Biosynthesis of Arginine and Polyamines
Chapter
25
NICOLAS GLANSDORFF
The biosynthesis of arginine has been a focus of interest for several laboratories during the last 40 years. At the time the operon concept arose (138), it was already clear that the arginine pathway offered a paradigm with which to study the molecular basis of control of gene expression at a higher level of complexity. Indeed, the arginine genes are not clustered on the chromosome in a single functional unit (14, 167) (Fig. 1). Furthermore, the occurrence of a precursor common to the biosynthesis of arginine and the pyrimidines—carbamoylphosphate—posed very early a question of basic physiological interest: what were the metabolic controls coordinating the production of this substance? Another important metabolic interconnection was disclosed with the discovery that arginine and ornithine are precursors in the biosynthesis of polyamines.
Investigations on arginine biosynthesis brought to light basic features of metabolic regulation. Interpreting a series of experiments suggesting that the synthesis of N-acetylornithinase was antagonized by arginine, Vogel (327) proposed the use of the term "repression" for a "relative decrease resulting from the exposure of cells to a given substance, in the rate of synthesis of a particular apoenzyme," a phenomenon for which a few examples had already been reported at the time (327). Experiments on the kinetics of ornithine carbamoyltransferase formation in batch cultures and in chemostats (109), in bradytrophic mutants, and in cells grown in arginine-free rich medium (228) established that not only added arginine but also endogenously produced amino acid sets the pace of enzyme synthesis in the pathway.
The first indications concerning the mechanism involved in this regulation came with the isolation of derepressed (argR) regulatory mutants in Escherichia coli (106, 191). The dominance of the argR + allele (194, 196) suggested that the argR product was an aporepressor, in keeping with the conclusion reached earlier for lacI (235). Considering the deviation from complete coordinate expression displayed by the scattered arginine genes (see Table 2), Maas (191) and Gorini et al. (106) proposed that a unique aporepressor could interact with a family of slightly dissimilar operators. For such physiological entities consisting of scattered genes controlled by the same regulatory molecule, Maas and Clark (194) coined the word "regulon."
Table 2Repression response in the arginine regulon of E. coli K-12 |
The research of the next 30 years has borne out this proposition. When methods bypassing the problems posed by the scattering of the genes in several functional units and the absence of suitable attachment sites to construct transducing phages were elaborated, the isolation of specific operator and promoter mutants became possible, and the way was paved toward the isolation of individual genes. These developments provided the tools for the demonstration in vivo and in vitro that regulation operates essentially at the level of gene transcription (see Level of Control and Nature of the Corepressor, below). The subsequent use of the newly developed cloning and sequencing methods led to the characterization of the arginine operator and of the argR gene product (see below). By a curious twist the results bearing out this "classical" type of mechanism came as a surprise to many, since there was no evidence for regulation by attenuation, in contrast with the conclusions reached in the meantime for many other amino acid pathways (352).
Genetic and molecular studies of the arginine regulon have yielded several findings of general interest: (i) the discovery of informational suppression by streptomycin (108); (ii) one of the two first cases of divergent operons expressed from two promoters facing each other over an internal control region (88, 141); (iii) the occurrence of gene amplification mechanisms reactivating silent genes by the formation of tandem or inverted repeats (19, 58, 62); (iv) the first unambiguous demonstration that a transposon (IS3) carries an outward promoter (59); (v) the organization of the control region of the carbamoylphosphate synthase operon (carAB) in tandem promoters that are respectively controlled by arginine and the pyrimidines (34, 253); (vi) the cryptic, inducible acetylornithine transaminase of E. coli, which might be related to a degradative pathway operating in other bacteria (15) (see below); (vii) the occurrence in E. coli K-12 of a duplicate ornithine carbamoyltransferase gene, argF, which appears to have been acquired laterally (103, 322); and (viii) the homology relating argF and argI to the structural gene (pyrB) for the catalytic subunit of aspartate carbamoyltransferase, one of the most elegant cases in support of the theory of enzyme recruitment (133, 324). The most significant advances of the last few years, to be discussed extensively below, concern the arginine repressor itself, its structure and mode of action in both E. coli and Salmonella typhimurium (official name, Salmonella enterica serovar Typhimurium) (see also reference 193a), the sequence analysis of all arg structural genes in E. coli and the resulting evolutionary inferences (see the sections describing specific enzymes, below), and the dual regulation of the carAB operon. One of the most exciting and recently discovered features of the system is the joint involvement of regulatory molecules such as the argR (or xerA) and the carP (or xerB) gene products in a totally unrelated cellular process: the site-specific recombination that resolves ColE1 dimers into monomers (see Arginine Repressor and Operator-Repressor Interactions, below, and The carAB Operon, a Dual-Control System, below). Reviews on arginine metabolism in procaryotes in general (71, 76) and in Bacillus subtilis in particular (17) have been published.
This section provides an overall picture of the pathways, their interconnections, the regulatory circuits involved, and the resulting interferences between them.
Arginine biosynthesis proceeds in eight steps, according to the scheme depicted in Fig. 1a, starting from glutamate. The first four intermediates are acetylated (330). Except for N-acetylornithine, they penetrate the membrane only weakly (mutants more permeable to N-acetylglutamate than the wild type have been repeatedly described). This N-acetylation prevents the cyclization which, in the proline pathway, leads from glutamate-γ-semialdehyde to Δ 1-pyrroline carboxylate. In members of the family Enterobacteriaceae and in Sulfolobus solfataricus (319), N-acetylornithine is deacylated by an acetylornithinase, whereas in other bacteria and in fungi, the acetyl group is recycled on glutamate by an ornithine-glutamate acetyltransferase. In some organisms, the same enzyme is endowed with acetylglutamate synthase and ornithine acetyltransferase activity (277, 279). Two strategies of feedback inhibition, by which arginine (or ornithine [277]) regulates the flow of metabolic intermediates through the pathway, correspond to these alternative fates of the acetyl group. In organisms with an acetyltransferase, N-acetylglutamokinase or the acetyltransferase itself is inhibited (83, 277, 314, 319); in some of these organisms, N-acetylglutamate synthase is inhibited as well (114). In contrast, in members of the Enterobacteriaceae, N-acetylglutamate synthase alone appears to be the target enzyme (336) (see below). From ornithine onward, the intermediates of the pathway appear to be common to all organisms investigated; ornithine and citrulline are taken up easily, whereas argininosuccinate is taken up poorly.
The synthesis of the eight enzymes is repressed by arginine to various extents in E. coli (106, 191, 336) (see below) and in S. typhimurium (3, 95, 160). A single regulatory gene (argR) accounts for this repression in both organisms (3, 95, 106, 160, 191).
Carbamoylphosphate is a precursor common to arginine and the pyrimidines. In both E. coli and S. typhimurium, it is produced by a single synthase, carbamoylphosphate synthase, with glutamine as the physiological amino group donor (1, 245, 248). This situation contrasts with the existence of separate enzymes specific for arginine and pyrimidine biosynthesis in Bacillus subtilis and fungi (17, 71, 76, 80, 81). The enzyme is inhibited by UMP; this inhibition is antagonized by ornithine and IMP (1, 10, 242). Since arginine controls the formation of ornithine through feedback inhibition of N-acetylglutamate synthase, the antagonistic effects of UMP and ornithine ensure a balanced distribution of carbamoylphosphate between the tributory pathways. In both organisms, carbamoylphosphate synthase is cumulatively repressed by arginine and the pyrimidines (1, 245, 248). The arginine repressor is involved in this control in both E. coli and S. typhimurium (4, 160, 246).
Pyrimidines interfere with the regulation of the arginine pathway at the level of carbamoylphosphate utilization. The growth of E. coli pyr mutants blocked after the aspartate carbamoyltransferase step in a chemostat limited by uracil results in a marked derepression of the arg pathway (100); this effect is not seen in pyrB mutants (lacking aspartate carbamoyltransferase). A similar effect has been observed in both E. coli and S. typhimurium: pyrimidine bradytrophic mutants are derepressed for both the pyrimidine and the arginine pathway (160; A. Piérard and N. Glansdorff, unpublished data). Conversely, adding uracil to wild-type cells growing in minimal medium results in partial derepression of arginine biosynthetic enzymes, presumably because a slightly excessive inhibition of carbamoylphosphate synthase by UMP diminishes the supply of carbamoylphosphate for arginine biosynthesis (107). Arginine represses the synthesis of aspartate carbamoyltransferase about twofold, even in an argR nonsense mutant. This is probably due to diversion of excess carbamoylphosphate toward the pyrimidine pathway when the pool of ornithine is low (244).
An interesting type of indirect suppression may result from an interaction between the arginine and proline pathways (137, 169): proAB auxotrophs of E. coli and S. typhimurium blocked in the conversion of glutamate into glutamate-γ-semialdehyde can revert to the Pro+ phenotype by the presence of mutations inactivating the argD gene, encoding N-acetylornithine-δ-transaminase. argD mutants display a leaky arginineless phenotype, probably because another transaminase takes over the missing function. When such cells are grown without extraneous arginine, the feedback inhibition of N-acetylglutamate synthase is lifted and the whole pathway is derepressed. Under those circumstances, enough N-acetylglutamate-γ-semialdehyde is produced and deacylated by the relatively nonspecific N-acetylornithinase to feed the proline pathway. In the presence of arginine, this indirect suppression is abolished and the cells return to the Pro– phenotype. Several authors (see N-Acetylglutamate Synthase, below) have taken advantage of this effect of arginine on argD proAB mutants to isolate derepressed mutants or strains in which N-acetylglutamate synthase is resistant to arginine.
Polyamine biosynthesis has been particularly well studied in E. coli (Fig. 1b). The diamine putrescine can be made either directly by decarboxylation of ornithine or indirectly by decarboxylation of arginine into agmatine followed by hydrolysis of agmatine into putrescine and urea by an agmatine ureohydrolase (218, 219). Urea is not degraded by E. coli and is actually an indication of the flux through the agmatine pathway (216). Putrescine and an aminopropyl group from enzymatically decarboxylated S-adenosylmethionine give rise to the triamine spermidine. The tetramine spermine is normally not found in E. coli. Cadaverine is produced by decarboxylation of lysine. Cadaverine and decarboxylated S-adenosylmethionine give rise to N-3-aminopropyl-1,5-diaminopentane.
Each of the amino acids arginine and ornithine is a substrate for two decarboxylases (298) (see below). In each case, the so-called biosynthetic and "constitutive" decarboxylase is the only one to be found in cells grown at physiological pH in minimal medium. At low pH, at high substrate concentration, and in rich media, distinct biodegradative decarboxylases are induced. These enzymes probably constitute a defense mechanism against acidity, since a mutant lacking arginine decarboxylase is not able to grow at low pH values (E. F. Becker, Jr., Fed. Proc. 26:812, 1967). Lysine decarboxylase had originally been considered to be exclusively an inducible enzyme, but evidence for another lysine decarboxylase activity, produced under normal growth conditions, has become available (339). However, it is not clear whether the activity observed is due to a lysine-specific decarboxylase or to ornithine decarboxylase (136).
In the presence of a concentration of arginine high enough to inhibit ornithine formation, putrescine is made via agmatine exclusively, whereas in unsupplemented medium, the route from ornithine is preferred (217). In minimal medium, putrescine and spermidine amount to about 30% of the arginine in the cell protein (302).
This section summarizes what we know about the enzymes involved in the arginine pathway of E. coli and S. typhimurium. In several instances, our understanding of enzymology and related observations on metabolite flow have been exploited to select mutants affected in the regulation of enzyme activity or enzyme synthesis. The rationale for these selection procedures is also presented in this section.
Table 1 gives the common and systematic names of each arginine biosynthetic enzyme and the reaction it catalyzes. The symbols and map positions of the corresponding gene(s) are reported in Fig. 1a.
Table 1Enzymes of arginine and polyamine biosynthesis |
Using resting cells, Vyas and Maas (336) showed that N -acetylglutamate synthase was inhibited by arginine. After Haas et al. (114) found that glycerol stabilized the enzyme in extracts of Pseudomonas aeruginosa, Leisinger and Haas (179) and Marvil and Leisinger (199) were able to examine the properties of the E. coli synthase. The enzyme consists of a single type with a subunit of molecular weight (MW) of 50,000; arginine and N-acetylglutamate stabilize or induce a hexameric form. Fifty percent inhibition is achieved with 0.02 mM l-arginine. N-Acetylglutamate synthase from S. typhimurium is also very sensitive to arginine (3). The analog O-(l-norvalyl-5)-isourea is as effective an inhibitor as arginine. Indospicin inhibits the enzyme to some extent.
Ennis and Gorini (89) showed that feedback inhibition efficiently controlled the flux of arginine precursors only when the level of arginine biosynthetic enzymes was kept low, a status that is normally achieved by repression in E. coli K-12 or W and is an intrinsic property of E. coli B (see The E. coli B Paradox, below). Derepressed (argR) mutants excrete arginine.
Desensitized argA mutants could be selected owing to the growth-inhibitory effect exerted by arginine on argD pro double mutants incubated in minimal medium without proline supplement (see General Outline, above). This inhibition is due to repression of the first five enzymes of the pathway and to feedback inhibition of N-acetylglutamate synthase. When applied in two steps, the selection first gives slow-growing argR mutants (137, 160), from which feedback-insensitive mutants can be obtained by further selection and screening for colonies able to excrete proline even in the presence of arginine (87). Alternatively, desensitized mutants can be detected directly in a population of mutagenized argD pro argR mutants as proline excreters (87).
The argA gene has been cloned, and its sequence has been determined (44). Curiously, the sequence shows no obvious similarity with that of the argJ gene from Neisseria gonorrhoeae (198) and Bacillus stearothermophilus (277), which encodes a protein with both acetylglutamate synthase and ornithine acetyltransferase activity.
N-Acetylglutamate and ATP are the substrates of N-acetylglutamokinase (332; see references 102 and 332 for reviews). From gel electrophoresis of the proteins synthesized by UV-irradiated cells infected with λdargECBH transducing phages, the molecular mass of the argB product was estimated to be around 29 kDa (184), which is in keeping with the value predicted from the argB nucleotide sequence (236). The argB gene of B. stearothermophilus is homologous, and the active enzyme was shown to be a dimer (277, 280). The corresponding sequences from Saccharomyces cerevisiae (31, 236), Schizosaccharomyces pombe (321), and Corynebacterium glutamicum (V. Sakanyan, personal communication) are also homologous.
The labile intermediate, N-acetylglutamylphosphate, has not been isolated in the free state, and it is not known whether the kinase and the subsequent N-acetylglutamylphosphate reductase, which are encoded by adjacent genes in both E. coli and S. typhimurium, form some kind of complex in vivo. In this respect, it is interesting that the two equivalent proteins in yeasts and Neurospora crassa are produced from a single unit of genetic expression (31, 139, 209, 337).
The enzyme N-acetylglutamylphosphate reductase catalyzes the reduction of N-acetylglutamylphosphate to the corresponding semialdehyde. It has been partially purified (333; see references 102 and 333 for reviews). From the analysis of UV-irradiated cells infected with λdargECBH transducing phages (184), the molecular mass of the argC gene product appears to range between 37 and 47 kDa. The value predicted from the sequence of the cloned gene is 35,911 Da (236). The cognate sequences from Saccharomyces cerevisiae (31, 236), Schizosaccharomyces pombe (321), B. stearothermophilus (277, 280), C. glutamicum (V. Sakanyan, personal communication), and Thermus aquaticus (M. Baetens, unpublished results from this laboratory) are clearly homologous.
N-Acetylornithine-δ-transaminase, the product of the argD gene, catalyzes the formation of N-acetylornithine and α-ketoglutarate from N-acetylglutamate semialdehyde and glutamate (331). The same enzyme also exhibits an ornithine-δ-transaminase activity, which probably plays no physiological role (25).
The transaminase has been purified and crystallized (26), with the objective of comparing it with the purified product of the argM gene. E. coli W (15) and E. coli K-12 (T. Eckhardt, Ph.D. thesis, Eidgenössische Technische Hoschschule, Zurich, Switzerland, 1975) indeed contain a cryptic function, ascribed to argM and unlinked to argD, which can be expressed by selecting for suppressors of argD mutants. The active argM enzyme is inducible by arginine; curiously, comparison of argR + and argR derivatives of argD argM + strains suggests that induction is mediated by the argR product itself (15). The respective MWs of the argM and argD transaminases of E. coli W are 61,000 and 119,000. These proteins do not cross-react immunologically, although both enzymes are composed of 31,000-Da subunits and their tryptic patterns are almost identical (26). The argM protein has ornithine-δ-transaminase activity as well (25). The data suggest a common origin for the two genes, but only the sequence of argD is known (122). It is not clear whether wild-type strains contain an inactive inducible transaminase or an inactive argM gene. The inducibility of argM suggests that this gene is a cryptic element of an arginine degradative pathway that is silent in the E. coli strains analyzed so far. In keeping with this idea, the argM enzyme is able to transaminate succinylornithine, an intermediate in a newly discovered catabolic pathway responsible for the breakdown of arginine in a number of bacteria including Klebsiella pneumoniae (formerly K. aerogenes MK53), another member of the Enterobacteriaceae (320). Clearly, our understanding of the mechanisms involved in the evolution of metabolic pathways in prokaryotes may gain much from further investigation of both transaminases and their genetic control.
E. coli argD and the yeast cognate biosynthetic gene are homologous (122). Interestingly, they are also homologous to the genes encoding ornithine aminotransferase in yeasts and animals (122). The origin of this homology can be understood in terms of enzyme recruitment (145, 353).
N-Acetylornithinase is characteristic of the so-called linear pathway for arginine biosynthesis. A superficially similar reaction has been found in organisms recycling the N-acetyl group, such as Saccharomyces cerevisiae and Thermus aquaticus; it could be ascribed to a carboxypeptidase (84). N-Acetylornithinase hydrolyzes N-acetylornithine into ornithine and acetate (334). It is dependent on Co2+ and a thiol compound (preferably glutathione) for maximal activity. The enzyme has been purified. It appears to be a dimer of 43-kDa subunits; the MW calculated from the gene nucleotide sequence is 42,320 (37, 204). It appears homologous to a lysine biosynthetic enzyme, succinyldiaminopimelate desuccinylase (the dapE product), which is again suggestive of enzyme recruitment (37). It also appears homologous to carboxypeptidase G2 from Pseudomonas spp. and to an aminoacylase from B. stearothermophilus (see the following section).
N-Acetylornithinase readily deacylates substrates other than N-acetylornithine, including N-acetylglutamate semialdehyde, N-acetylarginine, N-acetyl- and N-formylmethionine, and N-acetylhistidine (16, 334). The action of the enzyme on N-acetylglutamate semialdehyde explains the suppression of proAB mutants by argD mutations (see General Outline, above). Several interesting applications derive from the somewhat relaxed specificity of this enzyme. By using mutants resistant to N-acetylnorvaline, Kelker and Maas (157) obtained argR + revertants from derepressed mutants and argE auxotrophs from the wild type. Baumberg (16) obtained regulatory mutants with mutations in the arginine pathway by selecting for derivatives of his auxotrophs able to utilize N-acetylhistidine in the presence of ornithine and arginine. Sakanyan et al. (278) cloned the gene for an aminoacylase of B. stearothermophilus by complementing E. coli argE mutants. This strategy could be used to characterize other enzymes displaying aminoacylase activity.
With the collateral carbamoylphosphate synthase, this enzyme is the best known of the regulon. The reaction is usually observed in the forward direction, by far the more favored one, but arsenolytic cleavage of citrulline can be used to measure the reaction in the reverse direction (176).
A peculiarity of E. coli K-12 is that is contains two ornithine carbamoyltransferase genes (103), argF and argI, whose products interact to form a family of four trimeric isoenzymes (175). The argI gene or its equivalent is the only one to be found in other E. coli strains (140) or other members of the Enterobacteriaceae (177), S. typhimurium in particular (293). The kinetic parameters of the E. coli argF and argI isoenzymes are very similar, but the argF protein is much more thermolabile (177).
