![[Feedback]](/icons/banner/feedbackACT.gif)
![[For Subscribers]](/icons/banner/subscriberACT.gif)
![[Archive]](/icons/banner/archiveACT.gif)
![[Search]](/icons/banner/searchACT.gif)
![[Contents]](/icons/banner/tocACT.gif)
Microbiology and Molecular Biology Reviews, March 2005, p. 12-50, Vol. 69, No. 1
1092-2172/05/$08.00+0 doi:10.1128/MMBR.69.1.12-50.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
The Acetate Switch
Alan J. Wolfe*
Department of Microbiology and Immunology, Stritch School of Medicine, Loyola University Chicago, Maywood, Illinois
To succeed, many cells must alternate between life-styles that permit rapid growth in the presence of abundant nutrients and ones that enhance survival in the absence of those nutrients. One such change in life-style, the "acetate switch," occurs as cells deplete their environment of acetate-producing carbon sources and begin to rely on their ability to scavenge for acetate. This review explains why, when, and how cells excrete or dissimilate acetate. The central components of the "switch" (phosphotransacetylase [PTA], acetate kinase [ACK], and AMP-forming acetyl coenzyme A synthetase [AMP-ACS]) and the behavior of cells that lack these components are introduced. Acetyl phosphate (acetyl
P), the high-energy intermediate of acetate dissimilation, is discussed, and conditions that influence its intracellular concentration are described. Evidence is provided that acetylP influences cellular processes from organelle biogenesis to cell cycle regulation and from biofilm development to pathogenesis. The merits of each mechanism proposed to explain the interaction of acetylP with two-component signal transduction pathways are addressed. A short list of enzymes that generate acetylP by PTA-ACKA-independent mechanisms is introduced and discussed briefly. Attention is then directed to the mechanisms used by cells to "flip the switch," the induction and activation of the acetate-scavenging AMP-ACS. First, evidence is presented that nucleoid proteins orchestrate a progression of distinct nucleoprotein complexes to ensure proper transcription of its gene. Next, the way in which cells regulate AMP-ACS activity through reversible acetylation is described. Finally, the "acetate switch" as it exists in selected eubacteria, archaea, and eukaryotes, including humans, is described.
Definitions
To survive, many cells must switch from a physiological program that permits rapid growth in the presence of abundant nutrients to one that enhances survival in the absence of those nutrients. One such "switch" occurs when bacterial cells transit from a program of rapid growth that produces and excretes acetate (dissimilation) to a program of slower growth facilitated by the import and utilization (assimilation) of that excreted acetate (Fig. 1). This "acetate switch" occurs as cells deplete their environment of acetate-producing (acetogenic) carbon sources, e.g., D-glucose or L-serine, and begin to rely on their ability to scavenge for environmental acetate.
Throughout this review, I define the "acetate switch" physiologically as the moment when acetate dissimilation equals its assimilation. One observes this event experimentally as the peak accumulation of extracellular acetate (indicated by single-headed arrows in Fig. 1). Note that this physiological event cannot occur unless a molecular "switch" already has been "flipped" to express and activate the machinery responsible for acetate assimilation.
The Physiological "Switch": Three Examples
The "acetate switch" of Escherichia coli has been studied predominantly under three different growth conditions: in shake flask culture supplemented with D-glucose as the sole carbon source, in shake flask culture with a tryptone-based medium, and during high-cell-density glucose fermentation. The following simply describes the "switch" as it occurs under each regimen and does not attempt to explain the underlying mechanism(s). Such explanations will follow.
Shake flask culture: glucose.
Cells undergo an "acetate switch" during buffered growth on D-glucose (Fig. 1A). During exponential growth, cells consume the sugar and dissimilate acetate (195). Before they exhaust the sugar, however, the "switch" occurs and the cells coassimilate both acetate and the remaining sugar (325). This "switch" occurs just as the cells begin the transition to stationary phase, defined in this review as the moment that the cells begin to decelerate their growth.
Shake flask culture: tryptone.
During exponential growth on Bacto Tryptone broth (an unbuffered, essentially carbohydrate-free mixture of amino acids and small peptides), cells of E. coli consume amino acids in a strictly preferential order (Fig. 1B). First, they consume L-serine and then L-aspartate, while dissimilating acetate, which acidifies the unbuffered medium. As these cells begin the transition to stationary phase, they consume L-tryptophan while assimilating acetate. This consumption of acetate, combined with evolution of ammonia from amino acid metabolism, alkalinizes the medium. On entry into stationary phase, the cells consume a mixture of amino acids, two acetogenic (L-threonine and L-alanine) and one nonacetogenic (L-glutamate). The result is net acetate excretion, albeit to levels lower than those achieved during exponential growth. Because of the continued evolution of ammonia, the environment remains alkaline (78, 358, 359). Thus, the physiological switch can flip back and forth depending on the acetogenic nature of the amino acid(s) presently under consumption.
Glucose-fed high-cell-density fermentation.
The feeding of glucose in a nonlimiting manner to an aerobic fermentation (buffered at pH 7.0) results in an extended growth phase. This extended growth phase results in high cell density accompanied by excretion of large amounts of acetate. These glucose-fed fermentations begin with a glucose-consuming, acetogenic exponential phase during which oxygen is consumed and carbon dioxide evolves. Near the end of exponential growth, the fermentation pauses for a short interval. During this pause, cells halt the consumption of oxygen and the evolution of carbon dioxide. After about 30 min, fermentation reinitiates. Oxygen consumption and carbon dioxide evolution resume as the culture cometabolizes glucose and acetate. Despite buffering, a transient increase in pH accompanies the consumption of acetate (241).
Short History
The "acetate switch" possesses a rich past. Its components and intermediates were discovered and initially characterized in the 1940s and 1950s, during the effort to identify the "activated acetate." We now know this "activated acetate" as acetyl coenzyme A (acetyl-CoA), the high-energy intermediate that sits at the crossroads of central metabolism (Fig. 2) (for historical reviews, see references 32, 45, 231, and 404). During the next three decades, as researchers explored the fundamentals of molecular biology, studies of acetate metabolism faded from prominence, kept alive mostly by investigators concerned with fermentation. On occasion, general interest in the "switch" resurfaced transiently; however, it was not until the late 1980s and early 1990s that the "acetate switch" regained the spotlight. This renewed interest resulted primarily from the proposition that acetyl phosphate (acetylP), the high-energy intermediate of the dissimilation pathway, might function as a global signal (298, 463). Today, mounting evidence suggests that, indeed, acetylP plays such a role, regulating cellular processes as diverse as nitrogen assimilation, osmoregulation, flagellar biogenesis, pilus assembly, capsule biosynthesis, biofilm development, and pathogenicity (24, 25, 187, 273, 274, 291, 322, 343, 354, 355, 358, 473).
