Bacteria may alter cell morphology, the production of flagella, cell metabolism, gene transcription, and cell behavior in response to environmental fluctuations such as the availability of respiratory electron acceptors, the supply of carbon, nitrogen, or phosphate, changes in the osmolarity and temperature of the medium, and whether the microorganisms are growing in liquid or on a solid surface. In this way bacteria survive in and adapt to ever: changing environmental conditions. Indeed, bacteria can be found in essentially all environments where life is possible, and from this point of view they can be considered to be one of the most successful forms of life In addition, individual bacteria that undergo physiological or developmental responses to signaling molecules produced by other bacteria of their own type in the population may then go on to engage in cooperative behavior. Such cooperative behavior includes biofilm forma- tion, which is a common occurrence among bacteria, and fruiting body formation by myxobacteria. Underlying the adaptations and responses are sophisticated detection systems with which the bacterium continuously moni- tors the environment and transmits signals across its cell membrane to specific intracellular targets, which can be the transcriptional machinery, enzymes, or a cellular component such as the flagellum motor. This chapter describes some of these signaling circuits, and metabolic, behavioral, and developmental responses to the signals. 18 Microbial Development and Physiological Adaptation: Varied Responses to Environmental Cues and Intercellular Signals There are many signaling pathways that bacteria use to transmit environmental and intercellular signals to a cell target, be itcellular proteins (e.g,, flagellar motors) or transcription factors that regulate gene expression. In most cases the signal causes the cells to activate or inactivate a cytoplasmic transcription factor that regulates the transcription of a gene. Some genes are induced by activated transcription factors, and some genes are repressed. The activation or inactivation of the transcription factor itself as a result of the signaling pathway usually occurs in one of two ways: by covalent modification [e.g., phosphorylation (activation) or dephosphorylation (inactivation)} of the transcription factor or by a conformational change of the transcription factor upon binding acoinducer. Two-component systems, which will be described in Section 18.1, are the most com- mon type of signaling system. They include a histidine kinase (HK) protein that autophos: phorylates at a histidine residue in response to a signal, then transfers the phosphoryl group to an aspartate residue in a response regulator (RR) protein that regulates gene transcrip- tion in its phosphorylated form. There may be intermediate protein carriers that transfer the phosphoryl group from the HK to the RR. ‘Among the examples of two-component sys- tems discussed are the following: 1, The Arcsystem, which regulates gene trans- cription in response to oxygen supply 467 ne"468 "THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES 2. The Nar system, which regulates the trans- cription of genes required for nitrate and nitrite metabolism 3. The Che system, which regulates the rota- tion of flagellar motors (rather than gene transcription) 4, The RegB/RegA system in photosynthetic bacteria, which regulates the transcription of genes for the light-harvesting complex and the photoreaction center under anaer- obic conditions 5. ‘The Pho system, which regulates genes for phosphate assimilation when the supply of phosphate is limiting 6. The EnyZ/OmpR system, which regulates the expression of genes for porin synthesis The KdpABC system, which regulates expression of genes for potassium ion transport, 8. The Spo phosphorelay system for the regulation of sporulation genes in Bacillus subtilis 9. The CtrA system, which regulates gene expression during the Caulobacter cres- centus cell cycle 10. The Asg system in Myxococcus xanthus, which regulates gene expression during multicellular development 11. The PhoQ/PhoP system, which regulates gene expression important for the viru- lence of Salmonella 12. The Bvg system, which regulates the expression of virulence genes in Bordetella pertussis 13. The Agr system in Staphylococcus aureus, which regulates virulence gene expression 14. The Vir system in Agrobacterium tumefa- ciens, which regulates the expression of genes necessary for the transfer of plasmid DNA to plant cells Also described is the FNR (fumarate nitrate reductase) system. It regulates the expression of ‘genes required for fermentation and anaerobic respization but is nota two-component system. Other signaling systems discussed in this chap- ter include quorum-sensing systems, which are widespread in bacteria and are important for cell-to-cell signaling at high population 4. sities. Some of the quorum-sensing systems send signals to two-component signaling systems, and others do not. Quorum-sensing systems rely on the production of extracellular signal. ing molecules that accumulate in the external medium as the population density increases, The quorum-signaling molecules were first discovered as signals that induced lumines. cence in bioluminescent bacteria; and because the signal induced the gene that encoded the enzyme that made the signal in a positive feed. back loop, the quorum-sensing signals were originally called “autoinducers.” The detection by the producing bacteria of a threshold con- centration of the signaling molecule in the external medium lets the bacteria know how dense their own population is and stimulates adaptive changes in the bacteria when the cell population is dense enough for these changes to occur. In this sense, quorum sensing reflects a primitive form of multicellularity that has evolved among the bacteria, and as we shall see, is very important for the survival of bacteria in their natural habitat. Quorum sensing regulates different pro- cesses, depending upon the species of bacteria: bioluminescence, biofilm formation, fruiting body formation in myxobacteria, the produc: tion of virulence factors, and conjugation are among the processes shown to be so regulated. ‘The signaling molecules used by gram-positive bacteria are usually oligopeptides. The peptides generally work by binding to the externally exposed regions of specific membrane-bound two-component sensor kinases and activating the phosphorylation of the respective partner response regulator. In that way the peptides regulate gene transcription. The signaling molecules used by gram- negative bacteria are generally acylated homo- serine lactones (acyl-HSLs). Most of these function by binding to and activating a positive transcription factor in the cytoplasm, and in that way the signal serves as a coinducer. In some cases, the extracellular signals produced by bacteria derepress or repress the trans- cription of genes. This chapter will also give examples of other signaling molecules used by gram-positive and gram-negative bacteria These include a furanosyl borate diester calledMICROBIAL DEVELOPMENT AND PHYSIOLOGICAL ADAPTATION 469 AL2, used by both gram-positive and gram negative bacteria, and a subset of amino acids used in the developmental signaling pathway by Myxococcus xanthus, This chapter shows how the various sig- naling systems underlie adaptive metabolic changes (Sections 18.2~18.10); it also covers the production of virulence factors by patho- genic bacteria (Section 18.11), chemotaxis, aerotaxis, and photoresponses (Sections 18.12~ 18.14). Afteran introduction to bacterial devel- opment and to quorum sensing (Section 18.15), we discuss myxobacteria multicellular develop- ment (Section 18.16),cell-cycle-dependent DNA replication and transcription in Caulobacter (Section 18.17), Bacillus sporalation and com- petence (Sections 18.18 and 18.19), bioluminesc- ence (Section 18.20), LuxR/Luxl-like systems in nonluminescent bacteria (Section 18.21), and the formation of biofilms (Section 18.22). Since many of the signaling systems feed into two-component signaling systems, we begin with an introduction to two-component sys- tems (Section 18.1). 