The kinetic properties of the argI protein in E. coli W (176) and S. typhimurium (6) have been investigated extensively. The reaction displays a preferred sequence of substrate binding: carbamoylphosphate is the first to bind and Pi is the last to be released, as in the reaction catalyzed by aspartate carbamoyltransferase (256). In contrast to the E. coli isoenzymes, S. typhimurium ornithine carbamoyltransferase is moderately inhibited (58%) by arginine at relatively high concentrations (5 mM). This inhibition is unlikely to be of physiological importance, in contrast to the situation prevailing in Agrobacterium tumefaciens (326). Knight and Jones (165) have shown that the ornithine carbamoyltransferase of E. coli W displays a kinetically complex activation by orotate which may be of physiological value in coordinating the flow of metabolites through the arginine and pyrimidine pathways.
An extremely strong and specific inhibition of the reaction is exerted by the bisubstrate analog phosphonoacetylornithine (PALO) (239). E. coli is impervious to PALO but not to the oligopeptide Gly-Gly-PALO (238, 240), which penetrates via the oligopeptide permease, the product of the opp gene. PALO is then liberated by an intracellular peptidase and inhibits ornithine carbamoyltransferase. opp and, to a certain extent, argR mutants are resistant to Gly-Gly-PALO. These investigations may have therapeutic implications, since they provide an example of "illicit" uptake, by which a substance unable to penetrate the cell membrane does so when it is covalently associated with a pervasive carrier (238).
The relatively inefficient reverse reaction (i.e., phosphorolysis of citrulline) can be demonstrated in vivo; strains defective in carbamoylphosphate synthase grow very slowly on citrulline as a source of carbamoylphosphate for the pyrimidine pathway (289). This property was used to select for mutants with high ornithine carbamoyltransferase specific activity, such as argG bradytrophs (in the presence of citrulline, they accumulate this amino acid and at the same time exhibit derepression of ornithine carbamoyltransferase) and constitutive argF or argI mutants specifically derepressed by operator mutations or chromosomal rearrangements (146, 147, 178). The same strategy was applied to Saccharomyces cerevisiae to obtain cis-dominant constitutive ornithine carbamoyltransferase mutants (207).
The occurrence of hybrid argF-argI isoenzymes suggested that the constituent polypeptides were the homologous products of an ancestral duplication (175). Heteroduplex analysis of λ argF and λ argI transducing phages (163) and comparison based on parts of the cognate amino acid (99) and nucleotide (212, 252) sequences supported this idea. The full extent of the similarity can be appreciated now that the nucleotide sequences of argF (324) and argI (20) have been determined. The two genes display 86% amino acids and 71.8% nucleotides in common, not unlike the trpA genes of E. coli and S. typhimurium (85.1 and 72.5%, respectively). It seems that argF would have been inherited laterally from a related species, as suggested by the presence of two IS1 elements flanking that gene in E. coli K-12 (135, 354) and by codon usage analysis (322).
The comparison between ornithine carbamoyltransferase and the catalytic subunit of aspartate carbamoyltransferase is of considerable interest from the evolutionary point of view. The overall similarity is 35 to 40% in terms of amino acids and shows up in similarly structured domains, mainly in the polar moiety (the carbamoylphosphate binding domain) and helical regions joining the polar to the equatorial (or carboxy-terminal) domain (aspartate binding) of aspartate carbamoyltransferase (133, 324). The two proteins therefore appear to have a common origin. An ancestral, possibly ambiguous carbamoyltransferase gene (145, 353) could have been duplicated in near-tandem copies, which would have diverged in the course of evolution but have remained associated on the chromosome in certain organisms as suggested by the strong linkage between argI and pyrB in E. coli and S. typhimurium.
Very elegantly, Houghton et al. (134) have shown that fusing the ornithine carbamoyltransferase polar domain to the carboxy-terminal one of aspartate carbamoyltransferase produces a hybrid protein able to carbamoylate aspartate but not ornithine.
The sequences of a large number of ornithine carbamoyltransferase genes and the three-dimensional structure of P. aeruginosa ornithine carbamoyltransferase have been determined (V. Villeret, Ph.D. thesis, University of Liège, Liège, Belgium). All appear homologous whether they are involved in anabolism or catabolism (see reference 121 and references therein; unpublished data from our laboratory for Thermotoga maritima and Pyrococcus furiosus). Besides, comparison of their sequences with those of the few aspartate carbamoyltransferases analyzed up to now (343; unpublished data from our laboratory for Thermus aquaticus and Thermotoga maritima) reinforces the notion of a common, ancient origin for both carbamoyltransferases, as already suggested above.
Argininosuccinate synthase, encoded by the argG gene, catalyzes the conversion of citrulline, aspartate, and ATP into argininosuccinate, AMP, and PPi. The enzyme has been purified from mammals and from S. cerevisiae (127, 263). The yeast enzyme is a tetramer of identical MW 49,000 subunits, and the E. coli enzyme appears to consist of a basic polypeptide with a similar MW (48,000) as estimated from denaturing gels loaded with extracts of minicells producing a plasmid-encoded argG protein (222). This finding is in keeping with the E. coli argG nucleotide sequence (323), which predicts a polypeptide of 49,925 Da. The sequence is homologous to those from yeasts, methanogens, and humans.
Argininosuccinase, encoded by the argH gene, hydrolyzes argininosuccinate into arginine and fumarate. Little is known of the enzyme in the family Enterobacteriaceae (see references 76 and 335 for reviews). The purified mammalian argininosuccinase is a tetramer of 50,000-Da subunits (263). The E. coli enzyme appears similar, since extracts of UV-irradiated cells infected with a λ transducing phage carrying the argH gene display a 55,000-Da polypeptide when examined by electrophoresis under denaturing conditions on polyacrylamide gels (184). The sequence (27) bears high similarity to that of the yeast cognate gene and predicts a molecular mass of about 50 kDa.
In both E. coli and S. typhimurium, a single carbamoylphosphate synthase which utilizes glutamine as the natural amino group donor provides carbamoylphosphate for arginine and pyrimidine biosynthesis (1, 245, 248). For various reasons, this is presently the most interesting enzyme of the system; it therefore deserves extensive discussion.
The synthesis of carbamoylphosphate proceeds in four steps: (i) formation of enzyme-bound carboxyphosphate; (ii) reaction between this complex and glutamine; (iii) transfer of the amido group of glutamine to activated CO2 and formation of enzyme-bound carbamate; and (iv) phosphorylation of carbamate in the presence of a second molecule of ATP and liberation of the carbamoylphosphate formed (9).
Both in E. coli and in S. typhimurium, carbamoylphosphate synthase consists of two subunits, a heavy one (MW 130,000) able to catalyze the synthesis of carbamoylphosphate from HCO3 –, ATP, and NH3 (but not from glutamine) and a small one (MW 42,000) which carries the glutamine-binding site and displays a glutaminase activity in vitro (2, 309). The ammonium ion is a low-affinity nitrogen donor for the enzyme (150); the reaction with ammonium ion may become growth supporting in mutants deficient in glutaminase activity when the ammonium ion concentration in the medium is high (98, 206) or when the large subunit is produced in vast excess (206, 274). The small and large subunits are encoded by the adjacent carA and carB genes (2, 206). In both organisms, carA and carB (collectively named pyrA in older articles on Salmonella species) form an operon oriented from carA to carB; some nonsense carB mutants display curious antipolar effects as yet unexplained (69, 98). The complete sequences of the two genes in E. coli (229, 253) and that of carA in S. typhimurium (164) are known.
From glutamine bound to the small subunit, the amido group is transferred, probably as nascent NH3, to the site of the large subunit, which also accepts the ammonium ion as a nitrogen donor. Carbamoylphosphate synthase is thus basically similar to other amidotransferases involved in the synthesis of amino acids, purine, pyrimidines, or cofactors (120, 243, 261). It is possible that the evolution of this family of enzymes proceeded by the combination of a primordial glutaminase with various synthetases originally utilizing NH3 as the nitrogen donor (120, 243, 261).
There is a high degree of similarity (39% identical residues) between the amino- and carboxy-terminal moieties of the carB protein, and it has been suggested that carB results from the duplication of a smaller ancestral gene (229). As the enzyme is allosteric (see below), Nyunoya and Lusty (229) have suggested that the two halves of the carB molecule fold as separate domains capable of conformational interactions, each domain carrying one of the two ATP-binding sites recognized on carbamoylphosphate synthase (30, 259). Biochemical studies and the analysis of several mutants confirmed this view (112, 113; also see references 208 and 264 and references therein); moreover, they indicate (i) that the amino- and carboxy-terminal halves of the carB subunit are involved in the formation of carboxyphosphate and in the phosphorylation of carbamate, respectively (257), and (ii) that the two catalytic ATP sites interact with each other (112). A single mutation in a putative ATP- binding site of the carbamate phosphorylation domain may uncouple functional interactions between the two catalytic domains of the large subunit and also between the carbamate phosphorylation domain and the glutaminase (112, 190). Residues critical for catalytic action were identified by site-directed mutagenesis in both the carA and carB proteins (112, 190, 208, 210, 257).
Carbamoylphosphate synthase is a highly regulated enzyme (see below and reference 267). It is subject to feedback inhibition by UMP (1, 245). This inhibition is antagonized by IMP (10), an effect with probable significance for the coordination of purine and pyrimidine pathways, and, most importantly, it is antagonized by ornithine (242). Since arginine curtails ornithine formation by feedback inhibition of N-acetylglutamate synthase, the UMP-ornithine antagonism amounts to a double feedback by arginine and UMP. In the presence of ornithine, UMP exerts a partial inhibition; in the absence of ornithine, the inhibition is almost total. The sites of the known effectors of the enzyme are all on the large subunit (41, 309). All of these allosteric ligands act primarily on the binding of MgATP, and all appear to interact with the 20-kDa carboxy-terminal domain of carB (57, 275). In addition to its general toxic effects, cyanate, a well-known carbamoylating agent, appears to inhibit the synthase specifically (111).
Carbamoylphosphate synthase is thus controlled by several allosteric effectors belonging to different pathways. Anderson and Marvin (7, 8) have shown that the effectors affect the distribution of the enzyme among at least three conformational states, one of them stabilized by the positive effectors and favorable to substrate binding and another stabilized by the inhibitor UMP. Mutational alterations of carbamoylphosphate synthase are thus likely to result in a variety of phenotypes. In certain uracil-sensitive E. coli mutants such as strain P678M1 (G. Leclercq, Ph.D. thesis, Université Libre de Bruxelles, Brussels, Belgium, 1971), the enzyme displays a much higher apparent Km for ATP than does the wild-type enzyme, so that in the presence of UMP and even in the presence of ornithine, the activity is not high enough to support growth. The apparent Km for ATP is particularly high in the absence of ornithine. This observation provides an explanation for the partial sensitivity toward arginine also displayed by the mutant. A kinetic analysis of the effect of UMP, IMP, and ornithine on the partial reactions catalyzed by the E. coli synthase has been published (41).
supJ-hisT double mutants of S. typhimurium are also uracil sensitive (32). It is not yet known whether the primary effect of the mutation is structural or regulatory; an explanation in terms of attenuation control (32) now appears unlikely (see The carAB Operon, a Dual-Control System, below).
Other mutations at the car locus exhibit either a pseudo- arginineless phenotype or an arginine-sensitive one (4, 98, 206). From investigations with Salmonella typhimurium, Abdelal et al. (4) concluded that the arginine sensitivity of some car mutants was due to repression by arginine of ornithine carbamoyltransferase synthesis; indeed, in such mutants (but not in the wild type), an interaction between the synthase and the transferase would appear to be critical for proper assembly and function of the carA and carB gene products. These observations would of course be less surprising if the formation of multienzyme complexes between the two proteins were a normal feature of carbamoylphosphate metabolism (see Physiological and Evolutionary Considerations, below). According to the same authors (5), arginineless car mutants display this phenotype because arginine curtails the formation of N-acetylornithine, which appears to antagonize the maturation of the mutant synthase. More recently (119), the same group reported two remarkable observations that may be related to these findings: (i) a cold-sensitive car mutant of S. typhimurium affects an amino acid residue that appears to be critical for proper folding of the enzyme (indeed, the growth temperature affects the properties of the mutant synthase); and (ii) to a lesser extent, the characteristics of the wild-type enzyme also appear influenced by the growth temperature.
Gigot (97) observed that from a strain carrying a deletion in carB, it is possible to select derivatives growing in the presence of cyanate. Such mutants display an ammonia-dependent carbamoylphosphate synthase activity; surprisingly, the synthesis of the corresponding enzyme is repressible by uracil. The responsible gene has not yet been characterized.
Most of the structural genes of the arginine pathway have been identified and localized in E. coli and S. typhimurium by conventional selection and mapping techniques. argM, coding for the cryptic inducible acetylornithine transaminase, has not been localized with precision: it is carried by episome F'14 (266). Forward selection methods have also been devised for mutants blocked in steps A, B, C (68), and E (157). As in fungi (13, 220), car mutants could be recovered from a dihydroorotaseless (pyrC) strain by selecting for derivatives unable to accumulate the toxic intermediate carbamoylaspartate (D. Charlier, Master thesis, Vrije Universiteit Brussel, Brussels, Belgium, 1974). The toxicity of carbamoylaspartate is not understood; it has recently been observed in S. typhimurium as well (312).
With the exception of argF, which is a peculiarity of E. coli K-12, both E. coli and S. typhimurium display very similar arrangements of arg loci (Fig. 1a). The only genes to be clustered are argECBH. argECBH in E. coli was shown to be a divergent operon consisting of two arms, argE and argCBH (232, 258), with an internal operator region flanked by two convergent promoters (249, 250).
E. coli mutants affected in the regulatory gene argR were originally obtained by selecting for canavanine-resistant derivatives (106, 191). argR mutants of E. coli B have also been obtained by selecting for resistance to homoarginine (241), which does not affect the K-12 strain. Canavanine does not inhibit growth of S. typhimurium. argR mutants of both S. typhimurium and E. coli could also be isolated by indirect suppression of the proline auxotrophy induced by arginine in argD mutants (see N-Acetylglutamate Synthase, above). argR mutants of E. coli were also recovered as strains resistant to a mixture of 2-thiouracil and arginine (246), whose toxic effect (21) is now understood (see The carAB Operon, a Dual-Control System, below). This combination is useful to select for argR derivatives of arg auxotrophs; the use of auxotrophs also circumvents selection for permease mutants. Another useful method is the selection from histidine or methionine auxotrophs of derivatives able to use the cognate N-acetyl amino acid in the presence of ornithine or arginine (see N-Acetylornithinase, above). A last method of general interest for the selection of derepressed mutants involves selecting for derivatives of leaky mutants (a carA mutant was used [246]) that exhibit no further growth lag when transferred from minimal medium supplemented with arginine (and, in this case, uracil as well) to minimal medium. The method has been applied with success to isolate cis-dominant constitutive mutants of arg genes in Saccharomyces cerevisiae (139).
Because several arg genes are unlinked to each other, isolation of cis-acting regulatory mutants with mutations in the arginine regulon required ad hoc strategies.
Selecting for suppressors of polar argC or argB mutations gave rise to tandem duplications connecting argH or argB plus argH to other promoters or creating new ones at their novel joints (19, 58). The same approach provided a deletion (sup-102), which disclosed the existence of an internal operator region between argE and argC and of two flanking promoters facing each other, argEp and argCBHp (88).
Point mutations of the Oc (operator-constitutive) type were obtained by various means, including relief from repression of streptomycin-induced suppression of argI or argC nonsense mutations (141, 143). In the second case, the mutants were partially constitutive for both argE and argCBH, even though the mutations were mapped between argE and argC. The existence of a divergent operon was inferred from this finding, in keeping with the properties of the sup-102 deletion (see Structure of Control Regions, below). Other argECBH operator-constitutive mutants were obtained by localized mutagenesis (132), selecting for derivatives of his auxotrophs able to achieve high enough levels of N-acetylornithinase to deacetylate N-acetylhistidine even when the arginine regulon was repressed (38, 43, 67). This method proved particularly rewarding since it also provided mutations increasing the translational rate of argE (39). The same method gave apparently specific argE operator-constitutive mutants; these were later shown to harbor secondary promoters relatively insensitive to repression (73, 249) (see Structure of Control Regions, below).
cis-dominant mutants enhancing argE expression were also obtained by selecting for derivatives of the sup-102 deletion mutant able to utilize acetylornithine as the source of arginine. Some of these strains were of the operator-constitutive type and also affected argH; other were promoter-up mutations and involved the insertion of a mobile promoter such as IS3 (59, 249).
argF and argI operator mutants or unstable rearrangements causing argF constitutivity were isolated from a car deletion mutant as derivatives able to use citrulline as a source of carbamoylphosphate for pyrimidine biosynthesis (146, 178, 324) (see Ornithine Carbamoyltansferase, above). Some of these rearrangements are gene amplifications (146).
Last but not least, the construction of operon and protein fusion strains in which the lacZ gene was expressed from a foreign promoter (47) provided a general approach to isolate regulatory mutants that has been applied to the argECBH (22) and carAB operons (271).
This variety of structural and regulatory mutants provided the biological material that allowed subsequent cloning and sequence analysis of all the genes of the regulon (at least in E. coli) by a variety of approaches, which involved extensive applied lambdology (101, 162, 177, 200, 260), before the systematic use of plasmid cloning vectors could be generalized. The use of well- defined λdarg transducing phages was critical in establishing the reality of transcriptional repression, both in vivo and in vitro (see below).
Control of mRNA translation rather than modulation of mRNA synthesis had been advocated very early as a mechanism for the regulation of arg genes (201, 329, 335). When transducing phages carrying the argECBH genes became available, RNA-DNA hybridization allowed a direct test of these hypotheses. Rogers et al. (269) and Cunin and Glansdorff (75) showed that the amount of pulse-labeled argECBH hybridizable RNA varied considerably over the range of conditions investigated. Other studies confirmed the notion of transcriptional control in vivo (see reference 223 for argA, argF, and argI and reference 247 for carAB) and in vitro with argA, argECBH, argF, argI, and carAB DNA (77, 86, 247, 268, 286, 287). The kinetics of enzyme repression originally interpreted by Lavallé and De Hauwer (173, 174; see also reference 36) in terms of regulated translational arrest appear, rather, to reflect the progressive decay of enzyme accumulated before the onset of repression (247).
More precise investigations with low-background DNA probes for the two arms of the argECBH divergent operon established that the amplitude of variation for hybridizable RNA levels, even if considerable, remains three- to fourfold lower than for enzyme specific activities (72); Zidwick et al. were able to reproduce the phenomenon in vitro (355). Such a discrepancy, however, was not observed for the carAB operon: there, a close correspondence was obtained between enzyme and RNA levels (247).
One interpretation of the argECBH mRNA-enzyme discrepancy would have been attenuation control: in repression, the cells would contain a relatively higher proportion of leader mRNA not contributing to enzyme synthesis. However, Bény et al. (23), using several probes, showed that transcription of argCBH in a purified in vitro system was not restricted to a proximal leader sequence and that in vivo, no preferential transcription of operator-proximal sequences occurred under conditions of repression. Sequence data now available for all arg genes (61) provide no evidence for attenuation control.
Another explanation, advocated by Vogel and coworkers (202, 335), assumed the discrepancy to be due to a greater stability of the mRNA in derepression than in repression; indeed, restricting the arginine supply of an arginine auxotroph increased the chemical half-life of argECBH hybridizable RNA by a factor of 3 to 4. However, further estimates of the chemical half-life of argECBH mRNA (168) and analysis of the functional decay of argF, argI mRNA do not support the notion of a specific effect of arginine on the stability of the cognate mRNA (116).
At present, there is no completely satisfactory explanation for the argECBH mRNA-enzyme discrepancy, but the data remain compatible with the notion that in repressed cells, transcripts of a distal portion of the cluster, well beyond the control region, would be less abundant than those of more proximal segments, perhaps owing to rho-promoted premature transcription termination (355). In our opinion, however, the mRNA-enzyme discrepancy may result at least in part from another cause. In argC, there is a weak secondary argE promoter (argEp2) active under conditions of repression (see Structure of Control Regions, below); formation of translationally inactive RNA-RNA duplexes by mRNA molecules respectively initiated at argCp and at argEp2 could account for the relative excess of unproductive RNA present in repressed cells.