For several reasons, there also exists renewed interest in the acetate assimilation enzyme AMP-forming acetyl-CoA synthetase (AMP-ACS). First, AMP-ACS is a prototype for enzymes involved in the synthesis of fatty acids, some antibiotics, and certain anticancer drugs, as well as the degradation of pollutants (427). Second, AMP-ACS activity is regulated by an acetylation-deacetylation system homologous to that used by eukaryotes to control chromatin structure, silencing, mitochondrial signaling, and aging (72, 427). Third, the complex acs promoter that drives AMP-ACS expression in E. coli is fast becoming a model for how dynamic nucleoprotein complexes ensure that transcription occurs properly (33, 40, 62, 63).
Scope
The purpose of this review is to (re)introduce the "acetate switch," first giving a brief description of the acetate-rich colon, a key ecological niche for E. coli, and then explaining why, when, and how bacterial cells excrete acetate and other central metabolic intermediates. It will acquaint the reader with the enzyme components that comprise the molecular core of the "switch" and describe the behavior of mutants that lack some or all of those components. This review does not, however, summarize the structure-function relationships of these components. Next, emphasis shifts to acetylP, the high-energy intermediate of acetate dissimilation. The review describes how cells regulate the size of the acetylP pool, presents evidence that acetylP can act as a global signal that influences diverse cellular processes, and addresses the mechanism(s) by which acetylP might exert its influence. Next, it focuses on the molecular mechanisms that facilitate the "switch" from acetate dissimilation to acetate assimilation. These mechanisms regulate transcription from the complex acs promoter and the activity of AMP-ACS. Finally, it describes variants of the "acetate switch" found in other eubacterial species, selected archaea, and humans. Although this review does not exhaustively review the literature concerning acid and organic acid stress, the topic is addressed in passing. It also does not directly review efforts to metabolically engineer the "acetate switch," although much of the information provided will aid researchers interested in such endeavors.
In utero, the mammalian fetus is sterile (303). During and after birth, the human neonate becomes exposed to and colonized by large numbers of E. coli and Enterococcus organisms (108 to 1010 per g of contents). These rapidly growing, facultative bacteria metabolize the lactose present in breast milk to acetate and other short-chain fatty acids (SCFA, also known as volatile fatty acids), creating a reduced acidic environment favorable for colonization by the slower-growing, anaerobic acidophile Bifidobacterium. This acidophilic anaerobe eventually outcompetes E. coli, consumes most of the available sugar, and excretes large amounts of acetate and other SCFA (306).
Subsequent neonatal exposure to microbes through diet, and the competition between these microbes for limited nutrients and specific niches, cause the colonic microbiota to diversify (142, 143). The diversified adult colon hosts a complex flora, consisting of more than 50 genera and 400 species, generally attached as biofilms to particulate intestinal materials and embedded in the mucus layer that coats the colonic epithelial cells (colonocytes). This flora includes large numbers (1010 to 1011 per g) of anaerobes. It also contains a smaller number of facultative organisms, including E. coli, that maintain the reduced environment required by strict anaerobes. Some of these anaerobes, primarily members of the genera Bifidobacterium and Bacteroides, ferment dietary fiber—complex polysaccharides not digested and absorbed in the upper gut—to simple sugars and SCFA. Other anaerobes, mainly Peptostreptococcus and Fusobacterium, as well as certain facultative organisms, e.g., Enterococcus and E. coli, presumably cross-feed on those simple sugars and SCFA (101, 283, 284, 341).
In the colon, acetogenic sugars arise from diet, bacterial metabolism, or the host-secreted mucus (143, 352, 460). This mucus, which overlays colonocytes, is a complex gel of glycolipids, glycoproteins, and a variety of sugar residues, including N-acetylglucosamine, N-acetylgalactosamine, D-galactose, fucose, sialic acids, glucuronate, galacturonate, and gluconate (5). When stripped from mucus, these sugar residues provide carbon that permits rapid growth while contributing to the local accumulation of acetate and other SCFA. As in culture, the nature of the carbon source(s) dictates colonic pH, which can vary from 6 to 8 (14, 55, 102).
SCFA constitute approximately two-thirds of the colonic anion concentration (70 to 130 mM), mainly as acetate, propionate, and butyrate. Of these, acetate predominates. These SCFA, rapidly absorbed by the colonic mucosa, represent the primary energy source for colonocytes, hepatic cells, fat cells, and muscle cells (283, 284, 300, 313, 447). SCFA also perform functions of considerable significance to the health of the host (36, 37, 352, 397, 460). Finally, they enhance the virulence of certain enteric pathogens, including E. coli and other members of the Enterobacteriaceae, mostly by inducing protective mechanisms (39, 122, 255, 260, 270, 377).
|
WHY CELLS EXCRETE ACETATE
|
|---|
Acetogenesis, the excretion of acetate into the environment, results from the need to regenerate the NAD+ consumed by glycolysis and to recycle the coenzyme A (CoASH) required to convert pyruvate to acetyl-CoA. Since the tricarboxylic acid (TCA) cycle completes the oxidation of acetyl-CoA to carbon dioxide, acetogenesis occurs whenever the full TCA cycle does not operate or when the carbon flux into cells exceeds its capacity and that of other central metabolic pathways (78, 123, 126, 195, 196, 232, 263, 286, 386, 458, 477). Thus, acetate excretion occurs anaerobically during mixed-acid fermentation (54). It also occurs aerobically when growth on excess glucose (or other highly assimilable carbon sources) inhibits respiration (195, 196), a behavior called the bacterial Crabtree effect (98, 119, 280, 379). As a consequence of the Crabtree effect, as much as 15% of the glucose can be excreted as acetate (196). Although acetogenesis has long been considered simply the result of "overflow" metabolism (195, 196), a recent report raises the possibility that acetate excretion permits more rapid growth to higher cell densities primarily by providing the TCA cycle enzyme 2-ketoglutarate dehydrogenase (KGDH) (Fig. 3) with CoASH (124).
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 3. The pathways of central metabolism. For glycolysis, only some of the intermediates and enzymes of glycolysis are noted. PEP, phosphoenolpyruvate; Pyr, pyruvate; PFK, phosphofructokinase; TPIA, triosephosphate isomerase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PDHC, pyruvate dehydrogenase complex. For acetate metabolism: POXB, pyruvate oxidase; PTA-ACKA, phosphotransacetylase-acetate kinase pathway; ACS, AMP-forming acetyl-CoA synthetase. The dotted arrows denote the proposed PDHC bypass formed by POXB and AMP-ACS. For the TCA cycle: CS, citrate synthase; ACN, aconitase; IDH, isocitrate dehydrogenase; 2-KG, 2-ketoglutarate; KGDH, 2-ketoglutarate dehydrogenase; SCSC, succinyl-CoA synthetase complex; SDH, succinate dehydrogenase; FUMA, fumarase; MDH, malate dehydrogenase; OAA, oxaloacetate. FRD, fumarate reductase, expressed under anaerobic conditions, bypasses SDH. For the glyoxylate bypass: ICL, isocitrate lyase; MAS, malate synthase; IDHK/P, isocitrate dehydrogenase kinase/phosphatase. Underlines and dashed arrows denote enzymes and steps unique to the glyoxylate bypass. For gluconeogenesis: PPSA, PEP synthase; PCKA, pyruvate carboxykinase; MAEB and SFCA, malic enzymes. Boxes and double-lined arrows denote enzymes and steps unique to gluconeogenesis.
|
|
Limiting the Tricarboxylic Acid Cycle
The availability of oxygen and the nature and quantity of the carbon source dictate the status of the TCA cycle (Fig. 3) (6, 160, 420). In the absence of oxygen and under conditions that favor catabolite repression (e.g., excess glucose), E. coli cells do not induce the full TCA cycle (6, 320, 342). Instead, they operate a branched version, which forms succinyl-CoA by a reductive pathway and 2-ketoglutarate by an oxidative one (100, 168, 282, 420). This branched form of the TCA cycle does not generate energy; instead it functions biosynthetically, producing precursor metabolites. Thus, ATP must come from glycolysis (6) and substrate phosphorylation via the phosphotransacetylase (PTA)-acetate kinase (ACKA) pathway (61, 385, 441).
