decarboxylation because deuterium atoms would be lost to solvent. The origin of the remaining 20% of the methyl groups was obscured by the inclusion of a secondary carbon source in the growth medium (299).
Although FabH is well studied biochemically, genetic analyses of its function are complicated by the appearance of suppressor mutations. An E. coli strain carrying two chromosomal copies of fabH (one copy was inserted into a phage attachment site) and then disrupted the copy within the fab gene cluster. It was found that the second copy could not be removed (either by phage Int/Xis-catalyzed excision or by transduction with a closely linked Tn10). In contrast, if the fabH gene of the fab cluster was intact, then the second copy was readily deleted (168). However, in subsequent work using conditions that allowed prolonged incubation of plates, large colonies eventually appeared on a lawn of pinpoint colonies at a frequency consistent with a point mutation. These are bypass suppressors since the nonfunctional fabH gene remains. These suppressors are the subject of ongoing study. The "usual suspects" for suppressor mutations would be the other two 3-ketoacyl-ACP synthases, FabB and FabF, because of the known malonyl-ACP decarboxylase activity of these enzymes that produces acetyl-ACP. Moreover, FabB and FabF are known to condense acetyl-ACP with malonyl-ACP to form 3-ketobutyryl-ACP, the product of FabH. However, an activity that only catalyzes decarboxylation of malonyl-ACP would suffice, and since malonyl-ACP is an inherently unstable compound, this seems a relatively simple chemical task. It should be noted that the appearance of the fabH suppressors is of more than academic interest since FabH is thought to be an excellent target for antimicrobial and antimalarial drugs (295, 306). If FabH function can readily be bypassed, this will lower the efficacy of these drugs. The workers that constructed the Keio collection of E. coli null mutants reported fabH to be nonessential. However, upon transduction of the ▵fabH::kan lesion of this strain back into its parental strain, the tiny colony phenotype is seen. Therefore, the ▵fabH::kan strain of the Keio collection seems to have acquired a bypass suppressor mutation during its characterization. Other workers have reported that fabH is an essential gene (94). However, this report is based on the loss of fabH transposon insertions from a pool of transposon insertion mutants grown for many generations. Hence, the poor growth rate of fabH null mutants would result in their loss from the pool before suppressor mutations could arise. Finally, the malonyl-ACP decarboxylase activities of FabB and FabF involve the same active-site residues as the synthase reaction (2, 59, 204), and thus, elimination of this source of acetyl-ACP is problematical. Another possible source of acetyl-ACP would be transfer of acetyl-4'-PP from acetyl-CoA to apo ACP catalyzed by AcpS (see above). However, the intracellular level of AcpS activity seems far too low to provide the needed carbon flow.
The first reaction of each cycle is the condensation of an acyl thioester (acyl-ACP or for FabH, acetyl-CoA) with malonyl-ACP to form a 3-ketoacyl-ACP plus ACP. Three E. coli enzymes are known to catalyze this 3-ketoacyl-ACP synthase reaction. These enzymes were referred to as synthases I, II, and III but more recently have come to be called FabB, FabF, and FabH, respectively, after their gene names. The fabH and fabF genes are located within the fatty acid synthetic gene cluster (240), whereas fabB maps alone at a distant site. Siggard-Anderson and coworkers (264) reported a putative fourth KAS activity in E. coli and assigned an open reading frame to this activity. However, this report was based on a series of indirect inferences and was in error. The gene sequenced was the fabF gene as previously noted (196, 240), and the enzyme activity described seems likely to be a mixture of FabH with FabB and/or FabF. The three E. coli 3-ketoacyl-ACP synthases represent two classes of decarboxylating Claisen condensing enzymes (119, 300). The FabB and FabF enzymes have Cys-His-His active sites, whereas FabH has a Cys-His-Asn active site triad. These differences are reflected in the rest of the primary sequences of the proteins. FabB and FabF are about 37% identical, whereas alignment of either FabB or FabH with FabH gives only scattered alignments of very low quality. High-resolution crystal structures of all three enzymes are available (300).
The functional FabB is composed of two identical subunits (93) and contains both malonyl-ACP and fatty acyl-ACP binding sites (59). In the condensation reaction, the acyl group becomes covalently linked to the sulfhydryl of a cysteine residue of the enzyme (59). The acyl-enzyme undergoes condensation with malonyl-ACP to form 3-ketoacyl-ACP, CO2,holo-ACP, and free enzyme. Inhibition studies using cerulenin (see below) show the active-site cysteine to be Cys-163 (153).
Further investigation revealed the presence of an additional synthase activity in E. coli, FabF (58). Like FabB, FabF has a dimeric structure and is inhibited by cerulenin (283), although FabF is less sensitive to the antibiotic than is FabB. Both synthases are both capable of participating in saturated and unsaturated fatty acid synthesis. The enzymes have been shown, in vitro, to function similarly with all substrates except palmitoleoyl-ACP; palmitoleoyl-ACP is an excellent substrate for FabF, but not for FabB (78, 93). This observation is consistent with the role of FabF in the regulation of fatty acid composition of the membrane phospholipid in response to temperature (see below). Mutant strains having fabB missense or null mutations (49, 92), however, require unsaturated fatty acids for growth; therefore, in vivo FabB must catalyze a key reaction in unsaturated fatty acid synthesis that FabF cannot. This reaction is probably the elongation of cis-3-decenoyl-ACP, although this has not been shown experimentally. This step is the rate-limiting step in unsaturated fatty acid synthesis (33).
FabB overproduction has two known effects. First, overproduction of the enzyme overcomes its poor ability to elongate palmitoleoyl-ACP and an increased amount of cis-vaccenic acid is incorporated into phospholipid (65). The increase, however, has no effect on the temperature regulation of fatty acid composition (see below). Second, excess cellular FabB renders E. coli resistant to the antibiotic thiolactomycin (see below). In early studies, Alberts et al. observed that FabB catalyzed malonyl-ACP decarboxylation at high enzyme concentrations (2); this reaction has been explored in some detail (204). It was thought that this reaction offered the cell an alternative initiation pathway for fatty acid biosynthesis; the decarboxylation of malonyl-ACP results in the production of acetyl-ACP that can subsequently be used as a primer for chain elongation. With the isolation of thiolactomycin-resistant strains that have an altered FabB this hypothesis has been borne out in vivo (142). In the presence of the antibiotic, excess FabB appears to allow the cell to bypass the standard FabH initiation pathway by catalyzing the decarboxylation of malonyl-ACP to form the initiation primer, acetyl-ACP. Given the above observations, FabB appears to be the only E. coli 3-ketoacyl-ACP synthase absolutely required for growth (142, 278).
FabF was first detected as a peak of activity that was resolved from FabB by hydroxyapatite chromatography (58). The protein was purified to homogeneity and shown to differ from FabB by peptide mapping and antigenicity (93). The fabF gene was identified by its role in the temperature regulation of fatty acid composition (see below). Mutants lacking temperature control (called Cvc) lack FabF activity (92, 93). Reversion of a fabF mutation results in restoration of synthase II activity, cis-vaccenic acid synthesis, and temperature regulation (92, 93). FabF, therefore, was firmly established as playing an essential role in the thermal regulation of the fatty acid composition of E. coli (283).
Strains harboring a temperature-sensitive mutation in the fabB gene and an additional mutation in the fabF gene fail to synthesize any long-chain fatty acids at the nonpermissive temperature (92). Even supplementation with oleate, an unsaturated fatty acid that allows growth of fabB unsaturated fatty acid auxotrophs, fails to permit growth of a double fabB(Ts) fabF mutant at high temperature due to an inability to synthesize long-chain saturated fatty acids (92, 140). FabB and FabF are, therefore, the only E. coli synthase activities active in the synthesis of long-chain fatty acids. Both FabB and FabF can catalyze the synthesis of saturated fatty acids, but FabB is specifically required for the synthesis of unsaturated fatty acids. In addition, both enzymes play a major role in controlling fatty acid chain length in E. coli (see below).
The investigation of FabF was hampered by an inability to clone the intact fabF gene (196). One part of this puzzle was the finding that the gene encoding ACP, acpP, is adjacent to the fabF locus (240). Since overproduction of ACP is toxic to the cell (155, 240), this probably precluded isolation of the fabF gene from clone banks. However, directed cloning of fabF also failed, although the gene sequence was obtained by cloning chromosomal fragments and PCR products (196). The fabF gene was subsequently cloned in a very stringently controlled expression vector, although the clones were quite unstable (78). Upon induction of FabF expression from this expression plasmid fatty acid synthesis immediately ceased with a concomitant >80-fold increase in the intracellular concentration of malonyl-CoA (268). This inhibition was partially relieved by simultaneous overexpression of FabD, the enzyme catalyzing the conversion of malonyl-CoA to malonyl-ACP, which supports a model in which high levels of FabF block the access of FabB to FabD (268).
FabH is a dimeric protein of 33.5 kDa first detected as a condensation activity resistant to cerulenin both in vivo and in vitro (140). Although cerulenin blocks the synthesis of long-chain fatty acids, short-chain (C4 to C8) acids linked to ACP are accumulated both in vivo and in cell extracts. The fabH gene has been shown to encode KAS III. As noted above, the FabH sequence has no similarity to those of FabB or FabF, although there is good alignment with other enzymes known to catalyze condensation reactions. From the chain length of the acyl-ACPs produced and the behavior of fabB fabF double mutants mentioned above, it is clear that FabH does not participate in the progressive condensation steps of fatty acid synthesis that produce the long-chain acids found in the phospholipids and lipid A. However, the enzyme could produce the fatty acid synthetic intermediates used in lipoic acid (and perhaps biotin) synthesis. However, the high activity of the enzyme suggests that it does not exclusively function in this pathway (E. coli requires only a few hundred lipoic acid molecules per cell).
An open reading frame encoding a protein with strong similarities to several acetoacetyl-CoA reductases and particularly plant 3-ketoacyl-ACP reductases (>50 % identical residues) is found between the fabD and acpP genes in the fab gene cluster (240). This gene is designated as fabG and is obligatorily cotranscribed with the upstream genes (309, 310). Blocking fabG transcription blocked cell growth indicating that it is an essential gene (310). However, no mutants in this gene were available until the recent isolation of temperature-sensitive mutants of both E. coli and S. enterica (167). Strikingly, all of the mutations were located in or near the subunit interfaces of the FabG homotetramer suggesting that monomers and dimers of the enzyme are inactive. At the nonpermissive temperature fatty acids synthesis was blocked in the fabG(Ts) mutants following the initial condensation such that only four carbon acyl-ACP species accumulated. Thus, only a single NADPH-specific 3-ketoacyl-ACP reductase exists in E. coli and it functions with all acyl chain lengths (167). Crystal structures of FabG are available (300).
3-Hydroxyacyl-ACP dehydratase is not to be confused with the 3-hydroxydecanoyl-ACP dehydratase specifically required for introduction of the double bond of the unsaturated acids. Prior biochemical data gave a puzzling picture of this reaction. One group reported that this step is catalyzed by a single enzyme active with substrates of all chain lengths (18), whereas another laboratory reported the presence of three enzymes specific for short-, medium-, and long-chain-length substrates (209). More recent work with recombinant FabZ showed that the dehydratase efficiently catalyzed the dehydration of short-chain 3-hydroxyacyl-ACPs and long-chain saturated and unsaturated 3-hydroxyacyl-ACPs (118). The fabZ gene was isolated as a suppressor of a mutation in lipid A biosynthesis. The suppression is thought to be caused by increased intracellular levels of 3-hydroxymyristoyl-ACP (210). Unfortunately, the phenotype of the original mutant strain is too feeble for physiological studies. However, better mutants have recently been isolated (N. De lay, J. He, and J. E. Cronan, unpublished data). Although no structure of E. coli FabZ is presently available, several FabZ homologue structures have been reported, the most closely related of which is that of the P. aeruginosa protein (159). This FabZ shows sequence homology to FabA (which was a key to its discovery) and thus it was no surprise that the monomers have a hot-dog fold and the active sites lie at the dimer interface. FabZ has a Glu residue in place of the active-site Asp residue of FabA. This substitution was thought to play an important role in the differing specificities of the two enzymes. However, the discovery of a FabZ homologue that has FabA activity and has a Glu active-site residue (294) indicates that this simple hypothesis is incorrect.
Two forms of enoyl-ACP reductase, the last enzyme of the fatty acid cycle, were originally reported, one dependent on NADH and the other on NADPH (297). However, both activities were subsequently shown to be due to the same enzyme, FabI (14), the sole enoyl-ACP reductase of E. coli (114). The identification of the protein encoded by this gene, now called fabI, was the result of studying mutants of E. coli and S. enterica resistant to diazaborines, a class of potent antimicrobial agents that inhibit lipid synthesis (280). It has been shown that the plant enzyme expressed in a single copy from the fabI promoter can replace the fabI gene in E. coli fatty acid synthesis and give diazaborine-resistant growth (152). FabI is also the target of the widely used antimicrobial, triclosan (120, 205). FabI plays a role in the fatty acid synthesis cycle in that its action pulls the other reversible steps of the cycle (FabG, FabZ) to the right such that each cycle of fatty acid biosynthesis is completed (114). Several FabI crystal structures (300) and a useful fabI temperature-sensitive mutant (15) are available. It should be noted that E. coli FabI has been reported to be inhibited by palmitoyl-CoA (14). Although this has been ascribed a physiological role, it remains to be demonstrated that the observed inhibition is not due to the well-known detergent properties of long-chain acyl-CoAs.
A structure of 2-dodecenoyl-ACP cocrystallized with E. coli FabI has recently been reported (237). Unfortunately, the resolution of structure was such that it yielded little direct information and much of the proposed structure was based on model building and molecular dynamics simulations. Indeed, only parts of the ACP backbone and none of the side chains were visible in the X-ray structure and these were modeled in using the butyryl-ACP structure. Moreover, neither the prosthetic group nor the attached fatty acid was visible and hence these were also modeled into the structure. These manipulations plus the dynamic nature of ACP structure results in a quite speculative model. Given this caveat, the two proteins are reported to interact via a very small interface composed of electrostatic interactions between three acid residues of ACP helix 2 and three basic residues adjacent to the FabI substrate-binding loop. Although, this interface is similar in composition to that formed between ACP and AcpS it is much smaller and thus seems less stable. Although mutation of the implicated FabI basic residues did result in increased Michaelis constants for 2-dodecenoyl-ACP, some of these mutations had only modest effects and others did not conform to the purely electrostatic interactions proposed, in that substitution of alanine for a putatively key lysine residue had a greater effect than substitution with a glutamate residue. An interesting feature of the crystal structure is that only one of the two FabI monomers had an associated ACP suggesting that only one of the two subunits of the FabI homodimer is active at a time (237).
3-Hydroxydecanoyl-ACP dehydratase essentially extracts a major fraction of the 3-hydroxydecanoyl-ACP from the elongation cycle, introduces a cis double bond, and then, following action of FabB, these modified acyl chains are returned to the cycle for elongation to the long-chain unsaturated acids needed for phospholipid function (19). 3-Hydroxydecanoyl-ACP dehydratase specifically catalyzes the dehydration of 3-hydroxydecanoyl-ACP to a mixture of trans-2-decenoyl-ACP and cis-3-decenoyl-ACP (19). The reaction proceeds via the formation of trans-2-decenoyl-ACP as an enzyme-bound intermediate that can disassociate from the enzyme (19). When disassociation occurs, the trans-2 intermediate is reduced by an enoyl-ACP reductase and subsequently converted to saturated fatty acids as in the standard elongation cycle (33, 99, 118). Enzyme-bound trans-2-decenoyl-ACP, however, is isomerized to cis-3-decenoyl-ACP. The double bond is preserved and the cis-3 intermediate is elongated to the unsaturated fatty acids of E. coli, palmitoleic acid, and cis-vaccenic acid. The enzyme, a homodimer of 18-kDa subunits (176), is distinct from the elongation cycle FabZ dehydratase discussed above, although the two enzymes contain sequence similarities.
The first mutants isolated that were blocked in fatty acid biosynthesis, called fabA, lacked 3-hydroxydecanoyl-ACP dehydratase (265). These mutants are unable to synthesize unsaturated fatty acids but synthesize saturated fatty acids normally. In vitro, mutant fabA enzymes could form neither the cis-3 nor trans-2-decenoyl products (50). This finding, along with the observation that saturated fatty acid synthesis continues in vivo, indicated that another dehydratase was available for saturated fatty acid synthesis, the enzyme now called FabZ that catalyzes the formation of trans-2-decenoyl-ACP but cannot catalyze the isomerase reaction (118). In the absence of thermal regulation, the ratio of unsaturated to saturated fatty acids in E. coli is dependent on the levels of FabA and FabB. It was shown that overproduction of FabA in vivo failed to increase the level of unsaturated fatty acids but significantly increased the amount of saturated fatty acids incorporated into membrane phospholipids (33). This indicated that, although FabA is required for the synthesis of unsaturated fatty acids, the level of enzyme activity does not limit the rate of unsaturated fatty acid synthesis. Introduction of multiple copies of the fabB gene (encoding synthase I) reversed the effect of dehydratase overproduction resulting in wild-type fatty acid compositions (32). Thus, the step more likely to limit the rate of unsaturated fatty acid synthesis is the elongation of cis-3-decenoyl-ACP catalyzed by FabB. The levels of expression of the fabA and fabB genes, therefore, appear to establish a basal ratio of unsaturated to saturated fatty acid synthesis in the absence of thermal regulation. Modulation of the fatty acid composition of membrane phospholipid in response to temperature shift is discussed below.
FabA functions with acyl chain lengths from C4 to C12 in an in vitro fatty acid synthesis system reconstituted from purified enzymes (118). This is a considerably wider range of FabA substrates than originally reported in studies using synthetic model substrates (19). The reasons for this disagreement are unclear, but it should be noted that the ratio of products produced by the enzyme using acyl-ACP substrates (99) differs from that obtained using model substrates (19).
The mechanism of FabA has been thoroughly studied (19, 253) since it was the first enzyme for which a mechanism-activated ("suicide") inhibitor was described. This inhibitor, 3-decynoyl-N-acetylcysteamine, forms a covalent adduct with the active-site histidine residue resulting in loss of all of the partial reactions of the enzyme (19, 51, 176). The X-ray crystal structure of E. coli FabA was the first structure determined for an E. coli fatty acid synthetic protein. FabA was the first example of the "hot dog" fold in which a long central α-helix is wrapped by a six-stranded antiparallel β-sheet (176). FabA forms an unusually stable dimer and it seems that the dimerization is essential for activity since the two active sites are formed along the dimer interface with the critical His and Asp active-site residues being contributed by different monomers (176).
The genes encoding the enzymes catalyzing each of the steps of fatty acid synthesis are now known. The known genes are scattered about the genome with only two clusters, the minimal accBC operon (180, 181) and the fab cluster, containing the plsX fabH, fabD, fabG, acpP, and fabF genes (240, 309, 310). This latter clustering of genes, first observed by genetic means, seems to have functional significance since several genes are cotranscribed. However, this gene cluster is not transcribed as a classical operon, but rather multiple promoters are present and each gene is cotranscribed with at least one other gene (309, 310). Transcripts specific for single genes are also present (309, 310). Transcription of acpP is from at least two different promoters, and the major promoter has been mapped. Since ACP is such an abundant protein, it seemed possible that multiple promoters could be necessary to produce this mass of protein. However, the promoter located immediately upstream of the coding sequence was sufficient. Approximately 60% of the plsX transcripts were found to initiate at promoters located far upstream of the gene and to include the upstream rpmF ribosomal protein gene (309, 310). Indeed, a promoter located upstream of the plsX coding sequence (probably within rpmF) is sufficient for normal growth. The fabG gene is obligatorily cotranscribed with upstream genes, and fabD was found to have only a very weak promoter (51). The gene (rpmF) upstream of plsX encodes ribosomal protein L32, whereas the gene downstream of fabF is pabC, a gene of p-aminobenzoate synthesis, and thus the extent of the fab gene cluster is delimited. Cotranscription of PlsX with fatty acid synthetic genes seemed an enigma when first observed. However, the observation that the role of PlsX is in acyl chain metabolism (transacylation of fatty acid chains in preparation for an acyltransferase reaction) shows that it is functionally related to the downstream genes.
The regulation of E. coli acetyl-CoA carboxylase is unclear. The accB and accC genes are cotranscribed from a promoter located unusually far upstream of the accB gene (180, 181). Recent work has shown that this gene arrangement serves the purpose of preventing nonstoichiometry of the two subunits which disturbs both lipid synthesis and the regulation of biotin synthesis (1). Although RNA polymerase initiating transcription at this promoter traverses a region of DNA reported to attain a static curve (bent DNA) and a small ORF just upstream of accB, deletion of either or both of these features had no effect on accBC transcription (180). The major accA promoter lies within the coding sequence of the polC (dnaE) gene, although transcription through polC and perhaps other upstream genes also reads through the accA sequence (180). The significance of this transcription pattern is presently unclear. The accD gene is transcribed from a promoter located within the upstream dedA gene (180). Transcription of all four acc genes is under growth rate control, the rate of transcription decreasing with decreased growth rate (180). However, the situation is complex in that the accBC operon seems regulated by a mechanism different from that which regulates the accA and accD genes (180). Introduction of many copies of the accBC operon results in only modest increases (<3-fold) in accBC transcription and protein products, whereas similar experiments with the accA and D genes give the expected overproduction of transcription and translation products. Moreover, replacement of the accBC promoter with the heterologous lac promoter gives the expected overproduction of gene products (181). More recent work has shown that AccB functions to negatively regulate transcription of the accBC operon (143). Expression of a chimeric protein consisting of the N-terminal half of E. coli AccB and the C-terminal biotinylation domain of B. subtilis AccB down-regulated transcription of the E. coli accBC operon. A truncated form of AccB consisting of only the N-terminal 68 amino acids of E. coli AccB was sufficient to negatively regulate the accBC operon. In vivo bypass of acetyl-CoA carboxylase activity by expression of a malonyl-CoA synthase from Rhizobium trifolii allowed construction of strain deleted for the accA and accB genes. Unexpectedly, the ▵accB mutation could not be resolved from the ▵accA mutation. Transcription of the accBC operon in the ▵accB ▵accA strain continued well into stationary phase under growth conditions that normally resulted in greatly decreased transcription. These data support a model in which AccB acts as an autoregulator of accBC operon transcription the mechanism of which remains unclear (143).