Synthesis of acetylornithinase and argininosuccinase is enhanced by ppGpp, the chemical messenger of the stringent response (356). These investigations corrected an earlier report by Yang et al. (351) concluding that synthesis of acetylornithinase was inhibited by ppGpp and are in keeping with the positive effect of ppGpp observed on the synthesis of enzymes encoded by argA and argI (94, 156, 311). By performing in vitro experiments, Zidwick et al. (356) showed that this effect did not involve argR and found no evidence that it was exerted at the level of transcription. Further experiments from the same laboratory (M. G. Williams, Ph.D. thesis, University of Minnesota, Minneapolis, 1985) indicated that ppGpp acts at least in part by reducing the frequency of translational errors, in keeping with earlier observations (94). In contrast, ppGpp was shown to inhibit carAB expression partially (34); starvation for an amino acid, therefore, would not only turn off the synthesis of RNA but also affect the synthesis of RNA precursors.
It is now clear that arginine and not arginyl-tRNA is the corepressor of the arginine regulon. Hirschfield et al. (129) showed very early that mutants of arginyl-tRNA synthetase impaired in their charging ability and therefore accumulating arginine displayed reduced levels of ornithine carbamoyltransferase and remained repressible; no correlation was found between repression and the level of charging. Moreover, Leisinger and Vogel (180) and Celis and Maas (55) showed that the charging profiles of the five isoacceptor tRNAArgs appeared to be the same in repressed and derepressed cells. Subsequent reports on derepressed synthetase mutants or on derepression resulting from inhibition of the synthetase by arginine precursors (42) could not be confirmed or were shown to involve artifacts (see Arginyl-tRNA Synthetase, below).
Direct confirmation that arginine is able to play the role of corepressor came from in vitro experiments. In a purified system for transcription of the argECBH genes (233), addition of arginine and partially purified repressor free of tRNAArg synthetase provoked repression of hybridizable RNA synthesis (77). In similar experiments, Lissens et al. (185) found that free arginine and repressor repressed transcription of carAB in vitro to the same extent as in vivo. G. Bény (Ph.D. thesis, Vrije Universiteit Brussel, Brussels, Belgium, 1984) was unable to observe any effect of tRNAArg or of the synthase itself on repression of argECBH transcription in vitro. More recent work (61, 63, 181, 188) (see below) established that purified repressor specifically binds to arg and car operators in the presence of arginine and that in some genes at least, activated repressor prevents RNA polymerase binding by steric hindrance (61, 63).
Early experiments (105) suggesting that ornithine was able to counteract the repressive effect of arginine in a chemostat operated under conditions of partial derepression could not be reproduced. In chemostats in which the dilution rate was controlled with great accuracy, no effect of ornithine or citrulline could be observed (68; unpublished results from this laboratory). Furthermore, argG bradytrophs accumulating citrulline remain derepressed (178). Zidwick et al. (355) showed that l-ornithine, l-citrulline, and d-arginine had no effect on in vitro synthesis of argECBH mRNA and of two of the cognate enzymes.
Structure of Control Regions.
A feature common to all genes of the E. coli regulon, including argR and the carAB operon, is the presence of two partially conserved 18-bp sequences that display hyphenated symmetry and overlap the promoters to various extents (Fig. 2 and 3). These "ARG boxes" (249, 251) were first shown to constitute operator sites by examination of the nucleotide sequence changes in operator mutations in argECBH (73, 249), argF (324), and carAB (271). A consensus sequence could be derived from their comparisons; homologous ARG boxes are also present in front of the carAB and argR genes from S. typhimurium (188). All tandem ARG boxes are separated by 3 bp, except for argR, in which there is only 2 bp. In E. coli argG, a third ARG box is present 101 bp upstream from the pair of boxes found in the promoter region (61).
Except for E. coli argR and for the carAB operon, each expression unit appears to be transcribed mainly from one promoter region, sometimes from a close cluster of starting points. The argA promoter has not yet been identified experimentally, however. The control region of carAB contains a pair of tandem promoters: P2 is regulated by arginine and overlaps a pair of ARG boxes; P1, located 67 bp upstream, is regulated by pyrimidines (see The carAB Operon, a Dual-Control System, below). The argR gene is autoregulated; in E. coli, it is transcribed from two promoters: a constitutive one (P2), which accounts for about a third of the total repressor that can be produced, and P1, downstream, which overlaps a pair of ARG boxes (181). It is conceivable that the constitutive production of repressor is necessary to reach the maximal degree of repression observed in E. coli K-12 (308). In Salmonella species, only P1 is present (188).
The control region of the divergent argECBH operon deserves a special discussion. The first example of divergently arranged genes transcribed from closely spaced promoters was the CI-cro cluster of bacteriophage λ (306), but the bioABCFD and argECBH operons provided the first instances of control regions consisting of an internal operator flanked by two facing promoters. The properties of the deletion Δ sup102 clearly indicated this arrangement: the deletion destroyed part of argB and the whole of argC, abolished argE expression without affecting the gene itself, and made the expression of argH partly constitutive (88). This formal genetic evidence, joined to the discovery of point mutations derepressing the whole cluster and located between argE and argC (141), was fully confirmed at the sequence level (see below). This somewhat unexpected arrangement of controlling elements had been considered at an early stage as an unlikely possibility for the bio gene cluster (110); it was nevertheless proved to apply to that case as well (231). Other examples of divergent operons have been described since then; most of them, however, display back-to-back or overlapping promoters (18).
The biological significance of this face-to-face arrangement is not clear, although some speculations have been considered (18). Obviously, as the internal operator overlaps the two facing promoters to similar extents, we would expect coordinate expression of the two wings of the cluster. This is indeed the case for argE and argCBH, but the situation became clear only after a fine dissection of the regions adjacent to the control region was performed. The respective repression coefficients of argE and argCBH are at first sight markedly different: 9 and 23 in terms of mRNA, and 17 and 60 in terms of specific activities (72). This paradox was explained by the discovery of a weak secondary argE promoter (argEp2) located in argC 180 nucleotides upstream from the major transcription initiation site for argE. argE transcription initiated from argEp2 can be detected by S1 mapping in repressed cells (73; J. Piette, Ph.D. thesis, Vrije Universiteit Brussel, Brussels, Belgium, 1983) and accounts for about half the specific activity of enzyme E under these conditions (A. Boyen, unpublished data from this laboratory). When this background is considered, no difference remains between the individual repression responses of argE and argCBH (see Table 2), although they kept being recorded as uncoordinate on the basis of the early evidence (18). Expression from argEp2 appears relatively insensitive to repression, as confirmed by the partially constitutive acetylornithinase synthesis observed in "argEp2-up" mutants (73; Piette, Ph.D. thesis). These observations are in keeping with previous data on trp-lac fusions (265; see also The carAB Operon, a Dual-Control System, below) which suggest that steric hindrance between RNA polymerase and repressor is necessary to achieve efficient repression. Furthermore, results of experiments by Sens et al. (287) on the order of addition of S30 extracts from argR + and argR cells to in vitro-coupled translation-transcription systems suggest that steric hindrance is also sufficient to explain repression.
Do converging promoters interfere with each other? Initiation at either of the primary promoters overlapping the operator region probably occurs in an alternate fashion. Interferences between convergent promoters appear when the distance between them increases; derepression of argCp decreases transcription from argEp2 even when the latter is strengthened by up mutations (73; Piette, Ph.D. thesis). Likewise, derepression of argCp antagonizes expression from the outward promoter of IS3 inserted 190 nucleotides in front of the argCBH transcription start point (59). It is not clear whether these interferences result from collisions between polymerases, from interactions between complementary mRNA molecules initiated at argCp and argEp2, or from the unavailability of argEp2 when being traversed by frequent transcriptional waves initiated at argCp.
The sequence analysis of several mutations allowed a fine-structure dissection of the genetic determinants intertwined in the argECBH control region (73, 249) (Fig. 3). One of the most interesting results of this study concerns mutations that simultaneously depress the activity of the argCBH promoter and enhance the efficiency of argE translation (38, 39). The mutations are confined to the limits of argCp but also modify the sequence of the ribosome-binding sites for argE translation, between the Shine-Dalgarno box and the 5' end of argE mRNA. The sequence data strongly suggest that maximal translation efficiency depends directly on the composition of proximal RNA sequences not necessarily included in secondary structures. They support the notion of a proximal mRNA "consensus" sequence containing element interacting with ribosomal components (284).
argCp presents a paradox: the –35 region contains the consensus sequence TTGACA only 15 nucleotides away from the –10 region, yet it is not a weak promoter. It is possible that the sequence TTGTTG, 18 nucleotides from the Pribnow box, plays a role in polymerase-promoter interactions.
Arginine Repressor and Repressor-Operator Interactions.
After some early attempts that led to only partial purification (77, 158, 315, 317) but were nevertheless useful in the first in vitro analysis of arginine-mediated repression, the argR product was purified to homogeneity from E. coli K-12, E. coli B, and S. typhimurium from appropriate hyperproducing strains (181, 188, 308). A particularly gratifying feature of this molecule is that it can be precipitated by adding arginine at an early stage of the purification. In all three organisms, the arginine repressor was found to be a hexamer of identical subunits with molecular masses of about 17,000 Da. The functional significance of the single amino acid difference between E. coli K-12 and E. coli B repressors is discussed in the next section. There is 95% identity between the E. coli and S. typhimurium argR products (188). The hexameric form is the only one to have been observed in both organisms under a variety of conditions. Scrutinizing the sequence of the genes for potential secondary structures did not disclose any of the motifs recognized in various DNA-binding proteins. Crystals of the (E. coli K-12) arginine repressor have been obtained (C. Van Duyne and P. Zigler, unpublished experiments), and the analysis of the three-dimensional structure of the protein is in progress (193a). Mutational studies (45a, 193a, 308a) suggest the existence of an N-terminal DNA-binding site and of a C-terminal arginine-binding and oligomerization site.
The number of molecules of repressor per cell would appear to be around 500 to 600 in E. coli K-12 or E. coli B; autoregulation brings this number to 300 to 400 in the presence of excess arginine (308). The relative inefficiency of this repression appears to be due to the constitutive argR promoter P2 (see the previous section). This rather large number probably explains why "escape" synthesis of arg and car genes has been observed only when they were present in rather high copy number.
The basic features of the repressor-operator interactions have now been established. In the next paragraph of this section, I focus on E. coli K-12, for which data are available concerning all the control regions of the regulon (except argI), including several mutant derivatives, and S. typhimurium, in which interactions with the carAB operator have been investigated.
In vitro binding experiments analyzed by DNase I and hydroxyl radical footprinting on the operator sites of argA, argECBH, argD, argF, argG, argR and carAB in E. coli indicate that in all cases the repressor binds symmetrically to four consecutive helical turns (corresponding to pairs of boxes separated by 3 nucleotides, or by 2 nucleotides in argR) to one face of the DNA only (61, 63, 181, 188). This pattern was further confirmed by phosphate ethylation interference experiments on the E. coli argF operator (307) and more recently on the artificial fully symmetrical consensus operator sequence (Fig. 2) derived from the compilation of all E. coli ARG boxes and their inverse (Fig. 4) (338).
The close proximity to particular bases in the major and minor grooves of the E. coli ARG boxes was detected by purine methylation protection and premethylation interference experiments (61, 307). Positions resulting in strong interference signals are distributed symmetrically on both sides of the short spacer region separating two boxes and also within each box. Methylation protection and interference experiments on different operator sites highlight several guanines in the major groove (for numbers, see Fig. 2 and 4)—bp 4-15', 15-4', and 14-5'—whose functional importance is further emphasized by the single-substitution mutations with an Oc phenotype that have been obtained in vivo in argECBH, argF, and carAB (249, 271, 324). These positions also belong to the most highly conserved base pairs in the ARG box sequences. Several symmetrically located effects of A residues in the minor groove have also been noted (61, 188). Regarding S. typhimurium, the interpretation of depurination footprinting has emphasized possible interactions between the repressor and certain AT base pairs in the minor groove (188); however, no data obtained with mutants are as yet available. In the model presented by Lu et al. (188), the S. typhimurium arginine repressor is thought to interact with two AT-rich regions corresponding to two minor grooves facing the repressor molecule. As a consequence, repressor contacts would not be to the most highly conserved base pairs, whose role would be only to mark the boundaries of the AT-rich regions.
However, a recent study of E. coli K-12 arginine repressor binding to the consensus operator sequence by protection, premodification interference, and missing-contact probing techniques to purines and pyrimidines suggests contacts with specific bases in four consecutive major grooves and the two outermost minor grooves on one face of the helix (338) (Fig. 4). A more complete analysis of the S. typhimurium arginine repressor-operator interactions is therefore required before it can be concluded that the mode of binding of the Salmonella protein significantly differs from the one proposed for E. coli. At any rate, in merodiploids that combine totally nonfunctional argR mutations of S. typhimurium with an episome harboring the argR + allele of E. coli K-12, the K-12 repressor appears as effective as its Salmonella counterpart, an observation that implies conservation of regulatory elements in the two species (95). In similar diploids constructed with other argR alleles of Salmonella typhimurium, only partial repression was observed, perhaps because of the formation of partially defective hybrid repressor molecules (161).
Interestingly, the arg repressor of B. subtilis, also a hexameric molecule, was shown to repress E. coli genes in vivo and bind to arg operator DNA in vitro (79).
The stoichiometry of repressor-operator interactions has been estimated in E. coli by quantitative DNaseI footprinting (61) and by gel retardation with labeled repressor (307). The results of both methods indicate that one hexamer binds to a pair of boxes; each hyphenated ARG box would therefore contact two repressor subunits. Gel retardation experiments suggest that the stoichiometry is the same in Salmonella species (188). This mode of interaction would leave one site free: as a matter of fact, DNase I footprinting experiments performed with the argG control region, which contains a single ARG box 101 bp upstream from the standard tandem pair, suggest that this third site may become occupied if a loop is formed (61).
The study of interactions of the E. coli repressor with a variety of ad hoc constructions showed the importance of correct spacing between the two boxes (61). Reducing the 3-bp spacer to 2 bp already creates a partially constitutive phenotype (249). Increasing the spacer length showed that equivalent positions in tandem pairs of boxes must be properly aligned on the same face of the helix for efficient recognition to occur. The repressor binds to a single box but with a much higher Kd than to a pair (between 5- and 160-fold higher), the lower Kd being obtained with the box showing the best fit to the consensus (61, 307).
The configuration of the repressor-DNA complex is still a matter of conjecture; models have been discussed (61), taking into account that the repressor induces pronounced bending of argF operator DNA when binding to it (307). Until now, no other form of the arg repressor has been observed than the hexameric one, but it is not known whether in vivo, a hexameric form may not actually assemble on the DNA. This could explain the somewhat puzzling fact that an almost perfect linear correlation with a slope around 0.5 appears between apparent in vitro binding constants (Kd, all between 10–10 and 10–9 M) and the in vivo repression coefficients (Rc) when plotting the logarithm of Kd versus the logarithm of Rc (61). This cooperative pattern, already predicted by Berg (24) on the basis of a statistical sequence analysis, could arise if two repressor molecules were to bind cooperatively to adjacent boxes (which the stoichiometry experiments do not suggest) or if parts of the repressor were to assemble on operator DNA. On the other hand, the correlation observed could be a coincidence or, alternatively, could reflect the assembly of pairs of hexamers respectively bound at the ARG boxes present in the promoter region and at other sites, as yet undisclosed.
Not surprisingly, large differences in repression coefficient (Rc measured in vivo; Table 2) are not paralleled by comparable differences in Kd s (measured in vitro [61]). The extent of promoter-operator overlap could strongly influence the actual extent of repressibility (Fig. 3; Table 2). There is a rough correlation between Rc and extent of overlap (argF, –35 region in the overlap, Rc of 200) and less extensive ones (argCBH, argE, and carAB, –35 region out of the overlap, Rc from 50 to 60; argD, –35 region adjacent to the region covered by the repressor, Rc of 16). An actual overlap appears necessary for efficient repression; not only do the binding of RNA polymerase and the binding of repressor appear to be mutually exclusive in the E. coli carAB control region (63), but also it has been repeatedly observed that bound arginine repressor does not greatly hinder transcription initiated upstream of it, whether from the pyrimidine-specific promoter of carAB (63, 253) (see below), the secondary argEp2 promoter (73), or the outward promoter of an IS3 element (59). The differences observed between Kds and Rcs might also be accounted for, at least in part, by the relative positions of RNA polymerase and the repressor around the double helix.
As would be expected from a system controlled solely by transcriptional repression, physiological derepression of argR + cells achieved by limiting the arginine supply in various ways (68, 191, 228) leads to maximal enzyme levels very similar to those found in stringent argR mutants. This correlation contrasts with the situation encountered in tryptophan biosynthesis, in which starvation for tryptophan results, via the attenuation control mechanisms, in much higher levels than in trpR strains.
Since the intracellular concentration of repressor is quite high (10–6 M; i.e., around 500 molecules of hexamer per cell), the affinity for arginine is rather low (Kd of about 10–4 M), and the Kd of active repressor for operator DNA is between 10–9 and 10–10 M, the degreee of repression will mainly depend on the concentration of arginine, and even full repression will be achieved by a relatively small fraction of the total population of repressor molecules (308). The weak repressibility of argR expression would reflect the need to maintain a sufficient number of repressor molecules to ensure efficient repression (see also reference 159). On the other hand, as the arginine enzymes are still 80 to 90% repressed in cells growing in minimal medium without arginine, a moderate increase in repressor concentration may be necessary to maintain a sufficient level of active repressor.
Surprisingly, the arg repressor was found to fulfill an additional, unrelated function: in what appears to be a striking case of protein recruitment, and under the denomination of the xerA protein, it is involved in the resolution of ColE1 dimers into monomers by its ability to juxtapose two distant cer sites (similar to ARG boxes) in the presence of arginine, with the cutting and rejoining reactions being carried out in the vicinity by the xerC and xerD proteins (292). This joining of distant sites is reminiscent of the pattern observed when the arg repressor binds to the argG control region (61) (see above).
The E. coli B Paradox.
The regulation of arginine biosynthesis presents a paradox: while the K-12 strain exhibits potentially high and extensively repressible levels of enzyme, the B strain displays lower levels slightly inducible by arginine. Nevertheless, fully derepressed mutants with comparable specific activities could be obtained from both strains by selection for resistance to canavanine (106). Further work (142, 143, 152) disclosed that argR K and argR B were alleles and specified repressor molecules with different properties. E. coli B is slightly derepressed at low intracellular arginine concentrations; it is repressed at intermediate concentrations but becomes slightly derepressed again beyond a certain threshold. The B repressor was therefore postulated to exist in different interconvertible forms; excess arginine would favor a form with reduced affinity for operator sites (143, 152). The genetic aspects of this unifying account were confirmed by the extensive complementation studies of Kadner and Maas (148). In each combination of alleles (K-12–K-12 or K-12–B), the allele giving the lowest enzyme level in the presence of arginine, was found to be dominant. Sequence studies showed that the B and K-12 repressors differ by one amino acid residue only: a proline residue in K-12 is substituted by a leucine residue in B. The B protein is also a hexamer and exhibits physicochemical properties which are slightly different from those of its K-12 counterpart (308). Most significantly, however, DNA footprinting experiments revealed that in the presence of arginine, the K-12 repressor has greater affinity for argF operator DNA than that of B does, whereas in the absence of arginine, the reverse obtains (308). It therefore appears that free B repressor (but not free K repressor) is able to repress the transcription of arg genes under physiological conditions. When arginine is present, the B strain becomes only weakly repressed, because (i) the affinity of the repressor bound to arginine for operator DNA is lower in B than in K-12 and (ii) autoregulation of the synthesis of B repressor reduces the intracellular concentration of the free protein.
It is not known whether such different patterns of regulation as seen in K-12 and B reflect adaptation to different natural habitats. The K-12 strain has a much greater potential for enzyme synthesis and may therefore be better armed for survival under conditions of starvation for arginine (308). This possibility could be tested. Regarding feedback inhibition of N-acetylglutamate synthase, the intrinsically low level of enzyme present in strain B ensures efficient regulation.
The carAB Operon, a Dual-Control System.
Since transcription of the carAB operon is cumulatively repressed by arginine and the pyrimidines, the cognate control region is expected to be more complex than those of other genes of the arg regulon. Indeed, both in E. coli and in S. typhimurium, transcription of carAB is initiated at two tandem promoters (34, 164, 253) which differ in their regulatory properties (164, 253) (Fig. 3).