This branched version occurs because the absence of oxygen severely inhibits the expression of many TCA cycle enzymes, but most dramatically succinate dehydrogenase (SDH), the succinyl-CoA synthetase complex (SCSC), and KGDH. A more moderate inhibition occurs in the presence of excess glucose (160, 170, 193, 212, 335-339, 413).
In the absence of oxygen, the oxygen-sensitive global regulators ArcA and FNR mediate the repression of many TCA promoters, but most dramatically the sdh-suc operon, which encodes SDH, KDGH, and the SCSC (104, 105, 170, 293, 335, 339, 406). The mechanism that causes glucose repression, and thus the Crabtree effect, is less clear. Glucose represses sdh-suc indirectly through the action of EIICB(Glc), but the mechanism remains a mystery (439). The membrane-bound EIICB(Glc), part of the phosphoenolpyruvate (PEP):carbohydrate phosphotransferase system (PTS), phosphorylates and translocates glucose. This process causes dephosphorylation of EIICB(Glc) (351). Dephosphorylated EIICB(Glc) sequesters Mlc, a global repressor of genes that encode certain sugar-metabolizing enzymes and their uptake systems, including ptsG, the gene that encodes EIICB(Glc). Thus, translocation of glucose by EIICB(Glc) relieves Mlc-dependent repression, which permits expression of these glucose-activated genes. For a review, see reference 349.
For sdh-suc and other glucose-repressed promoters, however, the mechanism remains murky. The process does not involve Mlc directly (439), nor does it involve perturbations in the PEP pool (81, 82). Glucose translocation by EIICB(Glc) dephosphorylates EIIA(Glc), the immediate phosphoryl donor of the PTS. Dephosphorylated EIIA(Glc) causes inducer exclusion, by binding and inhibiting other sugar permeases (for reviews, see references 351 and 434). Dephosphorylation of EIIA(Glc) also may reduce cyclic AMP (cAMP) levels by reducing the activity of adenylate cyclase (351, 434), although this model has been disputed (239). However, despite the presence of putative DNA binding sites for the cAMP receptor protein (CRP) (472), it appears that cAMP-CRP does not control sdh-suc transcription directly (339). Apparently, another global carbon regulator, Cra (also known as FruR), also is not involved (339). Attempts to solve this puzzle are clearly warranted, especially given the negative effect exerted by acetate on the production of recombinant products during aerobic glucose-fed high-cell-density fermentations (15, 38, 174, 196, 219, 280, 332, 477).
Recycling Coenzyme A
The total CoA pool consists primarily of the nonesterified form (CoASH), and its thioesters acetyl-CoA, succinyl-CoA, and malonyl-CoA (214, 454). The size of the CoA pool is tightly regulated, remaining relatively constant, somewhere in the range of 100 to 500 µM (85, 454). This regulation occurs primarily through the utilization of pantothenate (the immediate CoASH precursor) and secondarily by degradation of CoASH (215, 216, 454). For reviews of CoASH biosynthesis, see references 41 and 213.
Because the CoA pool is limiting, its composition responds readily to the quality and quantity of the carbon source. This response is observed largely as a change in the ratio of acetyl-CoA to CoASH, whose concentrations vary inversely (84, 85). The nature of the carbon source affects this ratio. The addition to starved cells of assimilable carbon sources, e.g., D-glucose, causes this ratio to increase rapidly. In contrast, acetate, succinate, and nonassimilable sugars exert little or no effect on this ratio (85). This behavior explains, at least in part, why acetyl-CoA levels rise and then fall during growth on D-glucose and in tryptone broth (86, 360). The acetyl-CoA pool peaks during consumption of assimilable, acetogenic carbon sources and diminishes as the cells assimilate the previously excreted acetate (Fig. 1). Not surprisingly, this behavior correlates inversely with that of the TCA cycle, which becomes repressed during growth on D-glucose (6, 320, 342) and induced during growth on acetate (226, 329, 342).
Regenerating NAD+
Glycolysis oxidizes glucose to two molecules of pyruvate while generating only two ATP molecules (Fig. 3). To fulfill their demand for ATP, therefore, E. coli cells must consume large amounts of glucose. Oxidation of glucose, however, also produces two molecules of NADH, which corresponds to four reducing equivalents. Because NAD+ serves as a substrate for the glycoytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH), cells must reoxidize NADH to maintain glycolytic flux. In the absence of a functional TCA cycle, they achieve this by placing the reducing equivalents onto partially oxidized metabolic intermediates, predominantly D-lactate, succinate, formate and ethanol, which cells excrete into their environment along with acetate (Fig. 4). Whereas acetate excretion generates energy in the form of ATP, excretion of these other metabolic intermediates sacrifices energy to consume reducing equivalents (89, 157, 326, 432; for a review, see reference 137).
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 4. Pathways for the excretion of partially oxidized metabolites. Excreted metabolites are underlined. LDH, lactate dehydrogenase; PFL, pyruvate-formate lyase; PTA, phosphotransacetylase; ACKA, acetate kinase; ADH, alcohol dehydrogenase; FDO, aerobic formate dehydrogenase; FHL, formate-hydrogen lyase.
|
|
Environmental Influences
The composition of excreted fermentation products depends on a number of environmental factors, including the oxidation state of the carbon substrate (3), the extracellular oxidoreduction potential (380), and the pH of the external environment (54, 241). For example, the oxidation state dictates the amount of NADH to be recycled and therefore the composition of the excreted fermentation products. The oxidation of glucose (oxidation state = 0) into two molecules of pyruvate produces two NADH molecules, each corresponding to two reducing equivalents. A more reduced sugar alcohol (e.g., sorbitol; oxidation state = –1) produces three NADH, while a highly oxidized sugar acid (e.g., glucuronic acid; oxidation state = +2) yields no NADH. Thus, to recycle the larger amount of NADH formed during growth on the more reduced sorbitol, cells must excrete more of the highly reduced ethanol (oxidation state = –2). In contrast, cells growing on glucuronic acid are redox balanced and thus do not need to produce ethanol. Instead, they can convert more of their pyruvate to acetate (oxidation state = 0) (3, 54). Similarly, external pH influences the composition of excreted products. Near or above pH 7, the predominant products are acetate, ethanol, and formate, with moderate amounts of succinate (43). As the pH drops, cells produce lactate instead of acetate and formate (68) and convert the formate to H2 and CO2 (386). Since oxidized sugar acids (e.g., gluconate, glucuronate, and galacturonate) provide much of the carbon available in the colon (5, 341) and the pH of the adult colon generally ranges between 6 and 8 (14, 55, 102), acetate is the major component of the excreted fermentation products.