18.1 Introduction to Two- Component Signaling Systems In several systems, a signal transduction path- way exists, called a two-component system, Each two-component system includes a histi- dine kinase (HK) protein (often called a sensor kinase) that receives a signal and transmits it to a partner response regulator (RR) protein, sometimes via other proteins that take part in a phosphorelay.! The response regulator protein in turn transmits the signal to the target. This is summarized in Fig. 18.1. Histidine kinases have two domains, an input domain anda transmitter domain. Response te- gulators have two domains, a receiver domain and an output domain, See note 2 for a further discussion of domains in sensor kinases and response regulators. Specifically, the histidine kinase receives a signal at its input domain and autophosphorylates (using ATP as the phos- phoryl donor) ata histidine residue in its trans- miter domain, The transmitter domain is the carboxy-terminal region comprising approxi- mately 240 amino acids. The histidine kinase then transfers the phosphoryl group to an aspartate residue in the receiver domain of the partner response regulator protein, The receiver domain is the amino-terminal region comprising about 120 amino acids. This acti- vates the response regulator, which transmits the signal to its target via its ourput domain (Section 18.1.2 and Fig. 18.1). Mostofthe known cytoplasm cell membrane signal to target Fig. 18.1 Two-component regulatory systems. (1) A transmembrane histidine kinase (HK) is activated by a signal at its N-terminal domain. (2) The acti- vated protein autophosphorylates in the C-terminal domain. (3) The response regulator protein (RR), binds to the C-terminal end of the histidine kinase and the phosphoryl group is transferred from the histidine kinase to the response regulator, thus activating the latter. (4) The activated response regulator leaves the histidine kinase and stimulates its target. Shaded and stippled areas of the histidine kkinase and response regulator represent conserved amino acid sequences typical for the respective class of protein. The change in shape of the proteins represents a presumed conformational change. In some systems the histidine kinase is cytoplasmic and. detects signals within the cytoplasm.470 phosphorylated response regulators (RR-P) bind to DNA and stimulate or repress the trans cription of specific genes. Exceptions include P- CheB and P-CheY, which affect the chemotaxis machinery (Section 18,12). Histidine kinases may reside in the cell membrane (usually trans- membrane) or in the cytoplasm, although they are often in the membrane. The response regu- lators are in the cytoplasm. The signaling pathway also includes a phos phatase that dephosphorylates the response regulator, returning it to the nonstimulated state, where it once again can respond to the signal. The phosphatase may be the histidine kinase itself, the response regulator, or a separ- ate protein. Even though signaling systems that consist ofa histidine kinase and a response regulator protein are called “two-component” systems, often there are more than two proteins in the signal transduction pathway, since addi- tional proteins may exist that carry the phos phate from the histidine kinase to the response regulator protein, Proteins thatmake up the phos- phorelay pathway between the histidine kinase and the partner response regulator protein are called phosphotransferases. Two-component signaling systems involving histidine kinases also occur in the archaea, fungi, plants, slime molds, and presumably other eukaryotes.24 Tt must be emphasized that the histidine kinase need not be the first protein in the signal transduction pathway to respond to the signal. In other words, it need not be the sensor. In many systems, signals first interact with pro- tein(s) other than the histidine kinase, and the stimulus is relayed to the histidine kinas For example, in the E. coli chemotaxis system described in Section 18.12, the transmembrane proteins called chemoreceptors, or MCP pro- teins, are sensor proteins that respond to chemoeffectors, and as a consequence, change the activity of a cytoplasmic histidine kinase (CheA). Section 18.5 presents another example of an initial receiver of the signal that is not the histidine kinase occursin the PHO regulon con- trol system, which is repressed by inorganic phosphate. (A regulonisa set of noncontiguous genes or operons controlled by the same trans cription regulator.) The proteins that initially ind inorganic phosphate are in the phosphate transport system (Pts), which is believed to bind inorganic phosphate and then stimulate THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES the enzymatic activity of the membrane-boung PhoR histidine kinase. A third (and more com, plicated) example is found in the Net regulon which is repressed by ammonia (Section 18.4), The ammonia levels determine the concentra, tions of glutamine and o-ketoglutarate via the enzymes glutamine synthetase and glutamate synthase, respectively. (As discussed in Sec. tion 9.3-1, these enzymes together function to incorporate ammonia into glutamate and glutamine, which donate amino groups to other molecules during biosynthesis.) The q. ketoglutarate and glutamine in turn influence the activity of a bifunctional enzyme, utidydy} transferase-uridylyl-removing (UT-UR) enzyme, which modifies a signal transduction protein (Px), which in turn regulates the activity of a cytoplasmic histidine kinase (Ny). In this case the histidine kinase is indeed far removed from the initial signal, ammonia, Two-component signaling systems have been discovered in many bacteria, both gram negative and gram-positive, and have been implicated in a wide range of physiological responses"? These include nitrogen assimila- tion, outer membrane porin synthesis, chemo- taxis in E. coli and S. typhimurium, nitrogen fixation in Klebsiella and Rhizobium, sporula- tion in Bacillus, fruiting body formation in myxobacteria, oxygen regulation of gene expression in E. coli, the uptake of carboxylic acids in Rhizobium and Salmonella, the pro: duction of virulence factors by Salmonella and Bordetella, and bioluminescence in some Vibrio species. It is clearly a widespread and important signal transduction system that enables bacteria to adapt to changes in the external milieu. As mentioned, there is also evidence to suggest that similar systems occur in eukaryotes." 18.1.1 Components of two-component signaling systems Wenoted atthe outset thatthe “two-component” systems contain proteins of three types. 1. A histidine kinase (HK), sometimes called 3 sensor kinase. The histidine kinase receives a signal at its input domain and auto phosphorylates at a histidine residue in its transmitter domain:MICROBIAL DEVELOPMENT AND PHYSIOL HK + ATP ®"5HK-P + ADP 2. A partner response regulator (RR), also called a cognate response regulator, which is phosphorylated at an aspartate residue in its receiver domain by HK-P and sends a signal to its target (e.g., the genome or the flagella motor} via its output domain: RR+HK-P SRR Pye + HK However, not all signals result in increased synthesis of RR-P. See Section 18.1.2. Furthermore, as noted earlier, there may exist enzymes called phosphotransferases that carry the phosphate from the HK to the RR in what is referred to as a phosphorelay pathway. 3. A phosphatase, which inactivates the RR-P: RR-P+H,O 9 RR+P, ‘The phosphatase may be the histidine kinase, the response regulator, or a separate protein. As mentioned, some signals stimu late phosphatase activity and thus act as inhibitors or repressors rather than stimula tors or inducers. We will see some examples of this later. 