P2, the downstream promoter, is regulated specifically by arginine. P2 overlaps a tandem of ARG boxes as all other promoters regulated by the argR product. Binding of RNA polymerase and binding of the repressor at carAp2 were shown to be mutually exclusive in E. coli (63). P1, the upstream promoter, is specifically regulated by pyrimidines through a mechanism which is still unclear but involves some recently discovered elements (discussed below). In both species, it is clear that argR is not involved in P1 control (63, 188) and, most importantly, that transcription initiated at P1 is not blocked by P2-bound repressor (164, 253). This dual regulatory pattern explains why the combination (arginine plus thiouracil, a toxic pyrimidine analog) inhibits growth (21). P1 does not present the attenuation features found in the control region of the pyrBI operon (270, 313).
In broad outline, therefore, the carAB operon would function as follows. Both promoters would be active under conditions of derepression. Excess pyrimidines would inactivate P1, and excess arginine would repress P2. In the latter circumstance, P1 would be accessible but, to express the operon, the polymerase would have to displace the arginine repressor bound at P2; two mechanisms appear possible a priori: either the formation of an active RNA polymerase-DNA complex at P1 actually destabilizes P2-bound repressor by a direct protein-protein interaction (possibly involving pyrimidine-specific regulatory proteins and/or associated factors [see below]), or the very initiation of transcription at P1 renders the polymerase relatively insensitive to a downstream bound repressor, as in the case of the trp-lac fusion studied by Reznikoff et al. (265) or the argEp2 and IS3 outward promoters mentioned in Structure of Control Regions (above) (59, 73). The fact that the P1- and P2-associated transcription start sites are only 67 bp apart, as well as the mode of interaction between P1 and the polymerase (see below), would tend to support the first possibility.
Since it is clear that transcription initiated at P1 overrides arginine-mediated repression at P2, it may be not surprising that under strong pyrimidine limitation, carAB expression is only weakly repressed by arginine (63, 244). Moreover, at least in E. coli, repression by arginine at P2 is more intense when P1 is also repressed by pyrimidines (63); this can be easily understood if initiation at P1 facilitates expression at P2, leading to an apparent increase of repression at P2 when P1 becomes inhibited. In keeping with this interpretation, the enhancement of P2 repression by pyrimidines is not observed in an E. coli mutant in which P1 does not operate (63). The data are, however, not completely concordant in E. coli and S. typhimurium; in the latter organism, some P2 derepression is observed when CTP is limiting, even in a P1-deficient mutant (189); taking into account that pyrimidine nucleotides exert no effect on arginine repressor/carAB operator binding, the authors advocate a possible interaction between the arginine repressor and a specific pyrimidine regulatory protein (188).
The bulk of data available on P1 control give a rather complex and as yet incomplete vision of how pyrimidines exert their control on carAB expression. Old observations on "escape" synthesis of carbamoylphosphate synthase in E. coli suggested the existence of a pyrimidine-specific regulatory molecule able to repress carAB (101). Studies of mutants obtained more recently revealed the existence of a car-specific regulatory gene, carP, trans-dominant alleles of which lead to constitutive P1 expression (271). It was considered, therefore, that carP might encode an activator of P1, particularly since, in vitro, binding of RNA polymerase to P1 is weak, heparin sensitive, and shifted 20 bp upstream from the position expected for the formation of an open complex (63). Actually, the carP protein binds to car DNA, at least in vitro, in the absence of pyrimidine nucleotides (58a). It is possible, therefore, that in conjunction with additional factors serving as sensors for changes in the pyrimidine nucleotide pool, the carP protein acts as a repressor when pyrimidine is in excess and as an activator in the converse situation. To add to the complexity but also to the interest of the system, it should be mentioned that in Salmonella species, mutations affecting the RNA polymerase itself may lead either to enhanced levels of carbamoylphosphate synthase or to reduced levels and hyperrepressibility by uracil and/or arginine, with the rest of the arginine and pyrimidine regulons remaining unaffected (144, 225a). In Salmonella species also, the use-1 mutation (46) which results in superrepression of P1 by pyrimidines was recently shown to affect the gene for a minor arginyl-tRNA gene, an observation which remains unexplained (187). Finally, but maybe not exhaustively, three unexpected findings were recently brought to light: (i) in both species, the integration host factor (IHF) was shown to stimulate P1 activity in minimal medium but also to increase the repressibility of this promoter by pyrimidines—the site at which the IHF binds to exert these antagonistic effects was identified 300 bp upstream of the P1 transcription start (60); (ii) addition of pyrimidines significantly enhances the amount of IHF present in the cell, a phenomenon which is probably of more general significance (60); (iii) the carP gene was found to be identical with xerB, which, in conjuction with xerA (also known as argR; see Arginine Repressor and Repressor-Operator Interactions, above), xerC, and xerD, resolves ColE1 dimers into monomers (58a).
Even if no complete picture is as yet available, these results converge in suggesting that several proteins (IHF, carP [xerB], and perhaps others) interact, directly or indirectly, with RNA polymerase to modulate transcription initiation at P1. The carP (xerB) product displays an aminopeptidase activity (291), which is, however, not instrumental in the process (58a). From the known behavior of IHF in other systems, we would expect this factor to exert its antagonistic effects by modulating the binding or the interaction with the RNA polymerase of a specific regulatory molecule(s), including the carP protein. This seems to be confirmed by the recent observation (58b) that protection in vivo against Dam methylation of the GATC site 106 bp upstream of the start point of the pyrimidine-specific promoter in the E. coli and S. typhimurium carAB operons is correlated with down regulation of promoter activity by pyrimidines and requires the intact carP gene product and binding of IHF to its target site 200 bp upstream.
The physiological signification of a dual-promoter structure is, in the case of carAB, particularly clear. At the level of gene expression, it complements the regulatory effects exerted on the enzyme itself. It also constitutes an elegant alternative to the existence of independently controlled carbamoylphosphate synthase isoenzymes as in fungi and B. subtilis (76, 81).
The two genes of the carAB operon are separated by a short intercistronic space (229, 253), and translation of carA is initiated at the atypical UUG codon (341). UUG is a weak initiating codon; carAB is, nevertheless, efficiently translated thanks to the presence of a strong ribosome-binding site (340).
The molecular description of the arginine regulon is now well advanced, but a number of basic physiological questions remain unanswered or have not been addressed at all. The biological significance, if any, of the scattering of arginine genes in several units of expression which are not strictly coordinated is one of them. In principle, this pattern offers a greater flexibility than coordinated operon-type control. However, in bacilli (17, 76), most of the arginine biosynthetic genes form a tight cluster. It is true that in the latter case, there are two carbamoylphosphate synthases, one that is arginine controlled and encoded by a gene clustered with the arg loci, and one that is pyrimidine controlled. It may well be that the pattern of gene organization encountered in E. coli and S. typhimurium is, both qualitatively and quantitatively, related to the existence of only one carbamoylphosphate synthase in those organisms. At such a branching point, the system must be rather delicately poised; we have seen above (see Introduction) that even in the wild type, interference between the two tributary pathways does occur. Optimal functioning of such a system may be more readily achieved by individual fine tuning of the genes directly concerned (carAB, argF, argI, and pyrB) than by operon control. This necessity for fine tuning of genetic regulatory interactions has been illustrated further by the recent finding that expression of the argR gene from the strong tac promoter brings about the rather paradoxical result that formation of ornithine carbamoyltransferase becomes derepressed in the absence of arginine because of increased repression of carbamoylphosphate synthase (308). This experiment is also a good example of the experimental approaches that have been made possible as a result of the cloning and sequence analysis of all the genes of the regulon.
The most interesting challenges are, however, elsewhere. It would probably be misleading to try to interpret quantitatively the functioning of the pathway by viewing it exclusively as a succession of free enzymes. It is indeed likely that to function efficiently (which also includes meeting environmental challenges in an appropriate way), the pathway is organized as a channel, even if it is a leaky or an unstable one. In extreme thermophiles, in which carbamoylphosphate is too thermolabile to resist exposure to the aqueous phase and is moreover decomposed into cyanate, a nonspecific carbamoylating agent, channeling was recently shown to occur (174a, 319a). From the physiological point of view, it would be of considerable interest to investigate the occurrence of biosynthetic metabolic channels in mesophilic prokaryotes as well, considering, furthermore, that the structural basis for the specificity that would be expected to underlie such enzyme associations is a virtually unexplored domain.
Carbamoylphosphate synthase is probably the most obvious enzyme to investigate in this respect, especially since previous reports (4, 5) (see Arginine Biosynthetic Enzymes, above) indicate that the collateral ornithine carbamoyltransferase is involved in the proper assembly and function, perhaps even folding, of the synthase. It would be interesting to investigate the matter further, looking for other examples in which metabolically related enzymes could exhibit this type of mutual assistance.
Whereas the physiological constraints outlined above may have kept the genes for carbamoylphosphate synthase and the two tributary transferases under individual control, the clustering of genes into operons would be a general way of facilitating such mutual assistance by providing immediate proximity to the nascent enzymes; it would also facilitate assembly into multienzyme complexes. Actually, when considering that the most ancient lines of cellular descent appear to be extreme thermophiles, we may find several reasons to suggest that the clustering of genes into operons—which has never been satisfactorily explained—may have been primarily an ancestral adaptation to high temperatures, providing opportunities for mutual assistance in folding, for mutual stabilization, and for formation of multienzyme complexes channeling thermolabile substrates. After more than three billion years of evolution, we may of course expect extensive departures from this primitive situation, even in contemporary thermophiles; it may nevertheless be interesting to reconsider metabolic evolution from this point of view.
Other aspects of arginine metabolism in members of the Enterobacteriaceae deserve attention from the physiological standpoint. One is the remarkable chain of events that takes place when addition of arginine curtails the synthetic flow of ornithine, which is itself a precursor in the synthesis of hydroxamate siderophores (76). Another aspect is the growth inhibition which results from the accumulation of ornithine and possibly acetylornithine under certain circumstances (68). The basis of this effect has not yet been elucidated, but it has been used for the forward selection of arginine auxotrophs blocked at early steps of the pathway (68). Polyamines have been reported to counteract accumulation of ornithine by a mechanism that is as yet unexplained (48). Arginine was found to stimulate the growth of hemA mutants; mutants in which this stimulatory effect is abolished appear identical to previously described alu mutants defective for uptake of exogenous 5-aminolevulinic acid, a precursor in tetrapyrrole synthesis (325). This phenomenon also remains unexplained.
Arginyl-tRNA synthetase shares an unusual property with glutamyl- and glutaminyl-tRNA synthetases: it does not catalyze ATP-PPi exchange in the absence of tRNAArg (203, 234; also see reference 96 and references therein). Furthermore, no arginyl-adenylate intermediate accumulates, whether tRNAArg is present or not (70). It is not clear whether ATP, arginine, and tRNAArg coreact in one step by a concerted mechanism (186) or whether binding with tRNAArg is a prerequisite to activate the enzyme and make the formation of arginyl-adenylate possible (203). If the latter were the case, the aminoacyl-AMP would react immediately with the tRNAArg. It was shown (33) that in E. coli, arginyl-tRNA synthetase catalyzes ATP-PPi exchange in the presence of the analog tRNAArgc-c-2'dA (i.e., tRNA with 2'-deoxyadenosine at the 3' end), which cannot be acetylated. A classical two-step mechanism was therefore suggested.
The enzyme itself (α protein) is a monomer with a molecular mass of 60 to 70 kDa (65, 182), in keeping with the nucleotide sequence of the cognate gene (90); it is copurified with a 40-kDa β protein, whose function remains unknown. The two molecules can best be separated by two successive affinity elutions with a tRNA gradient after absorption of the protein preparation on a column of phosphocellulose (65). This procedure provides a 1,000-fold purified, 95% homogeneous enzyme which is more active than previously reported preparations. A somewhat improved purification procedure was published subsequently (182).
Apparent Km values for arginine in the aminoacylation reaction are in the micromolar range (64, 65, 182, 183), and there is agreement on a random substrate addition mechanism (65, 183, 234). The Km for tRNAArg, also in the micromolar range, appears higher than for other activating enzymes, even arginine-specific enzymes from other organisms (182, 183). If significant, this difference suggests that the actual concentration of tRNAArg may be more critical in E. coli than in other systems. Canavanine is also esterified by the synthase. The Km of canavanine is 4.10–4 M, and the V max is twice that found with arginine. Both canavanine and homoarginine inhibit the binding of arginine competitively, but homoarginine is not esterified (203, 211). Whether the enzyme is also inhibited by arginine precursors has been the subject of some debate. The synthetase of E. coli W and K-12 was claimed to be inhibited by argininosuccinate, ornithine, and citrulline (42). However, subsequent studies on partially purified extracts of E. coli W and K-12 (including the strain used by Brenchley and Williams [42]) showed that the enzyme remained insensitive to ornithine and citrulline up to 2 mM (64). At relatively high concentrations of argininosuccinate, the apparent inhibition exerted by this compound could be ascribed to isotopic dilution of the labeled arginine used in the assay by nonradioactive arginine produced from argininosuccinate by argininosuccinase present in the extract (64). When lower concentrations of argininosuccinate (1 μM) were used, the observations made by the two groups remain conflicting, and the discrepancy remains unresolved.
Mutants affected in arginyl-tRNA synthetase have been described. Canavanine-resistant (Canr) isolates of E. coli with both altered arginyl-tRNA synthetase activities and nonrepressible arginine biosynthetic enzymes have been reported (42). However, these observations could not be confirmed; double mutants may have been involved. In contrast, other Canr synthase mutants, with increased Km values for arginine or for ATP, were found to be fully repressible by arginine (128, 129); moreover, mutants with a block drastic enough to accumulate endogenous arginine showed enhanced repression. Hence, it was assumed that arginine and not arginyl-tRNA was the corepressor of the arginine regulon, an assumption validated by the results reported in the previous section. The mutants isolated by Hirschfield et al. (128, 129) provide evidence for a single arginine-activating enzyme in E. coli, encoded by the argS gene, which was cloned and sequenced (90). One argS mutation conferring resistance to canavanine and increasing the Km for ATP by five- to sixfold was shown to substitute a serine for an arginine, which is therefore suggested to be critical both for ATP binding and for the amino acid activation step (91).
Little is known about regulation of the synthesis of arginyl-tRNA synthase, but the availability of the cloned argS gene makes it possible to address this question directly. Like isoleucyl- and phenylalanyl-tRNA synthetases, the enzyme becomes permanently derepressed during cognate amino acid starvation (213, 225). In the case of phenylalanine-tRNA synthase, this phenomenon was correlated with the existence of an attenuation mechanism dependent on the concentrations of tRNAPhe (310). A positive correlation between growth rate and the steady-state concentration of several synthetases has been observed (224). The approximate number of molecules of arginyl-tRNA synthase per genome shifts from 192 when the cells use acetate as carbon source to 510 on glucose-supplemented medium and to 867 on rich medium. These values may be compared with the 1,500 molecules of ornithine carbamoyltransferase and 2,000 molecules of carbamoylphosphate synthase that the cell makes on minimal medium supplemented with glucose. These levels of the different synthetases remain in approximate balance with those of tRNA, elongation factors, and ribosomes.
Transport of arginine into E. coli and S. typhimurium is of the periplasmic type, common in gram-negative bacteria. Osmotic shock of wild-type E. coli abolishes transport activities and liberates three different periplasmic proteins binding arginine (345). One of them binds arginine, ornithine, and lysine (the LAO protein [56, 272, 273]) and is part of the so-called high-affinity system for arginine transport (Km 10–9 M for arginine); another binds arginine and ornithine (the AO protein [51]) and belongs to a lower-affinity system (Km 10–7 M for arginine); and the third appears to be specific for arginine and displays high affinity for it (272) (see below). The LAO system is repressible by lysine, and the AO system is repressible by arginine and ornithine (56). At arginine concentrations of 10–6 M or higher, the main route of entry appears to be through the lower-affinity system (117). In addition, an arginine-repressible acetylornithine permease was reported by Vogel (328); its relationships with the components involved in arginine transport are not known. Studies on arginine uptake and accumulation have taken considerable advantage of the fact that decarboxylation of this amino acid into agmatine is inhibited by aminooxyacetic acid (344).
The AO system has been thoroughly studied by Celis and coworkers (49, 50, 51, 52, 53, 54, 316). Large quantities of the AO-binding protein are produced by strains obtained by selecting for d-arginine utilization (50). These mutants display increased uptake of arginine and ornithine (no longer repressible by these amino acids) and are oversensitive to canavanine (51). The corresponding regulatory mutation was called abpR, whereas the structural gene, identified by nonsense mutations producing truncated versions of the AO-binding protein (51), was called abpS. Both types of mutations were mapped close to argA (52). It is not known whether the abpR mutations act in trans or in cis.
Celis and Urban also characterized a kinase able to phosphorylate both the AO- and LAO-binding proteins and mapped the corresponding structural gene (argK) close to serA by using an argK-defective, canavanine-resistant mutant (54, 316). The argK protein also displays ATPase activity and was shown to be essential for normal function of the transport system. Although the enzyme works in solution in vitro, it may be associated with the membrane in vivo; consistent with this notion is the fact that several other transport-defective mutants with mutations mapping in the same region (but not yet localized with precision) appear to be affected in membrane proteins (49, 56, 192, 193, 273, 285). These mutants have also been selected as resistant to canavanine. One of the mutations—originally called argP—results in reduced transport of arginine, ornithine, and lysine; its mutant displays strongly reduced ATPase and phosphorylating activity of the argK protein, although the Km for ATP appears unmodified (54). It is not yet known whether argP and argK are the same gene.
Recently, a cluster of genes called artPIQMJ (art meaning arginine transport) has been mapped around min 16, and the sequences have been determined (346). artJ encodes a protein binding l-arginine specifically and with high affinity; it could be the arginine-binding protein first described by Rosen (272). By their sequence similarity with the genes involved in the transport of lysine, arginine, ornithine, and histidine in S. typhimurium (see below), artQ and artM might encode transmembranous proteins and artP might encode a membrane-associated ATPase (346).
The situation in S. typhimurium differs somewhat from that in E. coli. The existence of a high-affinity, arginine-specific transport system and of another one common to lysine, arginine, and ornithine (LAO) was documented long ago (262). Homoserine and transhydroxyproline proved to be good inhibitors but not substrates of the arginine-specific system; homoserine was without effect in E. coli (49). The effect of these substances and the lack of effect of ε-N-methylhomoarginine in S. typhimurium suggest that the secondary nitrogen of arginine may act as a donor in hydrogen bond formation during the recognition process (262).
The LAO-binding protein of S. typhimurium was found to interact with the membrane-bound proteins involved in the histidine transport system, as originally brought to light by the study of hisP mutants; the latter, however, retain normal high-affinity, arginine-specific transport (170). The relationship between the LAO protein and the components of the histidine transport system has been analyzed in detail in S. typhimurium but remains to be determined in E. coli. The argT gene, encoding the Salmonella LAO protein, lies immediately upstream from the hisJQMP operon, which encodes, in order, a histidine-binding protein (clearly homologous to the argT protein), two transmembranous proteins, and an ATP-binding protein (126, 166). A similar gene complex exists in E. coli (227). The Salmonella LAO protein has been purified (226) and crystallized in the liganded and unliganded forms (151), and the three-dimensional structure has been determined (230).
From this concise survey, it appears that the data obtained with E. coli and S. typhimurium are complementary but also that several points remain to be clarified: what is the equivalent of the E. coli AO system in S. typhimurium, what is the relationship between E. coli argK kinase-ATPase with the proteins affected by E. coli argP mutations, what is the S. typhimurium equivalent of argK, and what is the S. typhimurium counterpart of the art system discovered in E. coli?
The biosynthesis of putrescine, spermidine, and cadaverine in E. coli is well known inasmuch as the cognate enzymes and structural genes have been characterized (40, 167, 193, 301, 349; also see below). It is clear that polyamines are required for optimal growth, although large variations of the polyamine pool can be present without exerting a pronounced effect on growth (298, 299). It is, however, very difficult to evaluate which cellular functions are most directly affected by their presence or their absence. The reader is referred to a recent review by Davis et al. (82) for a thorough discussion of the physiological aspects of regulation of polyamine biosynthesis in bacteria, fungi, and mammals. The present section deals with the biosynthetic and regulatory aspects of polyamine biosynthesis in E. coli. Volume 94 of Methods in Enzymology contains detailed information concerning polyamine biosynthetic and biodegradative enzymes, as well as mutant screening techniques (300). Table 1 gives the gene-enzyme relationship for the steps involved in polyamine biosynthesis.