Excretion Pathways
To excrete acetate (as well as formate and ethanol), E. coli cells must first decarboxylate pyruvate into acetyl-CoA. The conversion of pyruvate to acetyl-CoA can occur oxidatively under aerobic conditions and nonoxidatively under anaerobic conditions. Oxidative decarboxylation, a reaction catalyzed by the pyruvate dehydrogenase complex (PDHC), generates two additional NADH per glucose (Fig. 3 and 5). High concentrations of NADH inhibit PDHC activity (176). Thus, the PDHC does not operate under conditions, e.g., anaerobiosis, that do not favor the rapid reoxidation of NADH to NAD+. Note that anaerobiosis also represses transcription of the genes that encode the PDHC (362). Thus, oxidative decarboxylation functions primarily during respiratory metabolism, although some function may be retained during anaerobiosis (167, 223, 421).
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 5. Acetate activation pathways. PDHC, pyruvate dehydrogenase complex; POXB, pyruvate oxidase; PTA, phosphotransacetylase; ACKA, acetate kinase; ACS, AMP-forming acetyl-CoA synthetase; PPase, pyrophosphatase; TCA, tricarboxylic acid cycle; GB, glyoxylate bypass. The dotted arrows denote the proposed PDHC bypass formed by POXB and AMP-ACS.
|
|
During anaerobiosis, cells of E. coli instead decarboxylate pyruvate to acetyl-CoA and formate by means of pyruvate formate-lyase (PFL), which catalyzes a nonoxidative reaction (242) (Fig. 4). Because its activity in vitro depends on an oxygen-sensitive glycyl residue, PFL has long been thought to function only in the absence of oxygen. However, more recent reports suggest that PFL can function in vivo in the presence of some oxygen (4, 113). This may occur through YfiD, which functions as a substitute glycyl radical domain to repair oxygen-induced damage to the PFL glycyl radical (461). The fate of the formate formed by PFL depends on the environmental pH. At neutral pH, formate is excreted. As the environmental pH decreases, depending on the availability of oxygen, formate decomposes either to carbon dioxide by aerobic formate dehydrogenase (FDO) or to both carbon dioxide and dihydrogen by formate-hydrogen lyase (FHL) (1, 4, 386).
The resultant acetyl-CoA follows two alternative fates: either conversion to acetate or reduction to ethanol. The conversion of acetyl-CoA to acetate, catalyzed by the PTA-ACKA pathway, generates two ATP molecules per glucose but consumes no reducing equivalents (Fig. 4). The reduction of acetyl-CoA to ethanol, catalyzed by alcohol dehydrogenase (ADH), sacrifices energy but consumes reducing equivalents. Thus, by modulating the amount of ethanol and acetate it excretes, a cell can balance its requirement to regenerate NAD+ with its need for energy (for a review, see reference 54).
Acetate also can be excreted through the action of a third pyruvate-decarboxylating enzyme, pyruvate oxidase (POXB). Until recently, POXB had remained a mystery. This respiratory enzyme catalyzes the oxidative decarboxylation of pyruvate directly to acetate. The reaction produces carbon dioxide and reduces flavin adenine dinucleotide (FAD) (Fig. 3 and 5) (48, 49, 150). While the PDHC and PFL are considered essential enzymes, POXB generally has been regarded as nonessential and potentially wasteful (79, 80, 159). Mounting evidence, however, suggests that POXB provides energy and acetyl groups (as acetyl-CoA) under the microaerophilic conditions that prevail between exponential growth and stationary phase. Its transcription, dependent on 
s, is induced in the early stationary phase. Although it is expressed maximally during aerobic growth, some expression does occur during anaerobic growth (80). POXB null mutants grow less efficiently than their wild-type parent. When overexpressed or expressed constitutively, POXB can substitute for the PDHC, albeit less efficiently. Thus, it has been proposed that POXB contributes substantially to aerobic growth efficiency (2). How does POXB perform this function? After prolonged incubation (in the absence of acetate), mutants that lack the PDHC form microcolonies, whose development depends on a functioning POXB (79). The simplest model has POXB and AMP-ACS forming a pyruvate-to-acetyl-CoA bypass of PDHC (2) (Fig. 3 and 5), a model easily tested by the construction of mutants that lack both the PDHC and AMP-ACS. Some existing evidence, however, is consistent with POXB and AMP-ACS functioning at the same time: cells appear to coordinate the induction of acs and poxB transcripts (347), while poxB mutants exhibit reduced acs transcription (249).
Acetate as an Acid
The acetate that cells excrete into the environment presents those cells with a problem. Acetate, like other weak acids, is toxic (280, 390). In its undissociated or acidic form, this lipophilic weak acid easily permeates membranes, uncoupling the transmembrane pH gradient (34, 35, 233, 375). Once across the membrane, it dissociates into a proton and an anion (56, 233). The proton acidifies the cytoplasm, while the anion increases the internal osmotic pressure and interferes with methionine biosynthesis (381, 382, 388). However, acetate is only partially oxidized and thus is a potential source of both carbon and energy. Therefore, the ability of E. coli to perform the "acetate switch" (61, 251) permits it to solve its acetate problem in a rather creative manner: it removes this potential toxin from its environment by consuming it.
|
ACETATE ACTIVATION PATHWAYS
|
|---|
During aerobic growth of E. coli on acetogenic carbon sources, the switch from dissimilation to assimilation requires two acetate activation pathways. Dissimilation depends primarily on the PTA-ACKA pathway, while assimilation functions primarily through AMP-ACS.
Acetate Dissimilation: the PTA-ACKA Pathway
In E. coli, acetate dissimilation is catalyzed by the enzymes PTA [acetyl-CoA(CoA):Pi acetyltransferase; EC 2.7.2.1] (294) and ACKA (ATP:acetate phosphotransferase; EC 2.3.1.8) (264). PTA reversibly converts acetyl-CoA and inorganic phosphate to acetylP and CoASH, while ACKA reversibly converts acetylP and ADP to acetate and ATP (Fig. 5) (385). Thus, the PTA-ACKA pathway couples energy metabolism with those of carbon and phosphorus (463, 465). This pathway also can interconvert propionyl-CoA and propionate (385). Thus, it also functions in 
-ketobutyrate metabolism (457), degradation of fatty acids with odd numbers of carbons, the assimilation of propionate, and, in Salmonella, growth on 1,2-propandiol as a carbon and energy source (331).
Mutant phenotypes.