18.1.2 Signal transduction in two-component systems Figure 18.1 illustrates a simplified model of sig- nal transduction in two-component systems. In this model the signal stimulates phosphoryla- tion of the response regulator protein. (The ‘model will be modified later to accommodate differences between signaling systems, such as the inclusion of a separate sensor protein that receives the signal and transmits it to the histi dine kinase, and to show systems in which the signal actually results in less phosphorylation of the regulatory protein, hence ina suppression of a particular response rather than an activation.) In Fig. 18.1, the histidine kinase (HK) is depicted as a transmembrane protein com- Posed of three domains: an N-terminal domain that is presumed to be at the outer surface of the cell membrane and to bind to an external signaling ligand; a hydrophobic domain that is transmembrane; and a conserved C-terminal domain that is cytoplasmic. Some histidine CAL ADAPTATION, 471 kinases (e.g., NRy and CheA) are cytoplasmic proteins (Section 18.12.4). Signal transduction as depicted in Fig. 18.1 can be conveniently thought of as occurring in three steps. Step 1. In response to a stimulus at the N- terminal domain, the histidine kinase auto- phosphorylates at the C-terminal domain. The phosphoryl donor is ATP. Step 2. The phosphoryl group is transferred from the histidine kinase to its partner response regulator protein. All the response regulator proteins are related in having conserved amino acid sequences in the N- terminal domain (usually) that may bind to the conserved C-terminal region of the histi- dine kinases. Step 3. After phosphorylation, the response regulator becomes activated and changes the activity of its target. The effect is usually the stimulation or repression of gene transcrip- tion, As we shall see, some response regula- tor proteins (e.g., NarL) do both, depending. upon the target gene (Section 18.3). Other response regulators may have targets other than the genome. For example, in chemo- taxis, the phosphorylated derivatives of the response regulators CheY and CheB change the rotational direction of the flagella motors, and the extent of methylation of the chemoreceptor proteins in the membrane, respectively (Section 18.12.4). The response regulator proteins differ at their C-terminal domains, and this probably confers specific- ity with regard to their targets and activities. Not all two-component systems respond to the signal by increasing the phosphorylation of the response regulator protein. Sometimes the signal results in either dephosphorylation of the response regulator protein or inhibition of the phosphorylation of the response regula- tor protein. In either case, the level of RR-P falls rather than rises in response to the signal. The signal can result in the dephosphorylation of RR-P when the histidine kinases are bifunc- tional enzymes that can act either as kinases or as phosphatases when stimulated, depending upon the particular histidine kinase. For exam- ple, in the repression of the Ntr regulon by ammonia, the signal causes stimulation of the phosphatase activity of the histidine kinase,472 ‘THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES and as a consequence, represses gene trans- cription. In the presence of excess ammonia, the histidine kinase NRj acts as a phosphatase rather than as a kinase and inactivates the response regulator, NRj, which in its phos- phorylated form is a positive transcription factor (Section 18.4.1}. Another example may be the repression of the PHO regulon by inorganic phosphate, The histidine kinase, PhoR, appears to respond to excess inorganic phosphate by dephosphorylating the response regulator, PhoB (Section 18.5). In the case of chemotaxis to an attractant signal, the signal suppresses histidine kinase activity rather than stimulating the activity. This results in the for- mation of less RR-P (i.e., CheY-P, the protein that causes the cells to tumble and swim ran- domly: see Section 18.12.2), and consequently the cells swim smoothly toward the attractant. 18.1.3 Amino acid sequences define histidine kinases and response regulator proteins ‘The histidine kinases are defined by aconserved sequence of about 200 amino acids at the C. terminal end. The C-terminal domain isthe ste of the conserved histidine residue that becomes phosphorylated in response to a stimulus (Fig. 18.2). As indicated in Fig. 18.4, most of the known histidine kinases are transmem- brane, The C-terminal end is in the cytoplasm, where it interacts with the response regulator protein. The N-terminal end may be exposed on the extracellular membrane surface (the periplasm in gram-negative bacteria). As stated earlier, the amino acid sequence at the N- terminal domain varies with the different histidine kinases, presumably because they hy gr ophos: t noun Chea 3 on WR ——1—-s oe Phor 4} 2 1 env, = = —4-s-m— 0 Spetiy | 4» tos deta a tt om Pot ee a3 crea 4 ass Phew 4a m Vie a} $$ ee Phea ~--———- Une an ee} Fie == —_$_—___1—\—_a-m- ae Nerx. ——_ —s—___+—_sa— oo Deas. OH - aes Aarne memmensen- em} 0 3 Free er i Fig. 18.2 Structures of the histidine protein kinases. The histidine protein kinases are part of the two- component regulatory systems. They autophosphorylate ata histidine residue and then transfer the phosphate toa response regulator protein. Most of the histidine kinases are believed to be transmembrane proteins with an extracytoplasmic amino terminus that responds to stimuli. Exceptions are CheA, NRj, and FrzE, which are cytoplasmic. Hydrophobic domains that presumably span the membrane are indicated by the solid boxes at the amino-terminal end. Domain I, the cytoplasmic domain that includes the phosphorylated histidine residue, is indicated by the solid hours loss symbol. Regions Il and I (stippled and hatched boxes) represent regions in the carboxy-terminal domain where certain amino acids appear with high frequency at specific locations in the sequence. These are called conserved regions. For example, when the amino acid sequences are lined up for comparison, position 43 may be a glycine in all the proteins (totally conserved), whereas position 44 may be arginine in, say, 60% of the proteins, and so on (partially conserved). The open boxes at the extreme carboxy ends are homologous to response regulator domains at their amino-terminal end Lengths of proteins in amino acid residues, as predicted from nucleotide sequences, shown in right-most column. Source: Stock, J.B., A.J. Ninfa, and A. M. Stock. 1989, Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol. Rev. 53:450—490,‘MICROBIAL DEVELOPMENT AND PHYSIOLOGICAL ADAPTATION 473 respond to different stimuli. The response 10 the stimulus within the N-terminal region causes the enzyme to autophosphorylate the conserved histidine residue in the cytoplasmic domain (Fig. 18.1). The response regulator proteins are defined bya conserved amino-terminal domain of about 100 amino acids (Fig. 18.3]. The conserved amino-terminal end of the response regulator protein is thought to interact with the con- served carboxy end of the histidine kinase, becoming phosphorylated at an aspartate residue, The phosphate is eventually removed by a phosphatase, which may be the histidine Kinase, the response regulator protein, or per- haps a third protein. Since the carboxy ends of the histidine kinases and the amino ends of the different response regulator proteins are conserved, it is theoretically possible that a histidine kinase in one signaling system may activate a response regulator of a different system. This possible interaction between dif- ferent signaling systems would be an example of what has been termed “cross-regulation,” the regulation of a response regulator by a signal that comes from a source other than the cognate histidine kinase."