The accBC genes are also implicated in regulation at diverse growth temperatures, although this may only be an aspect of the growth rate control. Karow and coworkers studied a collection of null mutants (transposon insertions) unable to grow at elevated temperatures (151). One of these, called htrB, fails to grow at temperatures >33°C. Many second-site suppressors of htrB that allow growth at 42°C have been isolated, and one class of these suppressors maps in the accBC genes (151). These mutants are thought to decrease the function of AccB and/or biotin carboxylase since one such suppressor is an IS1 insertion into the DNA segment coding the leader mRNA. This insertion blocks transcription from the normal accBC promoter but provides weak transcription from an outward-reading IS1 promoter (151).
The fabB, fabI, and fabA genes are not linked to other lipid synthetic genes, and the first three genes seem to be transcribed from unique promoters as monocistronic mRNAs (24, 123, 124, 153). In contrast, fabZ lies within a complex cluster of genes involved in lipid A biosynthesis (210). Transcription of both fabA and fabB is positively regulated by FadR protein and negatively regulated by FabR protein.
FadR protein was first discovered as a repressor regulating the fatty acid degradation (fad) regulon of E. coli that includes genes of β-oxidation and fatty acid transport (24). Mutants having a defective fadR gene are constitutive for β-oxidation, and such mutants grow on short-chain (<C10) fatty acids that normally fail to induce the regulon (Chapter Two-Carbon Compounds and Fatty Acids as Carbon Sources). The function of FadR in fad gene regulation is straightforward and follows the E. coli lactose repressor paradigm. In cells growing in the absence of fatty acids, FadR binds to operator sites upstream of the fad gene coding sequences and represses transcription of these genes. Exogenous fatty acids enter the cell and are converted to acyl-CoA thioesters which bind to FadR. When complexed to acyl-CoA, FadR disassociates from the operators resulting in transcription of the fad regulon genes and hence β-oxidation (Chapter Two-Carbon Compounds and Fatty Acids as Carbon Sources). This view of FadR function in E. coli was expanded by the finding that FadR acts as a positive activator in the transcription of a fatty acid synthetic gene, fabA.
The first indication of the dual role of FadR in fatty acid metabolism was the finding that introduction of a fadR mutation into a conditional (temperature-sensitive) fabA mutant strain converted the strain to a nonconditional phenotype (219). Strains additionally defective in β-oxidation did not alter the fabA phenotype, indicating that the effect on fabA was due to the fadR mutation per se rather than to induction of the β-oxidation pathway (219). A second indication of a role for FadR in fatty acid synthesis was the increased sensitivity of fadR strains to a specific inhibitor of the FabA enzyme (33). Subsequent analyses showed that fadR strains had decreased unsaturated fatty acid contents relative to isogenic wild-type strains, indicating that functional FadR was necessary for either fabA gene expression and/or enzyme function. The former hypothesis was shown to be correct by Henry and Cronan (123, 124), who demonstrated that FadR is a positive transcriptional activator of fabA expression. The level of fabA gene expression is decreased 12-fold in a fadR null mutant. This regulation accounts for the decreased unsaturated fatty acid content of fadR strains as well as the unexpected double mutant and inhibitor results mentioned above. The explanation for these findings is that E. coli can tolerate the somewhat lowered unsaturated fatty acid content resulting either from decreased fabA gene expression or a partially defective FabA enzyme but not the greatly decreased unsaturated fatty acid content resulting from decreased expression of a partially defective enzyme.
In fadR null mutants the fabA gene is transcribed from two weak promoters of about equal strength, whereas in wild-type strains a 20-fold increase in transcription from the proximal promoter is seen (69, 123, 124). FadR binds to a 17-bp DNA sequence located in the −40 region of this promoter (69, 123, 124). This binding site is located at the position most often used by transcriptional activators of sigma 70 promoters. The DNA sequence is similar to those found within the promoters of two fad genes, fadBA and fadL, where FadR acts as a repressor. How can FadR protein act as both a transcriptional activator of fatty acid synthesis and a repressor of the fatty acid degradation regulon? Several other examples of activator proteins that also act as repressors (generally in autoregulation) are known. The distinction between these two roles usually depends on the location of the protein binding site relative to the transcriptional start. Most sigma 70-dependent promoters have their activator sites positioned such that the bound protein overlaps position −40, whereas DNA binding proteins act as repressors when positioned within a larger downstream region of the promoter, generally between −30 and +10 (36). This simple scheme readily explains FadR action. The binding site for fabA activation is centered at −40, while the fadBA and fadL binding sites (where FadR represses transcription) are centered at +9 and −17, respectively. Thus, by analogy to similar systems we expect that FadR binding to the fabA DNA aids RNA polymerase binding or action via protein-protein interactions. Likewise, FadR binding to the fad regulon operators would hinder the binding or action of RNA polymerase. Given this latter role it seems surprising that FadR fails to repress its own synthesis. It should be noted that in vitro experiments by DiRusso and coworkers (68, 69) demonstrated that purified FadR activates the proximal fabA promoter and represses the fadBA and fadL promoters.
Transcriptional activation of fabA gene expression is inhibited by fatty acids in vivo; this is due to decreased activity of the proximal promoter (123). Moreover, fatty acyl-CoAs inhibit the binding of FadR to the −40 region of the proximal promoter and the acyl chain lengths of the acyl-CoAs effective in FadR release from the DNA accurately reflect those of the fatty acids effective in decreasing fabA expression in vivo (123). A similar pattern was seen for the induction of the β-oxidation genes. Thus, FadR seems to monitor the intracellular concentration of long-chain acyl-CoA molecules and coordinately regulates fatty acid synthesis and oxidation in response to the levels of these compounds. FadR has recently been shown to bind acyl-CoA by direct techniques, and mutants defective in acyl-CoA binding have been isolated and characterized (238, 239). As expected, acyl-CoA is ineffective at removing these mutant proteins from the operator sites.
The amino acid sequence of FadR predicted a helix-turn-helix motif common to DNA-binding proteins and it was assigned to the winged helix-turn-helix family of regulatory proteins. The FadR binding site of fadBA possesses nearly perfect dyad symmetry, which suggested that FadR is homodimeric in structure, and this has been demonstrated by several techniques (it was first reported that FadR is monomeric). Crystal structures of FadR alone and in complexes with operator DNA or with myristoyl-CoA have been reported (285, 286, 307). Agreement among these structures is good and provides a straightforward and pleasing mechanism for FadR regulation. FadR is a variant type of helix-turn-helix motif called a winged helix motif, the wing generally being a small β-sheet that interacts nonspecifically with the minor groove of the operator DNA while the recognition helix of the helix-turn-helix motif makes sequence-specific interactions in the major groove. However, in the case of FadR the wing makes a sequence-specific interaction deep in the minor groove and uses only part of the recognition helix to form specific contacts. Binding of the long acyl chain of an acyl-CoA results in very significant reordering of the helices of a hydrophobic pocket present in the C-terminal domain of the protein, and this reordering is communicated by reorientation of the helix-turn-helix structure such that the two recognition helices move well away from the positions that allow binding to DNA (285, 286, 307). This, together with a shift in the orientation of the wing, results in loss of DNA binding by FadR.
The unraveling of the FadR regulation has given the first example of a regulatory protein that positively regulates the biosynthesis of a molecule and negatively regulates the catabolism of the same family of molecules (123, 124). FadR regulation of the fabA gene is also the first report of a repression system mediated by positive control (123). In the absence of fatty acyl-CoAs, FadR binding enables RNA polymerase to function in fabA transcription, whereas, in the presence of fatty acyl-CoAs, FadR is released from the DNA and fabA transcription decreases.
A plausible rationale for the physiological relevance of this regulatory system seems apparent. A major environment of E. coli, the colon, can be a rich source of unsaturated fatty acids. These fatty acids may be used for phospholipid precursors as well as an energy and carbon source. In such an environment, endogenous synthesis of unsaturated fatty acids would be unnecessary and inefficient. The presence of the fatty acyl-CoAs therefore results in decreased fabA expression and hence decreased endogenous synthesis of unsaturated fatty acids. If instead only saturated fatty acids are available, repression of fabA transcription is less efficient and the presence of the second FadR-independent promoter, together with residual function of the FadR-regulated promoter, allows a basal level of FabA to be produced. This allows synthesis, although decreased, of the unsaturated acids needed for functional phospholipids. When exogenous fatty acids are not available to the cell, FadR binds to the DNA increasing fabA transcription and the synthesis of endogenous unsaturated fatty acids. FadR regulation of fabB is regulated in a manner very similar to that of fabA (24). As expected, FadR specifically binds a site upstream of the fabB coding sequence that strongly resembles the FadR binding site of fabA. Moreover, strains having null mutations in fadR are hypersensitive to cerulenin, an antibiotic that at low concentrations specifically targets the FabB protein (24).
In addition to FadR, the fabA and fabB genes are regulated by a second transcription factor, FabR. The FabR binding site was discovered by bioinformatic analysis localized immediately downstream of the FadR binding site in the promoter regions of the fabA and fabB genes to which FabR specifically binds (203). They then used a DNA fragment containing the deduced binding site to isolate the protein they called FabR. FabR represses both fabA and fabB, although regulation of fabB transcription is more stringent (312). Regulation of the unsaturated fatty acid content by an FabR repressor requires the presence of an FadR activator and thus seems to antagonize the activator function of FadR. However, it remains unclear what the signal(s) that modulates FabR DNA binding is. Unlike FadR, which regulates the genes of β-oxidation and the glyoxylate shunt (Chapter Tricarboxylic Acid Cycle and Glyoxylate Bypass) in addition to fabA and fabB, the action of FabR is restricted only to the genes of unsaturated fatty acid biosynthesis.
The supply of ACP does limit fatty acid biosynthesis, since there is a significant pool of unacylated holo-ACP during logarithmic growth (141). In addition to unacylated ACP, there are significant pools of acetyl-ACP and malonyl-ACP (249), but acyl-ACPs of chain lengths of four carbons or longer are not detectable (249). It is also clear that E. coli can grow with appreciably reduced levels of ACP (141, 227). Recent work on the mechanisms coupling fatty acid synthesis to phospholipid synthesis indicates that an early step(s) in fatty acid synthesis is regulated by long-chain acyl-ACP levels (see below).
The physiological role(s) of the E. coli thioesterases remains unclear. We will discuss these enzymes to provide a background for their use in studying the regulation of lipid synthesis. It should be noted that the production of free fatty acids in E. coli upon expression of a thioesterase has become a standard means of assessing the specificities of plant acyl-ACP thioesterases (201, 291).
E. coli contains two well characterized enzymes that cleave the thioester bond of acyl-CoA molecules giving the free species of CoA and fatty acid. Both enzymes are much less active on palmitoyl-ACP than acyl-CoA due to the sequestration of the thioester bond by the ACP moiety (267). Thioesterase I is a protein of 20.5 kDa, encoded by the tesA gene, that cleaves acyl-CoAs of >C12 atoms and is unable to cleave 3-hydroxyacyl-CoA thioesters (9, 29). The deduced amino acid sequence of TesA has active-site residues arranged in a manner similar to those found in several mammalian thioesterases (29). The active site is also closely related to those of serine proteases consistent with covalent labeling and inhibition of TesA by serine esterase inhibitors (9, 29). A comparison of the tesA DNA sequence and that determined from the purified protein demonstrated that 26 amino acids are removed from the N terminus of the primary translation product indicating that TesA is a periplasmic enzyme (29). This prediction was confirmed by the demonstration that thioesterase I can be quantitatively released from E. coli cells by osmotic shock treatment (29). The tesA gene has been modified by deletion of the leader sequence such that the enzyme becomes trapped in the cytosol rather than exported to the periplasm (28). This results in a massive accumulation of free fatty acids (FFAs) in the medium due to hydrolysis of the growing fatty acyl chains from the acyl-ACP pool (see below). The cytosolic TesA also diverts acyl chains from ACP to an acyl-CoA-dependent pseudomonad pathway introduced into E. coli (164).
The substrate specificity of TesA measured in vitro predicts the accumulation of FFAs of chain length of >C12, with 16- and 18-carbon fatty acyl thioesters being the preferred substrates (9). Therefore, the large accumulation of C8 andC10 FFA observed in vivo upon trapping TesA in the cytosol (28, 43) was unexpected. However, at lower levels of cytosolic TesA activity (that seen upon overproduction of the wild-type enzyme) >5-fold lower levels of FFA were produced that consisted mainly of the 16- and 18-carbon acids expected from the in vitro specificity (30). It, therefore, seems that the high levels of cytosolic TesA activity allow the efficient hydrolysis of normally unfavored substrates, short-chain acyl-ACPs. It is interesting to note that, although TesA preferentially cleaves long-chain acyl thioesters, the enzyme also efficiently cleaves a dissimilar type of molecule, activated oxygen esters (30, 131). Hydrolysis of these molecules (synthetic substrates used in the assay of chymotrypsin) led to the conclusion that TesA was a protease ("protease I"), although the purified enzyme had not been shown to hydrolyze peptide bonds. A clue to this unusual specificity is that only the activated esters of nonpolar amino acids were hydrolyzed and thus hydrophobicity and a readily hydrolyzed ester (or thioester) bond seemed to be the determinants of substrate activity (30, 131). It may be that binding to the enzyme requires a critical hydrophobic surface area and that short-chain acyl-CoAs and activated esters of hydrophilic amino acids fall below this minimum. A precedent for this type of specificity is mammalian acyl-CoA binding protein which displays high binding affinity only toward long-chain (>C14) acyl-CoA molecules (166). The NMR structure of the complex of the protein with palmitoyl-CoA shows that only long-chain acyl-CoA molecules, but not short-chain acyl-CoA molecules can fill the hydrophobic binding site and allow the cooperative interactions required for tight binding (166). It should be noted that the LuxD thioesterase of bioluminescent bacteria that clearly acts as a thioesterase in supplying the myristoyl aldehyde consumed in light emission cleaves acyl p-nitrophenyl oxygen esters as well as acyl-CoA and acyl-ACP thioesters (173, 178). A more recent finding is that tesA is identical to pldC, the gene encoding lysophospholipase L1 (150). Therefore, it is clear that TesA can cleave some unactivated oxygen esters. The physiological significance of the lysophospholipase L1 activity of TesA is unclear since lysophospholipid cleavage has not been demonstrated in vivo and E. coli cells incorporate exogenously added lysophospholipids (130), although these are good substrates for TesA cleavage in vitro (71). Moreover, lysophospholipase action cannot account for the appearance of the short-chain fatty acids seen upon expression of the leaderless TesA species (28, 60) since such short-chain acids are not found in phospholipids.
A high-resolution crystal structure of TesA shows it to have a Ser10-Asp154-His157 active site and an oxyanion hole formed by Ser10, Gly44, and Asn73 (187). There also seems to be a groove in the protein surface that could accommodate an acyl chain. The kinetics of the enzyme has been studied in some detail and correlated with dynamic structural changes in the protein (174, 175, 282).
In contrast to TesA, thioesterase II (TesB) is a cytosolic tetrameric protein of a 32-kDa subunit encoded by the tesB gene (N1, B9). TesB cleaves acyl-CoAs of >C6 and 3-hydroxyacyl-CoAs but is unable to cleave acyl-panthetheine thioesters (20, 216). TesB lacks the active-site serine motif found in other thioesterases and is a homologue of a human thioesterase that functionally interacts with the product of the HIV-1 Nef gene (185, 186). The crystal structure of TesB shows the active site to have an unusual chemistry involving an Asp-Gln-Thr, catalytic triad (177). TesB has recently been shown to intercept 3-hydroxydecanoyl-CoA in vivo (318).
The physiological functions of thioesterases I and II are unknown and the presence of these enzymes remains an enigma. These enzymes have been assigned a role in phospholipid synthesis based on kinetic isotope effects (149, 211), but these data can be explained by differential isotope effects on the acyltransferases of phospholipid and lipid A synthesis rather than the acyl-ACP hydrolysis proposed. The chromosomal copies of both tesA and tesB have been disrupted to give null mutants (7, 29). Neither the tesA nor tesB null mutations affect cell growth. A tesA tesB double-null mutant strain also grows normally (29). This lack of a growth phenotype indicates that neither protein is essential for cell growth. However, it remains possible that the function of both enzymes can be replaced by another enzyme. Indeed, the tesB null mutant still retains some activity indicating the existence of a third thioesterase in E. coli, perhaps the YbgC protein (see below).
The E. coli thioesterases are most active on long-chain acyl-CoA substrates in vitro. However, detectable quantities of long-chain acyl-CoA molecules are only found in E. coli when β-oxidation is induced (growth on a >C12 fatty acid as sole C source). If the E. coli thioesterases present in such cells were active on acyl-CoAs, then β-oxidation would be inhibited. However, fatty acids are a good carbon source for E. coli, better than acetate (Chapter Two-Carbon Compounds and Fatty Acids as Carbon Sources). This remains the case even when TesA or TesB activities are overexpressed >10-fold (29). Thus, if the E. coli thioesterases cleave acyl-CoAs in vivo, then the acyl-CoA intermediates involved in β-oxidation must be somehow sequestered from the thioesterases. There is some evidence that these intermediates may be channeled between β-oxidation enzyme active sites (Chapter Two-Carbon Compounds and Fatty Acids as Carbon Sources), but acyl-CoAs are also acyl donors in the incorporation of exogenous fatty acids into the phospholipids of E. coli and regulate lipid metabolism by interaction with FadR protein (see above). It is difficult to see how acyl-CoAs could be shielded from TesA and TesB during all of these interactions. The high-resolution crystal structure of TesB showed that it was the first example of a new tertiary fold, called a "double hot dog," showing an internal repeat with a basic unit that is structurally similar to that of FabA. The catalytic site, inferred from the crystal structure and verified by site-directed mutagenesis, involves novel chemistry and includes Asp 204, Gln 278, and Thr 228, which synergistically activate a nucleophilic water molecule.
In the case of TesA, a rationale for the presence of a thioesterase in the periplasm of E. coli is not obvious. The usual explanation for periplasmic hydrolytic enzymes is to allow scavenging of portions of metabolically useful molecules. For example, phosphorylated metabolic intermediates can be hydrolyzed by periplasmic phosphatases to products suitable for transport across the cytoplasmic membrane. However, molecules such as acyl-CoAs and the activated oxygen esters hydrolyze spontaneously in aqueous solution, especially at basic pH values and, thus, enzymatic hydrolysis would not appear to be needed. This instability, together with the fact that acyl-CoAs are only found as metabolic intermediates, suggests that the primary role of thioesterase I might be to hydrolyze substrates other than acyl-CoAs. Reasonable candidates for alternative physiologically relevant substrates for thioesterase I are not obvious. Thioesterase I hydrolyzes only hydrophobic substrates, suggesting that any alternative substrate must also be nonpolar. Oxygen esters would seem the most reasonable alternate substrates, but the hydrophobic molecules abundant in nature (e.g., glycerides) form large micelles at very low concentrations. Such micelles should be unable to pass through the small pores of the E. coli outer membrane. Indeed, the primary function of the enterobacterial outer membrane is thought to be to protect the cytoplasmic membrane from surface agents such as lipid micelles. However, lysophospholipids form micelles only at fairly high concentrations and can enter E. coli (130). Moreover, a periplasmic phosphodiesterase, GlpQ (171), is available to further cleave the soluble lysophospholipase product (the G3P backbone plus the phospholipid head groups) to G3P. Hence, both the fatty acid and G3P moieties could be utilized. However, this does not seem to be the case. Upon supplementation of E. coli cultures with the lyso forms of either PE or phosphatidylcholine, the endogenous lipids are incorporated intact into diacyl lipids (130).
In growing cultures of E. coli fatty acid synthesis is tightly coupled to phospholipid synthesis. As discussed above the intracellular pools of fatty acid synthetic intermediates are very small (110, 249). This result suggests that fatty acid synthesis is coordinately regulated with or by phospholipid synthesis. Therefore, several groups were assayed for accumulation of fatty acid biosynthetic intermediates in the absence of phospholipid synthesis. However, widely conflicting results were obtained. Phospholipid synthesis was blocked by use of mutants to restrict the supply or utilization of G3P, the precursor required for the first step of phospholipid synthesis.
The earliest study (207) reported that no fatty acid synthetic intermediates such as free fatty acids (FFAs) accumulated. However, the strain used was proficient in β-oxidation and thus degradation of FFAs could explain the lack of accumulation seen. Indeed, when strains (fadE) blocked in β-oxidation were examined, it was reported that the rate of incorporation of [14C]acetate into FFA by glycerol-starved cells was the same as the rate of incorporation of this precursor into phospholipid when glycerol was supplied (54). However, later work showed that these labeling results were complicated by an unexpected shrinkage of the endogenous acetate (acetyl-CoA) pool of the glycerol-starved cells (38, 220). Thus, the specific activities of the cellular acetate pools utilized in fatty acid synthesis differed between the cultures starved for G3P and the unstarved control cultures. Therefore, equivalent rates of [14C]acetate incorporation did not denote equivalent rates of lipid synthesis.
A plausible explanation for the change in acetate pool size upon glycerol starvation is based on the finding that FadR, the repressor of the β-oxidation regulon, also downregulates the expression of the glyoxylate operon (Chapter Two-Carbon Compounds and Fatty Acids as Carbon Sources and Tricarboxylic Acid Cycle and Glyoxylate Bypass). Since the known regulatory activities of FadR are neutralized by fatty acyl-CoA binding, it seems probable that FFAs are produced and converted to acyl-CoAs via acyl-CoA synthetase (43). These acyl-CoAs would then neutralize FadR (43), resulting in increased production of the glyoxylate cycle enzymes, which would shrink the acetate pool. For these reasons, the choice of a precursor to measure lipid synthesis during glycerol starvation is problematic. Malonate fails to enter E. coli and the only direct precursor other than acetate, tritium-labeled water, is diluted by the water of the culture medium such that pulse labeling is essentially impossible. An alternative approach was to determine the level of fatty acyl-ACP molecules by labeling the protein moiety. Such experiments showed the accumulation of acyl-ACPs upon glycerol starvation (249). This result indicated that fatty acid synthesis did continue in the absence of phospholipid synthesis, but could provide no measure of any FFA produced by hydrolysis of acyl-ACPs. To deal with these constraints, an E. coli strain was constructed in which endogenous synthesis of acetate was blocked and the only fate of acetate was as a specific precursor of lipid synthesis (145). Use of this strain demonstrated that fatty acid synthesis is coupled to phospholipid synthesis and in the absence of phospholipid synthesis proceeds at only 10% to 20% of the rate seen during phospholipid synthesis (145). Acyl-ACPs accumulate under these conditions and are thought to feedback inhibit the fatty acid synthetic pathway accounting for the observed coupling (see below).