During shake flask growth, PTA-deficient mutants (pta or pta ackA) do not accumulate extracellular acetate (171, 172, 224, 360) (Table 1). In contrast, mutants that lack ACKA (ackA) accumulate small amounts of acetate (225, 360), which probably accumulates because the acetylP produced by PTA is labile at physiological pH (61). During high-cell-density fermentations, pta or ackA mutants each accumulate a fraction of the acetate accumulated by their wild-type parent (78, 93, 172, 224, 483). However, their specific acetate production differs: ackA mutants produce acetate like their wild-type parents, while pta mutants exhibit a considerably smaller specific acetate production (93). Some evidence suggests that the small amount of acetate produced by pta mutants does not result from POXB (78), suggesting the involvement of another acetate-producing pathway.
During aerobic growth on glucose, pta mutants excrete pyruvate, D-lactate, and L-glutamate instead of acetate, ethanol, formate, and dihydrogen (78, 118, 123, 224, 225, 482, 483). This resembles the normal behavior of wild-type cells exposed to low external pH, an environment that favors D-lactate excretion over that of acetate and formate. This behavior probably occurs because NADH-dependent lactate dehydrogenase (LDH) expression increases (68) while PTA expression and activity decreases (424, 437). Since wild-type cells exposed to high pH can tolerate a greater number of acid equivalents, they can up-regulate PTA expression (424), which favors the excretion of acetate (plus formate).
Excretion of D-lactate by pta mutants appears to replace excretion of ethanol as the primary mechanism for NAD+ regeneration. Excretion of L-glutamate is consistent with the observation that pta mutants up-regulate many TCA cycle genes (see below) and that aerobic growth on glucose represses KGDH (78, 439). The reduction in formate and dihydrogen levels suggests that the status of the PTA-ACKA pathway influences PFL activity (482), while the reduction in ethanol excretion supports the link between acetate (via the PTA-ACKA pathway) and ethanol (via ADH) as products of acetyl-CoA cleavage. Heterologous expression, in pta mutant cells, of enzymes or pathways that use acetyl-CoA as their substrate restores the wild-type pattern of fermentation products (with the exception of acetate) (78, 118, 482). These observations are consistent with the hypothesis that the PTA-ACKA pathway functions as an overflow pathway for excess carbon and that, in its absence, the cell uses alternative pathways to spill its excess carbon.
Cells that lack part or all of the PTA-ACKA pathway grow more slowly than their wild-type parents when grown aerobically in tryptone-based media, in defined minimal media supplemented with pyruvate, or in aerobic or anaerobic fermentations (61, 78, 93, 118, 172, 225, 240, 297, 466, 473, 482). In contrast, no defect occurs during growth on the nonacetogenic carbon source glycerol or the gluconeogenic carbon source succinate (78). The heterologous expression, in pta mutants cells, of enzymes or pathways that use acetyl-CoA as their substrate partially alleviates the growth defect (78, 118, 482). Thus, the growth defect might result from pyruvate accumulation, which would reduce the PEP/pyruvate ratio and, hence, lower the rate of substrate uptake by the PTS (78, 269, 340). Alternatively, it might be explained by a redox imbalance. Note that pta mutants cannot grow anaerobically on glucose unless they also lack ADH. These pta adh double mutants grow anaerobically on glucose by excreting the less reduced D-lactate via LDH instead of the highly reduced ethanol via ADH (Fig. 4). Furthermore, pta adh double mutants cannot grow anaerobically on sorbitol (a more reduced hexose than glucose) or glucuronate (a more highly oxidized hexose acid). Thus, the inability of pta mutants to grow anaerobically may result from redox imbalance (171). A similar imbalance might cause slowed aerobic growth of pta mutants, a problem that a heterologous acetyl-CoA-draining pathway could alleviate by depleting the acetyl-CoA pool and thus reducing the need to balance reducing equivalents.
Expression profile.
Slonczewski and coworkers (240) compared the proteome of wild-type cells to those lacking both PTA and ACKA. Similarly, Wolfe et al. (473) compared the transcriptomes of wild-type cells to those of mutants that lacked either ACKA alone or both PTA and ACKA. These analyses suggest that PTA-ACKA pathway mutants use a number of strategies in an attempt to cope with the loss of the energy-producing PTA-ACKA pathway. They increase the expression of TCA cycle enzymes, boost the expression of central glycolytic enzymes, and elevate the expression of a protein that facilitates PFL activity in the presence of oxygen (Table 1). In contrast, they do not increase the expression of the CoASH biosynthetic pathway or of enzymes involved in fatty acid biosynthesis, gluconeogenesis, or other fermentation pathways, including LDH and ADH. These mutants do, however, act as if they experience stress, especially that associated with the envelope and exposure to acid (240, 473). Indeed, their behavior resembles that of wild-type cells exposed to high acetate concentrations, an environment that results in down-regulated PTA expression (240).
Relative to their wild-type parents, both pta ackA and ackA mutants increase the steady-state transcript levels of genes that encode much of the TCA cycle (see Table S3 in the supplemental material of reference 473). They increase gltA (CS), icdA (IDH), sucD (a subunit of the SCSC), sucA (a subunit of KGDH), lpdA (a subunit of both the SCSC and PDHC), mdh, and b0725 (annotated as a hypothetical protein, but possibly only the 5' untranslated portion of the sdh-suc transcript) expression. This response suggests that PTA-ACKA pathway mutants attempt to compensate by using the TCA cycle to recycle CoASH and to generate ATP and is consistent with L-glutamate excretion. When coupled with the knowledge that both glucose starvation and anerobiosis increase the expression of the PTA-ACKA pathway while decreasing that of the TCA cycle enzymes and subunits IDH, SUCC, SUCB, SDHA, and LPDA (324), this response implies that elevated carbon flux through the PTA-ACKA pathway somehow decreases the expression of the TCA cycle. Because pta mutants survive glucose starvation poorly while ackA mutants survive about as well as their wild-type parents, Nyström proposed that acetylP might mediate this starvation response (324). However, heterologous expression in a pta mutant of an acetyl-CoA-draining pathway improves survival. Thus, Pan and coworkers have argued against Nyström's hypothesis (78). Interpretation of these studies remains problematic because of the intimate links between acetylP synthesis and acetyl-CoA levels. Future studies might benefit from a molecular and/or genetic system that uncouples this relationship, i.e., a system that permits acetylP synthesis without manipulating the acetyl-CoA pool or adding exogenous acetate, which can alter both the external and internal pH.
Both pta ackA and ackA mutants increase the steady-state level of the gapA transcript, which encodes GAPDH, the enzyme that catalyzes the NAD+-dependent oxidation of glyceraldehyde-3-P (Fig. 3). Similarly, pta ackA mutants increase the expression of triosephosphate isomerase, which catalyzes the interconversion of dihydroxyacetone phosphate and glyceraldehyde-3-P, just prior to the reaction catalyzed by GAPDH (Fig. 3) (240, 473). Note that both these mutants also up-regulate pfkB, which encodes the minor isoform of phosphofructokinase (473). Thus, pta ackA mutants may compensate by increasing carbon flux toward pyruvate. If so, then this compensation should increase the excretion of pyruvate and D-lactate, as observed (78, 482), and increase the amount of NADH.