® Cross-regulation is discussed in Section 18.5.1. chev —__—_ 129 Spoor — 103 wey SS eee $$ 09 Deto Sa eee ts nye SS eee —— Pata Serer a8 Feo Tr ss NEA ($3 ree TYR rere an Pho eo 220 ompR ———— 208 Area ——— ee 238 Phou-2 or ee Es vire ee 2st PhoP —— oe 28 Teto —— 224 ToxR a 234 Una =e 196 Fins ——— 204 Com. —— ss a Wert. — 216 Degu — ee zs wvrc-2 —S 220 che — a0 — 288 ——— 238 Sor 27 mm 192 Fig. 18.3. Structures ofthe response regulator proteins. Response regulator proteins are cytoplasmic proteins that are phosphorylated, or presumed to be phosphorylated, by the histidine kinase proteins. The phos- Phorylated regulator proteins transmit the signal to the genome or to some other cellular machinery (e.g. the flagellar motor). The amino-terminal region of the response regulator protein has conserved amino acids (open boxes). Approximately 20 to 30% of the amino acids are identical at corresponding posititions when the sequences are aligned. This is the region thought to interact with the carboxy domain of the histidine kinase, and to become phosphorylated. Other conserved regions are indicated by cross-hatched boxes. These are in the carboxy-terminal domain, which is thought to interact with. target molecules. For example, NRj, DetD, and NifA share a homologous carboxy-terminal domain that is thought to interact with one of the E. coli sigma factors, 0%. NR, and NifA have a common carboxy-terminal region that is thought to bind to DNA. One of the proteins, ToxR, spans the membrane, and the hydrophobic region is indicated by the Solid box. Lengths of the protein in amino acid residues shownas in Fig. 18.2, Source: Stock, J.B., A.J. Ninfa, and A. M, Stock. 1989, Protein phosphorylation and regulation of adaptive responses in bacteria. Microbiol, Rev, $3:450-490.474 ‘THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES 18.2 Responses by Facultative Anaerobes to Anaerobiosis A shift from an aerobic to an anaerobic atmosphere results in extensive changes in the metabolism of facultative anaerobes due to the repression of genes required for aerobic growth and the induction of genes necessary for anaer- obic growth." These adaptive responses, which have stimulated much interest, are described next. Most of the discussion focuses on what has been learned from studying E. coli and related bacteria. However, similar systems exist in other bacteria, The metabolic changes will be discussed first, followed by a description of the regulation of the relevant genes. Many genes that are responsive to anaerobiosis are globally regulated by two systems: the Arc (two-component) system, and the ENR (not two-component), system. Then we shall describe the regulation of the formate hhydrogen-lyase system, which also responds to anaerobiosis but is regulated by formate and a transcription activator, FhIA. The Regb/RegA system will also be described; thistwo-component system stimulates the tran. scription of genes for the light-harvesting complex and photoreaction center of certain photosyn- thetic bacteria under anaerobic incubation con- ditions, A fifth system, the (NarL/NarP/NarX/ NarQ) system, is a two-component system that regulates the response to nitrate and nitrite as electron acceptors under anaerobic conditions, is discussed in Section 18.3 18.2.1 Metabolic changes accompanying the shift to anaerobiosis During aerobic respiration, the citric acid pathway is cyclic and oxidative (Fig. 18.4A). However, in the absence of oxygen several important changes take place (Fig. 18.4B). These include the replacement of fumarase A and succinate dehydrogenase by fumarase B and fumarate reductase, respectively, and the repression of the synthesis of a-ketoglutarate dehydrogenase. As described in Section 8.10, the result of these enzymatic changes is the conversion of the oxidative citric acid cycle into a reductive noncyclic pathway. (Not all bacteria have a reductive citric acid pathway when they respire anaerobically. For example, when Pseudomonas stutzeri grows anerobically using nitrate as an electron acceptor, it oxidizes glucose completely to CO, and appears to have an oxidative citric acid cycle." Similarly, Paracoccus denitrificans has a complete citric acid cycle when it grows anaerobically, with nitrate as the electron acceptor.) Additionally, when E. coli is growing anaerobically, acetyl-CoA is no longer made by pyruvate dehydrogenase but rather by pyruvate-formate lyase. This is an advantage under anaerobic conditions because it decreases the amount of NADH that must be reoxidized, The acetyl-CoA thus formed is converted to acetate and ethanol, and the formate is con- verted to hydrogen gas and carbon dioxide via formate hydrogen-lyase. In E. coli there is also a decrease in synthesis of other enzymes used during aerobic growth (e.g. those of the slyoxylate cycle and fatty acid oxidation). Depending upon the presence of particular electron acceptors, major changes also take place in the respiratory pathway." Facultative anaerobes such as E. coli carry out aerobic respiration in the presence of oxygen; but in the absence of oxygen, they carry out anaerobic respiration by means of nitrate, fumarate, or some other electron acceptor (e.g., TMAO or DMSO, compounds that occur in nature and presumably exist in the intestine, where E. coli grows) There is a hierarchy of electron acceptors that are used: oxygen is the most preferred, with nitrate second, followed by fumarate and the other electron acceptors. The hierarchy par allels the maximum work that can be done when electrons travel over the electrode poten- tial gradientto the terminal acceptor. The work is proportional to the difference in electrode potential between the electron acceptor and donor. For example, the maximum work that can be done is greatest when oxygen is the elec- tron acceptor (Ey, = +0.82 V), less when nitrate is the electron acceptor (E’, = +0.42 V), and least when fumarate is the electron acceptor (Ej,=+0.03 V). As described in Section 4.4, the electrons are passed to these terminal electron acceptors from reduced quinone. For example, ubiquinone (Ey, = +0.1 V) transfers electrons from the reductant to either the cytochrome oxidase or the nitrate reductase module.8 ny won nxn 0 2euodsa aandepy TET ND “UT PU “s fondssHHOG “aanernoasy a emp TT 20] sondooo aomoaye ue yo so.Bage 98) UY SetNPas ONVURCONNH TL 2 siAanoe oseuadosp (yap aeze: Y FE Guas)seundoupafoporsoone sedan fa Bik : i 5 3 3 3 1qo,08uy (@) z ‘sesepKo E ‘ov E ‘ow 2 zZ : oo g fros vessf! Noms g ercon——amaly. dome ce sok fom 3s aNE1220exQ : Se) : Te g gL oe 100 amon 2 Sing 2, —476 THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES Menaquinone (E/, = ~0.074 V), rather than ubiquinone, transfers electrons to the fumarate reductase complex. (Fumarate reductase is at too low a potential to accept electrons from ubiguinone E. coli determines which electron acceptors will be used, in part by regulating the transcrip- tion of genes coding for the electron acceptors. Thus, in the presence of oxygen, the genes for. nitrate reductase, fumarate reductase, and the other reductases are repressed, and therefore only aerobic respiration can take place. (How- ever, see the discussion in Section 4.7.2 regard- ing denitrifying enzymes not sensitive to oxygen in some facultative anaerobes.) In the absence of oxygen but in the presence of nitrate, nitrate reductase genes are induced (by nitrate), but the genes for fumarate reductase and the other reductases are repressed by nitrate. Therefore, nitrate respiration takes place. In the absence of both oxygen and nitrate there is no longer any repression of fumarate reductase and so fumarate respiration takes place. Changes also take place within the aerobic respiratory pathway. When oxygen levels are high, E. coli uses cytochrome 0, encoded by the cyo operon, as the terminal oxidase. When oxygen becomes limiting (and during station- ary phase), cytochrome o is replaced by cyto- chrome d, encoded by the cyd operon. One advantage to this is that cytochrome d has a higher affinity for oxygen (K=0.23-0.38 uM) than has cytochrome 0 (K,,= 1-4-2.9 tM) In summary, then, oxygen represses the synthesis of the anaerobic reductases, ensuring that oxygen is used as the electron acceptor in air, Nitrate induces the synthesis of nitrate reductase and represses the synthesis of the other reductases, ensuring that nitrate is used as the electron acceptor in the presence of nitrate anaerobically, In the absence of any exogen- ously supplied electron acceptor, E. coli relies on fermentation as its major source of ATP. 18.2.2 Regulatory systems that govern gene expression accompanying the shift to anaerobiosis Four systems for the regulation of gene ex- pression by oxygen or nitrate in E. coli will be described, Two of these are activated by anaer- obiosis (ArcA/B and FNR), the third by