The main in vivo evidence for feedback inhibition by acyl-ACPs is that overexpression of either of the E. coli thioesterases results in relief of the inhibition of fatty acid synthesis engendered by glycerol starvation (145). Moreover, thioesterase overexpression also eliminated the accumulation of acyl-ACP species and the synthesis of fatty acids of abnormal length (145), indicating that the inhibition of fatty acid synthesis may be caused by the accumulation of acyl-ACP species. Restoration of fatty acid synthesis by thioesterase overproduction could result from either loss of the acyl-ACP per se or the increase in ACP concentration resulting from the cleavage of the acyl group. The latter explanation can be eliminated, since the ACP pools of the starved cells are not significantly depleted and overproduction of ACP failed to relieve inhibition of fatty acid synthesis (145).
The most straightforward model for the inhibition of fatty acid synthesis upon blockage of phospholipid synthesis is that long-chain acyl-ACP species accumulate and inhibit key fatty acid synthetic enzymes. Thioesterase overproduction would cleave the long-chain acyl-ACP species and thereby relieve the inhibition. A very similar (if not identical) feedback inhibition mechanism probably accounts for the dependence of fatty acid synthesis on cellular growth. Normally, E. coli ceases lipid synthesis upon entry into stationary phase (the residual synthesis can be attributed to the small fraction of growing cells present in stationary-phase cultures [8, 46]), but when high levels of thioesterase are present in the cytosol lipid synthesis continues as FFA synthesis, resulting in a very large increase in the total fatty acid produced per cell mass (28). A number of different thioesterases give this result. It was first seen upon expression of a cDNA encoding a novel thioesterase from the California bay tree, which resulted in the accumulation of a massive amount of lauric acid in the culture medium (291). Expression in E. coli of another sequence-related plant thioesterase gave a mixture of saturated and unsaturated long-chain fatty acids consistent with the different in vitro substrate specificity of the thioesterase (72). When localized in the cytosol, E. coli thioesterase I gives a mixture of fatty acid species (see above), and in this case labeling with radioactive acetate together with analysis of the radioactive products showed the excreted fatty acids are produced by cleavage of acyl-ACPs (28, 43). It should be noted that a second possible candidate for the regulatory molecule is acyl-CoA. In this scenario FFA generated by an endogenous thioesterase would be converted to acyl-CoA by the β-oxidation enzyme, acyl-CoA synthetase (FadD), and this acyl-CoA would downregulate fatty acid synthesis. However, fadD mutant strains producing the above thioesterases gave the same results as strains blocked elsewhere in β-oxidation or (in some cases) wild-type strains (43, 72, 291). Since fadD encodes the only acyl-CoA synthetic enzyme detectable in aerobic cultures in vivo (163, 212), a role for acyl-CoA can be ruled out.
Two general but distinct models can explain the release of fatty acid synthesis by thioesterase action. The first model is that proposed above, which states that the cleaved 'TesA substrate (acyl-ACP) normally acts as a feedback inhibitor of a fatty acid synthetic enzyme (s), whereas the second model proposes that the cleaved substrate is a transcriptional corepressor regulating the production of fatty acid synthetic enzymes (analogous to FadR system described above). To discriminate between these models the dependence of 'TesA action on subsequent mRNA synthesis was assayed. The first model states that FFA production should be independent of protein (hence mRNA) synthesis following production of 'TesA, whereas in the second model mRNA synthesis is necessary for the 'TesA effect.
To test these models the 'TesA expression was placed under control of a phage T7 promoter. Following a brief period of T7 RNA polymerase synthesis, E. coli mRNA synthesis was blocked by addition of rifampin, a specific inhibitor of E. coli RNA polymerase (but not the T7 enzyme). Thus, synthesis of any proteins other than 'TesA required for FFA production would be blocked. Induction of T7 RNA polymerase gave the expected increase in 'TesA (>100-fold increase in total cellular thioesterase activity), and significant production of FFA was observed when rifampin was added after only 5 min of T7 RNA polymerase induction. The level of FFA subsequently produced was more than half of the levels of cultures in which rifampin was omitted. The time allowed for synthesis of T7 RNA polymerase was consumed in polymerase synthesis and the subsequent synthesis of the first few molecules of 'TesA, and thus little (if any) time remained for protein synthesis in response to 'TesA production. Therefore, the release of inhibition of fatty acid synthesis is largely (if not entirely) an effect on enzymes present at the time of inhibition.
Stringent control of lipid synthesis during blockage of protein synthesis and carbon source shift-down results in coordinately decreased synthetic rates for fatty acid, phospholipid, and lipid A. This form of regulation is due to the rapid accumulation of the guanosine alarmones, guanosine-5'-triphosphate-3'-diphosphate and guanosine-5'-diphosphate-3'-diphosphate, collectively called (p)ppGpp, which inhibit the first step of phospholipid synthesis, the acylation of G3P (110). The effect has been attributed to inhibition of the PlsB although (p)ppGpp fails to inhibit PlsB-catalyzed acyl transfer from acyl-ACPs to G3P in vitro (56). Hence, the effects of (p)ppGpp on the newly discovered PlsX-PlsY acyltransferase system should be assessed. In any event it is clear that acyltransferase activity is the primary lipid synthetic target of the guanosine alarmones since the effects can be largely relieved by overexpression of PlsB. As expected from prior work (145) blocking acyltransfer to G3P resulted in inhibition of fatty acid synthesis and accumulation of long chain acyl-ACPs. Moreover, accumulation of guanosine alarmones had no effect on fatty acid synthesis in cells which expressed 'TesA (60). Therefore, the observed inhibition of fatty acid synthesis can be attributed to feedback inhibition of fatty acid synthetic enzymes by accumulated acyl-ACPs.
Finally, acyl-ACPs have been reported to inhibit of three fatty acid synthetic enzymes, FabI, FabH, and ACC, in vitro (60, 115, 116). In the case of FabI, the observed inhibition might be the simple product inhibition characteristic of all enzymes, since the enoyl-ACP reductase produces finished (fully reduced) acyl-ACPs. This would be a straightforward mode of inhibition. However, in the case of FabH the acyl-ACP species produced differ from those that inhibit the enzyme both in chain length (C4 versus C12 to C20) and in acyl chain oxidation state (3-keto versus fully reduced) and, thus, simple product inhibition seems unlikely. In the case of ACC neither partial reaction was inhibited by acyl-ACPs; only the overall reaction was inhibited. ACC was similarly inhibited by all acyl-ACP chain lengths tested (C6 to C20), although no inhibition was seen with the C16 thioester of spinach ACP (60). The redundant nature of fatty acid synthesis inhibition by acyl-ACPs indicates a coordinated shutdown of the pathway, which would make isolation of E. coli mutants defective in feedback inhibition problematical.
Thermal regulation of membrane fluidity is widely found in bacteria. At physiological temperatures, normal cell function requires a membrane bilayer in a largely fluid state. As growth temperatures are lowered, however, the membrane undergoes a reversible change from a fluid (disordered) to a nonfluid (ordered) state (42). In E. coli, the temperature at the point this transition occurs depends on the fatty acid composition of the membrane phospholipids. Marr and Ingraham (198) first noted that E. coli adjusts its fatty acid composition in response to lower growth temperature by increasing the amount of cis-vaccenic acid and decreasing the amount of palmitic acid incorporated into membrane phospholipid. The amount of palmitoleate incorporated, however, remains unchanged. Lower growth temperatures result in an increase in the number of diunsaturated phospholipids in the membrane. At 37°C, palmitic acid occupies position 1 of the phospholipid backbone, whereas palmitoleic acid is found only at position 2. As the growth temperature is lowered, cis-vaccenic acid competes with palmitic acid for position 1 of the newly synthesized phospholipids. This mechanism is thought to allow an organism to regulate the fluidity of its membrane to optimize membrane function at various growth temperatures.
The elucidation of the mechanism of thermal regulation in E. coli involved a number of independent observations. First, the finding that Cvc strains lacked both FabF and thermal regulation suggested that the elongation of palmitoleoyl-ACP played a role in thermal regulation. Another key observation was that the increased rate of cis-vaccenic acid synthesis characteristic of thermal regulation is evident within 30 s after temperature downshift (91). This indicated that neither mRNA nor protein synthesis is required for fatty acid composition adjustment; therefore, thermal regulation is exerted by a protein present at all temperatures but active only at low temperatures. It was subsequently shown that palmitoleoyl-ACP was not only an excellent substrate for FabF (78, 92, 93) but also, at lower temperatures, the relative activity on this substrate increased (78, 93). It was finally demonstrated that the Cvc phenotype and the lack of FabF are due to mutations in the same gene, fabF. Thus, FabF was firmly established as playing an essential role in the thermal regulation of fatty acid composition of E. coli.
Although it was known that fabF mutants lacked temperature control, the cause and effect relationship was not clear. It remained unclear whether only the presence of cis-vaccenate per se conferred thermal regulation or whether the synthesis of cis-vaccenate by FabF was required for the response. The observation that overproduction of synthase I produced an appreciable increase in the cis-vaccenic acid content of membrane phospholipids enabled this question to be addressed (65). A plasmid carrying the fabB gene was introduced into a fabF mutant and the increased cis-vaccenic acid content of cells overproducing FabB in a strain lacking FabF was found to be independent of growth temperature (65). Therefore, the sole enzyme responsible for thermal modulation of the fatty acid composition is FabF.
An interesting mutation, called Vtr, causes cells to overproduce cis-vaccenic acid at all temperatures (64). The Vtr mutation has been shown to be tightly linked to and perhaps allelic to the fabF gene (283). Efforts to detect a kinetic defect in the FabF of a Vtr mutant, however, were unsuccessful (283), and recent work has shown that both the fabF and acpP genes of a Vtr strain have wild-type coding sequences. It therefore seems that the Vtr mutation may effect the expression of these or other genes of the fab cluster. Further investigation of the genetic nature of the Vtr mutation will hopefully provide further insight into temperature regulation.
Palmitate, palmitoleate, and cis-vaccenate comprise the bulk of the fatty acids found in E. coli membranes. The 3-ketoacyl-ACP synthases play a major role in controlling the chain length and production of these fatty acids. In vitro substrate specificity experiments indicate that E. coli membrane phospholipids contain negligible levels of chain lengths >C18 because the precursor acyl-ACPs are poor substrates for elongation by the synthases (78, 93). Changes in cellular fatty acid composition when either the synthase enzymes are defective or overexpressed have given the most valuable insight into the function of the synthases. FabB deficiency leads to a lack of unsaturated fatty acids and therefore a deficiency in cis-vaccenate (92, 283). On the other hand, overexpression of FabB leads to the overproduction of cis-vaccenate (65). Thus, elevated FabB activity results in elongation of acyl-ACPs that are poor substrates for this enzyme. Mutants lacking FabF are unable to synthesize cis-vaccenate and therefore have a deficiency of membrane cis-vaccenate. The role of this enzyme in temperature control of fatty acid composition is reviewed above and is the major contribution of this enzyme to fatty acid composition. Mutants with severely impaired FabH activity are enriched in C18 fatty acids, whereas the FabH overexpression gives decreased average fatty acid chain length and the appearance of significant amounts of myristic acid in the phospholipids (277). This effect is attributed to an increased rate of fatty acid initiation that leads to a deficiency in malonyl-CoA (and/or malonyl-ACP) for the terminal elongation reactions. Overproduction of either malonyl-CoA:ACP transacylase (FabD) (197) or the NADH-dependent enoyl-ACP reductase (FabI) (152) have also been reported to give slight increases in the average chain lengths of the phospholipid fatty acids. However, the activity of the G3P acyltransferase system seems the most important regulator of acyl chain length. When phospholipid synthesis is slowed or arrested at the G3P acyltransferase step (by glycerol starvation of either a plsB mutant or a gpsA mutant), the fatty acids that accumulate have abnormally long chain lengths and can reach 20 to 24 carbon atoms (38, 54, 145). Conversely, overproduction of the PlsB acyltransferase results in a somewhat decreased average chain length represented mainly by an increase in myrisitic acid (38). Thus, competition among the rate of elongation by the synthase enzymes, the supply of malonyl-ACP, and the utilization of acyl-ACPs by the acyltransferases appear to be the most significant determinants of fatty acid chain length.
The first step of the phospholipid biosynthetic pathway is the synthesis of sn-glycerol 3-phosphate (G3P). G3P is a water-soluble intermediate that forms the scaffold of all phospholipid molecules and is the precursor to the polar head group of PG and the nonacylated glycerol moiety of CL. G3P can be synthesized by two routes. Growth on glycerol as the sole carbon source induces the enzymes of glycerol degradation (the glp operon). Glycerol kinase (encoded by glpK) is one of the enzymes induced and converts glycerol to G3P. During growth on carbon sources other than glycerol, G3P is made by the NADH-dependent reduction of the glycolytic intermediate, dihydroxyacetone phosphate (160, 161, 162). The enzyme catalyzing this reaction is called G3P synthase or the biosynthetic G3P dehydrogenase and is the product of the gpsA gene (32, 48). This enzyme is sensitive to allosteric feedback regulation by G3P (162) and this mechanism is responsible for the maintenance of a constant intracellular G3P concentration, as verified by the isolation of mutants resistant to feedback inhibition (13). These mutants were isolated as extragenic suppressors of the plsB mutants (see below) and contain elevated levels of G3P (13).
GpsA is the sole enzyme that produces G3P for phospholipid synthesis in cells grown on carbon sources other than glycerol since mutants lacking G3P synthase activity require exogenous G3P or glycerol for growth (32, 48, 160). In the absence of G3P, phospholipid synthesis is inhibited >95%. The residual 32Pi incorporation into phospholipid in gpsA mutants is attributed to phosphorylation of diacyglycerol resulting from PG turnover. The levels of G3P do not limit the rate of phospholipid biosynthesis. Not only are the G3P levels in wild-type strains maintained at a concentration of about 0.2 mM by feedback regulation of GpsA, but experimental manipulation of the G3P levels over a 10-fold range in vivo had no noticeable effect on the phospholipid biosynthetic rate or membrane phospholipid composition (95).
The next step is the transfer of the acyl chains of the acyl-ACP end products of fatty acid biosynthesis to G3P (Fig. 5). The first enzyme transfers fatty acids to position 1 of G3P whereas the second enzyme esterifies position 2 of the glycerol backbone. Like most phospholipids in nature, bacterial phospholipids have an asymmetric distribution of fatty acids between positions 1 and 2 of the G3P backbone that is controlled in part by the acyl chain specificity of the acyltransferases. The G3P acyltransferase systems are not considered to be a component of fatty acid biosynthesis per se; however, the activity of the G3P acyltransferases does affect both the chain length distribution of the fatty acids found in membrane phospholipids and indirectly determines the rate of fatty acid biosynthesis (see below).
The PlsB-catalyzed attachment of the first fatty acid to G3P to form monoacyl-G3P (lysophosphatidic acid) is much less conserved than is gpsA. Proteins homologous to that encoded by the E. coli plsB gene are restricted to the α- andγ-proteobacteria plus the mycobacteria. The original plsB mutants (12) required greatly elevated levels of G3P (either exogenously added or endogenously produced) for growth and the mutant enzyme showed a low activity and 10-fold increased Km defect for G3P (111). A perplexing finding was that the phenotype of these mutant strains depended on the presence of a mutation in a second unlinked gene, plsX (172). The role of the PlsX gene product was an enigma, but an important enigma since plsX is very widely conserved among bacteria (much more so than is plsB) and is invariably found adjacent to genes of fatty acid synthesis. This is the case in E. coli where plsX lies at the 5' end of the fab gene cluster. The data of Larson and coworkers (172) strongly suggested that PlsX was a second G3P acyltranferase. Another possibility was that PlsX played a role in determining intracellular G3P levels, but this seemed less likely because of the intimate connection between PlsX and fatty acid synthetic genes. Attempts to show G3P acyltranferase activity for PlsX were fruitless and it has very recently become clear why this was the case; the wrong acyl donors had been tested. PlsX is the first enzyme in a novel two-enzyme G3P acylation pathway that uses acyl phosphates rather than the usual acyl-ACPs or acyl-CoAs as the acyl donor (193). PlsX, a soluble enzyme, transfers the acyl groups of acyl-ACPs to phosphate. PlsX is a phosphate:acyl-ACP acyltransferase, the reaction is analogous the classical phosphotransacetylase reaction of two-carbon metabolism. Indeed, the function of PlsX was deduced by bioinformatic and structural comparisons of the PlsX and phosphotransacetylase proteins of B. subtilis (193). Transfer of the acyl group to phosphate from its thioester linkage to ACP retains the activated nature of the fatty acid carboxyl moiety since acyl phosphates are mixed anhydrides. A membrane-bound enzyme called PlsY then uses the acyl-phosphates produced by PlsX to acylate G3P. This acylation reaction is as energetically favorable as the PlsB reaction since acyl-phosphates are mixed anhydrides. The plsY gene was deduced by phylogenetic profiling of genes that showed covariance with plsX and in E. coli is the previously unassigned ygiH gene (193).
The discovery of the PlsX-PlsY pathway (193) plus subsequent genetic studies (308) finally allows explanation of the data of Larson and coworkers (172). First, plsB null mutants of E. coli are nonviable, whereas null mutants of plsX and plsY are fully viable (7, 308). Moreover, overproduction of the original ("Km") mutant PlsB protein complemented plsB null mutant strains (308) showing that expression of high levels of the protein compensates for its kinetic deficiencies (112). Hence, the conditional phenotype of the original plsB plsX mutant strains resulted from function of the mutant PlsB due to the presence of artificially increased intracellular G3P pools plus the lack of PlsX activity. When the original ("Km") plsB mutant allele was moved to a wild-type background, no phenotype was seen because the residual PlsB activity plus the PlsX-PlsY pathway performed the required acylation of G3P. It follows that plsX and plsY are nonessential genes in E. coli due to the presence of PlsB. Although the situation seems much more clear than before, additional studies seem indicated. One loose end is the use of chemically undefined media in some of the mutant analyses (308) that may have contributed low levels of G3P or glycerol. Another is the nature of the original plsX missense mutations. This uncertainty could be resolved by comparison with plsX null mutants. An unexplained result is that strains carrying null mutations in both plsX and plsY are nonviable unless the strain overproduced PlsB, neither the PlsB level produced by the chromosomal gene nor that from a plasmid carrying the plsB26 allele was sufficient for growth (308). However, in the presence of a wild-type chromosomal copy of plsB overexpression of the TesA thioesterase allowed growth of strains in which either plsX or plsY and expression of the other gene was decreased by use of an regulated ectopic promoter (308). These results suggest some interplay among PlsB, PlsX, PlsY, and/or their products or substrates (e.g., acyl-ACP) that should be studied more deeply. Finally, it would be interesting to test whether strains carrying the original plsB mutant allele plus a plsY null mutation have the same conditional phenotype as the original plsB plsX strains. E. coli therefore has two means to acylate G3P, the PlsX/Y and PlsB pathways, whereas most bacteria have only the former pathway and a few have only the latter pathway. It is not clear of what advantage (other than functional redundancy and perhaps removal of inhibitory metabolites) accrues from having the PlsX/Y pathway. However, PlsB provides a definite advantage, an acyltransferase capable of utilizing acyl-CoAs. This allows the incorporation of exogenous fatty acids to produce membrane phospholipid in lieu of the energy-expensive fatty acid synthesis pathway (193, 228).
The next step in phospholipid biosynthesis is catalyzed by 1-acyl-G3P acyltransferase (the plsC gene product), which acylates the product of the PlsB or PlsX/Y steps to form phosphatidic acid (Fig. 5) (34, 35). PlsC (which can utilize both acyl-ACP and acyl-CoA substrates) catalyzes the attachment of the second fatty acid to the G3P backbone to form the key intermediate, phosphatidic acid. Temperature-sensitive plsC mutants that accumulate 1-acyl-G3P at the nonpermissive temperature and lack 1-acyl-G3P acyltransferase activity in vitro were isolated (34). As expected, plsC is an essential gene (7, 94). The distribution of PlsC homologues among bacteria is similar to PlsB, but is even more restricted in that only a weak homologue is found in mycobacteria. However, some bacterial genomes have a PlsC homologue (or homologues), but lack a PlsB homologue. Although not widely distributed in bacteria, PlsC homologues are found in eukaryotes and expression of several plant and animal homologues has been shown to permit growth of a temperature-sensitive E. coli plsC mutant (21, 104, 298). PlsB and PlsC share common sequences and some of the conserved residues have been shown to be required for PlsB acyltransferase activity (111). There may be further complexity in G3P acylation enzymes since a Clostridial gene complemented E. coli plsB mutants but encoded a protein that more closely resembled plsC (109). The presence of phosphatases that specifically dephosphorylate phosphatidic acid and 1-acyl-G3P is known and mutants deficient in these enzymes have been isolated (132). The function of these phosphatases remains an enigma, but their potential involvement in the acylation of G3P is suggested by the observation that certain acyl-ACPs promote the dephosphorylation of G3P, 1-acyl-G3P, and phosphatidic acid (247). The pathway for the synthesis of the phospholipid species that are major membrane components is shown in Fig. 5. Phosphatidic acid is converted to CDP-diacylglycerol, which serves as an intermediate in the biosynthesis of all membrane phospholipids. PE, the most abundant membrane phospholipid, is synthesized by the exchange of serine for CMP catalyzed by PS synthase followed by PS decarboxylation to yielding PE. PG is synthesized by the exchange of G3P for CMP to form PGP, which is subsequently dephosphorylated to generate PG. CL is synthesized by the CL synthase-catalyzed condensation of two molecules of PG.