PTA-ACKA- and ACKA-deficient mutants increase the steady-state levels of the YfiD protein and its transcript (240, 473). It has been proposed that YfiD acts as a substitute glycyl radical domain used by oxygen-stressed cells to repair oxygen-induced damage to the glycyl radical of PFL (461). Consistent with this hypothesis, growth in the presence of pyruvate or under microaerobic conditions elevates the level of YfiD (50, 290), while the pyruvate-sensitive regulator PdhR and the oxygen-sensitive global regulators FNR and ArcA control yfiD transcription (290, 476). Cells also up-regulate YfiD when exposed to low pH (50, 476) or to the organic acids propionate (50), benzoate (476), or acetate (240). Up-regulation also occurs in cells that express the heterologous FNR protein HlyX (161) or a heterologous acetyl-CoA-draining pathway (175). Intriguingly, peptidoglycan-free L-forms up-regulate YfiD, which appears as a phosphoprotein (141). In contrast, cells entrapped in gels down-regulate YfiD (344), as do those that overproduce threonine and excrete less acetate (262) or lack the global regulator FlhDC or the aerotaxis receptor Aer (356). Since YfiD is induced in acidic and/or microaerobic environments, it has been proposed to prevent the accumulation of acidic fermentation products by facilitating carbon flux through PFL (476). It remains unclear why PTA-ACKA pathway mutants attempt to compensate for their reduced ability to process acetyl-CoA by increasing the efficiency of an enzyme complex (PFL) whose products include acetyl-CoA. Perhaps these cells drive PFL backwards, thereby producing pyruvate and D-lactate instead of formate, as proposed by Slonczewski and coworkers to explain why formate results in reduced expression of acetate-induced proteins (424).
pta mutants resist extreme acid stress better than their wild-type parents do, a behavior that depends on crl, which encodes a transcription factor required for most s-dependent phenomena (240). Consistent with their enhanced resistance to extreme acid stress, PTA-ACKA pathway mutants increase transcript and/or protein expression of Crl and a subset of s regulon members (240, 473). This subset also includes a hyperosmotic stress-induced protein (OsmY), a stationary-phase-induced nucleoid protein (Dps) that contributes to acid and oxidative stress tolerance, and two periplasmic chaperones (HdeA and HdeB) induced by extreme acid stress. It also includes PFKB and the tryptophan repressor binding protein WrbA.
Transcripts of other genes implicated in stress responses also accumulate, most notably those that encode key chaperones and heat shock proteins (DnaK, GroEL, GroES, and ClpB) (473). The levels of most of these transcripts and/or proteins also become elevated in cells exposed to benzoate (256) or in those that express a heterologous acetyl-CoA-draining pathway (175). Other up-regulated stress genes include yhiE (which encodes a hypothetical protein implicated in the response to acid stress) and slp (encoding an outer membrane protein induced by carbon starvation). They also include ivy (formerly known as yfkE) (240, 473), whose gene product inhibits vertebrate C-type lysozyme.
In addition to HdeA and HdeB, increased expression of several other genes and/or proteins suggests that PTA-ACKA pathway mutants experience stress that affects the ability of the cell to fold, assemble, and/or insert envelope proteins. Such mutants up-regulate rseA, which encodes a negative regulator of E, the sigma factor that responds specifically to periplasmic stress. They up-regulate RfbX, a putative O-antigen transporter. They also increase the expression of the outer membrane protease OmpT (240, 473). Finally, mutants that lack both PTA and ACKA, but not those that lack only ACKA, up-regulate the outer membrane porin OmpC (240, 473).
Several envelope- and/or stress-associated transcripts become down-regulated in PTA-ACKA pathway mutants (see Table S4 in the supplemental material of reference 473). These include genes that encode an outer membrane porin (OmpF), a component of the general secretory apparatus (SecG), a sugar-binding integral membrane component of a PEP-PTS system (GlvC, also known as PtiC), and a cytoplasmic, ribose binding component of the high-affinity ribose transport system (RbsD). They also include genes encoding a flagellar biosynthetic chaperone (FlgN), a hypothetical fimbrial chaperone (YqiH), and a capsule biosynthetic protein of unknown function (WcaM). Finally, they include genes that encode the protein repair enzyme isoaspartyl dipeptidase (Iad) and a cold shock protein (CspC) (240, 473). CspC stabilizes the rpoS transcript (346); thus, one might expect a decrease in the transcription of s-dependent genes. Instead PTA-ACKA pathway mutants increase transcription of these genes, which argues that CspC does not play a key regulatory role under these conditions. Note that the two-component sensor RcsC negatively regulates yqiH, glvC and cspC. In contrast, it increases the expression of ivy and osmY (129), genes up-regulated in PTA-ACKA pathway mutants.
Acetate Assimilation: AMP-Forming ACS
In E. coli, AMP-ACS (acetate:CoA ligase [AMP forming]; EC 6.2.1.1) catalyzes acetate assimilation. A member of the firefly luciferase superfamily (445, 446), AMP-ACS first converts acetate and ATP to the enzyme-bound intermediate acetyladenylate (acetyl-AMP) while producing pyrophosphate (Fig. 5). It then reacts acetyl-AMP with CoASH to form acetyl-CoA, releasing AMP (46, 87). For a review concerning structure-function relationships, see reference 427.
Although reversible in vitro, this reaction is irreversible in vivo because of the presence of intracellular pyrophosphatases (PPase). This high-affinity pathway (Km of 200 µM for acetate), therefore, functions only anabolically, scavenging for small amounts of environmental acetate (61, 251). The reversible PTA-ACKA pathway also can assimilate acetate, but only in relatively large concentrations, because the enzymes of this low-affinity pathway possess Km values for their substrates in the 7 to 10 mM range (61, 251).
Because acetate freely permeates the membrane in its undissociated form (56, 233, 375, 390), assimilation does not require a dedicated transport system. However, under certain circumstances acetate uptake is saturable, suggesting that such a system exists (224). Recently, Gimenez et al. reported the existence of an acetate permease (ActP, formerly YjcG) and provided evidence for the existence of a second acetate transporter. The authors propose that these systems play critical roles when cells scavenge for micromolar concentrations of acetate (153).
Mutant phenotypes.
Cells that lack AMP-ACS (acs) grow poorly on low concentrations of acetate (
2.5 mM) as their sole carbon and energy source (251). In contrast, cells defective for all or part of the reversible PTA-ACKA pathway grow poorly on higher concentrations of acetate (
25 mM) (61, 224, 251). Since growth on low concentrations of acetate depends on AMP-ACS while growth on high concentrations requires the PTA-ACKA pathway, mutants that lack both cannot grow on acetate at any concentration (251). The inability to grow on acetate at any concentration argues against compensation by other acetate-activating enzymes, such as propionyl-CoA synthetase, encoded by prpE (201), or the propionate kinases, encoded by tdcD in E. coli or Salmonella enterica (186) or by pduW in S. enterica (52, 331). PrpE can use acetate as an alternative substrate (201); thus, PrpE can restore growth on acetate to acs pta ackA mutants, but only when expressed from a high-copy-number plasmid and then only poorly (S. Kumari and A. J. Wolfe, unpublished data). Similarly, AMP-ACS can activate propionate; thus, cells must lack both AMP-ACS and PrpE (acs prpE) before they can no longer grow on low concentrations of propionate (201).