Both the biodegradative and the biosynthetic arginine decarboxylases have been well characterized. The biodegradative enzyme (93) was purified from E. coli B (29, 214) and shown to be a decamer with a molecular mass of 820 kDa. The biosynthetic decarboxylase (encoded by speA) is a tetramer (296 kDa) of a 70-kDa subunit (214, 348). The enzyme is located in the periplasm; the subunit is processed from a 74-kDa precursor (45). Because of this periplasmic location, the substrate arginine is channeled into putrescine before becoming mixed with the endogenous pool (303). The enzyme is inhibited by putrescine and spermidine, but, since the cognate Ki values are in the millimolar range and the enzyme is extracellular, these effects probably have no physiological significance. Arginine decarboxylase is inhibited by difluoromethylarginine(149) and by aminooxyacetic acid (344), a property that has been particularly useful for studies on arginine transport.
Agmatine ureohydrolase, which converts agmatine into putrescine and urea (219), has been purified from E. coli (283). It is a dimer of 38-kDa subunits. Inhibitory effects of ornithine and arginine were noted but are probably not physiologically significant. Together with arginine decarboxylase, this enzyme constitutes the so-called putrescine biosynthetic pathway II (215) (Fig. 1). Mutations in this pathway were isolated by screening for strains defective in the production of urea (215) or by looking for strains requiring putrescine for optimal growth in the presence of excess arginine, which curtails ornithine synthesis and therefore prevents formation of putrescine by ornithine decarboxylation (130, 193, 195). The first approach gave mutants only partially blocked in speA (the arginine decarboxylase gene) or speB (the ureohydrolase gene) and not requiring putrescine for optimal growth. The second approach delivered tight speA and speB mutants with a reduced growth rate in the absence of putrescine, spermidine, or spermine. speA and speB form a complex operon (294) (see below).
Pathway I leads directly from ornithine to putrescine via the biosynthetic ornithine decarboxylase (218). Like the biodegradative ornithine decarboxylase, the biosynthetic one is a dimer with a molecular mass of 160 kDa; on biochemical grounds, the two proteins appear evolutionarily related (11, 12, 214). The regulation of the biosynthetic enzyme appears at first sight surprisingly complex: it is inhibited by putrescine and spermidine, activated by GTP, and inhibited by ppGpp (299); however, in addition, polyamines increase the specific activity of inhibitory proteins called antizymes (124, 125, 171). Moreover, "antiantizymes," which bind to the antizyme and thus release the decarboxylase from the complex, have been described as well (171). In mammals, the antizyme would appear to facilitate degradation of the enzyme (82). It is not clear that all these regulatory effects are significant in E. coli: the Ki values for putrescine and spermidine are again in the millimolar range, and some studies suggest that the antizymes may not be functional in vivo (153). It should be noted that bacterial ornithine decarboxylases are quite different from their mammalian counterparts (82).
speC mutants define the gene for biosynthetic ornithine decarboxylase. They were isolated as derivatives of a speA strain requiring putrescine or spermidine in the absence of arginine (78). Their behavior suggests that putrescine can partially replace spermidine. speC appears homologous to speF (see below).
S-Adenosylmethionine decarboxylase was purified from E. coli (197, 342). It is a homohexamer with a molecular mass of 108 kDa. It requires covalently bound pyruvate for activity but not pyridoxal phosphate, and it is inhibited by decarboxylated S-adenosylmethionine. Unlike its mammalian counterpart, it is activated by Mg2+ rather than by putrescine(299). It is also feedback inhibited by spermidine but at much lower concentrations than ornithine decarboxylase is. Of all reported regulatory effects on the activity of polyamine biosynthetic enzymes in E. coli, it is this inhibition of S-adenosylmethionine decarboxylase by spermidine which would appear the most significant. It probably limits polyamine biosynthesis when the intracellular concentration of spermidine (most of which is normally bound to nucleic acids and phospholipids) becomes excessive. The concentration of putrescine is regulated largely by excretion of excess putrescine (see the discussion of transport systems, below).
The gene encoding S-adenosylmethionine decarboxylase is speD. Tight speD mutants grow at 75% of the wild-type rate and require spermidine for optimal growth (305, 350).
Pure spermidine synthase (or putrescine aminopropyltransferase), a dimer with a molecular mass of 72 kDa, was obtained from E. coli (35, 297). The enzyme also transfers the aminopropyl group to spermidine (to form spermine) or to cadaverine but much less efficiently than to putrescine; spermine is not normally found in E. coli, although small amounts were detected in speA speB speC mutants growing in the presence of spermidine (115). A putative transition state analog, S-adenosyl-1,8-diamino-3-thiooctane, inhibits the enzyme (237). The corresponding gene is speE, which forms an operon with speD (349).
Combinations of tight spe mutations have been used to create strains totally unable to synthesize putrescine and spermidine, in order to assess the physiological importance of these substances (115). A strain containing deletions affecting speA, speB, speC, and speD still grew at one-third the rate observed in the presence of spermidine. The cells were abnormal with respect to phage λ production, the mating ability of Hfr strains, and the adsorption of phage f2 (115). Putrescine was slightly less efficient than spermidine in restoring the growth rate. This partial requirement could be made absolute in a speA speB speC mutant by introducing an rpsL mutation (304), which affects the S12 ribosomal protein. Changes in ribosomal structure or conformation could therefore explain the stringent polyamine requirement. Indeed, polyamines are involved in ribosomal structure and in protein synthesis (296, 298, 299). The absolute requirement for polyamines displayed by the speA speB speC rpsL strain is also interesting because this strain should at first sight not be impaired in the synthesis of cadaverine, a putative substitute for the other diamines (but see below).
Under conditions of putrescine starvation, mutants unable to synthesize this compound produce detectable amounts of cadaverine (a product of lysine decarboxylation) and of its aminopropyl derivatives (115). Furthermore, exogenous cadaverine stimulates the growth of speB mutants depleted of polyamines by the addition of arginine (85). It was therefore conceivable (85; also see reference 339) that cadaverine would act as a substitute for other diamines. Data from Leifer (Z. Leifer, Ph.D. thesis, New York University Medical School, New York, N.Y., 1972) and Goldemberg (104) suggested that under physiological growth conditions, cadaverine would be produced by a lysine decarboxylase different from the inducible one studied by Sabo et al. (276; see also reference 28). Evidence for such an enzyme was provided by Wertheimer and Leifer (339), who showed that E. coli grown in minimal medium at neutral pH displays a lysine decarboxylase activity inhibited by putrescine and spermidine. Evidence for this substitute role of cadaverine is still inconclusive, however. Tabor et al. (295) constructed a speA speB speC speD strain also lacking inducible lysine decarboxylase activity (with a cadA genotype) and found this organism to be phenotypically identical to the cadA + parent, growing at a rate one-third of that found in the presence of polyamines. However, as noted by Wertheimer and Leifer (339), the identical growth rate of cadA and cadA + versions of the multiple spe mutants may reflect the fact that in the cadA + parent the cadaverine pool was already unusually low. The gene responsible for the putative biosynthetic lysine decarboxylase has not yet been characterized. In this respect, it should be noted that Igarashi et al.(136) made several observations concurring with the suggestion that cadaverine is actually formed by ornithine decarboxylase, which would then account for the alleged biosynthetic lysine decarboxylase. This would also be in keeping with the absolute requirement for polyamines of the abovementioned speA speB speC rpsL strain.
Of the so-called constitutive genes involved in the biosynthesis of polyamines, several have in fact been shown to be subject to metabolic control. However, observations made with different strains are conflicting, and their biological significance therefore remains unclear. The expression of speA and speC is partially repressed by putrescine (296, 302). According to Boyle’s group (282, 347), speA, speB, and speC are negatively and partially controlled by cyclic AMP via the mediation of the cyclic AMP receptor protein. From more recent data, this effect appears indirect and might involve an additional inhibitory protein (294). speD is probably repressed by spermidine (153). In contrast, working with strains different from those studied by Boyle and coworkers, Halpern and coworkers (118, 288) found that speB expression was subject to catabolite repression whereas speA and speC did not respond to different carbon sources. In addition, speB was stimulated by nitrogen limitation, which could actually override catabolite repression (288). speB can be induced by agmatine from a promoter internal to the speA speB operon (294); this accounts for previous claims that the two genes were not part of the same operon (115). Two- to threefold repression of speA by the purine repressor was reported recently (123).
Several systems are involved in the transport of polyamines in E. coli (221) (also see below). The potE protein actually excretes putrescine and is specific for this substrate (154); it functions as a putrescine/ornithine antiporter; its involvement in uptake of putrescine is probably small. Since accumulation of high internal concentrations of polyamines is detrimental to cell growth (82), the potE system exerts a physiologically important function in the adjustment of polyamine content. potE is part of an operon (mapping at min 16) which also includes speF (155), encoding an ornithine decarboxylase inducible at low pH: speF appears homologous to speC. Since most E. coli strains have been reported to lack the biodegradative ornithine decarboxylase mentioned above (11), the possible relationship between this enzyme and the speF-encoded protein remains to be determined.
A complex transport system for putrescine and spermidine (but active mainly with spermidine) is encoded by four genes—potA, potB, potC, and potD—clustered in that order into an operon and mapping at min 15 (92). The potD protein binds to putrescine and spermidine; it is the periplasmic component of the system. potA encodes a putative ATP-binding protein associated with the membrane. Still another periplasmic transport system, whose genetic components map at min 19, has been reported (254). The system is encoded by one operon clustering the genes potF, potG, potH, and potJ. The potF protein is a putrescine-specific periplasmic protein homologous to the potD protein.
Periplasmic polyamine transport is, at least in part, energy linked, but the respective parts played by ATP and the membrane potential remain to be clarified (82, 92). At any rate, sequestration of polyamines by binding to intracellular components is expected to favor global unidirectional transfer. Curiously, it would appear that streptomycin enters the E. coli cell via an inducible polyamine transport system (131). Further investigations of polyamine transport might help clarify some apparent contradictions: whereas Satishchandran and Boyle (282) found that none of their E. coli K-12 strains were able to utilize agmatine as a source of nitrogen, the reverse has been reported by Stalon and Mercenier (290) and Shaibe et al. (288) for other K-12 strains. A recent review on polyamine breakdown in microorganisms is available (172).
The network of reactions involved in arginine and polyamine biosynthesis is well established. Mechanisms controlling the activity and the synthesis of the relevant enzymes have been identified and studied in considerable detail. Recent research has revealed the molecular features of the mechanisms responsible for transcriptional control of enzyme synthesis by arginine. Interaction between a unique repressor molecule activated by arginine and a family of promoter regions containing related operators accounts for the quantitative variations observed from gene to gene in the repression response. It was surprising to identify the xerA gene (involved in plasmid recombination) as being identical to argR.
The physiological significance of the partially uncoordinated repression response remains a matter of debate, but it is clear that in terms of amplitude of expression, the gene for ornithine carbamoyltransferase (and that for aspartate carbamoyltransferase as well) stands out with respect to the other arg (and pyr) genes. When comparing different bacteria (76) (see above) two alternative patterns actually appear to emerge: either a large arg operon integrating an arginine-specific carbamoylphosphate synthase isoenzyme, or a scattered regulon with a car locus independently and subtly controlled by the two tributary pathways. Studies on possible metabolic channeling of biosynthetic intermediates and interactions between enzymes of the pathway could contribute much to our understanding of cell physiology.
Two peculiar genetic arrangements have been disclosed in the course of these studies: the converging promoters involved in the divergent transcription pattern of the argECBH cluster, and the tandem promoters of the carAB operon. The physiological significance of the latter structure is particularly clear. The pyrimidine-specific components of carAB control are not yet fully defined, but they include IHF and the carP gene, now recognized as being identical to xerB. As xerA (in its quality of argR) is also involved in arginine-specific control of carAB, a number of intriguing questions arise on the multifunctionality and the interrelationships of these functions.
Other observations of general interest made with the arginine system deserve further investigations, such as the influence exerted by mRNA sequences 5' to the Shine-Dalgarno box on the efficiency of translation (the case of mutations affecting argE translation), the role of cryptic genes in metabolic evolution (the argD-argM problem), and the formation of chromosomal rearrangements resulting in tandem or inverted repeats of the argE gene. Several examples of putative enzyme recruitment (homologous enzymes performing analogous functions) have been mentioned in this chapter. They provide excellent material for evolutionary studies on the genetic basis for substrate specificity and functional flexibility in modern enzymes.
Thanks are due to D. Charlier, J. Charlier, R. Cunin, and A. Piérard for critical reading of the manuscript, to R. Celis for communication of useful information, to P. Stalon for typing the manuscript, and to J. P. Ten Have and D. Charlier for preparing the figures. Work pursued in our laboratory was supported over the years by Belgian Research Foundations (FNRS-NFWO, FRFC-FKFO, IRSIA-IWONL) and by Concerted Actions between Brussels University and the Belgian State.
References
1. Abdelal, A., and J. L. Ingraham. 1969. Control of carbamoylphosphate synthesis in Salmonella typhimurium. J. Biol. Chem. 244:4033–4038.
2. Abdelal, A., and J. L. Ingraham. 1975. Carbamoylphosphate synthase from Salmonella typhimurium. Regulation, subunit composition and function of the subunits. J. Biol. Chem. 250:4410–4417.
3. Abdelal, A., and O. V. Nainan. 1979. Regulation of N-acetylglutamate synthesis in Salmonella typhimurium. J. Bacteriol. 137:1040–1042.
4. Abdelal, A. T. H., E. Griego, and J. L. Ingraham. 1976. Arginine-sensitive phenotype of mutations in pyrA of Salmonella typhimurium: role of ornithine carbamoyltransferase in the assembly of mutant carbamoylphosphate synthase. J. Bacteriol. 128:105–113.
5. Abdelal, A. T. H., E. Griego, and J. L. Ingraham. 1978. Arginine auxotrophic phenotype of mutation in pyrA of Salmonella typhimurium: role of N-acetylornithine in the maturation of mutant carbamoylphosphate synthase. J. Bacteriol. 134:528–536.
6. Abdelal, A. T. H., E. H. Kennedy, and O. Nainan. 1977. Ornithine transcarbamylase from Salmonella typhimurium: purification, subunit composition, kinetic analysis and immunological cross-reactivity. J. Bacteriol. 129:1387–1396.
7. Anderson, P. M. 1977. Binding of allosteric effectors to carbamoylphosphate synthase from Escherichia coli. Biochemistry 16:587–892.
8. Anderson, P. M., and S. V. Marvin. 1970. Effect of allosteric effectors and adenosine triphosphate on the aggregation and rate of inhibition by N-ethyl-maleimide on carbamoylphosphate synthase of Escherichia coli. Biochemistry 9:171–178.
9. Anderson, P. M., and A. Meister. 1966. Bicarbonate dependent cleavage of adenosine triphosphate and other reactions catalyzed by Escherichia coli carbamoylphosphate synthase. Biochemistry 5:3157–3163.
10. Anderson, P. M., and A. Meister. 1966. Control of Escherichia coli carbamoylphosphate synthase by purine and pyrimidine nucleotides. Biochemistry 5:3164–3167.
11. Appelbaum, D. M., J. C. Dunlop, and D. R. Morris. 1977. Comparison of the biosynthetic and biodegradative ornithine decarboxylases of Escherichia coli. Biochemistry 16:1580–1584.
12. Appelbaum, D. M., D. I. Sabo, E. H. Fischer, and D. R. Morris. 1975. Biodegradative ornithine decarboxylase of Escherichia coli. Purification, properties and pyridoxal 5'-phosphate binding site. Biochemistry 14:3675–36681.
13. Bach, M. L., and F. Lacroute. 1972. Direct selection techniques for the isolation of pyrimidine auxotrophs in yeast. Mol. Gen. Genet. 115:126–130.
14. Bachmann, B. 1990. Linkage map of Escherichia coli K-12, edition 8. Microbiol. Rev. 54:130–197.
15. Bacon, D. F., and H. J. Vogel. 1963. A regulatory gene simultaneously involved in repression and induction. Cold Spring Harbor Symp. Quant. Biol. 28:437–438.
16. Baumberg, S. 1970. Acetylhistidine as substrate for acetylornithinase: a new system for the selection of arginine regulation mutants in Escherichia coli. Mol. Gen. Genet. 106:162–173.
17. Baumberg, S., and U. Klingel. 1993. Biosynthesis of arginine, proline and related compounds, p. 299–306. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and Other Gram-Positive Bacteria: Biochemistry, Physiology, and Molecular Genetics. American Society for Microbiology, Washington, D.C.
18. Beck, C. F., and R. A. J. Warren. 1988. Divergent promoters, a common form of gene organization. Microbiol. Rev. 52:318–326.
19. Beeftink, F., R. Cunin, and N. Glansdorff. 1974. Arginine gene duplication in recombination proficient and deficient strains of Escherichia coli K12. Mol. Gen. Genet. 106:162–173.
20. Bencini, D. A., J. E. Houghton, T. A. Hoover, K. F. Folterman, J. R. Wild, and G. A. O’Donovan. 1983. The DNA sequence of argI from Escherichia coli K12. Nucleic Acids Res. 11:8509–8518.
21. Ben-Ishai, R., M. Lahav, and A. Zamir. 1964. Control of uracil synthesis in Escherichia coli. J. Bacteriol. 87:1436–1442.
22. Bény, G., A. Boyen, D. Charlier, W. Lissens, A. Feller, and N. Glansdorff. 1982. Promoter mapping and selection of operator mutants by using insertion of bacteriophage Mu in the argECBH divergent operon of Escherichia coli K-12. J. Bacteriol. 151:62–67.
23. Bény, G., R. Cunin, N. Glansdorff, A. Boyen, J. Charlier, and N. Kelker. 1982. Transcription of regions within the divergent argECBH operon of Escherichia coli: evidence for lack of an attenuation mechanism. J. Bacteriol. 151:58–61.
24. Berg, O. 1988. Selection of DNA binding sites by regulatory proteins. The LexA protein and the arginine repressor use different strategies for functional specificity. Nucleic Acids Res. 16:5089–5105.
25. Billheimer, J. T., H. N. Carnevale, T. Leisinger, T. Eckhardt, and E. E. Jones. 1976. Ornithine δ-transaminase activity in Escherichia coli: its identity with acetylornithine-δ-transaminase. J. Bacteriol. 127:1315–1323.
26. Billheimer, J. T., M. Y. Shen, H. N. Carnevale, H. R. Horton, and E. E. Jones. 1979. Isolation and characterization of acetylornithine-δ-transaminase of wild-type Escherichia coli W. Comparison with arginine-induced acetylornithine-δ-transaminase. Arch. Biochem. Biophys. 105:401–413.
27. Blattner, F. R., V. Burland, G. Plumkett III, H. J. Sofia, and D. L. Daniels. 1983. Analysis of the Escherichia coli genome. IV. DNA sequence of the region from 98.2 to 92.8 minutes. Nucleic Acids Res. 21:5408–5417.
28. Boeker, E. A., and E. H. Fischer. 1983. Lysine decarboxylase (Escherichia coli B). Methods Enzymol. 94:180–184.
29. Boeker, E. A., and E. E. Snell. 1972. Aminoacid decarboxylase, p. 217–253. In P. D. Boyer (ed.), The Enzymes, 3rd ed., vol. 6. Academic Press, Inc., New York.
30. Boettcher, B. R., and A. Meister. 1980. Covalent modification of the active site of carbamoylphosphate synthase by 5'-fluorosulfonylbenzoyladenosine. Direct evidence for two functionally different ATP-binding sites. J. Biol. Chem. 255:7129–7133.
31. Boonchird, C., F. Messenguy, and E. Dubois. 1991. Characterization of the yeast ARG5,6 gene: determination of its nucleotide sequence, localization of the functional domains and analysis of the control region. Mol. Gen. Genet. 226:154–166.
32. Bossi, L., T. Kohono, and J. R. Roth. 1983. Genetic characterization of the SufJ frameshift suppressor in Salmonella typhimurium. Genetics 103:31–42.
33. Bottu, G. 1983. The role of cations and of the tRNA 3'-terminal end in the activity of Escherichia coli arginyl-tRNA synthase. Arch. Int. Physiol. Biochim. 91:B6.
34. Bouvier, J., J. C. Patte, and P. Stragier. 1984. Multiple regulatory signals in the control region of the Escherichia coli carAB operon. Proc. Natl. Acad. Sci. USA 81:4139–4143.