nitrate and nitrite (NarL/NarP/NatX/NarQ system) the fourth is activated by anaerobiosis an formate and repressed by nitrate (the Fhia regulon). Photosynthetic bacteria havea regula, tory system, called the RegA/RegB system, which is activated under anaerobic condition, and controls the induction of photosynthetic genes, as decribed in Section 18.6.1. Acommon way to test the regulation of gene transcription is to construct a fusion between the promoter of the gene of interest and the gene for § galactosidase (lacZ). Then strains harboring the fusion are rested under different conditions to see how the test conditions affect the pro. duction of B-galactosidase. (A more detailed explanation of lacZ fusions is given later, in note 25.) The four systems in E. coli for the regulation of gene expression by oxygen or nitrate are as follows: 1. The Are (aerobic respiratory control, or anoxic redox control) system. This two- component, system represses under anaero- bic conditions the transcription of several genes that are expressed only during aerobic growth. In addition, it stimulates the trans- cription of a much smaller number of genes that are expressed during microaerophilic or anaerobic conditions. The Arcsystem func- tions during aerobic and anaerobic growth. 2. The FNR (fumarate nitrate reductase) system. This system stimulates the trans- cription of many genes that are required for fermentation and anaerobic respiraton, and represses the transcription of some genes that function only during aerobic growth. More than 75 genes are regulated by FNR in E. coli. The FNR system is active only during anaerobic growth; it is not a two- component system. 3. The NarL/NarPiNarX/NarQ system, also called the Nar system. This two-component system stimulates the transcription of the nitrate reductase and other genes required for nitrate and nitrite metabolism, and represses the transcription of the other ter minal reductase genes. The system is active only under anaerobic growth conditions in the presence of nitrate or nitrite. The Nar system is described in Section 18.3. 4. The FbIA regulon, also called the formate regulon. This systems consists of genes1 1 9 t ¢ 4 4 MICROBIAL DEVELOPMENT AND PHYSIOLOGICAL ADAPTATION 477 repressed by oxygen and nitrate and induced by formate. The regulon includes several genes including those for formate hydro genlyase, which is necessary to convert the fermentation end product formate to H and CO}. The Arc system The Arc system consists of a transmembrane sensor kinase (histidine kinase), ArcB, and a partner response regulator, ArcA. Mutations in the arc genes cause derepression of several genes and repression of a relatively short list of others, indicating that phosphorylated ArcA is both a repressor and an activator of gene transcription under anaerobic conditions. By analyzing these mutants, it has been possible to conclude that the Arc system is responsible for the following activities: 1, Anaerobic repression of genes for Citric acid cycle enzymes Glyoxylate cycle enzymes Several dehydrogenases for aerobic growth (eg, pyruvate dehydrogenase) Fatty acid oxidation enzymes Cytochrome o oxidase 2. Anaerobic induction of the gene for pyru- vate formate-lyase'” 3. Induction in low oxygen of the genes for cytochrome d oxidase and cobalamin syn- thesis! and (along with FNR), the pyruvate formate-lyase gene!” Interestingly, ArcA-P, but not ArcB, may also activate the transcription of the mating system genes, as explained in note 20, When oxygen levels are sufficiently low, the ArcB protein autophosphorylates, using ATP as the phosphoryl donor, and then transfers the phosphoryl group to ArcA (Fig, 18.5)222 How does ArcB detect changes in the levels of ‘oxygen? Apparently, itis not oxygen itself that is the signaling molecule. This conclusion has been reached from two lines of evidence: (1) the ‘Gi agié ana ghonyate eye enzymes| ‘9, Succinate dehyerogenase ‘2Owogtarate deryeogenase Fumarase A iocttae hace Pyvatedenycrogerase tclactate ooryeregenase ‘Amine acd dehyerogenase ‘SHycronyacy-CoA dehysrogenase Gpoctvome o onde Cyochvere doxidase Flue syntesis and ther sex acon Treen Fig. 18.5 The ArcA/ArcB regulatory system in E. coli. ArcBis a membrane protein activated by anoxia, per- haps by a reduced form of an electron carrier. The model postulates that the activated form of ArcB becomes Phosphorylated and then phosphorylates ArcA, which then becomes a repressor of aerobically expressed enzymes and an inducer of cytochrome d oxidase. It has been found that ArcA-P also activates the genes for cobalamin synthesis (cob genes) in Salmonella typhimurium. (See: Andersson, D. 1. 1992. Involvement of the Arc system in redox regulation of the cob operon in Salmonella typhimurium. Mol. Microbiol. 6:1491-1494,) ArcA also responds to CpxA, a membrane-bound sensor protein that is necessary for the Synthesis of the F-pilus and other sex factor functions. Source: Spiro, S., and J. R. Guest. 1991, Adaptive responses to oxygen limitation in Escherichia coli. Trends Biochem. Sci. 16:310-314478 ‘THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES level of expression of the sdh (succinate dehy- drogenase) operon in cells grown with different terminal electron acceptors and (2) the study of ‘mutants with deletions in the cytochrome oxi- dase genes, These experiments are described next. The level of expression of the sdh operon, which is under ArcA/ArcB control, varies with the midpoint potential of the electron acceptor. (The expression of the sdb operon is complex and is controlled by factors in addition to Arc\/ ArcB. See note 23 for further comment.) The level ofexpression is highest with oxygen, lower with nitrate, and lowest with fumarate. This parallels the midpoint potential of the electron acceptor, which is most positive for oxygen, less positive for nitrate, and least positive for fumarate. Instead of responding to the terminal electron acceptors fumarate, nitrate, and oxy- gen per se, the ArcB protein may respond to a reduced carbon compound in the cell, such as the reduced form of an electron transport car- rier (e.g., flavoprotein, quinone, cytochrome), or NADH, or some metabolic intermediate that might accumulate anaerobically. One way to test this hypothesis is to delete the cytochrome o and d genes and measure the expression of genes under the control of the ‘Are system. Deletion of the cytochrome oxi- dase genes inhibits electron transport at the terminal step and would be expected to have extensive metabolic consequences, including, for example, an increase in the ratio of reduced to oxidized forms of electron carriers, and the accumulation of metabolites that are formed in the absence of an exogenous terminal elec- tron acceptor. If this is the case, then deletion. ‘of the cytochrome oxidase genes would be expected to mimic the absence of oxygen. This proposition was tested, using cyo-lacZ and od-lacZ fusions as probes to monitor the expression of the cyo and cyd genes. (Gene fusions are explained in note 25.) When both the cyo and cyd genes were deleted and the cells were grown in air, cyo-lacZ expression was lowered and cyd-lacZ expression increased, as if oxygen were absent. (See note 26 for a description of the control experiments.) One interpretation of these experiments is that the reduced form of an electron transport carrier (e.g.,a quinone) may signal ArcB, which in turn autophosphorylates and then phos- phorylates ArcA (Fig. 18.5). The control of the phosphorylating activity of ArcB also appears to involve certain cellular metabolites, such a5 pyruvate, acetate, and p-lactate, that increase when oxygen is lacking. These products in. crease the autophosphorylation activity of ArcB in vitro. It has been demonstrated in vivo that p-lactate allosterically enhances the autophosphorylation activity of ArcB that has already been activated by anaerobiosis, rather than signaling inactive Arcb” Other intermedi ates may have similar effects. These and other aspects of the regulation of the ‘ArcA/ArcB system are discussed by Iuchi et al/2¥ and by Lynch and Lin. The ENR system When E, coli is shifted from aerobic to anaer- obic growth, a number of genes required to grow anaerobically are induced. At the same time, genes required for aerobic growth are repressed (Fig. 18.4). Part of the regulation is due to a proteincalled FNR, which is encoded by the fur gene. The FNR protein is a positive regulator of transcription for many genes that are expressed only during anaerobic growth, and a repressor for certain genes that are expressed only during aerobic growth (Fig. 18.6).