A key finding in the pathway for phospholipid synthesis was the discovery of an activated form of phosphatidic acid, CDP-diacylglycerol (25, 147, 279). This metabolically active intermediate constitutes only 0.05% of the total phospholipid pool (235). In E. coli conversion of phosphatidic acid to a mixture of CDP-diacylglycerol and dCDP-diacylglycerol is catalyzed by a single enzyme called CDP-diacylglycerol synthase (73) (Fig. 5). E. coli mutant strains severely deficient in this enzyme have lesions at a single genetic locus (cdsA) and some are temperature sensitive (88, 89, 90). Some of these mutants accumulate substantial amounts of phosphatidic acid (up to 5% of the total phospholipid) which may account for their increased sensitivity to erythromycin and elevated pH. Shifting of the growth medium from pH 7.0 to 8.5 triggers massive accumulation of phosphatidic acid (up to 25% of the total phospholipid) and inhibition of phospholipid synthesis (90). These data suggest that CDP-diacylglycerol synthase is present in large excess of the minimal amount of enzyme required to sustain phospholipid synthesis. An unlinked mutation that suppresses the phenotype of strains carrying a cdsA mutant allele has also been isolated (90), although the nature of this mutation and the specificity of its suppression ability are unknown. Null cdsA mutants are lethal (7, 94). CDP-diacylglycerol is the branch point in phospholipid synthesis. It reacts with either G3P or with serine to form PGP or PS, respectively. The presence of both ribo and deoxyribo forms of the liponucleotide could play a role in determining the relative rates of the synthesis of these two phospholipids if the respective synthases exhibit selectivity toward dCDP-diacylglycerol versus CDP-diacylglycerol. These species exist in roughly equal levels in vivo (89). However, both ribo- and deoxyriboliponucleotides are substrates for PS synthase in vitro and a change in the ratio of liponucleotides in vivo to 3.1 has no effect on the relative rates of PE and PG synthesis (89). Thus, the significance, if any, of the two liponucleotide forms remains to be determined.
The first step in the synthesis of PE is the condensation of CDP-diacylglycerol with serine to form PS catalyzed by PS synthase (288) (Fig. 5). PS synthase does not appear in the inner membrane fraction during standard cellular localization procedures, as the other enzymes of phospholipid synthesis do. The enzyme instead is found attached to ribosomes. This association is an artifact of cell disruption (190); the ribosomes act as an ion-exchange trap for the PS synthase, and after addition of CDP-diacylglycerol, PS synthase is dissociated from the ribosome and subsequently associates with and is activated by phospholipid vesicles containing CDP-diacylglycerol and certain other phospholipids (184, 189, 243). The PS synthase reaction proceeds via a ping-pong mechanism (234).
PS is a minor membrane constituent of E. coli since it is rapidly converted to PE by PS decarboxylase (Fig. 5). This inner membrane enzyme has been purified to homogeneity and found to have a pyruvate prosthetic group that participates in the decarboxylation reaction by forming a Schiff base with PS (252). The two nonidentical subunits of the mature enzyme are formed by cleavage of a proenzyme resulting in the conversion of Ser-254 to an amino-terminal pyruvate residue (76, 179). Protein processing is interrupted by mutational alteration of Ser-254 to either Cys-254 or Thr-254, but although the mutant proteins have only minimal activity, they are able to complement psd(Ts) mutants (179).
Mutants (psd) with a temperature-sensitive decarboxylase accumulate PS at the nonpermissive temperature (105, 106, 107). Despite the reduced levels of PE and the concomitant increase in PS levels, the mutants continue to grow for several hours after the shift to the nonpermissive temperature. Strains harboring psd clones overproduce the enzyme 30- to 50-fold (281). Under these conditions there is no effect on membrane phospholipid composition and only half of the enzyme remains associated with the inner membrane.
The psd gene is essential (7, 66, 94), indicating that the inability to synthesize PE is lethal. This was first seen with psd temperature-sensitive mutants that lyse at the nonpermissive temperature (105). However, as would be expected from the pathway (Fig. 1), these strains accumulate high levels of phosphatidylserine. This accumulation seems responsible for lysis since mutants lacking phosphatidylserine synthetase (PssA), the first enzyme of the PE pathway, do not lyse. The first pssA mutants were temperature-sensitive mutants (105) but, upon finding that growth under nonpermissive conditions could be supported by divalent cations, strains with pssA deletions could be constructed (66). These strains have been used extensively, although for unknown reasons they fail to grow on minimal media, a property that complicates some analyses. As discussed above the divalent ions are thought to stabilize nonlamellar structures that compensate for the lack of PE. If so, it is clear that the structural properties of PE rather than any chemical property of the molecule are required for cell growth, although there are perturbations in a large variety of important cellular functions, which have been discussed elsewhere (42, 77). Growth of these strains require CL synthesis as shown by the inability to introduce a null cls allele into either pssA::kan or psd::kan strains (54, 202). Thus, PE is essential for the polymorphic regulation of lipid structure. The physiological processes dependent on the formation of local regions of nonbilayer structures remain to be elucidated, but the process of cell division, the formation of contacts between inner and outer membranes, and the translocation of molecules across the membrane are viable candidates.
There is also a salvage pathway for the synthesis of PE from 2-acylglycerophosphoethanolamine (2-acyl-GPE). 2-Acyl-GPE may arise from the action of phospholipase A1 or as a by-product of protein acylation. It can also be taken up from the medium (128, 130). The resulting 2-acyl-GPE is recycled into PE by 2-acyl-GPE acyltransferase/acyl-ACP synthetase (the product of the aas gene). The acyltransferase was first recognized as an inner membrane protein called acyl-ACP synthetase (241). Acyl-ACP synthetase catalyzes the ligation of ACP to fatty acids in the presence of ATP, Mg2+, and high salt concentrations (241). The acyltransferase contains ACP as a bound subunit that acts as the intermediate acyl acceptor in the acyltransferase reaction (37, 246, 248). The high salt concentrations are required to dissociate the acyl-ACP intermediate from the enzyme, thereby uncovering the synthetase activity (37, 246, 248). Although the acyl-ACP synthetase reaction has proven extremely valuable in the preparation of acyl-ACPs that are substrates for other enzymes (257), the only reaction catalyzed by Aas in vivo is 2-acyl-GPE acyltransferase. This is consistent with the finding that exogenous fatty acids are not converted to acyl-ACP derivatives that can be utilized by the enzymes of fatty acid biosynthesis. The isolation of aas mutants supports the biochemical studies (129). The aas mutants are defective in both acyl-ACP synthetase and 2-acyl-GPE acyltransferase activities and accumulate both 2-acyl-GPE and acylphosphatidylglycerol. Acylphosphatidylglycerol accumulation was due to the transacylase activity of lysophospholipase L2 (the pldB gene product) since aas pldB double mutants accumulated 2-acyl-GPE, but not acylphosphatidylglycerol (129). Strains overexpressing the Aas gene overproduce both synthetase and acyltransferase enzyme activities (138). Analysis of the predicted amino acid sequence reveals extensive homology to synthetases that employ acyl-adenylates as intermediates as well as acyltransferases (138).
The first step in the synthesis of PG is the condensation of CDP-diacylglycerol with G3P to form PGP (Fig. 5). The reaction is analogous to the synthesis of PS with CMP as the other product. The protein has been purified to homogeneity by affinity chromatography on CDP-diacylglycerol-Sepharose (74). The reaction kinetics is consistent with a sequential bi-bi mechanism, and unlike PS synthase, PGP synthase fails to catalyze the hydrolysis of CDP-diacylglycerol (74). An additional substrate for the enzyme is the G3P analog 3,4-dihydroxybutyl-1-phosphonate, but the PGP analog synthesized from this substrate cannot be hydrolyzed by the PGP phosphatases (263). When cells are grown in the presence of this analog, the abnormal lipid accumulates and growth ceases. The conversion of 3,4-dihydroxybutyl-1-phosphonate to its PGP analog accounts for the increased sensitivity of E. coli to this analog in media containing elevated amounts of divalent cations (154).
The insertional inactivation of PGP synthase is lethal (7, 94, 108), but second-site suppressors arise that allow growth of pgsA null mutant strains. This area was reviewed in some detail elsewhere (42) and some progress has been made since that assessment concerning one class of suppressor mutations that involves the Rcs (Regulator of capsule synthesis) phosphorelay system (213, 214). The Rcs system activates genes required for capsular biosynthesis and copes with multiple stresses, while it represses other genes. This regulatory system affects many aspects of cell physiology, which may explain many of the diverse phenotypes observed for these strains (42).
The next step in the pathway, removal of the phosphate moiety from phosphatidylglycerol phosphate is not well understood since at least three different enzymes (PgpA, PgpB, and an unknown protein) catalyze this hydrolytic step (67). The pgpA and pgpB genes were isolated by in vitro activity assays. Based on the finding that PgpA specifically hydrolyzed PGP whereas the PgpB phosphatase also hydrolyzed phosphatidic acid, PgpA was thought to be the enzyme involved in PG biosynthesis (75). However, Funk and coworkers (86) disrupted both of these genes in a single strain, and although the respective phosphatase activities were reduced, PG synthesis was not impaired. Thus, neither of these phosphatases is required for PG synthesis, and the existence of at least one other phosphatase capable of operating in the PG biosynthetic pathway remains to be discovered. PgpB has recently been reported to also hydrolyze undecaprenyl pyrophosphate (79).
CL is formed by the condensation of two molecules of PG in a transesterification reaction (Fig. 5). This reaction is catalyzed in E. coli by the cls gene product, cardiolipin synthase (218, 275). This enzyme (which lacks a systematic name) is very widely conserved. Initially, CL was thought to be synthesized by the reaction of CDP-diacylglycerol with PG, the mechanism used by mammalian mitochondria. However, results of a series of physiological experiments indicated that unlike synthesis of other phospholipids, CL synthesis occurs under conditions of ATP depletion where synthesis of CDP-diacylglycerol was unlikely and these observations led to the identification of the enzyme catalyzing this reaction (126, 279). The physiological role of CL synthase was corroborated by the isolation of a mutant (cls) deficient in both the synthesis of CL and transesterification activity (218, 226). These mutants were initially thought not to have a growth phenotype, but subsequently, growth was shown to be impaired in mutants (pssA) deficient in PE synthesis (261). E. coli survives disruption of the cls gene, although the cells grow at a slower rate and to a lower density than the corresponding wild-type cells, indicating that CL may confer a growth or survival function (275). CL accumulation and CL synthase activity increase as the cells enter the stationary phase of growth (125, 275, 279), and CL is the most stable membrane phospholipid during prolonged incubation in stationary phase (125). The cls null mutants lose viability in stationary phase, supporting the idea that CL is important for long-term survival under nongrowing conditions (125). The strains are also unusually sensitive to the DNA gyrase inhibitor, novobiocin (276). The conclusion that CL is nonessential is complicated by the finding of residual CL in the cls null mutants (218, 275). The origin of the CL may be due to the activity of PS synthase. The cls null allele cannot be transferred to pssA mutant strains, suggesting that a low level of CL may be essential for viability. Amplification of CL synthase, in turn, leads to the overproduction of CL and to a decreased membrane potential and loss of viability (125). Therefore, E. coli tolerates rather large changes in the overall CL content, but the elimination or overproduction of CL leads to significant physiological imbalance. It is noteworthy that E. coli K-12 encodes two cls homologues, YmdC and YbhO. The latter protein has been expressed in E. coli and found to have cardiolipin synthase activity (103). Given this result it is curious that ybhO was reported unable to complement a cls mutation. However, it is not clear that the genome segment tested contained a functional promoter. It should be noted that ORFs encoding plausible homologues of most of the enzymes of phospholipid synthesis can be found in the E. coli genome, whereas homologues of the fatty acid synthetic enzymes are not found. For example, in addition to the cls homologues, yihC, ynbB, and ybjG (an undecaprenyl pyrophosphate phosphatase) encode proteins similar to PlsC, CdsA, and PgpA/B. E. coli accumulates phosphatidylmannitol and diphosphatidylmannitol when grown on 0.6 M mannitol and a number of other sugar alcohols can also be incorporated into phospholipid (262). The formation of these unusual analogs is increased when a pss mutant is shifted to the nonpermissive temperature. The formation of these lipids appears to be catalyzed by CL synthase, since cls strains fail to accumulate phosphatidylmannitol and a strain carrying a cloned cls gene overproduces these phospholipids. A pathway involving the reversal of the CL synthase reaction coupled with a lack of substrate specificity of CL synthase is proposed to account for these findings (262). Interestingly, cells that contain high levels phosphatidylmannitol and diphosphatidylmannitol grow almost normally, although the ability of these cells to respond to stress has not been evaluated.
The synthesis of cyclopropane fatty acids (CFAs) is more properly viewed as a postsynthetic modification since the substrate fatty acids are covalently bound components of phospholipid molecules located and functioning within the membrane (97). The reaction is a methylenation of these double bonds, the methylene donor being S-adenosylmethionine (SAM). Much is known about CFA synthesis (97), but the two most interesting questions remain. First, how does the soluble CFA synthase together with the soluble substrate, SAM, gain access to the phospholipids of the inner and outer membranes? Second, what role do these acids play that accounts for their synthesis by most bacterial species? Mutants of E. coli and S. enterica that completely lack CFA synthase activity (due to null mutations in the cfa gene) have been constructed (26, 96, 156), but grow and survive normally under virtually all conditions. The only exceptions to this finding are that cfa mutant strains of both organisms are more sensitive to acidic conditions (26, 156) and to freeze-thaw treatment than isogenic Cfa+ strains (96).
Recent efforts have focused on the purification and enzymology of CFA synthase and the regulation of cfa gene expression. CFA synthase has been purified to homogeneity (293) and modern techniques of protein purification have made the enzyme more tractable (39, 135). Together these data demonstrate that CFA synthase is a protein of 44 kDa encoded by the cfa gene; this has become the archetype of this class of enzymes.
The regulation of CFA formation is unusual. The bulk of CFA synthesis occurs as cultures enter the stationary phase of growth and there is a very sharp peak in CFA synthase activity at that time (292). The rise in activity is due to increased cfa transcription. Analyses of CFA gene transcription indicate the presence of two promoters of apparently equal strengths (292). The more distal promoter functions throughout the growth curve, whereas the proximal promoter is only active as cultures enter stationary phase. The proximal promoter requires a special sigma factor (sigma S) encoded by the rpoS gene (see chapter 90 in the last print predecessor of EcoSal). Indeed, the CFA content of rpoS strains is very low and transcription from the proximal promoter is absent in these strains (292). Despite this increase transcription CFA synthase levels are low in stationary-phase cultures, particularly in rpoS strains where activity declined precipitously in stationary phase. Therefore, the enzyme is unstable and is destroyed by proteolysis (27, 292).
The timing of CFA synthesis can now be explained (27, 292). In log-phase cultures only one promoter, P1, is active resulting in low CFA synthase activity. As cultures enter stationary phase RpoS (sigma S) is synthesized, which activates the proximal P2 promoter, resulting in increased levels of cfa transcription and CFA synthase. Moreover, as growth slows phospholipid synthesis becomes inhibited and, thus, the CFA synthase activity no longer encounters an expanding substrate pool (as is the case in exponentially growing cells). The increased CFA synthase activity can then efficiently convert the membrane phospholipids accumulated during log phase to their CFA derivatives. The instability of CFA synthase in stationary-phase cells results in little carryover of CFA synthetic capacity when exponential growth resumes upon dilution of stationary-phase cultures.
The instability observed for CFA synthase was unexpected since most E. coli proteins are stable (27). However, several regulatory proteins are known to be unstable and the turnover of these proteins is thought to play a role in various regulatory processes. A plausible role for the turnover of CFA synthase would be to prevent CFA synthesis upon resumption of exponential growth by stationary-phase cultures. However, the rationale for such a role is unclear, since high-level production of CFA by log-phase cultures fails to inhibit growth (26, 96, 156).
The enzymology of CFA synthase has been studied primarily from the standpoint of the SAM substrate (39, 100, 135). The most striking result is that the activity of CFA synthase requires bicarbonate, and is inhibited by borate, a planar trigonal molecule that mimics the structure of bicarbonate (40, 136). Substitutions of the conserved amino acids that act as ligands to the bicarbonate ion result in drastic losses in the activity of the protein, some of which can be remedied by addition of free bicarbonate (40, 136).
As mentioned above the literature on inhibitors of fatty acid synthesis has been the subject of many recent reviews (23, 121, 122, 306, 314). Since the most recent of our reviews two potentially very useful inhibitors of 3-ketoacyl-ACP synthases, platencin and platensimycin (295, 296), isolated from Streptomyces, have been reported by Merck. Bayer HealthCare has shown that two other natural products, moiramide B (85) and andrimid (84), are potent inhibitors of bacterial acetyl-CoA carboxylases. The two antibiotics were isolated from atypical sources, Pseudomonas fluorescens and an intracellular bacterial symbiont isolated from a brown plant hopper, respectively. Finally a fungal natural product (Cephalochromin) (317) is the first naturally occurring inhibitor of FabI.
Recent advances in proteomics have allowed survey of the interactions of lipid synthetic proteins with one another and with other cellular proteins mainly by various types of "pull-down" assays in which the bait protein is tagged. Although such data must be taken with more than a few grains of salt, they are interesting. It must also be kept in mind that a positive result does not demonstrate a direct interaction between two proteins. The interaction might be mediated through third-party proteins that were not detected for various technical reasons. This point is often ignored (194). With this caveat the recent data are provocative. The most striking result is that ACP is found in readily isolated complexes with a large number of proteins. Indeed, by this criterion, ACP is the most interactive of any discrete protein of E. coli and is one the four major interaction nodes of the organism; the others being the multiprotein complexes RNA polymerase, DNA polymerase, and the ribosome (22). Some of these interactions are clearly artifacts, such as the proposed interaction of ACP with SecA, a frequent contaminant of pull-down experiments in E. coli extracts, and the interpretation of other interactions suffer from massive overexpression of the bait protein. The most systematic and well controlled interaction study is that of Butland and coworkers (22), in which the tagged proteins were expressed in their normal chromosomal context, thereby ensuring native expression ratios and functionality of any essential proteins. These data have consistency with prior data in that proteins known to form complexes from biochemical studies (e.g., AccA and AccD, ACP and FabB) are detected as interacting. Many of the new interactions seem very reasonable. For example ACP was reported to interact with AcpS, Aas, FabZ, FabI, PlsB, etc. (22). Others are unexpected, such as SpoT, an enzyme that plays key roles in both the synthesis and degradation of the guanosine alarmones, ppGpp and pppGpp. As discussed above accumulation of (p)ppGpp results in blockage of lipid synthesis, which is overcome by PlsB overproduction (110). However, another opposing connection has been reported. Blockage of fatty acid synthesis (but not of phospholipid synthesis) was reported to result in accumulation of (p)ppGpp (256). The distinction between fatty acid synthesis and phospholipid synthesis is difficult to rationalize since (as discussed above) blocking phospholipid synthesis results in inhibition of fatty acid synthesis. The ACP-SpoT interaction has been recently investigated in some detail and the interaction has been interpreted as a regulatory switch that links fatty acid metabolism to the SpoT-dependent stress response (10). Interaction with ACP is reported to be specific to SpoT, the closely related RelA protein does not interact. Also, no interaction was seen between SpoT and a mutant ACP unable to accept the 4'-PP prosthetic group (10). The site of this putative interaction has been localized to a region of SpoT responsible for a shift of the protein to high levels of (p)ppGpp synthesis (10). Thus with the caveat of possible third-party proteins, a plausible case for a physiological role for a SpoT-ACP interaction can be made. Upon inhibition of fatty acid synthesis ACP or some ACP thioester would bind to SpoT and switch the protein into a (p)ppGpp synthetic mode. The accumulation of the alarmone would then shut down other synthetic pathways (10). However, we have no indication of what form of ACP must bind to SpoT in order to exert the regulatory response. ACP is present in a several hundred-fold molar excess over SpoT (10, 192) and most of the ACP is present as the nonacylated form, or as acetyl-ACP and malonyl-ACP (141, 249). Given the great excess of ACP over SpoT, if one or more of these forms was the regulatory molecule, SpoT would always be activated and thus no regulation would occur. Hence, if the ACP-SpoT interaction has physiological significance then the regulatory ACP form must be a minor species that accumulates only under conditions of inhibition of fatty acid synthesis. However, the form of ACP that is the putative regulatory form is not obvious. The form of inhibition tested was cerulenin inhibition. At the concentrations used this should result in accumulation of butyryl-ACP (C4) plus the forms present in unstressed cells (139, 140, 249). However, the older literature indicates that blocking the ACC reaction [by use of an accD(Ts) mutant] has the same effect on (p)ppGpp synthesis as cerulenin (256). A block at the ACC step should result in a complete lack of acylated ACP species (although this has not been experimentally determined) and thus nonacylated ACP should be the form that elicits the (p)ppGpp synthetic mode regulatory response. However, the huge excess of ACP over SpoT (discussed above) makes the nonacylated form seem a very unlikely regulatory ligand. We also must explain why inhibition of fatty acid synthesis engendered by blocking phospholipid synthesis is not equivalent to direct inhibition of the pathway (256).
Protein interaction studies have also led to the concept of a protein network for phospholipid synthesis that includes ACP, PlsB, PssA, and YbgC, a putative thioesterase (101). The putative thioesterase activity is postulated based on the activity on short-chain acyl-CoA thioesters demonstrated for a homologue from Haemophilus influenzae (319) having 53% sequence identity to the E. coli protein. Although the YbgC-ACP interaction had been identified in other studies, it is difficult to assess its role (if any) in E. coli lipid metabolism. Unlike most of the enzymes of E. coli lipid metabolism, YbgC is not essential for growth (7, 148). Thus, it seems unlikely that YbgC plays a key role in the coordination of lipid metabolism. It has been proposed that YbgC is required to terminate acyl chains by thioesterase action such that they could be transferred to lipids (101). This proposal seems speculative since there is no role for such a reaction and it would seem physiologically perverse to cleave an activated bond required for acyl transfer. Moreover, enzymatic activity of E. coli YbgC has not been demonstrated on any substrate much more on the long-chain acyl-ACPs postulated to be its substrates. Nonetheless, further study of YbgC seems indicated, particularly given the fine lines that can divide thioesterases from esterases and acyltransferases.