35. Bowan, W. H., C. W. Tabor, and H. Tabor. 1973. Spermidine biosynthesis. Purification and properties of polyamine transferase from Escherichia coli. J. Biol. Chem. 248:2480–2486.
36. Boy, E., J. Thèze, and J. C. Patte. 1973. Transient regulation of enzyme synthesis in Escherichia coli. Mol. Gen. Genet. 121:77–78.
37. Boyen, A., D. Charlier, J. Charlier, V. Sakanyan, I. Mett, and N. Glansdorff. 1992. Acetylornithine deacetylase, succinyldiaminopimelate desuccinylase and carboxypeptidase G2 are evolutionarily related. Gene 116:1–6.
38. Boyen, A., D. Charlier, M. Crabeel, R. Cunin, S. Palchaudhuri, and N. Glansdorff. 1978. Studies on the control region of the bipolar argECBH operon of Escherichia coli. 1. Effects of regulatory mutations and IS2 insertions. Mol. Gen. Genet. 161:185–196.
39. Boyen, A., J. Piette, R. Cunin, and N. Glansdorff. 1982. Enhancement of translation efficiency in Escherichia coli by mutations in a proximal domain of messenger RNA. J. Mol. Biol. 162:715–720.
40. Boyle, S. M., G. D. Markham, E. W. Hafner, J. M. Wright, H. Tabor, and C. W. Tabor. 1984. Expression of the cloned genes encoding the putrescine biosynthetic enzymes and methionine adenosyltransferase of Escherichia coli (speA, speB, speC, and metK). Gene 30:129–136.
41. Braxton, B. L., L. S. Mullins, F. M. Rauschel, and G. D. Reinhart 1992. Quantifying the allosteric properties of Escherichia coli carbamylphosphate synthase: determination of thermodynamic linked-function parameters in an ordered kinetic mechanism. Biochemistry 31: 2309–2316.
42. Brenchley, J. E., and L. S. Williams. 1975. Transfer RNA involvement in the regulation of enzyme synthesis. Annu. Rev. Microbiol. 29:251–274.
43. Bretscher, A. P., and S. Baumberg. 1976. Divergent transcription of the argECBH cluster of Escherichia coli K12. Mutations which alter the control of enzyme synthesis. J. Mol. Biol. 102:205–220.
44. Brown, K., P. W. Finch, I. D. Hickson, and P. T. Emmerson. 1987. Complete nucleotide sequence of the Escherichia coli argA gene. Nucleic Acids Res. 15:10586.
45. Buch, J. K., and S. M. Boyle. 1985. Biosynthetic arginine decarboxylase in Escherichia coli is synthesized as a precursor and located in the cell envelope. J. Bacteriol. 163:522–527.
45a. Burke, M., F. Merican, and D. J. Sherrat. 1994. Mutant Escherichia coli arginine repressor proteins that fail to bind l-arginine yet retain ability to bind their normal DNA binding sites. Mol. Microbiol. 13:609–618.
46. Bussey, L. B., and J. L. Ingraham. 1982. A regulatory gene (use) affecting the expression of pyrA and certain other pyrimidine genes. J. Bacteriol. 151:144–152.
47. Casadaban, M., and S. N. Cohen. 1979. Lactose genes fused to exogenous promoters in one step using Mu-lac bacteriophage as in vivo probe for transcriptional control sequences. Proc. Natl. Acad. Sci. USA 76:4520–4533.
48. Cataldi, A. A., and I. D. Algranati. 1989. Polyamines and regulation of ornithine biosynthesis in Escherichia coli. J. Bacteriol. 171:1998–2002.
49. Celis, R. 1977. Properties of an Escherichia coli K-12 mutant defective in the transport of arginine and ornithine. J. Bacteriol. 130:1234–1243.
50. Celis, R. 1977. Independent regulation of transport and biosynthesis in Escherichia coli K-12. J. Bacteriol. 130:1244–1252.
51. Celis, R. 1981. Chain-terminating mutants affecting a periplasmic binding protein involved in the active transport of arginine and ornithine in Escherichia coli. J. Biol. Chem. 256:773–779.
52. Celis, R. 1982. Mapping of two loci affecting the synthesis and structure of a periplasmic protein involved in arginine and ornithine transport in Escherichia coli. J. Bacteriol. 151:1314–1319.
53. Celis, R. 1984. Phosphorylation in vivo and in vitro of the arginine-ornithine periplasmic transport protein of Escherichia coli. Eur. J. Biochem. 143:403–411.
54. Celis, R. 1990. Mutant of Escherichia coli K-12 with defective phosphorylation of two periplasmic transport proteins. J. Biol. Chem. 265:1787–1793.
55. Celis, R., and W. K. Maas. 1971. Studies on the mechanism of repression of arginine biosynthesis in Escherichia coli. IV. Further studies of repression on the role of arginine transfer RNA in repression of the enzymes of arginine biosynthesis. J. Mol. Biol. 62:179–188.
56. Celis, R., H. J. Rosenfeld, and W. K. Maas. 1973. Mutants of Escherichia coli. K-12 defective in the transport of basic amino acids. J. Bacteriol. 116:619–626.
57. Cervera, J., F. Conejerolara, J. Ruizsanz, M. L. Galisteo, P. L. Mateo, C. J. Lusty, and V. Rubio. 1993. The influence of effectors and subunit interactions on Escherichia coli carbamoylphosphate synthase studied by differential scanning calorimetry. J. Biol. Chem. 268:12504–12511.
58. Charlier, D., M. Crabeel, R. Cunin, and N. Glansdorff. 1979. Tandem and inverted repeats of arginine genes in Escherichia coli K12. Mol. Gen. Genet. 174:75–88.
58a. Charlier, D., G. Hassanzadeh , A. Kholti, D. Gigot, A. Piérard, and N. Glansdorff. 1995. carP, involved in pyrimidine regulation of the Escherichia coli carbamoylphosphate synthetase operon encodes a sequence-specefic DNA-binding protein identical to XerB and PepA, also required for resolution of colE1 multimers. J. Mol. Biol. 250:392–406.
58b. Charlier, D., N. Huysveld, and N. Glansdorff. On the role of the Escherichia coli integration host factor (IHF) in repression at a distance of the pyrimidine specific promoter P1 of the carAB operon. Biochimie 76:1041–1051.
59. Charlier, D., J. Piette, and N. Glansdorff. 1982. IS3 can function as a mobile promoter in E. coli. Nucleic Acids Res. 10:5935–5948.
60. Charlier, D., M. Roovers, D. Gigot, N. Huysveld, A. Piérard, and N. Glansdorff. 1993. Integration host factor (IHF) modulates the expression of the pyrimidine-specific promoter of the carAB operons of Escherichia coli K12 and Salmonella typhimurium LT2. Mol. Gen. Genet. 237:273–286.
61. Charlier, D., M. Roovers, F. Van Vliet, A. Boyen, R. Cunin, Y. Nakamura, N. Glansdorff, and A. Piérard. 1992. Arginine regulon of Escherichia coli K-12. A study of repressor-operator interactions and of in vitro binding affinities versus repression. J. Mol. Biol. 226:367–386.
62. Charlier, D., Y. Severne, M. Zafarullah, and N. Glansdorff. 1983. Turn-on of inactive genes by promoter recruitment in Escherichia coli: inverted repeats resulting in artificial divergent operons. Genetics 105:469–488.
63. Charlier, D., G. Weyens, M. Roovers, J. Piette, C. Bocquet, A. Piérard, and N. Glansdorff. 1988. Molecular interactions in the control region of the carAB operon encoding Escherichia coli carbamoylphosphate synthase. J. Mol. Biol. 204:867–877.
64. Charlier, J., and E. Gerlo. 1976. Arginyl-tRNA synthase from Escherichia coli. Influence of arginine biosynthetic precursors on the charging of arginine acceptor tRNA with [14C]arginine. Eur. J. Biochem. 70:137–145.
65. Charlier, J., and E. Gerlo. 1979. Arginyl-tRNA synthetase from Escherichia coli K12. Purification, properties and sequence of substrate addition. Biochemistry 18:3171–3178.
66. Cleary, M. L., R. T. Garvin, and E. James. 1977. Synthesis of the Escherichia coli K12 isoenzymes of ornithine transcarbamylase performed in vitro. Mol. Gen. Genet. 157:155–165.
67. Crabeel, M., D. Charlier, A. Boyen, R. Cunin, and N. Glansdorff. 1974. Mutant selection in the control region of argECBH bipolar operon of Escherichia coli. Arch. Intern. Physiol. Biochim. 82:973–974.
68. Crabeel, M., D. Charlier, R. Cunin, A. Boyen, N. Glansdorff, and A. Piérard . 1974. Accumulation of arginine precursors in Escherichia coli: effects on growth, enzyme repression, and application to the forward selection of arginine auxotrophs. J. Bacteriol. 123:898–904.
69. Crabeel, M., D. Charlier, G. Weyens, A. Feller, A. Piérard, and N. Glansdorff. 1980. Use of gene cloning to determine polarity of an operon: genes carAB of Escherichia coli. J. Bacteriol. 143:921–925.
70. Craine, J., and A. Peterkovsky. 1975. Evidence that arginyladenylate is not an intermediate in the arginyl-tRNA synthetase reaction. Arch. Biochem. Biophys. 168: 343–350.
71. Cunin, R. 1983. Regulation of arginine biosynthesis in prokaryotes, p. 53–79. In K. Herman and R. Somerville (ed.), Biotechnology Series 3: Amino Acid Biosynthesis and Genetic Regulation. Addison-Wesley, New York.
72. Cunin, R., A. Boyen, P. Pouwels, N. Glansdorff, and M. Crabeel. 1975. Parameters of gene expression in the bipolar argECBH operon of E. coli K12. The question of translational control. Mol. Gen. Genet. 140:51–60.
73. Cunin, R., T. Eckhardt, J. Piette, A. Boyen, A. Piérard, and N. Glansdorff 1983. Molecular basis for modulated regulation of gene expression in the regulon of Escherichia coli K-12. Nucleic Acids Res. 11:5007–5019.
74. Cunin, R., D. Elseviers, G. Sand, G. Freundlich, and N. Glansdorff. 1969. On the functional organization of the argECBH cluster of genes in Escherichia coli K12. Mol. Gen. Genet. 106:32–47.
75. Cunin, R., and N. Glansdorff. 1971. Messenger RNA from arginine and phosphoenolpyruvate carboxylase genes in argR + and argR – strains of Escherichia coli K12. FEBS Lett. 18:135–137.
76. Cunin, R., N. Glansdorff, A. Piérard, and V. Stalon. 1986. Biosynthesis and metabolism of arginine in bacteria. Microbiol. Rev. 50:314–352.
77. Cunin, R., N. Kelker, A. Boyen, H. Lang-Yang, G. Zubay, N. Glansdorff, and W. K. Maas. 1976. Involvement of arginine in in vitro repression of transcription of arginine genes C, B and H in Escherichia coli K12. Biochem. Biophys. Res. Commun. 69:377–382.
78. Cunningham-Rundles, S., and W. K. Maas. 1975. Isolation, characterization, and mapping of Escherichia coli mutants blocked in the synthesis of ornithine decarboxylase. J. Bacteriol. 124:791–799.
79. Czaplewski, L., A. North, M. Smith, S. Baumberg, and P. Stockley. 1992. Purification and initial characterization of AhrC, the regulator of arginine metabolism genes in Bacillus subtilis. Mol. Microbiol. 6:267–275.
80. Davis, R. 1983. Arginine synthesis in eukaryotes, p. 81–102. In K. Herman and R. Somerville (ed.), Biotechnology Series 3: Amino Acid Biosynthesis and Genetic Regulation. Addison-Wesley, New York.
81. Davis, R. 1986. Compartmental and regulatory mechanisms in the arginine pathways of Neurospora crassa and Saccharomyces cerevisiae. Microbiol. Rev. 50:280–313.
82. Davis, R. H., D. R. Morris, and P. Coffino. 1992. Sequestered end products and enzyme regulation: the case of ornithine decarboxylase. Microbiol. Rev. 56:280–290.
83. De Deken, R. H. 1963. Biosynthèse de l’arginine chez la levure. I. Le sort de la N-α-acétylornithine. Biochim. Biophys. Acta 78:606–616.
84. Degryze, E. 1974. Evidence that yeast acetylornithinase is a carboxypeptidase. FEBS Lett. 43:285–288.
85. Dion, A. S., and S. S. Cohen. 1972. Polyamine stimulation of nucleic acid synthesis in an uninfected and phage-infected polyamine auxotroph of Escherichia coli K12. Proc. Natl. Acad. Sci. USA 69:213–217.
86. Dohi, M., A. Kikuchi, and L. Gorini. 1978. Some regulation profiles of ornithine transcarbamylase synthesis in vitro. J. Biochem. 84:1401–1409.
87. Eckhardt, T., and T. Leisinger. 1975. Isolation and characterization of mutants with a feedback resistant N-acetylglutamate synthase in Escherichia coli K12. Mol. Gen. Genet. 258:225–232.
88. Elseviers, D., R. Cunin, N. Glansdorff, S. Baumberg, and E. Ashcroft. 1972. Control regions within the argECBH gene cluster of Escherichia coli K12. Mol. Gen. Genet. 117:349–366.
89. Ennis, H. L., and L. Gorini. 1961. Control of arginine biosynthesis in strains of Escherichia coli not repressible by arginine. J. Mol. Biol. 3:439–446.
90. Eriani, G., G. Dirheimer, and J. Gangloff. 1989. Isolation and characterization of the gene coding for Escherichia coli arginyl-tRNA synthetase. Nucleic Acids Res. 17:5725–5735.
91. Eriani, G., G. Dirheimer, and J. Gangloff. 1990. Structure-function relationship of arginyl-tRNA synthetase from isolation and characterization of the argS mutation MA5002. Nucleic Acids Res. 18:1475–1479.
92. Furuchi, T., K. Kashiwagi, H. Kobayashi, and K. Igarashi. 1991. Characteristics of the gene for a spermidine and putrescine transport system that maps at 15 min on the Escherichia coli chromosome. J. Biol. Chem. 266:20928–20933.
93. Gale, E. F. 1946. The bacterial amino acid decarboxylases. Adv. Enzymol. 6:1–32.
94. Gallant, J. A. 1979. Stringent control in E. coli. Annu. Rev. Genet. 13:393–415.
95. Gardner, M. M., D. D. Hennig, and R. A. Kelln. 1983. Control of arg gene expression in Salmonella typhimurium by the arginine repressor from Escherichia coli K12. Mol. Gen. Genet. 189:458–462.
96. Gerlo, E., W. Freist, and J. Charlier. 1982. Arginyl-tRNA synthetase from Escherichia coli K12: specificity with regard to ATP analogs and their magnesium complexes. Hoppe-Seyler’s Z. Physiol. Chem. 363:S165–S173.
97. Gigot, D. 1990. Recruitment of an alternative carbamoylphosphate synthase in a carB deletion mutant of Escherichia coli. Arch. Int. Physiol. Biochim. 98:B128.
98. Gigot, D., M. Crabeel, A. Feller, D. Charlier, W. Lissens, N. Glansdorff, and A. Piérard. 1980. Patterns of polarity in the Escherichia coli carAB gene cluster. J. Bacteriol. 143:914–920.
99. Gigot, D., N. Glansdorff, C. Legrain, A. Piérard, V. Stalon, W. Konigsberg, I. Caplier, A. D. Strosberg, and G. Hervé. 1977. Comparison of the N-terminal sequences of aspartate and ornithine carbamoyltransferases of Escherichia coli. FEBS Lett. 81:28–32.
100. Glansdorff, N., S. Bourgeois, and J. M. Wiame. 1962. Interaction de régulation enzymatique entre les biosynthèses de l’arginine et des pyrimidines chez Escherichia coli. Arch. Int. Physiol. Biochim. 70:149–151.
101. Glansdorff, N., C. Dambly, S. Palchaudhuri, M. Crabeel, A. Piérard, and P. Halleux. 1976. Isolation and heteroduplex mapping of a λ transducing phage carrying the structural gene for Escherichia coli K-12 carbamoylphosphate synthase: regulation of enzyme synthesis in lysogens. J. Bacteriol. 127:302–308.
102. Glansdorff, N., and G. Sand. 1965. Coordination of enzyme synthesis in the arginine patway of Escherichia coli K12. Biochim. Biophys. Acta 108:308–311.
103. Glansdorff, N., G. Sand, and G. Verhoef. 1967. The dual genetic control of ornithine transcarbamylase synthesis in Escherichia coli K-12 Mutat. Res. 4:743–751.
104. Goldemberg, S. H. 1980. Lysine decarboxylase mutants of Escherichia coli: evidence for two enzyme forms. J. Bacteriol. 141:1428–1431.
105. Gorini, L. 1960. Antagonism between substrate and repressor controlling the formation of a biosynthetic enzyme. Proc. Natl. Acad. Sci. USA 46:682–690.
106. Gorini, L., W. Gundersen, and M. Burger. 1961. Genetics of regulation of enzyme synthesis in the arginine biosynthetic pathway of Escherichia coli. Cold Spring Harbor Symp. Quant. Biol. 26:173–182.
107. Gorini, L., and S. M. Kalman. 1963. Control by uracil of carbamoylphosphate synthesis in Escherichia coli. Biochim. Biophys. Acta 69:335–360.
108. Gorini, L., and E. Kataja. 1964. Phenotypic repair by streptomycin of defective genotypes in E. coli. Proc. Natl. Acad. Sci. USA 51:487–493.
109. Gorini, L., and W. K. Maas. 1957. The potential for the formation of a biosynthetic enzyme in Escherichia coli. Biochim. Biophys. Acta 25:208–209.
110. Guha, A., Y. Saturen, and W. Szybalski. 1971. Divergent orientation of transcription from the biotin locus of Escherichia coli. J. Mol. Biol. 56:53–62.
111. Guilloton, M., and F. Karst. 1987. Cyanate specifically inhibits arginine biosynthesis in Escherichia coli K12: a case of by-product inhibition. J. Gen. Microbiol. 133:655–665.
112. Guillou, F., M. Liao, A. Garcia-Espana, and C. J. Lusty. 1992. Mutational analysis of carbamylphosphate synthase. Substitution of Glu841 leads to loss of functional coupling between the two catalytic domains of the synthase subunit. Biochemistry 31:1656–1664.
113. Guillou, F., S. R. Rubino, R. S. Markovitz, D. M. Kinney, and C. J. Lusty. 1989. Escherichia coli carbamoylphosphate synthase: domains of glutaminase and synthase subunit interaction. Proc. Natl. Acad. Sci. USA 86:8304–8308.
114. Haas, D., V. Kurer, and T. Leisinger. 1972. N-acetylglutamate synthase of Pseudomonas aeruginosa. An assay in vitro and feedback inhibition by arginine. Eur. J. Biochem. 31:290–295.
115. Hafner, E. W., C. W. Tabor, and H. Tabor. 1979. Mutants of Escherichia coli that do not contain 1,4-diaminobutane (putrescine) or spermidine. J. Biol. Chem. 254:12419–12426.
116. Hall, B. G., and J. A. Gallant. 1973. On the rate of messenger decay during amino acid starvation. J. Mol. Biol. 73:121–124.
117. Hallshall, D. M. 1975. Overproduction of lysine by mutant strains of Escherichia coli with defective transport systems. Biochem. Genet. 13:109–124.
118. Halpern, S. Y., and E. Metzer. 1989. Utilization of l-arginine, l-ornithine, agmatine and putrescine as the major source of nitrogen and its control in Escherichia coli K12, p. 85–95. In U. Bachrach and Y. M. Meiner (ed.), The Physiology of Polyamines, vol. II. CRC Press, Inc., Boca Raton, Fla.
119. Han, B. D., W. G. Nolan, H. P. Hopkins, R. T. Jones, J. L. Ingraham, and A. T. Abdelal. 1990. Effect of growth temperature on folding of carbamoylphosphate synthase of Salmonella typhimurium and a cold-sensitive derivative. J. Bacteriol. 172:5089–5096.
120. Hartman, S. C. 1973. Relationship between glutamine amidotransferase and glutaminase, p. 319–330. In S. Prusiner and E. R. Stadman (ed.), The Enzymes of Glutamine Metabolism. Academic Press, Inc., New York.