°! There is no evid- ence that FNR is phosphorylated or is part of a two-component regulatory system, and it is discussed here in the context of a protein involved in aerobic/anaerobic regulation of gene expression, Genes whose expression requires FNR include those coding for the anaerobic respiratory enzymes fumarate reductase (frdABCD) and nitrate reductase (narGHJ1), as well as several other enzymes, including pyruvate formate lyase (pfl genes), formate dehydrogenase-N (the respiratory formate dehydrogenase, fdnGHI), aspartase (asp), anaerobic fumarase B (fim), and glycerol-3-phosphate dehydrogenase (gipA). In agreement with the role of activator of gene expression, mutations in the fur gene result in an inability to grow anaerobically on fumarate or nitrate as electron acceptors. FNR is also a negative regulator for several aerobi- cally expressed genes, including those for cyto: chrome o oxidase (cyoABCDE), cytochrome d oxidase (cydAB), succinate dehydrogenase (sdhCDAB), and superoxide dismutase (sodA). Thus, FNR is a global regulator of geneMICROBIAL DEVELOPMENT AND PHYSIOLOGIC: PNR, Nn 479 Racine cytochrome o oxidase ¥ ENR (autoregulation } NADH dehydrogenase IL ssparginase I sparse DMSO: TMNO reductase ormate detydrogenase fumarate reducase state reductase pyruvate formate-lyase Fig. 18.6 ENR is a transcription regulator protein during anaerobic growth. In the absence of oxygen, FNR becomes activated to become an inducer for many anaerobically expressed genes and a repressor for certain aerobically expressed genes. It has been speculated that FNR might become activated by the reduction of bound ferric ion, causing a conformational changein the protein. This s not an example of atwo-component regulatory system. expression during anaerobic growth, It should be pointed out, however, that many of the genes regulated by FNR are also regulated by the ArcA/ArcB system and the Nar system Although FNR regulates gene activity during anaerobic growth, studies have shown that the FNR proteinin E. coliis present in comparable amounts in both aerobically and anaerobi- cally growing cells. However, it is believed to be largely in an inactive state during aerobic growth So, how does E. coli regulate the activity of FNR? The answer entails an iron-sulfur cluster in FNR. Active FNR is a homodimer of an iron-sulfur protein with an oxygen-labile [4Fe-4S] cluster in each monomer. Upon exposure to oxygen the [4Fc-48] cluster is oxidized and can even be lost from the protein. (See note 34 for a further explanation.) When this happens a fraction of the FNR loses its iron clusters and becomes an apoprotein.3>” Both the protein bearing an oxidized cluster and the apoprotein are inactive. They bind with low affinity to DNA and do not stimulate transcription. Both the oxidation of the iron-sulfur clusters and their loss are reversible, and under anaerobic condi- tions the Fe-S clusters are restored, However, FNR expression need not be regulated exactly the same in all bacteria. For example, whereas in £, coli the transcription of frr seems to be similar under both aerobic and anaerobic growth conditions, and the amounts of active ENR increase anaerobically solely at the post-translational level, transcription of far is strongly stimulated in B. subtilis during oxygen limitation, The anaerobic activation of transcription of fr in B. subtilis requires resDE, @ two-component system (resD is a response regulator gene and resE is a histidine kinase gene). The activity of FNR in B. subtilis may also be regulated by anaerobiosis in a fashion similar to its regulation in E. coli An interesting parallel exists between the FNR protein and another transcriptional regulator, the cAMP receptor protein, CRP (cyclic AMP receptor protein, also called the catabolite activator protein, CAP). The CRP. protein isa positive transcriptional regulator for catabolite-sensitive genes (ie, genes repressed by glucose). CRP, in response to binding to cAMP, binds to specific sites on the promoter region of the target gene to activate transcrip- tion. A comparison between the nucleotide- derived amino acid sequences of FNR and CRP reveals that FNR and CRP are very similar in structure: both have a DNA-binding domain that allows the dimer to bind and a nucleotide- binding domain (although numerous attempts to show specific binding of nucleotides to FNR have failed). This has led to the suggestion that FNR and CRP are examples of a family of proteins that regulate transcription at the pro- moter region of target genes. Indeed, there have been several reports of proteins resembling ENR in bacteria other than the enterics, and these FNR-like proteins take part in the regula- tion of a variety of metabolic activities. (See note 39 and refs, 40 and 41.)480 ‘THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES In summary, ArcA and FNR are activated by anaerobiosis and regulate the enzymological changes that accompany the shift from aerobic to anaerobic growth. For example, activated ArcA (ArcA-P) primarily represses several aerobically expressed genes, including pyruvate dehydrogenase, succinate dehydrogenase, c- keroglutarate dehydrogenase, and cytochrome o oxidase (Fig.18.5). It is also a positive regu- lator for some genes expressed when oxygen is low or lacking, including the cytochrome d oxidase gene and the pyruvate-formate lyase gene. Similarly, activated FNR represses sev- eral genes normally expressed during aerobic growth, including the genes for cytochrome © oxidase, succinate dehydrogenase, and pyruvate dehydrogenase, and is required for the induction of many genes during anaerobic growth, including pyruvate-formate lyase, fumarate reductase, and nitrate reductase (cf. Figs. 18.5 and 18.6. Also, see note 23.) Ir is important to point out that in several cases, FNR and ArcA regulate the same gene. For example, they are both positive regulators for the pyruvate-formate lyase gene. Some- times they have opposite effects on the same gene. For example, ArcA is a positive regulator for the cytochrome d oxidase gene, whereas FNR js a negative regulator for that gene. Multiple controls of the same gene are a com- mon theme in bacterial physiology. However, a note of caution must be introduced when one is attempting to extract from physiological studies of mutants conclusions about coregula~ tion by ArcA and FNR. This is because FNR has been reported to stimulate anaerobic arcA expression when this activity was monitored via arcA—lacZ fusions.*! Regulation of the formate hydrogen-lyase pathway ‘When enterobacteria such as E. coli are grown anaerobically, formic acid is converted to FI, and CO; via the enzyme complex formate hydrogen- lyase (Figs. 14.7 and 18.4B). Formate hydro- gen-lyase consists of two enzymes, formate dehydrogenase H and hydrogenase 3, whose genes are repressed by oxygen and nitrate, and induced by formate. Induction also requires an acidic pH in the external medium, which probably means that the cells make formate hydrogen-lyase to prevent the medium fro, becoming too acidic owing to the production of formic acid from pyruvate (the pyruvate formate-lyase reaction). (E. coli actually makes three distinct formate dehydrogenases ang three different hydrogenases. See note 42 for 4 discussion of this point.) The genes for formate hydrogen-lyase are part of a regulon that has two names: the Fh, regulon and the formate regulon. The regulon is controlled by a transcription factor called FhIA and by formate, which is a coactivator of FHIA. (See ref. 43 for a review; see note 44 for a further description of the regulon.) Under aerobic conditions, formate levels are kept low, and because of this the regulon is not induced, ‘Two different mechanisms keep the formate levels low under aerobic conditions. The syn. thesis of formate from pyruvate is prevented under aerobic conditions because the pf gene, which codes for pyruvate formate-lyase, which in turn synthesizes acetyHCoA and formate from pyruvate and CoASH, requires active ENR as a positive regulator. Since oxygen pre- vents activation of FNR, oxygen represses the pfl gene, thus preventing formate synthesis, Oxygen and nitrate also lower the levels of formate in the cell because they induce the synthesis of the respiratory formate dehydro- genases FDH-O and FDH-N, which oxidize formate to CO). 