Our knowledge of lipid synthesis in E. coli is by far the most detailed of any organism and this information has been extremely useful in understanding the pathways of other organisms, particularly those of other bacteria and plants. However, the pathways of lower eukaryotes and mammals have also profited. Indeed, an important advance is the realization that the typical type II bacterial fatty acid synthetic pathway is not limited to bacteria. Several protozoan parasites including the malarial parasite, P. falciparum, use a fatty acid synthetic pathway that is remarkably similar to that of E. coli to provide the precursor for lipoic acid synthesis (41, 254, 274). A parallel picture arises in mitochondria. In yeast mitochondria a complete type II pathway has been demonstrated both genetically and biochemically. Knockouts of any of these genes results in defective lipoic acid synthesis (269). The situation in mammalian mitochondria is similar although less well defined because of the lack of genetics. However, it should be noted that there are bacteria that deviate from the E. coli paradigm. In some cases this is because these bacteria synthesize lipids that are markedly different from those of E. coli (e.g., the branched chain acids of the bacilli). On the other hand, there are bacteria that synthesize lipids very similar to those of E. coli by using markedly different proteins. Three examples should be mentioned. First, in the case of unsaturated fatty acid synthesis genomic analyses indicate that the classical E. coli pathway of anaerobic unsaturated fatty acid synthesis is not widely distributed among bacteria. Indeed, only the α− andγ-proteobacteria have the proteins of this pathway. Other bacteria including many pathogens make unsaturated fatty acids under anaerobic conditions but lack recognizable homologues of the key enzymes (FabA and FabB) of the E. coli pathway. Thus far, Streptococcus pneumoniae has been shown to introduce the cis double bond an isomerase of unrelated sequence that does one of the FabA reactions (FabA is also a dehydratase) (199). In contrast Enterococcus faecalis has proteins that are homologues of FabZ and FabF but perform the roles of FabA and FabB, respectively (294). Moreover, P. aeruginosa has the classical FabA-FabB pathway that provides the bulk of unsaturates, but also has two oxygen-requiring desaturase pathways, one of which uses endogenous phospholipid acyl chains, whereas the other desaturase uses exogenous saturated fatty acids (315). The second exception to the E. coli paradigm is enoyl-ACP reductase, the enzyme that catalyzes the last step of the elongation cycle, formation of a saturated acyl-ACP by an NAD(P)H-dependent reduction of the enoyl-ACP double bond. In E. coli this reaction is catalyzed by FabI. However, although FabI homologues are widely distributed among bacteria and are found even in P. falciparum, at least three additional enoyl-ACP reductase isozymes (FabK, FabL, and FabV) are present in other bacteria (200, 311). Most of these enzymes were discovered by virtue of their resistance to triclosan (113, 205), a synthetic compound used in antibacterial hand soaps and a large variety of other everyday products to which E. coli FabI is extremely sensitive. The diversity of the bacterial enoyl-ACP reductases relative to the lack of structural and mechanistic diversity seen in the other enzymes of the FAS II elongation cycle argued that naturally occurring compounds exist that selectively inhibit one or another of these enzymes. This hypothesis was recently confirmed by the discovery of natural enoyl-ACP reductase inhibitors of fungal origin that specifically target FabI (Cephalochromin) (317) and FabK (Atromentin and Leucomelone) (317). The final exception to the E. coli paradigm is in phospholipid synthesis. As discussed above PlsB is found mainly in the γ-proteobacteria. Most other bacteria use the PlsX-PlsY acyltransferase in place of PlsB (193).
References
1. Abdel-Hamid, A. M., and J. E. Cronan. 2007. Coordinate expression of the acetyl coenzyme A carboxylase genes, accB and accC, is necessary for normal regulation of biotin synthesis in Escherichia coli. J. Bacteriol. 189:369–376.[PubMed] [CrossRef]
2. Alberts, A. W., R. M. Bell, and P. R. Vagelos. 1972. Acyl carrier protein. XV. Studies of β-ketoacyl-acyl carrier protein synthetase. J. Biol. Chem. 247:3190–3198.[PubMed]
3. Alberts, A. W., and P. R. Vagelos. 1972. Acyl-CoA carboxylases. In P. D. Boyer (ed.), The Enzymes. Academic Press, New York, NY.
4. Arthur, C. J., A. Szafranska, S. E. Evans, S. C. Findlow, S. G. Burston, P. Owen, I. Clark-Lewis, T. J. Simpson, J. Crosby, and M. P. Crump. 2005. Self-malonylation is an intrinsic property of a chemically synthesized type II polyketide synthase acyl carrier protein. Biochemistry 44:15414–15421.[PubMed] [CrossRef]
5. Arthur, C. J., A. E. Szafranska, J. Long, J. Mills, R. J. Cox, S. C. Findlow, T. J. Simpson, M. P. Crump, and J. Crosby. 2006. The malonyl transferase activity of type II polyketide synthase acyl carrier proteins. Chem. Biol. 13:587–596.[PubMed] [CrossRef]
6. Attwood, P. V., and J. C. Wallace. 2002. Chemical and catalytic mechanisms of carboxyl transfer reactions in biotin-dependent enzymes. Acc. Chem. Res. 35:113–120.[PubMed] [CrossRef]
7. Baba, T., T. Ara, M. Hasegawa, Y. Takai, Y. Okumura, M. Baba, K. A. Datsenko, M. Tomita, B. L. Wanner, and H. Mori. 2006. Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2:2006–0008. (Online. doi: 10.1038/msb4100050.) [CrossRef]
8. Barbin, A., F. Peypoux, and G. Michel. 1970. Composition phospholipidique de mutants thermosensibles d'Escherichia coli K 12. Biochim. Biophys. Acta 218:453–462.[PubMed]
9. Barnes, E. M., Jr., and S. J. Wakil. 1968. Studies on the mechanism of fatty acid synthesis. XIX. Preparation and general properties of palmityl thioesterase. J. Biol. Chem. 243:2955–2962.[PubMed]
10. Battesti, A., and E. Bouveret. 2006. Acyl carrier protein/SpoT interaction, the switch linking SpoT-dependent stress response to fatty acid metabolism. Mol. Microbiol. 62:1048–1063.[PubMed] [CrossRef]
11. Beckett, D. 2007. Biotin sensing: universal influence of biotin status on transcription. Annu. Rev. Genet. 41:443–464.[PubMed] [CrossRef]
12. Bell, R. M. 1975. Mutants of Escherichia coli defective in membrane phospholipid synthesis. Properties of wild type and Km defective sn-glycerol-3-phosphate acyltransferase activities. J. Biol. Chem. 250:7147–7152.[PubMed]
13. Bell, R. M., and J. E. Cronan, Jr. 1975. Mutants of Escherichia coli defective in membrane phospholipid synthesis. Phenotypic suppression of sn-glycerol-3-phosphate acyltransferase Km mutants by loss of feedback inhibition of the biosynthetic sn-glycerol-3-phosphate dehydrogenase. J. Biol. Chem. 250:7153–7158.[PubMed]
14. Bergler, H., S. Fuchsbichler, G. Hogenauer, and F. Turnowsky. 1996. The enoyl-[acyl-carrier-protein] reductase (FabI) of Escherichia coli, which catalyzes a key regulatory step in fatty acid biosynthesis, accepts NADH and NADPH as cofactors and is inhibited by palmitoyl-CoA. Eur. J. Biochem. 242:689–694.[PubMed] [CrossRef]
15. Bergler, H., P. Wallner, A. Ebeling, B. Leitinger, S. Fuchsbichler, H. Aschauer, G. Kollenz, G. Hogenauer, and F. Turnowsky. 1994. Protein EnvM is the NADH-dependent enoyl-ACP reductase (FabI) of Escherichia coli. J. Biol. Chem. 269:5493–5496.[PubMed]
16. Best, E. A., and V. C. Knauf. 1993. Organization and nucleotide sequences of the genes encoding the biotin carboxyl carrier protein and biotin carboxylase protein of Pseudomonas aeruginosa acetyl coenzyme A carboxylase. J. Bacteriol. 175:6881–6889.[PubMed]
17. Bilder, P., S. Lightle, G. Bainbridge, J. Ohren, B. Finzel, F. Sun, S. Holley, L. Al-Kassim, C. Spessard, M. Melnick, M. Newcomer, and G. L. Waldrop. 2006. The structure of the carboxyltransferase component of acetyl-CoA carboxylase reveals a zinc-binding motif unique to the bacterial enzyme. Biochemistry 45:1712–1722.[PubMed] [CrossRef]
18. Birge, C. H., D. F. Silbert, and P. R. Vagelos. 1967. A β-hydroxydecanoyl-ACP dehydrase specific for saturated fatty acid biosynthesis in E. coli. Biochem. Biophys. Res. Commun. 29:808–814.[PubMed] [CrossRef]
19. Bloch, K. 1971. β-Hydroxydecanoyl thioester dehydrase, p. 441–464. In P. D. Boyer (ed.), The Enzymes, 3rd ed. Academic Press, New York, NY.
20. Bonner, W. M., and K. Bloch. 1972. Purification and properties of fatty acyl thioesterase I from Escherichia coli. J. Biol. Chem. 247:3123–3133.[PubMed]
21. Brown, A. P., J. Coleman, A. M. Tommey, M. D. Watson, and A. R. Slabas. 1994. Isolation and characterization of a maize cDNA that complements a 1-acyl sn-glycerol-3-phosphate acyltransferase mutant of Escherichia coli and encodes a protein which has similarities to other acyltransferases. Plant Mol. Biol. 26:211–223.[PubMed] [CrossRef]
22. Butland, G., J. M. Peregrin-Alvarez, J. Li, W. Yang, X. Yang, V. Canadien, A. Starostine, D. Richards, B. Beattie, N. Krogan, M. Davey, J. Parkinson, J. Greenblatt, and A. Emili. 2005. Interaction network containing conserved and essential protein complexes in Escherichia coli. Nature 433:531–537.[PubMed] [CrossRef]
23. Campbell, J. W., and J. E. Cronan, Jr. 2001. Bacterial fatty acid biosynthesis: targets for antibacterial drug discovery. Annu. Rev. Microbiol. 55:305–332.[PubMed] [CrossRef]
24. Campbell, J. W., and J. E. Cronan, Jr. 2001. Escherichia coli FadR positively regulates transcription of the fabB fatty acid biosynthetic gene. J. Bacteriol. 183:5982–5990.[PubMed] [CrossRef]
25. Carter, J. R., Jr. 1968. Cytidine triphosphate: phosphatidic acid cytidyltransferase in Escherichia coli. J. Lipid Res. 9:748–754.[PubMed]
26. Chang, Y. Y., and J. E. Cronan, Jr. 1999. Membrane cyclopropane fatty acid content is a major factor in acid resistance of Escherichia coli. Mol. Microbiol. 33:249–259.[PubMed] [CrossRef]
27. Chang, Y. Y., J. Eichel, and J. E. Cronan, Jr. 2000. Metabolic instability of Escherichia coli cyclopropane fatty acid synthase is due to RpoH-dependent proteolysis. J. Bacteriol. 182:4288–4294.[PubMed] [CrossRef]
28. Cho, H., and J. E. Cronan, Jr. 1995. Defective export of a periplasmic enzyme disrupts regulation of fatty acid synthesis. J. Biol. Chem. 270:4216–4219.[PubMed] [CrossRef]
29. Cho, H., and J. E. Cronan, Jr. 1993. Escherichia coli thioesterase I, molecular cloning and sequencing of the structural gene and identification as a periplasmic enzyme. J. Biol. Chem. 268:9238–9245.[PubMed]
30. Cho, H., and J. E. Cronan, Jr. 1994. "Protease I" of Escherichia coli functions as a thioesterase in vivo. J. Bacteriol. 176:1793–1795.[PubMed]
31. Choi-Rhee, E., and J. E. Cronan. 2003. The biotin carboxylase-biotin carboxyl carrier protein complex of Escherichia coli acetyl-CoA carboxylase. J. Biol. Chem. 278:30806–30812.[PubMed] [CrossRef]
32. Clark, D., V. Lightner, R. Edgar, P. Modrich, J. E. Cronan, Jr., and R. M. Bell. 1980. Regulation of phospholipid biosynthesis in Escherichia coli. Cloning of the structural gene for the biosynthetic sn-glycerol-3-phosphate dehydrogenase. J. Biol. Chem. 255:714–717.[PubMed]
33. Clark, D. P., D. DeMendoza, M. L. Polacco, and J. E. Cronan, Jr. 1983. β-Hydroxydecanoyl thio ester dehydrase does not catalyze a rate-limiting step in Escherichia coli unsaturated fatty acid synthesis. Biochemistry 22:5897–5902.[PubMed] [CrossRef]
34. Coleman, J. 1990. Characterization of Escherichia coli cells deficient in 1-acyl-sn-glycerol-3-phosphate acyltransferase activity. J. Biol. Chem. 265:17215–17221.[PubMed]
35. Coleman, J. 1992. Characterization of the Escherichia coli gene for 1-acyl-sn-glycerol-3-phosphate acyltransferase (plsC). Mol. Gen. Genet. 232:295–303.[PubMed]
36. Collado-Vides, J., B. Magasanik, and J. D. Gralla. 1991. Control site location and transcriptional regulation in Escherichia coli. Microbiol. Rev. 55:371–394.[PubMed]
37. Cooper, C. L., L. Hsu, S. Jackowski, and C. O. Rock. 1989. 2-Acylglycerolphosphoethanolamine acyltransferase/acyl-acyl carrier protein synthetase is a membrane-associated acyl carrier protein binding protein. J. Biol. Chem. 264:7384–7389.[PubMed]
38. Cooper, C. L., S. Jackowski, and C. O. Rock. 1987. Fatty acid metabolism in sn-glycerol-3-phosphate acyltransferase (plsB) mutants. J. Bacteriol. 169:605–611.[PubMed]
39. Courtois, F., C. Guerard, X. Thomas, and O. Ploux. 2004. Escherichia coli cyclopropane fatty acid synthase. Eur. J. Biochem. 271:4769–4778.[PubMed] [CrossRef]
40. Courtois, F., and O. Ploux. 2005. Escherichia coli cyclopropane fatty acid synthase: is a bound bicarbonate ion the active-site base? Biochemistry 44:13583–13590.[PubMed] [CrossRef]
41. Crawford, M. J., N. Thomsen-Zieger, M. Ray, J. Schachtner, D. S. Roos, and F. Seeber. 2006. Toxoplasma gondii scavenges host-derived lipoic acid despite its de novo synthesis in the apicoplast. EMBO J. 25:3214–3222.[PubMed] [CrossRef]
42. Cronan, J. E. 2003. Bacterial membrane lipids: where do we stand? Annu. Rev. Microbiol. 57:203–224.[PubMed] [CrossRef]
43. Cronan, J. E., Jr. 1997. In vivo evidence that acyl coenzyme A regulates DNA binding by the Escherichia coli FadR global transcription factor. J. Bacteriol. 179:1819–1823.[PubMed]
44. Cronan, J. E., Jr. 1978. Molecular biology of bacterial membrane lipids. Annu. Rev. Biochem. 47:163–189.[PubMed] [CrossRef]
45. Cronan, J. E., Jr. 1982. Molecular properties of short chain acyl thioesters of acyl carrier protein. J. Biol. Chem. 257:5013–5017.[PubMed]
46. Cronan, J. E., Jr. 1968. Phospholipid alterations during growth of Escherichia coli. J. Bacteriol. 95:2054–2061.[PubMed]
47. Cronan, J. E., Jr. 2001. The biotinyl domain of Escherichia coli acetyl-CoA carboxylase. Evidence that the "thumb" structure is essential and that the domain functions as a dimer. J. Biol. Chem. 276:37355–37364.[PubMed] [CrossRef]
48. Cronan, J. E., Jr., and R. M. Bell. 1974. Mutants of Escherichia coli defective in membrane phospholipid synthesis: mapping of sn-glycerol 3-phosphate acyltransferase Km mutants. J. Bacteriol. 120:227–233.[PubMed]
49. Cronan, J. E., Jr., C. H. Birge, and P. R. Vagelos. 1969. Evidence for two genes specifically involved in unsaturated fatty acid biosynthesis in Escherichia coli. J. Bacteriol. 100:601–604.[PubMed]
50. Cronan, J. E., Jr., and E. P. Gelmann. 1973. An estimate of the minimum amount of unsaturated fatty acid required for growth of Escherichia coli. J. Biol. Chem. 248:1188–1195.[PubMed]
51. Cronan, J. E., Jr., W. B. Li, R. Coleman, M. Narasimhan, D. de Mendoza, and J. M. Schwab. 1988. Derived amino acid sequence and identification of active site residues of Escherichia coli β-hydroxydecanoyl thioester dehydrase. J. Biol. Chem. 263:4641–4646.[PubMed]
52. Cronan, J. E., Jr., and C. O. Rock. 1994. The presence of linoleic acid in Escherichia coli cannot be confirmed. J. Bacteriol. 176:3069–3071.[PubMed]
53. Cronan, J. E., Jr., and G. L. Waldrop. 2002. Multi-subunit acetyl-CoA carboxylases. Prog. Lipid Res. 41:407–435.[PubMed] [CrossRef]
54. Cronan, J. E., Jr., L. J. Weisberg, and R. G. Allen. 1975. Regulation of membrane lipid synthesis in Escherichia coli. Accumulation of free fatty acids of abnormal length during inhibition of phospholipid synthesis. J. Biol. Chem. 250:5835–5840.[PubMed]
55. Cronan, J. E., and C. O. Rock. 1996. Biosynthesis of membrane lipids, p. 612–636. In R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, vol. 1. ASM Press, Washington, DC.
56. Cronan, J. E., and C. O. Rock. 1987. Biosynthesis of membrane lipids, p. 3474–3497. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, vol. 1. American Society for Microbiology, Washington, DC.
57. Cronan, J. E., and P. R. Vagelos. 1972. Metabolism and function of the membrane phospholipids of Escherichia coli. Biochim. Biophys. Acta 265:25–60.[PubMed]
58. D'Agnolo, G., I. S. Rosenfeld, and P. R. Vagelos. 1975. Multiple forms of β-ketoacyl-acyl carrier protein synthetase in Escherichia coli. J. Biol. Chem. 250:5289–5294.[PubMed]
59. D'Agnolo, G., I. S. Rosenfeld, and P. R. Vagelos. 1975. β-Ketoacyl-acyl carrier protein synthetase. Characterization of the acyl-enzyme intermediate. J. Biol. Chem. 250:5283–5288.[PubMed]
60. Davis, M. S., J. Solbiati, and J. E. Cronan, Jr. 2000. Overproduction of acetyl-CoA carboxylase activity increases the rate of fatty acid biosynthesis in Escherichia coli. J. Biol. Chem. 275:28593–28598.[PubMed] [CrossRef]
61. De Lay, N. R., and J. E. Cronan. 2006. A genome rearrangement has orphaned the Escherichia coli K-12 AcpT phosphopantetheinyl transferase from its cognate Escherichia coli O157:H7 substrates. Mol. Microbiol. 61:232–242.[PubMed] [CrossRef]
62. De Lay, N. R., and J. E. Cronan. 2006. Gene-specific random mutagenesis of Escherichia coli in vivo: isolation of temperature-sensitive mutations in the acyl carrier protein of fatty acid synthesis. J. Bacteriol. 188:287–296.[PubMed] [CrossRef]
63. De Lay, N. R., and J. E. Cronan. 2007. In vivo functional analyses of the type II acyl carrier proteins of fatty acid biosynthesis. J. Biol. Chem. 282:20319–20328.[PubMed] [CrossRef]
64. de Mendoza, D., J. L. Garwin, and J. E. Cronan, Jr. 1982. Overproduction of cis-vaccenic acid and altered temperature control of fatty acid synthesis in a mutant of Escherichia coli. J. Bacteriol. 151:1608–1611.[PubMed]
65. de Mendoza, D., A. Klages Ulrich, and J. E. Cronan, Jr. 1983. Thermal regulation of membrane fluidity in Escherichia coli. Effects of overproduction of β-ketoacyl-acyl carrier protein synthase I. J. Biol. Chem. 258:2098–2101.[PubMed]
66. DeChavigny, A., P. N. Heacock, and W. Dowhan. 1991. Sequence and inactivation of the pss gene of Escherichia coli. Phosphatidylethanolamine may not be essential for cell viability. J. Biol. Chem. 266:10710.