121. Hatziloukas, E., and N. Panopoulos. 1992. Origin, structure, and regulation of argK, encoding the phaseolotoxin-resistant ornithine carbamoyltransferase in Pseudomonas syringae pv. phaseolicola, and functional expression of argK in transgenic tobacco. J. Bacteriol. 174:5895–5909.
122. Heimberg, H., A. Boyen, M. Crabeel, and N. Glansdorff. 1990. Escherichia coli and Saccharomyces cerevisiae acetylornithine aminotransferase: evolutionary relationship with ornithine aminotransferase. Gene 90:69–78.
123. He, B., K. Y. Choi, and H. Zalkin. 1993. Regulation of Escherichia coli glnB, prsA, and speA by the purine repressor. J. Bacteriol. 175:3598–3606.
124. Heller, J. S., W. F. Fong, and E. S. Canellakis. 1976. Induction of a protein inhibitor to ornithine decarboxylase by the end products of its reaction. Proc. Natl. Acad. Sci. USA 73:1858–1862.
125. Heller, J. S., R. Rostomily, D. A. Kyriadis, and E. S. Canellakis. 1983. Regulation of polyamine biosynthesis in Escherichia coli by basic proteins. Proc. Natl. Acad. Sci. USA 80:5181–5184.
126. Higgins, F. C., and G. F. L. Ames. 1981. Two periplasmic transport proteins which interact with a common membrane receptor show extensive homology: complete nucleotide sequence. Proc. Natl. Acad. Sci. USA 78:6038–6042.
127. Hilger, F., J. P. Simon, and V. Stalon. 1979. Yeast argininosuccinate synthase. Purification, structural and kinetic properties. Eur. J. Biochem. 94:153–163.
128. Hirschfield, I. N., and H. P. J. Bloemers. 1969. The biochemical characterization of two mutant arginyl transfer ribonucleic acid synthetases from Escherichia coli K12. J. Biol. Chem. 244:2911–2916.
129. Hirschfield, I. N., R. H. De Deken, P. C. Horn, D. A. Hopwood, and W. K. Maas. 1968. Studies on the mechanism of repression of arginine biosynthesis in Escherichia coli. III. Repression of enzymes of arginine biosynthesis in arginyl-tRNA synthetase mutants. J. Mol. Biol. 35:83–93.
130. Hirschfield, I. N., H. J. Rosenfeld, Z. Leifer, and W. K. Maas. 1970. Isolation and characterization of a mutant of Escherichia coli blocked in the synthesis of putrescine. J. Bacteriol. 101:725–730.
131. Höltje, J. V. 1978. Streptomycin uptake via an inducible polyamine transport system in Escherichia coli. Eur. J. Biochem. 86:345–351.
132. Hong, J., and B. N. Ames. 1979. Localized mutagenesis of any specific small region of the bacterial chromosome. Proc. Natl. Acad. Sci. USA 68:3158–3162.
133. Houghton, J. E., D. A. Bencini, G. A. O’Donovan, and J. R. Wild. 1984. Protein differentiation: a comparison of aspartate transcarbamoylase and ornithine transcarbamoylase from Escherichia coli K12. Proc. Natl. Acad. Sci. USA 81:4864–4868.
134. Houghton, J. E., G. A. O’Donovan, and J. R. Wild. 1989. Reconstruction of an enzyme by domain substitution effectively switches substrate specificity. Nature (London) 338:172–174.
135. Hu, M., and R. C. Deonier. 1981. Mapping of IS1 elements flanking the argF region of the Escherichia coli K-12 chromosome. Mol. Gen. Genet. 181:222–229.
136. Igarashi, K., K. Kashiwagi, H. Hamasaki, A. Miura, T. Kakegawa, S. Hirose, and S. Matsuzaki. 1986. Formation of compensatory polyamine by Escherichia coli polyamine-requiring mutants during growth in the absence of polyamines. J. Bacteriol. 166:128–134.
137. Itikama, H., S. Baumberg, and H. J. Vogel. 1968. Enzymic basis for a genetic suppression: accumulation and deacylation of N-acetylglutamine-γ-semialdehyde in enterobacterial mutants. Biochim. Biophys. Acta 159:547–550.
138. Jacob, F., and J. Monod. 1961. Genetic regulatory mechanism in the synthesis of proteins. J. Mol. Biol. 3:318–356.
139. Jacobs, P., J. C. Jauniaux, and M. Grenson. 1980. A cis-dominant regulatory mutation linked to the argB-argC gene cluster in Saccharomyces cerevisiae. J. Mol. Biol. 139:691–704.
140. Jacoby, G. A. 1971. Mapping the gene determining ornithine transcarbamylase and its operator in Escherichia coli B. J. Bacteriol. 108:645–651.
141. Jacoby, G. A. 1972. Control of the argECBH cluster in Escherichia coli. Mol. Gen. Genet. 177:337–348.
142. Jacoby, G. A., and L. Gorini. 1967. Genetics of control of the arginine pathway in Escherichia coli. J. Mol. Biol. 24:41–50.
143. Jacoby, G. A., and L. Gorini. 1969. A unitary account of the repression mechanism of arginine biosynthesis in Escherichia coli. I. The genetic evidence. J. Mol. Biol. 39:73–87.
144. Jensen, K. F., J. Neuhardt, and L. Schack. 1982. RNA polymerase involvement in the regulation of expression of Salmonella typhimurium pyr genes. Isolation and characterization of a fluorouracil-resistant mutant with high, constitutive expression of the pyrB and pyrE genes due to a mutation in rpoBC. EMBO J. 1:69–74.
145. Jensen, R. A. 1976. Enzyme recruitment in evolution of new functions. Annu. Rev. Microbiol. 30:409–425.
146. Jessop, A. P., and C. Clugston. 1985. Amplification of the argF region in strain Hfr P4X of E. coli K12. Mol. Gen. Genet. 201:347–350.
147. Jessop, A. P., and N. Glansdorff. 1980. Genetic factors affecting recovery of nonpoint mutations in the region of a gene coding for ornithine transcarbamylase: involvement of both the F factor in its chromosomal state and the recA gene. Genetics 96:779–799.
148. Kadner, R., and W. K. Maas. 1971. Regulatory gene mutations affecting arginine biosynthesis in Escherichia coli. Mol. Gen. Genet. 111:1–14.
149. Kallo, A., P. P. McCann, and P. Berg. 1981. dl-α-(Difluoromethyl)arginine: a potent enzyme-activated irreversible inhibitor of bacterial arginine decarboxylase. Biochemistry 20:3163–3166.
150. Kalman, S. M., P. H. Duffield, and T. Brzozowski. 1965. Identity in Escherichia coli of a carbamoylphosphokinase and an activity which catalyzes amino group transfer from glutamine to ornithine in citrulline biosynthesis. Biochem. Biophys. Res. Commun. 18:530–537.
151. Kang, C. H., S. Gokcen, and G. F.-L. Ames. 1992. Crystallization and preliminary X-ray studies of the liganded lysine, arginine, ornithine-binding protein from Salmonella typhimurium. J. Mol. Biol. 225:1123–1125.
152. Karlström, O., and L. Gorini. 1969. A unitary account of the repression mechanisms of arginine biosynthesis in Escherichia coli. II. Application to the physiological evidence. J. Mol. Biol. 3:89–94.
153. Kashiwagi, K., and K. Igarashi. 1988. Adjustment of polyamine contents in Escherichia coli. J. Bacteriol. 170:3131–3135.
154. Kashiwagi, K., S. Miyamoto, F. Suzuki, H. Kobayashi, and K. Igarashi. 1992. Excretion of putrescine by the putrescine-ornithine antiporter encoded by the potE gene of Escherichia coli. Proc. Natl. Acad. Sci. USA 89:4529–4533.
155. Kashiwagi, K., T. Suzuki, F. Suzuki, T. Furuchi, H. Kobayashi, and K. Igarashi. 1991. Coexistence of the genes for putrescine transport protein and ornithine decarboxylase at 16 min on the Escherichia coli chromosome. J. Biol. Chem. 266:20922–20927.
156. Kelker, N., and T. Eckhardt. 1977. Regulation of argA operon expression in Escherichia coli K12: cell-free synthesis of β-galactosidase under argA control. J. Bacteriol. 132:67–72.
157. Kelker, N., and W. K. Maas. 1974. Selection of genetically repressible (argR +) strains of E. coli K-12 from genetically derepressed (argR –) mutants using acetylnorvaline. Mol. Gen. Genet. 132:131–135.
158. Kelker, N., W. K. Maas, H. L. Yang, and G. Zubay. 1976. In vitro synthesis and repression of argininosuccinase in Escherichia coli K-12: partial purification of the arginine repressor. Mol. Gen. Genet. 144:17–20.
159. Kelley, R. L., and C. Yanofsky. 1982. trp aporepressor production is controlled by autogenous regulation and inefficient translation. Proc. Natl. Acad. Sci. USA 79:3120–3124.
160. Kelln, R. A., and G. A. O’Donovan. 1976. Isolation and partial characterization of an argR mutant of Salmonella typhimurium. J. Bacteriol. 128:528–535.
161. Kelln, R. A., and V. L. Zak. 1978. Arginine regulon control in Salmonella typhimurium-Escherichia coli hybrid merodiploid. Mol. Gen. Genet. 161:333–335.
162. Kikuchi, A., D. Elseviers, and L. Gorini. 1975. The isolation and characterization of λ transducing phages for argF, argI and adjacent genes. J. Bacteriol. 122:727–742.
163. Kikuchi, A., and L. Gorini. 1975. Similarity of genes argF and argI. Nature (London) 256:621–623.
164. Kilstrup, M., C. D. Lu, A. Abdelal, and J. Neuhard. 1988 Nucleotide sequence of the carA gene and regulation of the carAB operon in Salmonella typhimurium. Eur. J. Biochem. 176:421–429.
165. Knight, D. M., and E. E. Jones. 1977. Regulation of Escherichia coli ornithine transcarbamylase by orotate. J. Biol. Chem. 252:5928–5930.
166. Kraft, R., and L. A. Leinwand. 1987. Sequence of the complete P protein gene and part of the M protein gene from the histidine transport operon of Escherichia coli and Salmonella typhimurium. Nucleic Acids Res. 15:8568.
167. Kröger, H., R. Wahl, and P. Rice. 1993. Compilation of DNA sequence of Escherichia coli (update 1993). Nucleic Acids Res. 21:2973–3000.
168. Krzyzek, R. A., and P. Rogers. 1976. Effect of arginine on the stability and size of argECBH messenger ribonucleic acid in Escherichia coli. J. Bacteriol. 126:365–376.
169. Kuo, T., and B. A. D. Stocker. 1969. Suppression of proline requirement of proA and proAB deletion mutants in Salmonella typhimurium by mutation to arginine requirement. J. Bacteriol. 98:593–598.
170. Kustu, S. G., and G. F. Ames. 1973. The hisP protein, a known histidine transport component in Salmonella typhimurium, is also an arginine transport component. J. Bacteriol. 116:107–113.
171. Kyriakidis, D. A., J. S. Heller, and E. S. Canellakis. 1978. Modulation of ornithine decarboxylase in Escherichia coli by positive and negative effectors. Proc. Natl. Acad. Sci. USA 75:4699–4703.
172. Large, P. J. 1992. Enzymes and pathways of polyamine breakdown in microorganisms. FEMS Microbiol. Rev. 8:249–262.
173. Lavallé, R. 1970. Regulation at the level of translation in the arginine pathway of E. coli K-12. J. Mol. Biol. 51:449–451.
174. Lavallé, R., and G. De Hauwer. 1970. Tryptophan messenger translation in Escherichia coli. J. Mol. Biol. 51:435–437.
174a. Legrain, C., M. Demarez, N. Glansdorff, and A. Piérard. 1995. Ammonia-dependent synthesis and metabolic channeling of carbamoyl phosphate in the hyperthermophilic archaeon Pyroccus furiosus.Microbiology 141:1093-1099
175. Legrain, C., P. Halleux, V. Stalon, and N. Glansdorff. 1972. The dual genetic control of ornithine carbamoyltransferase in Escherichia coli: a case of bacterial hybrid enzymes. Eur. J. Biochem. 27:93–102.
176. Legrain, C., and V. Stalon. 1976. Ornithine carbamoyltransferase from Escherichia coli W: purification, structure and steady-state kinetic analysis. Eur. J. Biochem. 63:289–301.
177. Legrain, C., V. Stalon, and N. Glansdorff. 1976. Escherichia coli ornithine carbamoyltransferase isoenzymes: evolutionary significance and the isolation of λargF and λargI transducing phages. J. Bacteriol. 128:35–38.
178. Legrain, C., V. Stalon, N. Glansdorff, D. Gigot, A. Piérard, and M. Crabeel. 1976. Structural and regulatory mutations allowing utilization of citrulline or carbamoylaspartate as a source of carbamoylphosphate in Escherichia coli. J. Bacteriol. 128:39–48.
179. Leisinger, T., and D. Haas. 1975. N-acetylglutamate synthase of Escherichia coli: regulation of synthesis and activity by arginine. J. Biol. Chem. 250:1690–1693.
180. Leisinger, T., and H. J. Vogel. 1969. Repression by arginine in Escherichia coli: a comparison of arginyl transfer RNA profiles. Biochim. Biophys. Acta 182:572–578.
181. Lim, D., J. D. Oppenheim, T. Eckhardt, and W. K. Maas. 1987. Nucleotide sequence of the argR gene of Escherichia coli K12 and isolation of its product, the arginine repressor. Proc. Natl. Acad. Sci. USA 84:6697–6701.
182. Lin, S. X., J. P. Shi, X. D. Cheng, and Y. L. Wang. 1988. Arginyl-tRNA synthase from Escherichia coli, purification by affinity chromatography, properties, and steady-state kinetics. Biochemistry 27:6343–6348.
183. Lin, S. X., Q. Wang, and Y. L. Wang. 1988. Interactions between Escherichia coli arginyl-tRNA synthase and its substrates. Biochemistry 27:6348–6353.
184. Linn, T., M. Goman, and J. Scaife. 1979. Lambda transducing bacteriophage carrying deletion of the argECBH-rpoBC region of the Escherichia coli chromosome. J. Bacteriol. 140:479–489.
185. Lissens, W., R. Cunin, N. Kelker, N. Glansdorff, and A. Piérard. 1980. In vitro synthesis of Escherichia coli carbamoylphosphate synthase: evidence for participation of the arginine repressor in cumulative repression. J. Bacteriol. 141:58–66.
186. Loftfield, R. B., and E. A. Elgner. 1969. Mechanism of action of amino acid transfer ribonucleic acid ligases. J. Biol. Chem. 244:1746–1754.
187. Lu, C. D., and A. T. Abdelal. 1993. The Salmonella typhimurium uracil-sensitive mutation use is in argU and encodes a minor arginine-tRNA. J. Bacteriol. 175:3897–3899.
188. Lu, C. D., J. E. Houghton, and A. T. Abdelal. 1992. Characterization of the arginine repressor from Salmonella typhimurium and its interactions with the carAB operator. J. Mol. Biol. 225:11–24.
189. Lu, C. D., M. Kilstrup, J. Neuhard, and A. Abdelal. 1989. Pyrimidine regulation of tandem promoters for carAB in Salmonella typhimurium. J. Bacteriol. 171:5436–5442.
190. Lusty, C. J., and M. Liao. 1993. Substitution of Glu841 by lysine in the carbamate domain of carbamoyl phosphate synthase alters the catalytic properties of the glutamine subunits. Biochemistry 32:1278–1284.
191. Maas, W. K. 1961. Studies on repression of arginine biosynthesis in Escherichia coli. Cold Spring Harbor Symp. Quant. Biol. 26:183–191.
192. Maas, W. K. 1965. Genetic defects affecting an arginine permease and repression of arginine synthesis in Escherichia coli. Fed. Proc. 24:1239–1242.
193. Maas, W. K. 1972. Mapping of genes involved in the synthesis of spermidine in Escherichia coli. Mol. Gen. Genet. 119:1–9.
193a. Maas, W. K. 1994. The arginine repressor of Escherichia coli. Microbiol. Rev. 58:631–640.
194. Maas, W. K., and A. J. Clark. 1964. Studies on the mechanism of repression of arginine biosynthesis in Escherichia coli. II. Dominance of repressibility in diploids. J. Mol. Biol. 8:365–370.
195. Maas, W. K., Z. Leifer, and J. Pointdexter. 1970. Studies with mutants blocked in the synthesis of polyamines. Ann. N. Y. Acad. Sci. 171:957–967.
196. Maas, W. K., R. Maas, J. M. Wiame, and N. Glansdorff. 1964. Studies on the mechanism of repression of arginine biosynthesis in Escherichia coli. I. Dominance of repressibility in zygotes. J. Mol. Biol. 8:359–364.
197. Markham, G. D., C. W. Tabor, and H. Tabor. 1983. S-adenosylmethionine decarboxylase (Escherichia coli). Methods Enzymol. 94:228–230.
198. Martin, P. R., and M. H. Mulks. 1992. Sequence analysis and complementation studies of the argJ encoding ornithine acetyltransferase from Neisseria gonorrhoeae. J. Bacteriol. 174:2694–2701.
199. Marvil, D. K., and T. Leisinger. 1977. N-acetylglutamate synthase of Escherichia coli. Purification, characterization, and molecular properties. J. Biol. Chem. 252:3295–3303.
200. Mazaitis, A., S. Palchaudhuri, N. Glansdorff, and W. K. Maas. 1976. Isolation and characterization of dargECBH transducing phages and heteroduplex analysis of the argECBH cluster. Mol. Gen. Genet. 143:185–196.
201. McLellan, W., and H. J. Vogel. 1970. Translation repression in the arginine system of Escherichia coli. Proc. Natl. Acad. Sci. USA 67:1703–1707.
202. McLellan, W., and H. J. Vogel. 1973. Stability of argECBH messenger RNA under arginine excess or restriction. Biochem. Biophys. Res. Commun. 55:1385–1389.
203. Mehler, A. H., and S. K. Mitra. 1967. The activation of arginyl transfer ribonucleic acid synthetase by transfer ribonucleic acid. J. Biol. Chem. 242:5495–5499.
204. Meinnel, T., E. Schmitt, Y. Mechulam, and S. Blanquet. 1992. Structural and biochemical characterization of the Escherichia coli argE gene product. J. Bacteriol. 174:2323–2331.
205. Meng, S. Y., and G. N. Bennet. 1992. Nucleotide sequence of the Escherichia coli cad operon: a system for neutralization of low extracellular pH. J. Bacteriol. 174:2659–2669.
206. Mergeay, M., D. Gigot, J. Beckman, N. Glansdorff, and A. Piérard. 1974. Physiology and genetics of carbamoylphosphate synthase in Escherichia coli K-12. Mol. Gen. Genet. 133:299–316.
207. Messenguy, F. 1976. Regulation of arginine biosynthesis in Saccharomyces cervisiae: isolation of cis-dominant constitutive mutant for ornithine carbamoyltransferase synthesis. J. Bacteriol. 128:49–55.
208. Miles, B. W., S. M. Mareay, L. E. Post, D. J. Post, S. H. Chang, and F. M. Raushel. 1993. Differential roles for three conserved histidine residues within the large subunit of carbamoylphosphate synthase. Biochemistry 32:232–240.
209. Minet, M., J. C. Jauniaux, P. Thuriaux, M. Grenson, and J. M. Wiame. 1979. Organization and expression of a two-gene cluster in arginine biosynthesis of Saccharomyces cerevisiae. Mol. Gen. Genet. 168:299–308.
210. Miran, S. G., S. H. Chang, and F. M. Rauschel. 1991. Role of four conserved histidine residues in the amidotransferase domain of carbamoyl phosphate synthase. Biochemistry 30:7901–7907.
211. Mitra, S. K., and A. H. Mehler. 1967. The arginyl transfer ribonucleic acid synthetase of Escherichia coli. J. Biol. Chem. 242:5490–5494.
212. Moore, S., R. Garvin, and E. James. 1981. Nucleotide sequence of the argF regulatory region of Escherichia coli K12. Gene 16:119–132.
213. Morgan, S. D., and D. Söll. 1978. Regulation of the biosynthesis of amino acids: tRNA ligases and tRNA. Prog. Nucleic Acid Res. Mol. Biol. 21:181–207.
214. Morris, D. R., and E. A. Boeker. 1983. Biosynthetic and biodegradative ornithine and arginine decarboxylases from Escherichia coli. Methods Enzymol. 94:125–134.