18.3 Response to Nitrate and Nitrite: The Nar Regulatory System E, coli can be grown anaerobically by using nitrate (NO3) as the electron acceptor. The pro- cess is called nitrate respiration and is reviewed in Section 4.7.1. (See the subsection entitled Anaerobic respiratory chains.) In fact, when F. coli is given a choice of electron acceptors such as nitrate, nitrite, or fumarate under anaerobic conditions, it will utilize the nitrate. This prop- erty may be advantageous because nitrate has a more positive redox potential than the alter- native electron acceptors, and therefore more energy is potentially available from electron transport when nitrate is the electron acceptor. Nitrate is preferentially used as an electron acceptor during anaerobic respiration because it induces the transcription of genes resulting inMICROBIAL DEVELOPMENT AND PHYSIOLOGICAL ADAPTATION 481 the synthesis of a membrane-bound nitrate reductase and represses the transcription of genes encoding the other reductases (e.g. fumarate reductase, DMSO/TMNO reductas formate-dependent nitrite reductase). The formate-dependent nitrite reductase operon nif (nitrite reduction by formate) encodes a periplasmic nitrite reductase that catalyzes nitrite reduction via formate as the electron donor. A Ap is produced.* (See note 46, which lists the various genes in these operons.) When nitrite is the electron acceptor, it induces trans cription of genes encoding both the cytoplasmic and periplasmic nitrite reductases. ‘After a brief summary of the pathway for nitrate reduction (Section 18.3.1), we discuss the enzymes involved (Section 18.3.2). A sim- plified overview of the Nar system and how it regulates the transcription of genes required to use nitrate and nitrite as electron acceptors during anaerobic respiration (Section 18.3.3) is followed by a more detailed account of the ‘Nar regulatory system (Section 18.3.4). Then ‘we consider the role of integration host factor (IH) in the regulation of gene transcription by the Nar system (Section 18.3.5) and the biochemical and genetic evidence for some of the conclusions regarding how the genes are regulated by the Nar system (Section 18.3.6). 18.3.1 Pathway of nitrate reduction ‘The nitrate (NO5) is first reduced to nitrite (NO}) which, in turn, is reduced to ammonia (NH;). A total of eight electrons is required: two to reduce nitrate to nitrite, and six to re- duce nitrite to ammonia, The enzymes involved are described in Section 18.3.2. Part of the ammonia is assimilated into cellular nitro- geneous compounds, and part is excreted into the medium. Thus, under anaerobic conditions E, colican use nitrate both as an electron accep- tor during anaerobic respiration and 2s.a source of cellular nitrogen. 18.3.2 The enzymes involved The two major enzymes for the reduction of nitrate to ammonia during nitrate respir- ation in E. coli are a membrane-bound nitrate reductase that reduces nitrate to nitrite (a two- electron transfer) and a cytoplasmic NADH- linked nitrite reductase that reduces nitrite to ammonia (a six-electron transfer). Nitrate reductase is also called respiratory nitrate reductase and is encoded by the narG operon (narGH]1). It translocates one proton per elec- tron from the cytoplasm to the periplasm dur- ing quinol oxidation andis therefore a coupling site. This topic, which was discussed in Section 4.7.1, is reviewed in ref. 47. The role of the cytoplasmic NADH-linked nitrite reductase, which is encoded by the iB operon (nirBDC), is to remove nitrite (which is toxic) produced from nitrate, and to regenerate NAD*. As mentioned earlier, E. coli also makes a periplasmic nitrite reductase involved in anaer- obic respiration, cytochrome cs.,, encoded by the nrfA operon (nrfABCDEEG). It is also called respiratory nitrite reductase and is a coupling site, but the mechanism is not under- stood. The enzyme is induced by nitrite and repressed by nitrate. Fora review of nitrate and nitrite respiration, see ref. 48, 18.3.3 The Nar system and gene regulation, an overview For the following discussion, refer to Fig. 18.7. The responses to both nitrate and nitrite are regulated by dual two-component signaling systems that consist of two membrane-bound sensor kinases (NarX and NarQ) and two cytoplasmic response regulator proteins (NarL. and NarP), rather than one sensor kinase and one response regulator protein.” The systems are NarX/NarL and NarQ/NarP. (Note 53 explains how the nar genes were discovered.) The biochemical evidence for this is summa- rized later. The result of this regulation is that in the presence of excess nitrate there is pre- ferential synthesis of enzymes that use nitrate as an electron acceptor (ie, nitrate reductase), and in the presence of excess nitrite there is pre ferential synthesis of enzymes that use nitrite as an electron acceptor (i.e., respiratory nitrite reductase, cytochrome Cs). Both nitrate and nitrite induce the forma- tion of a cytoplasmic nitrite reductase, and this enzyme probably serves to prevent the accumu lation of toxic nitrite in the cytoplasm. Nitrate stimulates transcription of the respiratoryie 482, ‘THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES E c 3 some ofthe reputed genes : sitet reductase (nar) xg [fost evgee Ge) q F ) | oitate excretion (narK) F A oa ytoplasmic nitrite reductase (nirB) ‘ Nex, ate ea, fumarate reductase (r4ABCD) Noy > (e) | DMSOPTMAO reetse(dnsABC) nur : apr © peracid fafa) 4 Nar-® ' : Nap : periplasmic nitrate reductase (nap) ve [Penis rite elves (uA) ‘i ) 1 cytoplasmic nitsite reductase (nirB) : fomate dehydrogenase (6x6) B Na : NaQ 18 20 E BtKS i ATP (NOs Nak ' f J i ADP 105 Nar-® 4 i NaQ-® Nap fy Fig, 18.7 Model for niteate/nirite control of anaerobic gene expression. The membrane-bound histidine kinase sensor proteins are NarQ and NarX. The cytoplasmic response regulator proteins are NarL and NarP. Nitrate stimulates the phosphorylation of NarL and NarP via NarX (A) and NarQ (B). Nitrite stimulates che le phosphorylation of Narl. and NatP via NarQ, but causes the dephosphorylation of NarL-P by stimulating the phosphatase activity of NarX. The result is that nitrate increases the amounts of NarL-P and nitrite l lowers the amounts of NarL-P. (C} Some of the genes regulated by the response regulator proteins. In the presence of nitrate, the higher levels of Narl-P induce the narG operon, which encodes respiratory nitrate reductase, and repress the nrfA operon, which encodes respiratory nitrite reductase. The n7fA operon is induced, however, in the presence of nitrite. Thisis because (1) nitrite stimulates the phosphorylation of Nar? via NarQ, and NarP-P stimulates transcription of the nfA operon, and (2) the lower concentrations of NarL- a P caused by nitrite actually induce the nrfA operon. The second effect has been explained by a second site for z the nrfA operon (an activation site) that has a high affinity for NarL-P, which stimulates transcription when bound to this site. See: Darwin, A. JK. L. Tyson, J. W. Busby, and V. Stewart. 1997. Differential regulation by the homologous response regulators Narl. and NarP of Escherichia coli K12 depends on the DNA binding site arrangement, Mol. Microbiol. 25:583-595. Kk nitrate reductase gene (and inhibits transcrip- amounts of the response regulator Nar?