67. Dillon, D. A., W. I. Wu, B. Riedel, J. B. Wissing, W. Dowhan, and G. M. Carman. 1996. The Escherichia coli pgpB gene encodes for a diacylglycerol pyrophosphate phosphatase activity. J. Biol. Chem. 271:30548–30553.[PubMed] [CrossRef]
68. DiRusso, C. C., T. L. Heimert, and A. K. Metzger. 1992. Characterization of FadR, a global transcriptional regulator of fatty acid metabolism in Escherichia coli. Interaction with the fadB promoter is prevented by long chain fatty acyl coenzyme A. J. Biol. Chem. 267:8685–8691.[PubMed]
69. DiRusso, C. C., A. K. Metzger, and T. L. Heimert. 1993. Regulation of transcription of genes required for fatty acid transport and unsaturated fatty acid biosynthesis in Escherichia coli by FadR. Mol. Microbiol. 7:311–322.[PubMed] [CrossRef]
70. Doerrler, W. T., H. S. Gibbons, and C. R. Raetz. 2004. MsbA-dependent translocation of lipids across the inner membrane of Escherichia coli. J. Biol. Chem. 279:45102–45109.[PubMed] [CrossRef]
71. Doi, O., and S. Nojima. 1975. Lysophospholipase of Escherichia coli. J. Biol. Chem. 250:5208–5214.[PubMed]
72. Dormann, P., T. A. Voelker, and J. B. Ohlrogge. 1995. Cloning and expression in Escherichia coli of a novel thioesterase from Arabidopsis thaliana specific for long-chain acyl-acyl carrier proteins. Arch. Biochem. Biophys. 316:612–618.[PubMed] [CrossRef]
73. Dowhan, W. 1997. CDP-diacylglycerol synthase of microorganisms. Biochim. Biophys. Acta 1348:157–165.[PubMed]
74. Dowhan, W. 1992. Phosphatidylglycerophosphate synthase from Escherichia coli. Methods Enzymol. 209:313–321.[PubMed] [CrossRef]
75. Dowhan, W., and C. R. Funk. 1992. Phosphatidylglycerophosphate phosphatase from Escherichia coli. Methods Enzymol. 209:224–230.[PubMed] [CrossRef]
76. Dowhan, W., and Q. X. Li. 1992. Phosphatidylserine decarboxylase from Escherichia coli. Methods Enzymol. 209:348–359.[PubMed] [CrossRef]
77. Dowhan, W., E. Mileykovskaya, and M. Bogdanov. 2004. Diversity and versatility of lipid-protein interactions revealed by molecular genetic approaches. Biochim. Biophys. Acta 1666:19–39.[PubMed] [CrossRef]
78. Edwards, P., J. S. Nelsen, J. G. Metz, and K. Dehesh. 1997. Cloning of the fabF gene in an expression vector and in vitro characterization of recombinant fabF and fabB encoded enzymes from Escherichia coli. FEBS Lett. 402:62–66.[PubMed] [CrossRef]
79. El Ghachi, M., A. Derbise, A. Bouhss, and D. Mengin-Lecreulx. 2005. Identification of multiple genes encoding membrane proteins with undecaprenyl pyrophosphate phosphatase (UppP) activity in Escherichia coli. J. Biol. Chem. 280:18689–18695.[PubMed] [CrossRef]
80. Fall, R. R. 1976. Stabilization of an acetyl-coenzyme A carboxylase complex from Pseudomonas citronellolis. Biochim. Biophys. Acta 450:475–480.[PubMed]
81. Fall, R. R., and P. R. Vagelos. 1973. Acetyl coenzyme A carboxylase. Proteolytic modification of biotin carboxyl carrier protein. J. Biol. Chem. 248:2078–2088.[PubMed]
82. Fischl, A. S., and E. P. Kennedy. 1990. Isolation and properties of acyl carrier protein phosphodiesterase of Escherichia coli. J. Bacteriol. 172:5445–5449.[PubMed]
83. Flugel, R. S., Y. Hwangbo, R. H. Lambalot, J. E. Cronan, Jr., and C. T. Walsh. 2000. Holo-(acyl carrier protein) synthase and phosphopantetheinyl transfer in Escherichia coli. J. Biol. Chem. 275:959–968.[PubMed] [CrossRef]
84. Freiberg, C., N. A. Brunner, G. Schiffer, T. Lampe, J. Pohlmann, M. Brands, M. Raabe, D. Habich, and K. Ziegelbauer. 2004. Identification and characterization of the first class of potent bacterial acetyl-CoA carboxylase inhibitors with antibacterial activity. J. Biol. Chem. 279:26066–26073.[PubMed] [CrossRef]
85. Freiberg, C., J. Pohlmann, P. G. Nell, R. Endermann, J. Schuhmacher, B. Newton, M. Otteneder, T. Lampe, D. Habich, and K. Ziegelbauer. 2006. Novel bacterial acetyl coenzyme A carboxylase inhibitors with antibiotic efficacy in vivo. Antimicrob. Agents Chemother. 50:2707–2712.[PubMed] [CrossRef]
86. Funk, C. R., L. Zimniak, and W. Dowhan. 1992. The pgpA and pgpB genes of Escherichia coli are not essential: evidence for a third phosphatidylglycerophosphate phosphatase. J. Bacteriol. 174:205–213.[PubMed]
87. Fuqua, C., and E. P. Greenberg. 2002. Listening in on bacteria: acyl-homoserine lactone signalling. Nat. Rev. Mol. Cell. Biol. 3:685–695.[PubMed] [CrossRef]
88. Ganong, B. R., J. M. Leonard, and C. R. Raetz. 1980. Phosphatidic acid accumulation in the membranes of Escherichia coli mutants defective in CDP-diglyceride synthetase. J. Biol. Chem. 255:1623–1629.[PubMed]
89. Ganong, B. R., and C. R. Raetz. 1982. Massive accumulation of phosphatidic acid in conditionally lethal CDP-diglyceride synthetase mutants and cytidine auxotrophs of Escherichia coli. J. Biol. Chem. 257:389–394.[PubMed]
90. Ganong, B. R., and C. R. Raetz. 1983. pH-sensitive CDP-diglyceride synthetase mutants of Escherichia coli: phenotypic suppression by mutations at a second site. J. Bacteriol. 153:731–738.[PubMed]
91. Garwin, J. L., and J. E. Cronan, Jr. 1980. Thermal modulation of fatty acid synthesis in Escherichia coli does not involve de novo enzyme synthesis. J. Bacteriol. 141:1457–1459.[PubMed]
92. Garwin, J. L., A. L. Klages, and J. E. Cronan, Jr. 1980. β-Ketoacyl-acyl carrier protein synthase II of Escherichia coli. Evidence for function in the thermal regulation of fatty acid synthesis. J. Biol.Chem. 255:3263–3265.[PubMed]
93. Garwin, J. L., A. L. Klages, and J. E. Cronan, Jr. 1980. Structural, enzymatic, and genetic studies of β-ketoacyl-acyl carrier protein synthases I and II of Escherichia coli. J. Biol. Chem. 255:11949–11956.[PubMed]
94. Gerdes, S. Y., M. D. Scholle, J. W. Campbell, G. Balazsi, E. Ravasz, M. D. Daugherty, A. L. Somera, N. C. Kyrpides, I. Anderson, M. S. Gelfand, A. Bhattacharya, V. Kapatral, M. D'Souza, M. V. Baev, Y. Grechkin, F. Mseeh, M. Y. Fonstein, R. Overbeek, A. L. Barabasi, Z. N. Oltvai, and A. L. Osterman. 2003. Experimental determination and system level analysis of essential genes in Escherichia coli MG1655. J. Bacteriol. 185:5673–5684.[PubMed] [CrossRef]
95. Goelz, S. E., and J. E. Cronan, Jr. 1980. The positional distribution of fatty acids in Escherichia coli phospholipids is not regulated by sn-glycerol 3-phosphate levels. J. Bacteriol. 144:462–464.[PubMed]
96. Grogan, D. W., and J. E. Cronan, Jr. 1986. Characterization of Escherichia coli mutants completely defective in synthesis of cyclopropane fatty acids. J. Bacteriol. 166:872–877.[PubMed]
97. Grogan, D. W., and J. E. Cronan, Jr. 1997. Cyclopropane ring formation in membrane lipids of bacteria. Microbiol. Mol. Biol. Rev. 61:429–441.[PubMed]
98. Guchhait, R. B., S. E. Polakis, P. Dimroth, E. Stoll, J. Moss, and M. D. Lane. 1974. Acetyl coenzyme A carboxylase system of Escherichia coli. Purification and properties of the biotin carboxylase, carboxyltransferase, and carboxyl carrier protein components. J. Biol. Chem. 249:6633–6645.[PubMed]
99. Guerra, D. J., and J. A. Browse. 1990. Escherichia coli β-hydroxydecanoyl thioester dehydrase reacts with native C10 acyl-acyl-carrier proteins of plant and bacterial origin. Arch. Biochem. Biophys. 280:336–345.[PubMed] [CrossRef]
100. Guianvarc'h, D., T. Drujon, T. E. Leang, F. Courtois, and O. Ploux. 2006. Identification of new inhibitors of Escherichia coli cyclopropane fatty acid synthase using a colorimetric assay. Biochim. Biophys. Acta 1764:1381–1388.[PubMed]
101. Gully, D., and E. Bouveret. 2006. A protein network for phospholipid synthesis uncovered by a variant of the tandem affinity purification method in Escherichia coli. Proteomics 6:282–293.[PubMed] [CrossRef]
102. Gully, D., D. Moinier, L. Loiseau, and E. Bouveret. 2003. New partners of acyl carrier protein detected in Escherichia coli by tandem affinity purification. FEBS Lett. 548:90–96.[PubMed] [CrossRef]
103. Guo, D., and B. E. Tropp. 2000. A second Escherichia coli protein with CL synthase activity. Biochim. Biophys. Acta 1483:263–274.[PubMed]
104. Hanke, C., F. P. Wolter, J. Coleman, G. Peterek, and M. Frentzen. 1995. A plant acyltransferase involved in triacylglycerol biosynthesis complements an Escherichia coli sn-1-acylglycerol-3-phosphate acyltransferase mutant. Eur. J. Biochem. 232:806–810.[PubMed] [CrossRef]
105. Hawrot, E., and E. P. Kennedy. 1975. Biogenesis of membrane lipids: mutants of Escherichia coli with temperature-sensitive phosphatidylserine decarboxylase. Proc. Natl. Acad. Sci. USA 72:1112–1116.[PubMed] [CrossRef]
106. Hawrot, E., and E. P. Kennedy. 1976. Conditional lethal phosphatidylserine decarboxylase mutants of Escherichia coli. Mapping of the structural gene for phosphatidylserine decarboxylase. Mol. Gen. Genet. 148:271–279.[PubMed] [CrossRef]
107. Hawrot, E., and E. P. Kennedy. 1978. Phospholipid composition and membrane function in phosphatidylserine decarboxylase mutants of Escherichia coli. J. Biol. Chem. 253:8213–8220.[PubMed]
108. Heacock, P. N., and W. Dowhan. 1987. Construction of a lethal mutation in the synthesis of the major acidic phospholipids of Escherichia coli. J. Biol. Chem. 262:13044–13049.[PubMed]
109. Heath, R. J., H. Goldfine, and C. O. Rock. 1997. A gene (plsD) from Clostridium butyricum that functionally substitutes for the sn-glycerol-3-phosphate acyltransferase gene (plsB) of Escherichia coli. J. Bacteriol. 179:7257–7263.[PubMed]
110. Heath, R. J., S. Jackowski, and C. O. Rock. 1994. Guanosine tetraphosphate inhibition of fatty acid and phospholipid synthesis in Escherichia coli is relieved by overexpression of glycerol-3-phosphate acyltransferase (plsB). J. Biol. Chem. 269:26584–26590.[PubMed]
111. Heath, R. J., and C. O. Rock. 1998. A conserved histidine is essential for glycerolipid acyltransferase catalysis. J. Bacteriol. 180:1425–1430.[PubMed]
112. Heath, R. J., and C. O. Rock. 1999. A missense mutation accounts for the defect in the glycerol-3-phosphate acyltransferase expressed in the plsB26 mutant. J. Bacteriol. 181:1944–1946.[PubMed]
113. Heath, R. J., and C. O. Rock. 2000. A triclosan-resistant bacterial enzyme. Nature 406:145–146.[PubMed] [CrossRef]
114. Heath, R. J., and C. O. Rock. 1995. Enoyl-acyl carrier protein reductase (fabI) plays a determinant role in completing cycles of fatty acid elongation in Escherichia coli. J. Biol. Chem. 270:26538–26542.[PubMed] [CrossRef]
115. Heath, R. J., and C. O. Rock. 1996. Inhibition of β-ketoacyl-acyl carrier protein synthase III (FabH) by acyl-acyl carrier protein in Escherichia col. J. Biol. Chem. 271:10996–11000.[PubMed] [CrossRef]
116. Heath, R. J., and C. O. Rock. 1996. Regulation of fatty acid elongation and initiation by acyl-acyl carrier protein in Escherichia coli. J. Biol. Chem. 271:1833–1836.[PubMed] [CrossRef]
117. Heath, R. J., and C. O. Rock. 1995. Regulation of malonyl-CoA metabolism by acyl-acyl carrier protein and β-ketoacyl-acyl carrier protein synthases in Escherichia coli. J. Biol. Chem. 270:15531–15538.[PubMed] [CrossRef]
118. Heath, R. J., and C. O. Rock. 1996. Roles of the FabA and FabZ β-hydroxyacyl-acyl carrier protein dehydratases in Escherichia coli fatty acid biosynthesis. J. Biol. Chem. 271:27795–27801.[PubMed] [CrossRef]
119. Heath, R. J., and C. O. Rock. 2002. The Claisen condensation in biology. Nat. Prod. Rep. 19:581–596.[PubMed] [CrossRef]
120. Heath, R. J., J. R. Rubin, D. R. Holland, E. Zhang, M. E. Snow, and C. O. Rock. 1999. Mechanism of triclosan inhibition of bacterial fatty acid synthesis. J. Biol. Chem. 274:11110–11114.[PubMed] [CrossRef]
121. Heath, R. J., S. W. White, and C. O. Rock. 2002. Inhibitors of fatty acid synthesis as antimicrobial chemotherapeutics. Appl. Microbiol. Biotechnol. 58:695–703.[PubMed] [CrossRef]
122. Heath, R. J., S. W. White, and C. O. Rock. 2001. Lipid biosynthesis as a target for antibacterial agents. Prog. Lipid Res. 40:467–497.[PubMed] [CrossRef]
123. Henry, M. F., and J. E. Cronan, Jr. 1992. A new mechanism of transcriptional regulation: release of an activator triggered by small molecule binding. Cell 70:671–679.[PubMed] [CrossRef]
124. Henry, M. F., and J. E. Cronan, Jr. 1991. Escherichia coli transcription factor that both activates fatty acid synthesis and represses fatty acid degradation. J. Mol. Biol. 222:843–849.[PubMed] [CrossRef]
125. Hiraoka, S., H. Matsuzaki, and I. Shibuya. 1993. Active increase in cardiolipin synthesis in the stationary growth phase and its physiological significance in Escherichia coli. FEBS Lett. 336:221–224.[PubMed] [CrossRef]
126. Hirschberg, C. B., and E. P. Kennedy. 1972. Mechanism of the enzymatic synthesis of cardiolipin in Escherichia coli. Proc. Natl. Acad. Sci. USA 69:648–651.[PubMed] [CrossRef]
127. Hitchman, T. S., J. Crosby, K. J. Byrom, R. J. Cox, and T. J. Simpson. 1998. Catalytic self-acylation of type II polyketide synthase acyl carrier proteins. Chem. Biol. 5:35–47.[PubMed] [CrossRef]
128. Homma, H., M. Nishijima, T. Kobayashi, H. Okuyama, and S. Nojima. 1981. Incorporation and metabolism of 2-acyl lysophospholipids by Escherichia coli. Biochim. Biophys. Acta 663:1–13.[PubMed]
129. Hsu, L., S. Jackowski, and C. O. Rock. 1991. Isolation and characterization of Escherichia coli K-12 mutants lacking both 2-acyl-glycerophosphoethanolamine acyltransferase and acyl-acyl carrier protein synthetase activity. J. Biol. Chem. 266:13783–13788.[PubMed]
130. Hsu, L., S. Jackowski, and C. O. Rock. 1989. Uptake and acylation of 2-acyl-lysophospholipids by Escherichia coli. J. Bacteriol. 171:1203–1205.[PubMed]
131. Ichihara, S., Y. Matsubara, C. Kato, K. Akasaka, and S. Mizushima. 1993. Molecular cloning, sequencing, and mapping of the gene encoding protease I and characterization of proteinase and proteinase-defective Escherichia coli mutants. J. Bacteriol. 175:1032–1037.[PubMed]
132. Icho, T. 1988. Membrane-bound phosphatases in Escherichia coli: sequence of the pgpB gene and dual subcellular localization of the pgpB product. J. Bacteriol. 170:5117–5124.[PubMed]
133. Issartel, J. P., V. Koronakis, and C. Hughes. 1991. Activation of Escherichia coli prohaemolysin to the mature toxin by acyl carrier protein-dependent fatty acylation. Nature 351:759–761.[PubMed] [CrossRef]
134. Ito, K., M. Nakanishi, W. C. Lee, Y. Zhi, H. Sasaki, S. Zenno, K. Saigo, Y. Kitade, and M. Tanokura. 2008. Expansion of substrate specificity and catalytic mechanism of azoreductase by X-ray crystallography and site-directed mutagenesis. J. Biol. Chem. 283:13889–13896.[PubMed] [CrossRef]
135. Iwig, D. F., A. T. Grippe, T. A. McIntyre, and S. J. Booker. 2004. Isotope and elemental effects indicate a rate-limiting methyl transfer as the initial step in the reaction catalyzed by Escherichia coli cyclopropane fatty acid synthase. Biochemistry 43:13510–13524.[PubMed] [CrossRef]
136. Iwig, D. F., A. Uchida, J. A. Stromberg, and S. J. Booker. 2005. The activity of Escherichia coli cyclopropane fatty acid synthase depends on the presence of bicarbonate. J. Am. Chem. Soc. 127:11612–11613.[PubMed] [CrossRef]
137. Jackowski, S., H. H. Edwards, D. Davis, and C. O. Rock. 1985. Localization of acyl carrier protein in Escherichia coli. J. Bacteriol. 162:5–8.[PubMed]
138. Jackowski, S., L. Hsu, and C. O. Rock. 1992. 2-Acylglycerophosphoethanolamine acyltransferase/acyl-[acyl-carrier- protein] synthetase from Escherichia coli. Methods Enzymol. 209:111–117.[PubMed] [CrossRef]
139. Jackowski, S., C. M. Murphy, J. E. Cronan, Jr., and C. O. Rock. 1989. Acetoacetyl-acyl carrier protein synthase. A target for the antibiotic thiolactomycin. J. Biol. Chem. 264:7624–7629.[PubMed]
140. Jackowski, S., and C. O. Rock. 1987. Acetoacetyl-acyl carrier protein synthase, a potential regulator of fatty acid biosynthesis in bacteria. J. Biol. Chem. 262:7927–7931.[PubMed]
141. Jackowski, S., and C. O. Rock. 1983. Ratio of active to inactive forms of acyl carrier protein in Escherichia coli. J. Biol. Chem. 258:15186–15191.[PubMed]
142. Jackowski, S., Y. M. Zhang, A. C. Price, S. W. White, and C. O. Rock. 2002. A missense mutation in the fabB β-ketoacyl-acyl carrier protein synthase I) gene confers thiolactomycin resistance to Escherichia coli. Antimicrob. Agents Chemother. 46:1246–1252.[PubMed] [CrossRef]
143. James, E. S., and J. E. Cronan. 2004. Expression of two Escherichia coli acetyl-CoA carboxylase subunits is autoregulated. J. Biol. Chem. 279:2520–2527.[PubMed] [CrossRef]
144. Jayasuriya, H., K. B. Herath, C. Zhang, D. L. Zink, A. Basilio, O. Genilloud, M. T. Diez, F. Vicente, I. Gonzalez, O. Salazar, F. Pelaez, R. Cummings, S. Ha, J. Wang, and S. B. Singh. 2007. Isolation and structure of platencin: a FabH and FabF dual inhibitor with potent broad-spectrum antibiotic activity. Angew. Chem. Int. Ed. Engl. 46:4684–4688.[PubMed] [CrossRef]
145. Jiang, P., and J. E. Cronan, Jr. 1994. Inhibition of fatty acid synthesis in Escherichia coli in the absence of phospholipid synthesis and release of inhibition by thioesterase action. J. Bacteriol. 176:2814–2821.[PubMed]
146. Jitrapakdee, S., and J. C. Wallace. 2003. The biotin enzyme family: conserved structural motifs and domain rearrangements. Curr. Protein Pept. Sci. 4:217–229.[PubMed] [CrossRef]
147. Kanfer, J., and E. P. Kennedy. 1964. Metabolism and function of bacterial lipids. II. biosynthesis of phospholipids in Escherichia coli. J. Biol. Chem. 239:1720–1726.[PubMed]
148. Kang, Y., T. Durfee, J. D. Glasner, Y. Qiu, D. Frisch, K. M. Winterberg, and F. R. Blattner. 2004. Systematic mutagenesis of the Escherichia coli genome. J. Bacteriol. 186:4921–4930.[PubMed] [CrossRef]
149. Kaplan, A., and P. D. Boyer. 1968. Catalysis of water oxygen and of acetate incorporation into fatty acids by Escherichia coli fatty acid synthetase. J. Biol. Chem. 243:4077–4082.[PubMed]
150. Karasawa, K., K. Yokoyama, M. Setaka, and S. Nojima. 1999. The Escherichia coli pldC gene encoding lysophospholipase L(1) is identical to the apeA and tesA genes encoding protease I and thioesterase I, respectively. J. Biochem. (Tokyo) 126:445–448.[PubMed]
151. Karow, M., O. Fayet, and C. Georgopoulos. 1992. The lethal phenotype caused by null mutations in the Escherichia coli htrB gene is suppressed by mutations in the accBC operon, encoding two subunits of acetyl coenzyme A carboxylase. J. Bacteriol. 174:7407–7418.[PubMed]
152. Kater, M. M., G. M. Koningstein, H. J. Nijkamp, and A. R. Stuitje. 1994. The use of a hybrid genetic system to study the functional relationship between prokaryotic and plant multi-enzyme fatty acid synthetase complexes. Plant Mol. Biol. 25:771–790.[PubMed] [CrossRef]
153. Kauppinen, S., M. Siggaard-Anderson, and P. van Wettstein-Knowles. 1988. β-Ketoacyl-ACP synthase I of Escherichia coli: nucleotide sequence of the fabB gene and identification of the cerulenin binding residue. Carlsberg Res. Commun. 53:357–370.[PubMed] [CrossRef]
154. Ke, L., R. Engel, and B. E. Tropp. 1992. The phosphonic acid analog of phosphatidylglycerol phosphate: influence on Escherichia coli growth and physiology. Biochim. Biophys. Acta 1128:250–257.[PubMed]
155. Keating, D. H., M. R. Carey, and J. E. Cronan, Jr. 1995. The unmodified (apo) form of Escherichia coli acyl carrier protein is a potent inhibitor of cell growth. J. Biol. Chem. 270:22229–22235.[PubMed] [CrossRef]
156. Kim, B. H., S. Kim, H. G. Kim, J. Lee, I. S. Lee, and Y. K. Park. 2005. The formation of cyclopropane fatty acids in Salmonella enterica serovar Typhimurium. Microbiology 151:209–218.[PubMed] [CrossRef]
157. Kim, Y., and J. H. Prestegard. 1989. A dynamic model for the structure of acyl carrier protein in solution. Biochemistry 28:8792–8797.[PubMed] [CrossRef]
158. Kim, Y., and J. H. Prestegard. 1990. Refinement of the NMR structures for acyl carrier protein with scalar coupling data. Proteins 8:377–385.[PubMed] [CrossRef]
159. Kimber, M. S., F. Martin, Y. Lu, S. Houston, M. Vedadi, A. Dharamsi, K. M. Fiebig, M. Schmid, and C. O. Rock. 2004. The structure of (3R)-hydroxyacyl-acyl carrier protein dehydratase (FabZ) from Pseudomonas aeruginosa. J. Biol. Chem. 279:52593–52602.[PubMed] [CrossRef]
160. Kito, M., M. Lubin, and L. I. Pizer. 1969. A mutant of Escherichia coli defective in phosphatidic acid synthesis. Biochem. Biophys. Res. Commun. 34:454–458.[PubMed] [CrossRef]
161. Kito, M., and L. I. Pizer. 1969. Phosphatidic acid synthesis in Escherichia coli. J. Bacteriol. 97:1321–1327.[PubMed]
162. Kito, M., and L. I. Pizer. 1969. Purification and regulatory properties of the biosynthetic L-glycerol 3-phosphate dehydrogenase from Escherichia coli. J. Biol. Chem. 244:3316–3323.[PubMed]
163. Klein, K., R. Steinberg, B. Fiethen, and P. Overath. 1971. Fatty acid degradation in Escherichia coli. An inducible system for the uptake of fatty acids and further characterization of old mutants. Eur. J. Biochem. 19:442–450.[PubMed] [CrossRef]
164. Klinke, S., Q. Ren, B. Witholt, and B. Kessler. 1999. Production of medium-chain-length poly(3-hydroxyalkanoates) from gluconate by recombinant Escherichia coli. Appl. Environ. Microbiol. 65:540–548.[PubMed]
165. Kol, M. A., D. W. Kuster, H. A. Boumann, H. de Cock, A. J. Heck, B. de Kruijff, and A. I. de Kroon. 2004. Uptake and remodeling of exogenous phosphatidylethanolamine in E. coli. Biochim. Biophys. Acta 1636:205–212.[PubMed]
166. Kragelund, B. B., J. Knudsen, and F. M. Poulsen. 1999. Acyl-coenzyme A binding protein (ACBP). Biochim. Biophys. Acta 1441:150–161.[PubMed]
167. Lai, C. Y., and J. E. Cronan. 2004. Isolation and characterization of β-ketoacyl-acyl carrier protein reductase (fabG) mutants of Escherichia coli and Salmonella enterica serovar Typhimurium. J. Bacteriol. 186:1869–1878.[PubMed] [CrossRef]
168. Lai, C. Y., and J. E. Cronan. 2003. β-Ketoacyl-acyl carrier protein synthase III (FabH) is essential for bacterial fatty acid synthesis. J. Biol. Chem. 278:51494–51503.[PubMed] [CrossRef]
169. Lambalot, R. H., A. M. Gehring, R. S. Flugel, P. Zuber, M. LaCelle, M. A. Marahiel, R. Reid, C. Khosla, and C. T. Walsh. 1996. A new enzyme superfamily—the phosphopantetheinyl transferases. Chem. Biol. 3:923–936.[PubMed] [CrossRef]
170. Lambalot, R. H., and C. T. Walsh. 1997. Holo-[acyl-carrier-protein] synthase of Escherichia coli. Methods Enzymol. 279:254–262.[PubMed] [CrossRef]
171. Larson, T., M. Ehrmann, and W. Boos. 1883. Periplasmic glycerophosphodiester phosphodiesterase of Escherichia coli, a new enzyme of the glp regulon. J. Biol. Chem. 258:5428–5432.