In contrast to null mutants of acs, those that lack actP grow about as well as their wild-type parent on all concentrations of acetate tested. Gimenez et al. explain this puzzling result by noting that the lowest concentration tested (2.5 mM) is 50 times higher than the Km for ActP (5.4 µM). Since AMP-ACS possesses a Km of 200 µM for acetate, the authors propose that ActP functions with AMP-ACS to scavenge acetate at considerably lower concentrations than those tested (153). The observation that acs mutants compete poorly with their wild-type parent during prolonged starvation (A. J. Wolfe, unpublished data) is consistent with this proposal. Presumably, acs cells cannot scavenge for the small amounts of acetate released into the environment by dying cells. If so, then AMP-ACS (and perhaps ActP) should play critical competitive roles in any environment depleted of primary carbon sources.
When grown aerobically in shake flask culture, acs mutant cells grow as rapidly and accumulate as much acetate as their wild-type parents do (249, 251), exhibiting only a slight reduction in total biomass (A. J. Wolfe and C. M. Beatty, unpublished data). This reduction probably occurs because they cannot assimilate the previously excreted acetate even though they possess a fully functionally PTA-ACKA pathway (249, 251). Although reversible, the PTA-ACKA pathway generally does not participate in the assimilation of excreted acetate by cells grown in shake flask culture. This inability to participate in acetate assimilation arises because the extracellular acetate concentration generally does not exceed 1 mM and the enzymes of this low-affinity pathway possess Km values for their substrates in the 7 to 10 mM range (61, 251). In contrast, the PTA-ACKA pathway contributes to acetate assimilation during high-cell-density fermentation, probably because such growth results in considerably larger concentrations of extracellular acetate (93, 241, 455).
In glucose-controlled high-cell-density fermentation, the behavior of the acs mutant is puzzling. It grows initially with the same maximum specific growth rate as its wild-type parent. Growth slows, however, after a few hours and the culture achieves less than half the final biomass attained by the wild-type parent. Intriguingly, acs mutants accumulate only about a quarter as much extracellular acetate (1.8 g/liter, compared to 6.9 g/liter for the wild-type), while their specific production of acetate falls to about half. These results may implicate AMP-ACS in the control of carbon flux through the PTA-ACKA pathway (93). They also may indicate that AMP-ACS controls the expression and/or activity of the glyoxylate bypass (GB) of the TCA cycle (Fig. 3). During high-cell-density fermentations, the acs mutant achieves extracellular acetate concentrations high enough to permit assimilation by the low-affinity PTA-ACKA pathway (93). If the absence of AMP-ACS permits increased carbon flux through the GB, then the AMP-ACS-deficient mutant might accumulate extracellular acetate to a level lower than predicted by its specific rate of production. Indeed, two predictions of this model appear likely. First, a derivative of E. coli K-12 (JM109) accumulates considerably more extracellular acetate than does a derivative of E. coli B (BL21), although both produce acetate at the same specific rate (456). The GB appears to provide much of the difference. Flux and transcription analyses show that JM109 possesses a relatively inactive GB. In contrast, the GB of BL21 is considerably more active (323, 347). Second, acs mutants exhibit substantially elevated aceBAK transcription (Wolfe and Beatty, unpublished). Efforts to elucidate the relationship between the expression of AMP-ACS and that of the GB should help us understand why cells undergo the Crabtree effect.
Glyoxylate bypass and gluconeogenesis.
To assimilate acetate through either AMP-ACS or the reversible PTA-ACKA pathway, E. coli cells also require the GB (to replenish the TCA cycle) and gluconeogenesis (to provide sugar phosphates). Using two unique enzymes, isocitrate lyase (ICL) and malate synthase (MAS), induced during growth on acetate or fatty acids, the GB bypasses the two oxidative steps in the TCA cycle that evolve CO2 (Fig. 3). By avoiding the loss of two carbons at the expense of energy production, the GB permits the net accumulation of four-carbon biosynthetic precursors (e.g., succinate) during growth on two-carbon substrates (e.g., acetate) (for reviews, see references 90, 95, 100, and 246). Cells balance energy and biosynthetic needs, using exquisitely sensitive biochemical and genetic switches to control the flow of isocitrate through the TCA cycle and the GB. The biochemical switch modulates the activities of ICL and IDH (258). Whereas the phosphorylation of both ICL and IDH channels the flow of isocitrate into the GB, the dephosphorylation of both enzymes favors flow through the TCA cycle (203, 257). The genetic switch controls the aceBAK operon, which encodes ICL, MAS, and IDHK/P, the enzyme that reversibly phosphorylates IDH (44, 149). It involves repression by the GB regulator ICLR and activation by the nucleoid protein IHF and the global regulator Cra (also known as FruR) (reviewed in reference 95). ICLR also disassembles those transcription complexes that form, thereby augmenting its negative effect upon aceBAK transcription (478). For gluconeogenesis, PEP carboxykinase (PCKA) converts oxaloacetate (OAA) to PEP and the malic enzymes (encoded by sfcA and maeB) plus PEP synthase (PPSA) convert malate to PEP (329) (Fig. 3). Other gluconeogenic enzymes and reversible glycolytic enzymes drive the PEP toward glucose-6-phosphate. Note that Cra also activates genes that encode gluconeogenesis enzymes (319, 366, 367, 389).
Expression profile.
Relative to growth on glucose, wild-type cells grown on high concentrations of acetate (42 or 84 mM) up-regulate the steady-state levels of the transcripts and proteins of the following enzymes and pathways: AMP-ACS, the GB, the TCA cycle, and gluconeogenesis. In contrast, these cells down-regulate the transcripts and proteins of the PTA-ACKA pathway, nonacetate carbon transport systems, and glycolytic enzymes (226, 328, 329, 342, 451). Expression profiling of acs mutants has not been reported.
|
REGULATION OF THE ACETYLP POOL
|
|---|
The concentration of the intracellular acetylP pool should depend on the following factors: expression and activity of the pathway components, and the availability of pathway substrates.
Expression and Activity of the PTA-ACKA Pathway
In E. coli and S. enterica, ackA and pta are contiguous (254, 466, 479). In E. coli, this operon produces two transcripts, one that encodes both ackA and pta and a second that encodes only pta (224). On the basis of genetic evidence, each transcript appears to result from a distinct promoter, one positioned upstream of ackA and another upstream of pta (466). In S. enterica, in contrast, the evidence suggests a single transcript driven by a single promoter (254).