-P. tion of the respiratory nitrite reductase gene) because nitrate causes increased amounts of the H response regulator NarL-P, and NarL-P stimu i lates the expression of narG, the operon that f encodes nitrate reductase, and inhibits expres- if sion of nrfA, the operon that encodes respir- } atory nitrite reductase. On the other hand, nitrite stimulates expression of the respiratory t nitrite reductase operon and lowers the ex- i pression of the nitrate reductase operon. This is because nitrite lowers the amounts of the Ke response regulator NarL-P (hence depressing transcription of narG) and increases the The higher concentrations of NarP-P stimulate expression of the respiratory nitrite reductase operon (nrfA). The lower concentrations of NarL-P also stimulate expression of n7fA, and this has been explained by an activation site for the nrfA operon that has a high affinity for Narl-P.° 18.3.4 The roles of NarX and NarQ Evidence supporting this section’s descriptions of NarX and NarQ do is provided in Sec- tion 18.3.6.MICROBIAL DEVELOPMENT AND PHYSIOLOGICAL ADAPTATION 483 NarX In the presence of nitrate, NarX phosphory- lates NarL. However, very importantly, in the presence of nitrite, NarX dephosphorylates NarL-P (Fig. 18.7A, reaction 1). This is why there is less NarL-P in the presence of excess nitrite, As a consequence, the nitrate reductase operon (narG) is stimulated by nitrate but not by nitrite, and the periplasmic nitrite reductase operon (7trfA) is repressed by nitrate but not by nitrite. The levels of NarP-P, on the other hand, are the same for both nitrate and nitrite, and therefore the differential effects of nitrate and nitrite on gene transcription depend upon NarL-P and not NarP-P. There are several other examples of a signaling molecule lower- ing the levels of a phosphorylated response regulator protein by stimulating its dephos- phorylation. Another example of a signal that does this is Py, which represses the glnALG NarQ In contrast to the differential effect of nitrate and nitrite on the sensor kinase, NarX activity with respect to the response regulator Narl. (nitrate stimulates the kinase activity, whereas nitrite stimulates the phosphatase activity), the response of NarQ to either nitrate or nitrite is the same: thatis, phosphorylation of NarL and NarP (Fig. 18.7B). Therefore, it is the NarX/ NarL couple thats responsible for the differen- tial response to nitrate and nitrite. 18.3.5 The regulatory role of IHF We know that THF (integration host factor) activates as well as represses genes. See the discussion of the nucleoid in Section 1.2.6 for a description of IHF and its role in DNA bending and the regulation of gene activity, as well operon by stimulating the dephosphorylation as Sections 10.2.5, 18.4.2, and 18.7.2 for more of NR,-P by NR; (Fig. 18.8) discussion of IHF. The transcription of the A f re a “fg 8} b+ NR, NR, ATP NR, P, XR Nk +P ADP + RP HO | s on m une [sania] pom Kor unre Wet Grant] ac — PUMP HO 4 Gi Fig. 18.8 Model for transcriptional regulation of ghiALG. The protein products of the operon are glutamine synthetase (GS) a histidine protein kinase/phosphatase, nitrogen regulator Il, (NR,), and a response regula- tor (NR, also called NtrC). (A) Under ammonia-limiting conditions, NRy phosphorylates NR,, forming NR,-P. The increased levels of NRr-P stimulate a large increase in transcription rate from glnAp2 (labeled ‘Ap2), which uses sigma 54 RNA polymerase. As the levels of NR,-P increase, other Ntr operons are also stimulated. Under conditions of excess ammonia, Py levels rise, and NR, is stimulated by Py to decrease Phosphorylation of NR, (not shown), and to dephosphorylate NR,-P. Asa consequence, transcription from glnAp2 stops and sigma 70 RNA polymerase transcribes from promoters gin Ap! and glnLp (labeled Ap? Lp) ata low rate. Transcription from these promoters is further regulated by repression by NR,, Not shown isthe effect of a-ketoglutarate, which is to inhibit the activity of P, on NR, resulting in more NR--P. (B) The levels of Py are increased by glutamine (signaling high ammonia). The bifunctional enzyme uridylyltrans- ferase/uridydyl removal (UT/UR) enzyme, which can both add and remove UMP from Py, is inhibited by glutamine in adding UMP to Py, and is stimulated by glutamine in removing UMP from Pi-UMP. The net result is that mote Pyis generated. For a description model, see: Ninfa, A. J., and P. Jiang. 2005. Pj signal transduction proteins: sensors of a-ketoglutarate that regulate nitrogen metabolism. Curr. Opin. Microbiol. 8:168-173,484 “THE PHYSIOLOGY AND BIOCHEMISTRY OF PROKARYOTES nitrate reductase gene (narG) and the nitrite extrusion protein gene (narK), which catalyzes the electrogenic excretion of nitrite, also is stimulated by integration host factor. This indicates that IHF brings a loop of DNA to which NARL-P is attached close to the pro- moter of these genes. (See Fig. 10.27 in Chapter 10.) 18.3.6 Some biochemical and genetic evidence for roles of NarX, NarQ, NarL, and NarP ‘There exist biochemical data in support of the model that NarX and NarQ are kinases that autophosphorylate and transfer the phos- phoryl group to the response regulator proteins. Ithas been demonstrated with purified proteins that NarX and NarQ can autophosphorylate, using ATP as the phosphoryl donor, and trans- fer the phosphoryl group to NarL.** NarX and NarQ can also negatively regulate the NarL protein by dephosphorylating Narl-P, although NarX is more active than NarQ in this respect” NarX and NarQ have two transmembrane regions near the N-terminal end and a region in between, exposed to the periplasm, Mutations in the periplasmic region of NatX or NarQ resultin mutants thateither have lost the ability to respond to nitrate and nitrite or behave as if nitcate or nitrive were present even in its absence. The mutants were analyzed by means of narG-lacZ, and frdA-lacZ, fusions. (Note 25 gave a description of gene fusions.) From these data ithas been concluded that nitrate and nitrite are detected in the periplasmic region and that the signal is transduced across the cell membrane to the response regulator proteins Both NarL and NarP have been demonstrated to bind to specific sites in target operons.® 18.4 Response to Nitrogen Supply: The Nir Regulon Bacteria use inorganic and organic nitrogenous compounds as a source of nitrogen for growth. ‘The inorganic nitrogen is in the form of ammo- nia, nitrate, or dinitrogen gas, and the organic nitrogen is in the form of amino acids, urea, and other nitrogenous compounds. Ultimately, all these nitrogenous compounds are either reduced or catabolized by the bacteria to ammonia, The enteric bacteria have two enzy. matic reactions for the assimilation of ammo. nia; the glutamine synthetase reaction and the glutamate dehydrogenase reaction. Glutamine synthetase 1-Glutamate + ATP +NH, glutamine + ADP +P, Glutamate dehydrogenase o-Ketoglutarate + NADPH +H*+NH, > L-glutamate + NADP* +H,0 A third reaction which will become important for the discussion that follows, is the con version of o-ketoglutarate to L-glutamate by the enzyme glutamate synthase. : Glutamate synthase a-Ketoglutarate + glutamine + NADPH + H* 2 1-glutamate + NADP* The three foregoing reactions, which were described in Section 9.3.1, are crucial because glutamate supplies nitrogen to approximately 85% of the nitrogen-containing molecules in the cell, and glutamine supplies the remaining 15%. When the supply of ammonia is restricted, glutamine synthetase is the main reaction for the assimilation of ammonia because the enzyme has a low K,, for ammonia. On the other hand, glutamate dehydrogenase has @ high K, for ammonia (about 1 mM) and oper- ates only under conditions of high external ammonia concentrations. When the ammonia supply is adequate, then ammonia is the pre- ferred source of nitrogen and it represses genes required for the assimilation of other nitro genous compounds, However, when ammonia in the growth medium becomes limiting, genes required for ammonia transport and ammonia production from external nitrogen sources are induced. Furthermore, under limiting ammo- nia conditions, glutamine synthetase becomes ‘more active and the transcription of the gene for glutamine synthetase is also stimulated, resulting in more glutamine synthetase t0 scavenge the small amounts of ammonia for the synthesis of glutamine and glutamate.