172. Larson, T. J., D. N. Ludtke, and R. M. Bell. 1984. sn-Glycerol-3-phosphate auxotrophy of plsB strains of Escherichia coli: evidence that a second mutation, plsX, is required. J. Bacteriol. 160:711–717.[PubMed]
173. Lawson, D. M., U. Derewenda, L. Serre, S. Ferri, R. Szittner, Y. Wei, E. A. Meighen, and Z. S. Derewenda. 1994. Structure of a myristoyl-ACP-specific thioesterase from Vibrio harveyi. Biochemistry 33:9382–9388.[PubMed] [CrossRef]
174. Lee, L. C., Y. L. Lee, R. J. Leu, and J. F. Shaw. 2006. Functional role of catalytic triad and oxyanion hole-forming residues on enzyme activity of Escherichia coli thioesterase I/protease I/phospholipase L1. Biochem. J. 397:69–76.[PubMed] [CrossRef]
175. Lee, L. C., Y. C. Liaw, Y. L. Lee, and J. F. Shaw. 2007. Enhanced preference for pi-bond containing substrates is correlated to Pro110 in the substrate-binding tunnel of Escherichia coli thioesterase I/protease I/lysophospholipase L(1). Biochim. Biophys. Acta 1774:959–967.[PubMed]
176. Leesong, M., B. S. Henderson, J. R. Gillig, J. M. Schwab, and J. L. Smith. 1996. Structure of a dehydratase-isomerase from the bacterial pathway for biosynthesis of unsaturated fatty acids: two catalytic activities in one active site. Structure 4:253–264.[PubMed] [CrossRef]
177. Li, J., U. Derewenda, Z. Dauter, S. Smith, and Z. S. Derewenda. 2000. Crystal structure of the Escherichia coli thioesterase II, a homolog of the human Nef binding enzyme. Nat. Struct. Biol. 7:555–559.[PubMed] [CrossRef]
178. Li, J., R. Szittner, Z. S. Derewenda, and E. A. Meighen. 1996. Conversion of serine-114 to cysteine-114 and the role of the active site nucleophile in acyl transfer by myristoyl-ACP thioesterase from Vibrio harveyi. Biochemistry 35:9967–9973.[PubMed] [CrossRef]
179. Li, Q. X., and W. Dowhan. 1990. Studies on the mechanism of formation of the pyruvate prosthetic group of phosphatidylserine decarboxylase from Escherichia coli. J. Biol. Chem. 265:4111–4115.[PubMed]
180. Li, S. J., and J. E. Cronan, Jr. 1993. Growth rate regulation of Escherichia coli acetyl coenzyme A carboxylase, which catalyzes the first committed step of lipid biosynthesis. J. Bacteriol. 175:332–340.[PubMed]
181. Li, S. J., and J. E. Cronan, Jr. 1992. The gene encoding the biotin carboxylase subunit of Escherichia coli acetyl-CoA carboxylase. J. Biol. Chem. 267:855–863.[PubMed]
182. Li, S. J., C. O. Rock, and J. E. Cronan, Jr. 1992. The dedB (usg) open reading frame of Escherichia coli encodes a subunit of acetyl-coenzyme A carboxylase. J. Bacteriol. 174:5755–5757.[PubMed]
183. Lim, K. B., C. R. Walker, L. Guo, S. Pellett, J. Shabanowitz, D. F. Hunt, E. L. Hewlett, A. Ludwig, W. Goebel, R. A. Welch, and M. Hackett. 2000. Escherichia coli α-hemolysin (HlyA) is heterogeneously acylated in vivo with 14-, 15-, and 17-carbon fatty acids. J. Biol. Chem. 275:36698–36702.[PubMed] [CrossRef]
184. Linde, K., G. Gröbner, and L. Rilfors. 2004. Lipid dependence and activity control of phosphatidylserine synthase from Escherichia coli. FEBS Lett. 575:77–80.[PubMed] [CrossRef]
185. Liu, L. X., N. Heveker, O. T. Fackler, S. Arold, S. Le Gall, K. Janvier, B. M. Peterlin, C. Dumas, O. Schwartz, S. Benichou, and R. Benarous. 2000. Mutation of a conserved residue (D123) required for oligomerization of human immunodeficiency virus type 1 Nef protein abolishes interaction with human thioesterase and results in impairment of Nef biological functions. J. Virol. 74:5310–5319.[PubMed] [CrossRef]
186. Liu, L. X., F. Margottin, S. Le Gall, O. Schwartz, L. Selig, R. Benarous, and S. Benichou. 1997. Binding of HIV-1 Nef to a novel thioesterase enzyme correlates with Nef-mediated CD4 down-regulation. J. Biol. Chem. 272:13779–13785.[PubMed] [CrossRef]
187. Lo, Y. C., S. C. Lin, J. F. Shaw, and Y. C. Liaw. 2003. Crystal structure of Escherichia coli thioesterase I/protease I/lysophospholipase L1: consensus sequence blocks constitute the catalytic center of SGNH-hydrolases through a conserved hydrogen bond network. J. Mol. Biol. 330:539–551.[PubMed] [CrossRef]
188. Lobo, S., G. Florova, and K. A. Reynolds. 2001. A Streptomyces collinus thiolase with novel acetyl-CoA:acyl carrier protein transacylase activity. Biochemistry 40:11955–11964.[PubMed] [CrossRef]
189. Louie, K., Y. C. Chen, and W. Dowhan. 1986. Substrate-induced membrane association of phosphatidylserine synthase from Escherichia coli. J. Bacteriol. 165:805–812.[PubMed]
190. Louie, K., and W. Dowhan. 1980. Investigations on the association of phosphatidylserine synthase with the ribosomal component from Escherichia coli. J. Biol. Chem. 255:1124–1127.[PubMed]
191. Lowe, P. N., and S. Rhodes. 1988. Purification and characterization of [acyl-carrier-protein] acetyltransferase from Escherichia coli. Biochem. J. 250:789–796.[PubMed]
192. Lu, P., C. Vogel, R. Wang, X. Yao, and E. M. Marcotte. 2007. Absolute protein expression profiling estimates the relative contributions of transcriptional and translational regulation. Nat. Biotechnol. 25:117–124.[PubMed] [CrossRef]
193. Lu, Y. J., Y. M. Zhang, K. D. Grimes, J. Qi, R. E. Lee, and C. O. Rock. 2006. Acyl-phosphates initiate membrane phospholipid synthesis in Gram-positive pathogens. Mol. Cell. 23:765–772.[PubMed] [CrossRef]
194. Mackay, J. P., M. Sunde, J. A. Lowry, M. Crossley, and J. M. Matthews. 2007. Protein interactions: is seeing believing? Trends Biochem. Sci. 32:530–531.[PubMed] [CrossRef]
195. Maekawa, T., K. Yanagihara, and E. Ohtsubo. 1996. Specific nicking at the 3' ends of the terminal inverted repeat sequences in transposon Tn3 by transposase and an E. coli protein ACP. Genes Cells 1:1017–1030.[PubMed] [CrossRef]
196. Magnuson, K., M. R. Carey, and J. E. Cronan, Jr. 1995. The putative fabJ gene of Escherichia coli fatty acid synthesis is the fabF gene. J. Bacteriol. 177:3593–3595.[PubMed]
197. Magnuson, K., W. Oh, T. J. Larson, and J. E. Cronan, Jr. 1992. Cloning and nucleotide sequence of the fabD gene encoding malonyl coenzyme A-acyl carrier protein transacylase of Escherichia coli. FEBS Lett. 299:262–266.[PubMed] [CrossRef]
198. Marr, A. G., and J. L. Ingraham. 1962. Effect of temperature on the composition of fatty acids in Escherichia coli. J. Bacteriol. 84:1260–1267.[PubMed]
199. Marrakchi, H., K. H. Choi, and C. O. Rock. 2002. A new mechanism for anaerobic unsaturated fatty acid formation in Streptococcus pneumoniae. J. Biol. Chem. 277:44809–44816.[PubMed] [CrossRef]
200. Massengo-Tiasse, R. P., and J. E. Cronan. 2008. Vibrio cholerae FabV defines a new class of enoyl-acyl carrier protein reductase. J. Biol. Chem. 283:1308–1316.[PubMed] [CrossRef]
201. Mayer, K. M., and J. Shanklin. 2007. Identification of amino acid residues involved in substrate specificity of plant acyl-ACP thioesterases using a bioinformatics-guided approach. BMC Plant Biol. 7:1. [CrossRef]
202. McAllister, K. A., R. B. Peery, and G. Zhao. 2006. Acyl carrier protein synthases from gram-negative, gram-positive, and atypical bacterial species: biochemical and structural properties and physiological implications. J. Bacteriol. 188:4737–4748.[PubMed] [CrossRef]
203. McCue, L., W. Thompson, C. Carmack, M. P. Ryan, J. S. Liu, V. Derbyshire, and C. E. Lawrence. 2001. Phylogenetic footprinting of transcription factor binding sites in proteobacterial genomes. Nucleic Acids Res. 29:774–782.[PubMed] [CrossRef]
204. McGuire, K. A., M. Siggaard-Andersen, M. G. Bangera, J. G. Olsen, and P. von Wettstein-Knowles. 2001. β-Ketoacyl-[acyl carrier protein] synthase I of Escherichia coli: aspects of the condensation mechanism revealed by analyses of mutations in the active site pocket. Biochemistry 40:9836–9845.[PubMed] [CrossRef]
205. McMurry, L. M., M. Oethinger, and S. B. Levy. 1998. Triclosan targets lipid synthesis. Nature 394:531–532.[PubMed] [CrossRef]
206. Mercer, A. C., and M. D. Burkart. 2007. The ubiquitous carrier protein—a window to metabolite biosynthesis. Nat. Prod. Rep. 24:750–773.[PubMed] [CrossRef]
207. Mindich, L. 1972. Control of fatty acid synthesis in bacteria. J. Bacteriol. 110:96–102.[PubMed]
208. Misra, A., S. K. Sharma, N. Surolia, and A. Surolia. 2007. Self-acylation properties of type II fatty acid biosynthesis acyl carrier protein. Chem. Biol. 14:775–783.[PubMed] [CrossRef]
209. Mizugaki, M., A. C. Swindell, and S. J. Wakil. 1968. Intermediate- and long-chain β-hydroxyacyl-ACP dehydrases from E. coli fatty acid synthetase. Biochem. Biophys. Res. Commun. 33:520–527.[PubMed] [CrossRef]
210. Mohan, S., T. M. Kelly, S. S. Eveland, C. R. Raetz, and M. S. Anderson. 1994. An Escherichia coli gene (fabZ) encoding (3R)-hydroxymyristoyl acyl carrier protein dehydrase. Relation to fabA and suppression of mutations in lipid A biosynthesis. J. Biol. Chem. 269:32896–32903.[PubMed]
211. Monson, K. D., and J. M. Hayes. 1980. Biosynthetic control of the natural abundance of carbon 13 at specific positions within fatty acids in Escherichia coli. Evidence regarding the coupling of fatty acid and phospholipid synthesis. J. Biol. Chem. 255:11435–11441.[PubMed]
212. Morgan-Kiss, R. M., and J. E. Cronan. 2004. The Escherichia coli fadK (ydiD) gene encodes an anaerobically regulated short chain acyl-CoA synthetase. J. Biol. Chem. 279:37324–37333.[PubMed] [CrossRef]
213. Nagahama, H., T. Oshima, H. Mori, K. Matsumoto, and H. Hara. 2007. Hyperexpression of the osmB gene in an acidic phospholipid-deficient Escherichia coli mutant. J. Gen. Appl. Microbiol. 53:143–151.[PubMed] [CrossRef]
214. Nagahama, H., Y. Sakamoto, K. Matsumoto, and H. Hara. 2006. RcsA-dependent and -independent growth defects caused by the activated Rcs phosphorelay system in the Escherichia coli pgsA null mutant. J. Gen. Appl. Microbiol. 52:91–98.[PubMed] [CrossRef]
215. Nakanishi, M., C. Yatome, N. Ishida, and Y. Kitade. 2001. Putative ACP phosphodiesterase gene (acpD) encodes an azoreductase. J. Biol. Chem. 276:46394–46399.[PubMed] [CrossRef]
216. Narasimhan, M. L., J. L. Lampi, and J. E. Cronan, Jr. 1986. Genetic and biochemical characterization of an Escherichia coli K-12 mutant deficient in acyl-coenzyme A thioesterase II. J. Bacteriol. 165:911–917.[PubMed]
217. Nenortas, E., and D. Beckett. 1996. Purification and characterization of intact and truncated forms of the Escherichia coli biotin carboxyl carrier subunit of acetyl-CoA carboxylase. J. Biol. Chem. 271:7559–7567.[PubMed] [CrossRef]
218. Nishijima, S., Y. Asami, N. Uetake, S. Yamagoe, A. Ohta, and I. Shibuya. 1988. Disruption of the Escherichia coli cls gene responsible for cardiolipin synthesis. J. Bacteriol. 170:775–780.[PubMed]
219. Nunn, W. D., K. Giffin, D. Clark, and J. E. Cronan, Jr. 1983. Role for fadR in unsaturated fatty acid biosynthesis in Escherichia coli. J. Bacteriol. 154:554–560.[PubMed]
220. Nunn, W. D., D. L. Kelly, and M. Y. Stumfall. 1977. Regulation of fatty acid synthesis during the cessation of phospholipid biosynthesis in Escherichia coli. J. Bacteriol. 132:526–531.[PubMed]
221. Oefner, C., H. Schulz, A. D'Arcy, and G. E. Dale. 2006. Mapping the active site of Escherichia coli malonyl-CoA-acyl carrier protein transacylase (FabD) by protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 62:613–618.[PubMed] [CrossRef]
222. Osborn, M. J., N. C. Jones, and M. Schindler. 1978. Incorporation of lipid vesicles by Salmonella. Ann. N. Y. Acad. Sci. 308:215–225.[PubMed] [CrossRef]
223. Paoletti, L., Y. J. Lu, G. E. Schujman, D. de Mendoza, and C. O. Rock. 2007. Coupling of fatty acid and phospholipid synthesis in Bacillus subtilis. J. Bacteriol. 189:5816–5824.[PubMed] [CrossRef]
224. Parris, K. D., L. Lin, A. Tam, R. Mathew, J. Hixon, M. Stahl, C. C. Fritz, J. Seehra, and W. S. Somers. 2000. Crystal structures of substrate binding to Bacillus subtilis holo-(acyl carrier protein) synthase reveal a novel trimeric arrangement of molecules resulting in three active sites. Structure 8:883–895.[PubMed] [CrossRef]
225. Perham, R. N. 2000. Swinging arms and swinging domains in multifunctional enzymes: catalytic machines for multistep reactions. Annu. Rev. Biochem. 69:961–1004.[PubMed] [CrossRef]
226. Pluschke, G., Y. Hirota, and P. Overath. 1978. Function of phospholipids in Escherichia coli. Characterization of a mutant deficient in cardiolipin synthesis. J. Biol. Chem. 253:5048–5055.[PubMed]
227. Polacco, M. L., and J. E. Cronan, Jr. 1981. A mutant of Escherichia coli conditionally defective in the synthesis of holo-[acyl carrier protein]. J. Biol. Chem. 256:5750–5754.[PubMed]
228. Polacco, M. L., and J. E. Cronan, Jr. 1977. Mechanism of the apparent regulation of Escherichia coli unsaturated fatty acid synthesis by exogenous oleic acid. J. Biol. Chem. 252:5488–5490.[PubMed]
229. Powell, G. L., M. Bauza, and A. R. Larrabee. 1973. The stability of acyl carrier protein in Escherichia coli. J. Biol. Chem. 248:4461–4466.[PubMed]
230. Powell, G. L., J. Elovson, and P. R. Vagelos. 1969. Acyl carrier protein. XII. Synthesis and turnover of the prosthetic group of acyl carrier protein in vivo. J. Biol. Chem. 244:5616–5624.[PubMed]
231. Prescott, D. J., and P. R. Vagelos. 1972. Acyl carrier protein. Adv. Enzymol. Relat. Areas Mol. Biol. 36:269–311.[PubMed] [CrossRef]
232. Qiu, X., and C. A. Janson. 2004. Structure of apo acyl carrier protein and a proposal to engineer protein crystallization through metal ions. Acta Crystallogr. D Biol. Crystallogr. 60:1545–1554.[PubMed] [CrossRef]
233. Rabinowitch, H. D., D. Sklan, D. H. Chace, R. D. Stevens, and I. Fridovich. 1993. Escherichia coli produces linoleic acid during late stationary phase. J. Bacteriol. 175:5324–5328.[PubMed]
234. Raetz, C. R., G. M. Carman, W. Dowhan, R. T. Jiang, W. Waszkuc, W. Loffredo, and M. D. Tsai. 1987. Phospholipids chiral at phosphorus. Steric course of the reactions catalyzed by phosphatidylserine synthase from Escherichia coli and yeast. Biochemistry 26:4022–4027.[PubMed] [CrossRef]
235. Raetz, C. R., and E. P. Kennedy. 1973. Function of cytidine diphosphate-diglyceride and deoxycytidine diphosphate-diglyceride in the biogenesis of membrane lipids in Escherichia coli. J. Biol. Chem. 248:1098–1105.[PubMed]
236. Raetz, C. R., C. M. Reynolds, M. S. Trent, and R. E. Bishop. 2007. Lipid A modification systems in gram-negative bacteria. Annu. Rev. Biochem. 76:295–329.[PubMed] [CrossRef]
237. Rafi, S., P. Novichenok, S. Kolappan, X. Zhang, C. F. Stratton, R. Rawat, C. Kisker, C. Simmerling, and P. J. Tonge. 2006. Structure of acyl carrier protein bound to FabI, the FASII enoyl reductase from Escherichia coli. J. Biol. Chem. 281:39285–39293.[PubMed] [CrossRef]
238. Raman, N., P. N. Black, and C. C. DiRusso. 1997. Characterization of the fatty acid-responsive transcription factor FadR. Biochemical and genetic analyses of the native conformation and functional domains. J. Biol. Chem. 272:30645–30650.[PubMed] [CrossRef]
239. Raman, N., and C. C. DiRusso. 1995. Analysis of acyl coenzyme A binding to the transcription factor FadR and identification of amino acid residues in the carboxyl terminus required for ligand binding. J. Biol. Chem. 270:1092–1097.[PubMed] [CrossRef]
240. Rawlings, M., and J. E. Cronan, Jr. 1992. The gene encoding Escherichia coli acyl carrier protein lies within a cluster of fatty acid biosynthetic genes. J. Biol. Chem. 267:5751–5754.[PubMed]
241. Ray, T. K., and J. E. Cronan, Jr. 1976. Activation of long chain fatty acids with acyl carrier protein: demonstration of a new enzyme, acyl-acyl carrier protein synthetase, in Escherichia coli. Proc. Natl. Acad. Sci. USA 73:4374–4378.[PubMed] [CrossRef]
242. Reuter, K., M. R. Mofid, M. A. Marahiel, and R. Ficner. 1999. Crystal structure of the surfactin synthetase-activating enzyme Sfp: a prototype of the 4'-phosphopantetheinyl transferase superfamily. EMBO J. 18:6823–6831.[PubMed] [CrossRef]
243. Rilfors, L., A. Niemi, S. Haraldsson, K. Edwards, A. S. Andersson, and W. Dowhan. 1999. Reconstituted phosphatidylserine synthase from Escherichia coli is activated by anionic phospholipids and micelle-forming amphiphiles. Biochim. Biophys. Acta 1438:281–294.[PubMed]
244. Rock, C. O., and J. E. Cronan, Jr. 1981. Acyl carrier protein from Escherichia coli. Methods Enzymol. 71(Pt C):341–351. [CrossRef]
245. Rock, C. O., and J. E. Cronan, Jr. 1979. Re-evaluation of the solution structure of acyl carrier protein. J. Biol. Chem. 254:9778–9785.[PubMed]
246. Rock, C. O., and J. E. Cronan, Jr. 1979. Solubilization, purification, and salt activation of acyl-acyl carrier protein synthetase from Escherichia coli. J. Biol. Chem. 254:7116–7122.[PubMed]