In E. coli, the steady-state levels of ackA and pta transcripts and their gene products ACKA and PTA, as well as their enzymatic activity, vary in response to diverse environmental factors. Oxygen tension is a major influence. Under anaerobic conditions, the levels of ACKA and PTA increase 8- and 10-fold (324) but their activities increase only 2- to 3-fold (61). In S. enterica, ackA transcription increases in response to anaerobiosis, but only about 2-fold (254). As described previously, the environmental pH and the presence of acetate both exert sizeable effects. During aerobic growth, the steady-state level of PTA decreases as the environment acidifies and increases as the alkalinity rises (424). In contrast, during anaerobic growth, the steady-state level of ACKA increases as the environment becomes more acidic (220). During growth on or after exposure to high concentrations of acetate, the levels of PTA protein and of pta and ackA transcripts decrease about two- to threefold (240, 328, 329, 451). Temperature also exerts an effect: as it increases, so too does PTA activity. In contrast, ackA transcription decreases (360). The richness of the medium also plays a role: the steady-state levels of both PTA and ACKA, and of ackA transcripts, rise three- to fivefold in glucose-supplemented rich medium relative to glucose-supplemented minimal medium (324, 440). A similar effect occurs on starvation: ACKA and PTA levels increase 1.5- and 3-fold, respectively (324). The nature of the carbon source influences ACKA and PTA activities, but only marginally; they double during growth on pyruvate, D-lactate, or D-gluconate relative to growth on all other carbon sources tested, including D-glucose and acetate. The increased PTA activity probably results from the ability of pyruvate to activate the enzyme (437). In contrast, the carbon source exerts no apparent effect on the level of transcript or protein in either E. coli (324) or S. enterica (254). Finally, NADH, ADP, and ATP inhibit PTA activity (437). Thus, a warm, alkaline, anaerobic, and nutrient-rich environment generally favors increased pathway expression and/or activity while a cool, acidic, aerobic, and acetate-rich environment does not.
Availability of Pathway Substrates
The balance between intracellular acetyl-CoA and extracellular acetate probably represents the most important influence on the intracellular acetylP pool. This should be especially true when the concentrations of extracellular phosphate, the total intracellular CoA pool, and the energy charge in ATP and ADP remain constant (196, 297). Under these conditions, the steady-state concentration of acetylP should vary with the substrate (acetyl-CoA) and the product (acetate) and should be influenced by the same environmental factors that affect either. These factors include the acetogenic nature and amount of the carbon source, the availability of oxygen or nitrogen, the glycolytic flux, and the environmental pH and temperature.
Measurements of the AcetylP Pool
Direct measurements of acetylP in bacterial lysates have been reported on only a few occasions. Two biochemical approaches have been used: one that relies on two-dimensional thin-layer chromatography (53), and one that uses a coupled assay (208).
Prüß and Wolfe, using an adaptation of Hunt's assay, observed a strong correlation between the pools of acetylP and acetyl-CoA during growth in tryptone broth (Fig. 1B). Immediately following dilution of overnight cultures into fresh tryptone broth, acetylP was undetectable (5 µM). The concentrations of both metabolic intermediates increased throughout the earliest stage of exponential growth, achieving a peak of 300 to 400 µM. The concentrations of both acetyl-CoA and acetylP decreased progressively, reaching about 20 µM by entry into stationary phase; that of acetylP remained low well into stationary phase (360). These results agree with measurements of acetyl-CoA levels in E. coli grown aerobically on glucose (86, 438). When one compares the consumption of amino acids with the accumulation of acetate and its activated derivatives, the following becomes apparent: acetyl-CoA and acetylP levels peak during the transition from L-serine to L-aspartate consumption, while extracellular acetate peaks during the transition from L-aspartate to L-tryptophan consumption. The reaccumulation of extracellular acetate during stationary phase occurs as cells cometabolize acetogenic amino acids, e.g. L-threonine and L-alanine, with those that require the TCA cycle, e.g., L-glutamate.
Using the two-dimensional thin-layer chromatography approach, McCleary and Stock showed that acetylP levels correlate with the acetogenic nature of the carbon source. Since the identity or quantity of the carbon source imposes only small effects on pathway activity and none on expression, these observations support the proposition that carbon flux exerts a major influence over the size of the acetylP pool. If the sole carbon source was either pyruvate or D-glucuronate, cells grown in low-phosphate defined medium, and harvested at midexponential phase, possessed high concentrations of acetylP (>1,000 µM). If the carbon source was D-glucose, they accumulated moderate amounts (300 µM). If the carbon source was D-fructose, glycerol, or fumarate, the levels were low or undetectable (50 µM). Relative to cells harvested during exponential growth, those harvested during early stationary phase increased their acetylP pool three- to fourfold, but only if grown on D-glucose or D-fructose. If grown on pyruvate, in contrast, the pool decreased by two- to threefold (297). Note that cells grown on pyruvate behave quite similarly to cells grown in tryptone broth, i.e., elevated levels of acetylP during exponential growth and reduced levels during stationary phase (297, 360). This should not be surprising, since cells convert L-serine, the preferred carbon source in tryptone broth (359), to pyruvate by means of an energy-independent transaminase reaction (334).
Prüß and Wolfe also demonstrated a correlation between incubation temperature and the intracellular acetylP pool. At or below 34°C, they could not detect acetylP; above that temperature, the concentration increased. The influence of temperature became more obvious in mutant cells that lacked ACKA; acetylP could be detected at 34°C, and its concentration rose dramatically up to 40°C (360). These results are consistent with the observations that extracellular acetate correlates with temperature (250) and can be readily explained by reduced ackA transcription coupled with increased PTA activity (360). Using a similar rationale, one might predict elevated acetylP levels during growth on glycolytic intermediates at alkaline pH. Such an environment should favor synthesis of PTA over that of ACKA.
Indirect measures also have been performed, using acetylP-dependent reporter systems. The best example interconnects nitrogen assimilation and acetylP. In the absence of NRII, acetylP acts as the predominate source of phosphoryl groups for the activation of the NRI and, hence, the transcription of glnA (Fig. 6). Since glnA encodes the enzyme glutamine synthetase (GS), one can monitor acetylP concentrations indirectly by measuring GS activity. Using this assay, Atkinson and Ninfa demonstrated that the status of a cell's acetylP pool depends on its ability to modify GS and regulate its activity. An adenylyltransferase (ATase) inactivates GS; thus, one would expect ATase-deficient cells to exhibit elevated GS activity. Instead, cells that lack both ATase and NRII display reduced GS activity. Since this activity depends on PTA, the ATase must regulate acetylP levels. This regulation probably occurs indirectly by controlling flux through acetyl-CoA. Because L-glutamate, a GS substrate, is derived from 2-ketoglutarate, elevated GS activity may deplete TCA cycle intermediates. This would draw acetyl-CoA into the TCA cycle, diminish flux through the PTA-ACKA pathway, and hence reduce acetylP accumulation. In contrast, reduced GS activity should favor elevated acetylP levels (20, 127). Direct measurements should be performed to test these conclusions and those derived from other acetylP-sensitive reporter systems.
P POOL
P Pool
P AS A GLOBAL SIGNAL
P Work?
P Signal?
P In Vivo
P-Forming Enzymes
Top
Next