247. Rock, C. O., S. E. Goelz, and J. E. Cronan, Jr. 1981. Phospholipid synthesis in Escherichia coli. Characteristics of fatty acid transfer from acyl-acyl carrier protein to sn-glycerol 3-phosphate. J. Biol. Chem. 256:736–742.[PubMed]
248. Rock, C. O., and S. Jackowski. 1985. Pathways for the incorporation of exogenous fatty acids into phosphatidylethanolamine in Escherichia coli. J. Biol. Chem. 260:12720–12724.[PubMed]
249. Rock, C. O., and S. Jackowski. 1982. Regulation of phospholipid synthesis in Escherichia coli. Composition of the acyl-acyl carrier protein pool in vivo. J. Biol. Chem. 257:10759–10765.[PubMed]
250. Roujeinikova, A., C. Baldock, W. J. Simon, J. Gilroy, P. J. Baker, A. R. Stuitje, D. W. Rice, A. R. Slabas, and J. B. Rafferty. 2002. X-ray crystallographic studies on butyryl-ACP reveal flexibility of the structure around a putative acyl chain binding site. Structure 10:825–835.[PubMed] [CrossRef]
251. Roujeinikova, A., W. J. Simon, J. Gilroy, D. W. Rice, J. B. Rafferty, and A. R. Slabas. 2007. Structural studies of fatty acyl-(acyl carrier protein) thioesters reveal a hydrophobic binding cavity that can expand to fit longer substrates. J. Mol. Biol. 365:135–145.[PubMed] [CrossRef]
252. Satre, M., and E. P. Kennedy. 1978. Identification of bound pyruvate essential for the activity of phosphatidylserine decarboxylase of Escherichia coli. J. Biol. Chem. 253:479–483.[PubMed]
253. Schwab, J., and J. Henderson. 1990. Enzyme-catalyzed allylic rearrangements. Chem. Rev. 90:1203–1245. [CrossRef]
254. Seeber, F., A. Aliverti, and G. Zanetti. 2005. The plant-type ferredoxin-NADP+ reductase/ferredoxin redox system as a possible drug target against apicomplexan human parasites. Curr. Pharm. Des. 11:3159–3172.[PubMed] [CrossRef]
255. Serre, L., E. C. Verbree, Z. Dauter, A. R. Stuitje, and Z. S. Derewenda. 1995. The Escherichia coli malonyl-CoA:acyl carrier protein transacylase at 1.5-Å resolution. Crystal structure of a fatty acid synthase component. J. Biol. Chem. 270:12961–12964.[PubMed] [CrossRef]
256. Seyfzadeh, M., J. Keener, and M. Nomura. 1993. spoT-dependent accumulation of guanosine tetraphosphate in response to fatty acid starvation in Escherichia coli. Proc. Natl. Acad. Sci. USA 90:11004–11008.[PubMed] [CrossRef]
257. Shanklin, J. 2000. Overexpression and purification of the Escherichia coli inner membrane enzyme acyl-acyl carrier protein synthase in an active form. Protein Expr. Purif. 18:355–360.[PubMed] [CrossRef]
258. Sharma, A. K., S. K. Sharma, A. Surolia, N. Surolia, and S. P. Sarma. 2006. Solution structures of conformationally equilibrium forms of holo-acyl carrier protein (PfACP) from Plasmodium falciparum provides insight into the mechanism of activation of ACPs. Biochemistry 45:6904–6916.[PubMed] [CrossRef]
259. Sharpe, P. L., and N. L. Craig. 1998. Host proteins can stimulate Tn7 transposition: a novel role for the ribosomal protein L29 and the acyl carrier protein. EMBO J. 17:5822–5831.[PubMed] [CrossRef]
260. Shen, Y., C. Y. Chou, G. G. Chang, and L. Tong. 2006. Is dimerization required for the catalytic activity of bacterial biotin carboxylase? Mol. Cell. 22:807–818.[PubMed] [CrossRef]
261. Shibuya, I., C. Miyazaki, and A. Ohta. 1985. Alteration of phospholipid composition by combined defects in phosphatidylserine and cardiolipin synthases and physiological consequences in Escherichia coli. J. Bacteriol. 161:1086–1092.[PubMed]
262. Shibuya, I., S. Yamagoe, C. Miyazaki, H. Matsuzaki, and A. Ohta. 1985. Biosynthesis of novel acidic phospholipid analogs in Escherichia coli. J. Bacteriol. 161:473–477.[PubMed]
263. Shopsis, C. S., R. Engel, and B. E. Tropp. 1974. The inhibition of phosphatidylglycerol synthesis in Escherichia coli by 3,4-dihydroxybutyl-1-phosphonate. J. Biol. Chem. 249:2473–2477.[PubMed]
264. Siggaard-Andersen, M., M. Wissenbach, J. A. Chuck, I. Svendsen, J. G. Olsen, and P. von Wettstein-Knowles. 1994. The fabJ-encoded β-ketoacyl-[acyl carrier protein] synthase IV from Escherichia coli is sensitive to cerulenin and specific for short-chain substrates. Proc. Natl. Acad. Sci. USA 91:11027–11031.[PubMed] [CrossRef]
265. Silbert, D. F., and P. R. Vagelos. 1967. Fatty acid mutant of E. coli lacking a β-hydroxydecanoyl thioester dehydrase. Proc. Natl. Acad. Sci. USA 58:1579–1586.[PubMed] [CrossRef]
266. Solbiati, J., A. Chapman-Smith, and J. E. Cronan, Jr. 2002. Stabilization of the biotinoyl domain of Escherichia coli acetyl-CoA carboxylase by interactions between the attached biotin and the protruding "thumb" structure. J. Biol. Chem. 277:21604–21609.[PubMed] [CrossRef]
267. Spencer, A. K., A. D. Greenspan, and J. E. Cronan, Jr. 1978. Thioesterases I and II of Escherichia coli. Hydrolysis of native acyl-acyl carrier protein thioesters. J. Biol. Chem. 253:5922–5926.[PubMed]
268. Subrahmanyam, S., and J. E. Cronan, Jr. 1998. Overproduction of a functional fatty acid biosynthetic enzyme blocks fatty acid synthesis in Escherichia coli. J. Bacteriol. 180:4596–4602.[PubMed]
269. Tehlivets, O., K. Scheuringer, and S. D. Kohlwein. 2007. Fatty acid synthesis and elongation in yeast. Biochim. Biophys. Acta 1771:255–270.[PubMed]
270. Therisod, H., and E. P. Kennedy. 1987. The function of acyl carrier protein in the synthesis of membrane-derived oligosaccharides does not require its phosphopantetheine prosthetic group. Proc. Natl. Acad. Sci. USA 84:8235–8238.[PubMed] [CrossRef]
271. Therisod, H., A. C. Weissborn, and E. P. Kennedy. 1986. An essential function for acyl carrier protein in the biosynthesis of membrane-derived oligosaccharides of Escherichia coli. Proc. Natl. Acad. Sci. USA 83:7236–7240.[PubMed] [CrossRef]
272. Thomas, J., and J. E. Cronan. 2005. The enigmatic acyl carrier protein phosphodiesterase of Escherichia coli: genetic and enzymological characterization. J. Biol. Chem. 280:34675–34683.[PubMed] [CrossRef]
273. Thomas, J., D. J. Rigden, and J. E. Cronan. 2007. Acyl carrier protein phosphodiesterase (AcpH) of Escherichia coli is a non-canonical member of the HD phosphatase/phosphodiesterase family. Biochemistry 46:129–136.[PubMed] [CrossRef]
274. Thomsen-Zieger, N., J. Schachtner, and F. Seeber. 2003. Apicomplexan parasites contain a single lipoic acid synthase located in the plastid. FEBS Lett. 547:80–86.[PubMed] [CrossRef]
275. Tropp, B. E. 1997. Cardiolipin synthase from Escherichia coli. Biochim. Biophys. Acta 1348:192–200.[PubMed]
276. Tropp, B. E., L. Ragolia, W. Xia, W. Dowhan, R. Milkman, K. E. Rudd, R. Ivanisevic, and D. J. Savic. 1995. Identity of the Escherichia coli cls and nov genes. J. Bacteriol. 177:5155–5157.[PubMed]
277. Tsay, J. T., W. Oh, T. J. Larson, S. Jackowski, and C. O. Rock. 1992. Isolation and characterization of the β-ketoacyl-acyl carrier protein synthase III gene (fabH) from Escherichia coli K-12. J. Biol. Chem. 267:6807–6814.[PubMed]
278. Tsay, J. T., C. O. Rock, and S. Jackowski. 1992. Overproduction of β-ketoacyl-acyl carrier protein synthase I imparts thiolactomycin resistance to Escherichia coli K-12. J. Bacteriol. 174:508–513.[PubMed]
279. Tunaitis, E., and J. E. Cronan, Jr. 1973. Characterization of the cardiolipin synthetase activity of Escherichia coli envelopes. Arch. Biochem. Biophys. 155:420–427.[PubMed] [CrossRef]
280. Turnowsky, F., K. Fuchs, C. Jeschek, and G. Hogenauer. 1989. envM genes of Salmonella typhimurium and Escherichia coli. J. Bacteriol. 171:6555–6565.[PubMed]
281. Tyhach, R. J., E. Hawrot, M. Satre, and E. P. Kennedy. 1979. Increased synthesis of phosphatidylserine decarboxylase in a strain of Escherichia coli bearing a hybrid plasmid. Altered association of enzyme with the membrane. J. Biol. Chem. 254:627–633.[PubMed]
282. Tyukhtenko, S. I., A. V. Litvinchuk, C. F. Chang, Y. C. Lo, S. J. Lee, J. F. Shaw, Y. C. Liaw, and T. H. Huang. 2003. Sequential structural changes of Escherichia coli thioesterase/protease I in the serial formation of Michaelis and tetrahedral complexes with diethyl p-nitrophenyl phosphate. Biochemistry 42:8289–8297.[PubMed] [CrossRef]
283. Ulrich, A. K., D. de Mendoza, J. L. Garwin, and J. E. Cronan, Jr. 1983. Genetic and biochemical analyses of Escherichia coli mutants altered in the temperature-dependent regulation of membrane lipid composition. J. Bacteriol. 154:221–230.[PubMed]
284. Vagelos, P. R., and A. R. Larrabes. 1967. Acyl carrier protein. IX. Acyl carrier protein hydrolase. J. Biol. Chem. 242:1776–1781.[PubMed]
285. van Aalten, D. M., C. C. DiRusso, and J. Knudsen. 2001. The structural basis of acyl coenzyme A-dependent regulation of the transcription factor FadR. EMBO J. 20:2041–2050.[PubMed] [CrossRef]
286. van Aalten, D. M., C. C. DiRusso, J. Knudsen, and R. K. Wierenga. 2000. Crystal structure of FadR, a fatty acid-responsive transcription factor with a novel acyl coenzyme A-binding fold. EMBO J. 19:5167–5177.[PubMed] [CrossRef]
287. Van Den Bosch, H., J. R. Williamson, and P. R. Vagelos. 1970. Localization of acyl carrier protein in Escherichia coli. Nature 228:338–341.[PubMed] [CrossRef]
288. Vance, J. E. 2003. Molecular and cell biology of phosphatidylserine and phosphatidylethanolamine metabolism. Prog. Nucleic Acid Res. Mol. Biol. 75:69–111.[PubMed] [CrossRef]
289. Verwoert, I. I., E. C. Verbree, K. H. van der Linden, H. J. Nijkamp, and A. R. Stuitje. 1992. Cloning, nucleotide sequence, and expression of the Escherichia coli fabD gene, encoding malonyl coenzyme A-acyl carrier protein transacylase. J. Bacteriol. 174:2851–2857.[PubMed]
290. Verwoert, I. I., E. F. Verhagen, K. H. van der Linden, E. C. Verbree, H. J. Nijkamp, and A. R. Stuitje. 1994. Molecular characterization of an Escherichia coli mutant with a temperature-sensitive malonyl coenzyme A-acyl carrier protein transacylase. FEBS Lett. 348:311–316.[PubMed] [CrossRef]
291. Voelker, T. A., and H. M. Davies. 1994. Alteration of the specificity and regulation of fatty acid synthesis of Escherichia coli by expression of a plant medium-chain acyl-acyl carrier protein thioesterase. J. Bacteriol. 176:7320–7327.[PubMed]
292. Wang, A. Y., and J. E. Cronan, Jr. 1994. The growth phase-dependent synthesis of cyclopropane fatty acids in Escherichia coli is the result of an RpoS(KatF)-dependent promoter plus enzyme instability. Mol. Microbiol. 11:1009–1017.[PubMed] [CrossRef]
293. Wang, A. Y., D. W. Grogan, and J. E. Cronan, Jr. 1992. Cyclopropane fatty acid synthase of Escherichia coli: deduced amino acid sequence, purification, and studies of the enzyme active site. Biochemistry 31:11020–11028.[PubMed] [CrossRef]
294. Wang, H., and J. E. Cronan. 2004. Functional replacement of the FabA and FabB proteins of Escherichia coli fatty acid synthesis by Enterococcus faecalis FabZ and FabF homologues. J. Biol. Chem. 279:34489–34495.[PubMed] [CrossRef]
295. Wang, J., S. Kodali, S. H. Lee, A. Galgoci, R. Painter, K. Dorso, F. Racine, M. Motyl, L. Hernandez, E. Tinney, S. L. Colletti, K. Herath, R. Cummings, O. Salazar, I. Gonzalez, A. Basilio, F. Vicente, O. Genilloud, F. Pelaez, H. Jayasuriya, K. Young, D. F. Cully, and S. B. Singh. 2007. Discovery of platencin, a dual FabF and FabH inhibitor with in vivo antibiotic properties. Proc. Natl. Acad. Sci. USA 104:7612–7616.[PubMed] [CrossRef]
296. Wang, J., S. M. Soisson, K. Young, W. Shoop, S. Kodali, A. Galgoci, R. Painter, G. Parthasarathy, Y. S. Tang, R. Cummings, S. Ha, K. Dorso, M. Motyl, H. Jayasuriya, J. Ondeyka, K. Herath, C. Zhang, L. Hernandez, J. Allocco, A. Basilio, J. R. Tormo, O. Genilloud, F. Vicente, F. Pelaez, L. Colwell, S. H. Lee, B. Michael, T. Felcetto, C. Gill, L. L. Silver, J. D. Hermes, K. Bartizal, J. Barrett, D. Schmatz, J. W. Becker, D. Cully, and S. B. Singh. 2006. Platensimycin is a selective FabF inhibitor with potent antibiotic properties. Nature 441:358–361.[PubMed] [CrossRef]
297. Weeks, G., and S. J. Wakil. 1968. Studies on the mechanism of fatty acid synthesis. 18. Preparation and general properties of the enoyl acyl carrier protein reductases from Escherichia coli. J. Biol. Chem. 243:1180–1189.[PubMed]
298. West, J., C. K. Tompkins, N. Balantac, E. Nudelman, B. Meengs, T. White, S. Bursten, J. Coleman, A. Kumar, J. W. Singer, and D. W. Leung. 1997. Cloning and expression of two human lysophosphatidic acid acyltransferase cDNAs that enhance cytokine-induced signaling responses in cells. DNA Cell Biol. 16:691–701.[PubMed]
299. White, R. H. 1980. Stoichiometry and stereochemistry of deuterium incorporated into fatty acids by cells of Escherichia coli grown on [methyl-2H3]acetate. Biochemistry 19:9–15.[PubMed] [CrossRef]
300. White, S. W., J. Zheng, Y. M. Zhang, and C. O. Rock. 2005. The structural biology of type II fatty acid biosynthesis. Annu. Rev. Biochem. 74:791–831.[PubMed] [CrossRef]
301. Woehlke, G., and P. Dimroth. 1994. Anaerobic growth of Salmonella typhimurium on L(+)- and D(−)-tartrate involves an oxaloacetate decarboxylase Na+ pump. Arch. Microbiol. 162:233–237.[PubMed] [CrossRef]
302. Woehlke, G., K. Wifling, and P. Dimroth. 1992. Sequence of the sodium ion pump oxaloacetate decarboxylase from Salmonella typhimurium. J. Biol. Chem. 267:22798–22803.[PubMed]
303. Worsham, L. M., L. Earls, C. Jolly, K. G. Langston, M. S. Trent, and M. L. Ernst-Fonberg. 2003. Amino acid residues of Escherichia coli acyl carrier protein involved in heterologous protein interactions. Biochemistry 42:167–176.[PubMed] [CrossRef]
304. Worsham, L. M., K. G. Langston, and M. L. Ernst-Fonberg. 2005. Thermodynamics of a protein acylation: activation of Escherichia coli hemolysin toxin. Biochemistry 44:1329–1337.[PubMed] [CrossRef]
305. Worsham, L. M., M. S. Trent, L. Earls, C. Jolly, and M. L. Ernst-Fonberg. 2001. Insights into the catalytic mechanism of HlyC, the internal protein acyltransferase that activates Escherichia coli hemolysin toxin. Biochemistry 40:13607–13616.[PubMed] [CrossRef]
306. Wright, C. W. 2007. Recent developments in naturally derived antimalarials: cryptolepine analogues. J. Pharm. Pharmacol. 59:899–904.[PubMed] [CrossRef]
307. Xu, Y., R. J. Heath, Z. Li, C. O. Rock, and S. W. White. 2001. The FadR.DNA complex. Transcriptional control of fatty acid metabolism in Escherichia coli. J. Biol. Chem. 276:17373–17379.[PubMed] [CrossRef]
308. Yoshimura, M., T. Oshima, and N. Ogasawara. 2007. Involvement of the YneS/YgiH and PlsX proteins in phospholipid biosynthesis in both Bacillus subtilis and Escherichia coli. BMC Microbiol. 7:69. [CrossRef]
309. Zhang, Y., and J. E. Cronan, Jr. 1996. Polar allele duplication for transcriptional analysis of consecutive essential genes: application to a cluster of Escherichia coli fatty acid biosynthetic genes. J. Bacteriol. 178:3614–3620.[PubMed]
310. Zhang, Y., and J. E. Cronan, Jr. 1998. Transcriptional analysis of essential genes of the Escherichia coli fatty acid biosynthesis gene cluster by functional replacement with the analogous Salmonella typhimurium gene cluster. J. Bacteriol. 180:3295–3303.[PubMed]
311. Zhang, Y. M., Y. J. Lu, and C. O. Rock. 2004. The reductase steps of the type II fatty acid synthase as antimicrobial targets. Lipids 39:1055–1060.[PubMed] [CrossRef]
312. Zhang, Y. M., H. Marrakchi, and C. O. Rock. 2002. The FabR (YijC) transcription factor regulates unsaturated fatty acid biosynthesis in Escherichia coli. J. Biol. Chem. 277:15558–15565.[PubMed] [CrossRef]
313. Zhang, Y. M., and C. O. Rock. 2008. Membrane lipid homeostasis in bacteria. Nat. Rev. Microbiol. 6:222–233.[PubMed] [CrossRef]
314. Zhang, Y. M., S. W. White, and C. O. Rock. 2006. Inhibiting bacterial fatty acid synthesis. J. Biol. Chem. 281:17541–17544.[PubMed] [CrossRef]
315. Zhang, Y. M., K. Zhu, M. W. Frank, and C. O. Rock. 2007. A Pseudomonas aeruginosa transcription factor that senses fatty acid structure. Mol. Microbiol. 66:622–632.[PubMed] [CrossRef]
316. Zheng, C. J., M. J. Sohn, and W. G. Kim. 2006. Atromentin and leucomelone, the first inhibitors specific to enoyl-ACP reductase (FabK) of Streptococcus pneumoniae. J. Antibiot. (Tokyo) 59:808–812.[PubMed] [CrossRef]
317. Zheng, C. J., M. J. Sohn, S. Lee, Y. S. Hong, J. H. Kwak, and W. G. Kim. 2007. Cephalochromin, a FabI-directed antibacterial of microbial origin. Biochem. Biophys. Res. Commun. 362:1107–1112.[PubMed] [CrossRef]
318. Zheng, Z., Q. Gong, T. Liu, Y. Deng, J. C. Chen, and G. Q. Chen. 2004. Thioesterase II of Escherichia coli plays an important role in 3-hydroxydecanoic acid production. Appl. Environ. Microbiol. 70:3807–3813.[PubMed] [CrossRef]
319. Zhuang, Z., F. Song, B. M. Martin, and D. Dunaway-Mariano. 2002. The YbgC protein encoded by the ybgC gene of the tol-pal gene cluster of Haemophilus influenzae catalyzes acyl-coenzyme A thioester hydrolysis. FEBS Lett. 516:161–163.[PubMed] [CrossRef]
Module3.6.4