215. Morris, D. R., and C. M. Jorstad. 1970. Isolation of conditionally putrescine-deficient mutants of Escherichia coli. J. Bacteriol. 101:731–737.
216. Morris, D. R., and K. L. Koffron. 1967. Urea production and putrescine biosynthesis in Escherichia coli. J. Bacteriol. 94:1516–1519.
217. Morris, D. R., and K. L. Koffron. 1969. Putrescine biosynthesis in Escherichia coli. J. Biol. Chem. 244:6094–6099.
218. Morris, D. R., and A. B. Pardee. 1965. A biosynthetic ornithine decarboxylase in Escherichia coli. Biochem. Biophys. Res. Commun. 20:697–702.
219. Morris, D. R., and A. B. Pardee. 1966. Multiple pathways of putrescine biosynthesis in Escherichia coli. J. Biol. Chem. 241:3129–3135.
220. Motta, R. 1967. Méthodes de sélection de mutants uracile exigeants au locus ur1 de Coprinus radiatus. C. R. Acad. Sci. (Paris) 204:654–657.
221. Munro, G. F., C. A. Bell, and M. Lederman. 1974. Multiple transport components for putrescine in Escherichia coli. J. Bacteriol. 118:952–963.
222. Nakamura, V., and H. Uchida. 1983. Isolation of conditionally lethal amber mutations affecting synthesis of the nusA protein of Escherichia coli. Mol. Gen. Genet. 190:196–203.
223. Natter, W., D. Sens, and E. James. 1977. Metabolism of arginine-specific messenger ribonucleic acids in Escherichia coli K-12. J. Bacteriol. 131:214–223.
224. Neidhardt, F. C., P. L. Bloch, S. Pedersen, and S. Reeh. 1977. Chemical measurements of steady-state levels of aminoacyl transfer ribonucleic acid synthetase in Escherichia coli. J. Bacteriol. 129:378–387.
225. Neidhardt, F. C., J. Parker, and W. G. McKeever. 1975. Function and regulation of aminoacyl-tRNA synthetase in prokaryotic and eukaryotic cells. Annu. Rev. Microbiol. 29:215–250.
225a. Neuhard, J., K. F. Jensen, and E. Stauning. 1982. Salmonella typhimurium mutants with altered expression of the pyrA gene due to changes in RNA polymerase. EMBO J. 1:1141–1149.
226. Nikaido, K., and G. F.-L. Ames. 1992. Purification and characterization of the periplasmic lysine-, arginine-, ornithine-binding protein (LAO) from Salmonella typhimurium. J. Biol. Chem. 267:20706–20712.
227. Nonet, M. L., C. C. Marvel, and D. R. Tolan. 1987. The hisT-purF region of the Escherichia coli K-12 chromosome. J. Biol. Chem. 262:12209–12217.
228. Novick, R. P., and W. K. Maas. 1961. Control by endogenously synthesized arginine of the formation of ornithine transcarbamylase in Escherichia coli. J. Bacteriol. 81:236–240.
229. Nyunoya, H., and C. J. Lusty. 1983. The carB gene of Escherichia coli: a duplicated gene coding for the large subunit of carbamoylphosphate synthethase. Proc. Natl. Acad. Sci. USA 80:4629–4633.
230. Oh, B. H., C. H. Kang, H. De Bondt, S. H. Kim, K. Nikaido, A. Joski, and G. Ames. 1994. The bacterial periplasmic histidine-binding protein. Structure/function analysis of the ligand-binding site and comparison with related proteins. J. Biol. Chem. 269:4135–4143.
231. Otsuka, A., and J. Abelson. 1978. The regulatory region of the biotin operon in Escherichia coli. Nature (London) 276:689–694.
232. Panchal, C. J., S. N. Baghee, and A. Guha. 1974. Divergent orientation of transcription from the argECBH operon of Escherichia coli. J. Bacteriol. 117:675–680.
233. Pannekoek, H., R. Cunin, A. Boyen, and N. Glansdorff. 1975. In vitro transcription of the bipolar argECBH operon in Escherichia coli K12. FEBS Lett. 51:143–145.
234. Papas, T. S., and A. Peterkofsky. 1972. A random sequential mechanism for arginyl transfer ribonucleic acid synthase of Escherichia coli. Biochemistry 11:4602–4608.
235. Pardee, A. B., F. Jacob, and J. Monod. 1959. The genetic control and cytoplasmic expression of "inducibility" in the synthesis of β-galactosidase by Escherichia coli. J. Mol. Biol. 1:165–178.
236. Parsot, C., A. Boyen, G. N. Cohen, and N. Glansdorff. 1988. Nucleotide sequence of Escherichia coli argB and argC genes: comparison of N-acetylglutamate kinase and N-acetylglutamate-γ-semialdehyde dehydrogenase with homologous and analogous enzymes. Gene 68:275–283.
237. Pegg, A. E., A. J. Bitoni, P. P. McCann, and J. K. Coward. 1983. Inhibition of bacterial aminopropyltransferase by S-adenosyl-1,8-diamino-3-thiooctane and by dicyclohexamine. FEBS Lett. 155:192–196.
238. Penninckx, M. 1980. "Illicit" uptake of antimetabolites: potential use in antimicrobial chemotherapy. Trends Pharmacol. Sci. 1:271–272.
239. Penninckx, M., and D. Gigot. 1978. Synthesis and interaction with Escherichia coli l-ornithine carbamoyltransferase of two potential transition state analogs. FEBS Lett. 88:94–96.
240. Penninckx, M., and D. Gigot. 1979. Synthesis of a peptide from N-δ-(phosphonoacetyl)- l-ornithine. Its antibacterial effect through the specific inhibition of Escherichia coli carbamoyltransferase. J. Biol. Chem. 254:6392–6396.
241. Peyru, G. M., and W. K. Maas. 1967. Inhibition of Escherichia coli B by homoarginine. J. Bacteriol. 94:712–718.
242. Piérard, A. 1966. Control of the activity of Escherichia coli carbamoylphosphate synthase by antagonistic allosteric effectors. Science 154:1572–1573.
243. Piérard, A. 1983. Evolution des systèmes de synthèse et d’utilisation du carbamoylphosphate, p. 55–61. In G. Hervé (ed.), L’Evolution des Protéines. Masson, Paris.
244. Piérard, A., N. Glansdorff, D. Gigot, M. Crabeel, P. Halleux, and L. Thiry. 1976. Repression of Escherichia coli carbamoylphosphate synthase: relationship with enzyme synthesis in the arginine and pyrimidine pathways. J. Bacteriol. 127:291–301.
245. Piérard, A., N. Glansdorff, M. Mergeay, and J. M. Wiame. 1965. Control of the biosynthesis of carbamoylphosphate in Escherichia coli. J. Mol. Biol. 14:23–36.
246. Piérard, A., N. Glansdorff, and J. Yashphe. 1972. Mutations affecting uridine monophosphate pyrophosphorylase or the argR gene in Escherichia coli. Effects on carbamoylphosphate and pyrimidine biosynthesis and on uracil uptake. Mol. Gen. Genet. 118:235–245.
247. Piérard, A., W. Lissens, P. Halleux, R. Cunin, and N. Glansdorff. 1980. Role of transcriptional regulation and enzyme inactivation in the synthesis of Escherichia coli carbamoylphosphate synthase. J. Bacteriol. 141:382–385.
248. Piérard, A., and J. M. Wiame. 1964. Regulation and mutation affecting a glutamine-dependent formation of carbamoylphosphate in Escherichia coli. Biochem. Biophys. Res. Commun. 15:76–81.
249. Piette, J., R. Cunin, A. Boyen, D. Charlier, M. Crabeel, F. Van Vliet, N. Glansdorff, C. Squires, and C. L. Squires. 1982. The regulatory region of the divergent argECBH operon in Escherichia coli K12. Nucleic Acids Res. 10:8031–8048.
250. Piette, J., R. Cunin, M. Crabeel, A. Boyen, N. Glansdorff, C. Squires, and C. L. Squires. 1980. Nucleotide sequence of the control region of the argECBH bipolar operon in Escherichia coli. Arch. Int. Physiol. Biochim. 88:B242–B243.
251. Piette, J., R. Cunin, M. Crabeel, and N. Glansdorff. 1981. The regulatory region of the argF gene of Escherichia coli. Arch. Int. Physiol. Biochim. 89:B127–B128.
252. Piette, J., R. Cunin, F. Van Vliet, D. Charlier, M. Crabeel, Y. Ota, and N. Glansdorff. 1982. Homologous control sites and DNA transcription starts in the related argF and argI genes of Escherichia coli K12. EMBO J. 1:853–857.
253. Piette, J., H. Nyunoya, C. Lusty, R. Cunin, G. Weyens, M. Crabeel, D. Charlier, N. Glansdorff, and A. Piérard. 1984. DNA sequence of the carA gene and the control region of carAB: tandem promoters, respectively controlled by arginine and the pyrimidines, regulate the synthesis of carbamoylphosphate synthase in Escherichia coli K12. Proc. Natl. Acad. Sci. USA 81:4134–4138.
254. Pistocchi, R., K. Kashiwagi, S. Migamoto, E. Nukui, Y. Sadataka, H. Kobayashi, and K. Igarashi. 1993. Characteristics of the operon for a putrescine transport system that maps at 19 minutes on the Escherichia coli chromosome. J. Biol. Chem. 268:146–152.
255. Popkin, P. S., and W. K. Maas. 1980. Escherichia coli regulatory mutation affecting lysine transport and lysine decarboxylase. J. Bacteriol. 141:485–492.
256. Porter, R. W., M. O. Modebe, and G. R. Stark. 1969. Aspartate transcarbamylase. Kinetic studies of the catalytic subunit. J. Biol. Chem. 244:1846–1859.
257. Post, L. E., D. J. Post, and F. M. Raushel. 1990. Dissection of the functional domain of Escherichia coli carbamoyl phosphate synthase by site-directed mutagenesis. J. Biol. Chem. 265:7742–7747.
258. Pouwels, P., R. Cunin, and N. Glansdorff. 1974. Divergent transcription of the argECBH cluster of genes in Escherichia coli K-12. J. Mol. Biol. 83:421–424.
259. Powers, S. G., and A. Meister. 1978. Mechanism of the reaction catalyzed by carbamoylphosphate synthase. Binding of ATP to the two functionally different ATP sites. J. Biol. Chem. 253:800–803.
260. Press, R., N. Glansdorff, P. Miner, J. De Vries, R. Kadner, and W. K. Maas. 1971. Isolation of transducing particles of bacteriophage that carry different regions of the Escherichia coli genome. Proc. Natl. Acad. Sci. USA 68:795–798.
261. Prusiner, S. 1973. Glutaminase of Escherichia coli. Properties, regulation and evolution, p. 293–318. In S. Prusiner and E. R. Stadtman (ed.), The Enzymes of Glutamine Metabolism. Academic Press, Inc., New York.
262. Quay, S., and H. N. Christensen. 1974. Basis of transport discrimination of arginine from other basic amino acids in Salmonella typhimurium. J. Biol. Chem. 249:7011–7017.
263. Ratner, S. 1976. Enzymes of arginine and urea synthesis, p. 181–220. In S. Grisolia, R. Bagena, and F. Mayor (ed.), The Urea Cycle. Academic Press, Inc., New York.
264. Raushel, F. M., B. W. Miles, L. E. Post, and D. J. Post. 1992. Mutational analysis of two putative domains within the large subunit of carbamoyl phosphate synthase from Escherichia coli. Bioorg. Med. Chem. Lett. 2:319–322.
265. Reznikoff, W. S., J. H. Miller, J. G. Scaife, and J. R. Beckwith. 1969. A mechanism for repressor action. J. Mol. Biol. 43:201–213.
266. Riley, M., and N. Glansdorff. 1983. Cloning the Escherichia coli K-12 argD gene specifying acetylornithine δ-transaminase. Gene 24:335–339.
267. Robin, J. P., B. Penverne, and G. Hervé. 1989. Carbamoyl phosphate biosynthesis and partition in pyrimidine and arginine pathways of Escherichia coli. In situ properties of carbamoylphosphate synthase, ornithine transcarbamylase and aspartate transcarbamylase in permeabilized cells. Eur. J. Biochem. 183:519–528.
268. Rogers, P., T. M. Kaden, and M. Thoth. 1975. Repression of arg mRNA synthesis by l-arginine in cell-free extracts of Escherichia coli. Biochem. Biophys. Res. Commun. 65:1284–1291.
269. Rogers, P., R. Krzyzek, T. M. Kaden, and E. Arfman. 1971. Effect of arginine and canavanine on arginine messenger RNA synthesis. Biochem. Biophys. Res. Commun. 44:1220–1226.
270. Roof, W. D., K. F. Folterman, and J. R. Wild. 1983. The organization and regulation of the pyrBI operon in E. coli includes a rho-dependent attenuator sequence. Mol. Gen. Genet. 187:391–400.
271. Roovers, M., D. Charlier, A. Feller, D. Gigot, F. Holemans, W. Lissens, A. Piérard, and N. Glansdorff. 1988. carP, a novel gene regulating the transcription of the carbamoylphosphate synthase operon of Escherichia coli. J. Mol. Biol. 204:857–865.
272. Rosen, B. P. 1971. Basic amino acid transport in Escherichia coli. J. Biol. Chem. 246:3653–3662.
273. Rosen, B. P. 1973. Basic amino acid transport in Escherichia coli: properties of canavanine-resistant mutants. J. Bacteriol. 116:627–635.
274. Rubino, S. D., H. Nyuoya, and C. J. Lusty. 1987. In vivo synthesis of carbamyl phosphate from NH3 by the large subunit of Escherichia coli carbamyl phosphate synthase. J. Biol. Chem. 262:4382–4386.
275. Rubio, V., J. Cervera, C. J. Lusty, E. Bendala, and H. G. Britton. 1991. Domain structure of the large subunit of Escherichia coli carbamoyl phosphate synthase. Location of the binding site for the allosteric inhibitor UMP in the COOH-terminal domain. Biochemistry 30:1068–1075.
276. Sabo, D. L., E. A. Boeker, B. Bryers, H. Waron, and E. H. Fischer. 1974. Purification and physical properties of inducible Escherichia coli lysine decarboxylase. Biochemistry 13:662–670.
277. Sakanyan, V., D. Charlier, C. Legrain, A. Kochikyan, I. Mett, A. Piérard, and N. Glansdorff. 1993. Primary structure, partial purification and regulation of key enzymes of the acetyl cycle of arginine biosynthesis in Bacillus stearothermophilus: dual function of ornithine acetyltransferase. J. Gen. Microbiol. 139:393–402.
278. Sakanyan, V., L. Desmarez, C. Legrain, D. Charlier, I. Mett, A. Kochikyan, A. Savchenko, A. Boyen, P. Falmagne, A. Piérard, and N. Glansdorff. 1993. Gene cloning, sequence analysis, purification, and characterization of a thermostable aminoacylase from Bacillus stearothermophilus. Appl. Environ. Microbiol. 59:3878–3888.
279. Sakanyan, V., A. Kochikyan, I. Mett, C. Legrain, D. Charlier, A. Piérard, and N. Glansdorff. 1992. A re-examination of the pathway for ornithine biosynthesis in a thermophilic and two mesophilic Bacillus species. J. Gen. Microbiol. 138:125–130.
280. Sakanyan, V., C. Legrain, D. Charlier, A. Kochikyan, N. K. Osina, and N. Glansdorff. 1993. N-Acetylglutamate-5-phosphotransferase of thermophilic bacterium Bacillus stearothermophilus: nucleotide sequence and enzyme characterisation. Genetika 29:556–570. (In Russian.)
281. Sanderson, K. E., and J. R. Roth. 1988. Linkage map of Salmonella typhimurium. Edition VII. Microbiol. Rev. 52:485–532.
282. Satishchandran, C., and S. M. Boyle. 1984. Antagonistic transcriptional regulation of the putrescine biosynthetic enzyme agmatine ureohydrolase by cyclic AMP and agmatine in Escherichia coli. J. Bacteriol. 154:552–559.
283. Satishchandran, C., and S. M. Boyle. 1986. Purification and properties of agmatine ureohydrolase, a putrescine biosynthetic enzyme in Escherichia coli. J. Bacteriol. 165:843–848.
284. Scherer, G. F., M. D. Walkinshaw, S. Arnott, and D. J. Morré. 1980. E. coli ribosomes have regions with signal character in both the leader and protein coding segments. Nucleic Acids Res. 8:3895–3907.
285. Schwartz, J. H., W. K. Maas, and E. J. Simon. 1959. An impaired concentrating mechanism for amino acids in mutants of Escherichia coli resistant to l-canavanine and d-serine. Biochim. Biophys. Acta 32:582–583.
286. Sens, D., W. Natter, R. T. Garvin, and E. James. 1977. Transcription of the argF and argI genes of the arginine biosynthetic regulon of E. coli K12 performed in vitro. Mol. Gen. Genet. 155:7–18.
287. Sens, D., W. Natter, R. T. Garvin, and E. James. 1977. In vitro transcription of the Escherichia coli K-12 argA, argE, and argCBH operons. J. Bacteriol. 130:642–655.
288. Shaibe, E., E. Metzer, and Y. S. Halpern. 1985. Control of utilization of l-arginine, l-ornithine, agmatine, and putrescine as nitrogen sources in Escherichia coli K-12. J. Bacteriol. 163:938–942.
289. Shepherdson, M., and A. B. Pardee. 1960. Production and crystallization of aspartate transcarbamylase. J. Biol. Chem. 235:3233–3237.
290. Stalon, V., and A. Mercenier. 1984. l-Arginine utilization by Pseudomonas species. J. Gen. Microbiol. 130:6669–6676.
291. Stirling, C. J., S. D. Collows, J. F. Collins, G. Szatmari, and D. J. Sherrat. 1989. xerB, an Escherichia coli gene required for plasmid ColE1 site-specific recombination, is identical to pepA, encoding aminopeptidase A, a protein with substantial similarity to bovine lens leucine aminopeptidase. EMBO J. 8:1623–1627.
292. Stirling, C. J., G. Szatmari, G. Stewart, C. H. Smith, and D. J. Sherrat. 1988. The arginine repressor is essential for plasmid-stabilizing site-specific recombination at the ColE1 cer locus. EMBO J. 7:4389–4395.
293. Syvanen, J. L., and J. R. Roth. 1972. The structural genes for ornithine transcarbamylase in Salmonella typhimurium and Escherichia coli. J. Bacteriol. 110:69–76.
294. Szumanski, M. B., and S. M. Boyle. 1992. Influence of cyclic AMP, agmatine, and a novel protein encoded by a flanking gene on speB (agmatine ureohydrolase) in Escherichia coli. J. Bacteriol. 174:758–764.
295. Tabor, C. W., E. W. Hafner, and H. Tabor. 1980. Construction of an Escherichia coli strain unable to synthesize putrescine, spermidine, or cadaverine: characterization of two genes controlling lysine decarboxylase. J. Bacteriol. 144:952–956.
296. Tabor, C. W., and H. Tabor. 1976. 1,4-Diaminobutane (putrescine), spermidine, and spermine. Annu. Rev. Biochem. 45:285–306.
297. Tabor, C. W., and H. Tabor. 1983. Putrescine aminopropyltransferase. Methods Enzymol. 94:265–269.
298. Tabor, C. W., and H. Tabor. 1984. Polyamines. Annu. Rev. Biochem. 53:749–790.
299. Tabor, C. W., and H. Tabor. 1985. Polyamines in microorganisms. Microbiol. Rev. 49:81–99.
300. Tabor, C. W., H. Tabor, and E. W. Hafner. 1983. Mass screening for mutants in the biosynthetic pathway for polyamines in Escherichia coli: a general method for mutants in enzymatic reactions producing CO2. Methods Enzymol. 94:83–90.
301. Tabor, C. W., H. Tabor, E. W. Hafner, G. D. Markham, and S. M. Boyle. 1983. Cloning of the Escherichia coli genes for the biosynthetic enzymes for polyamines. Methods Enzymol. 94:117–124.
302. Tabor, H., and C. W. Tabor. 1969. Formation of 1,4-diaminobutane and of spermidine by an ornithine auxotroph of Escherichia coli grown under limiting ornithine or arginine. J. Biol. Chem. 244:2286–2292.
303. Tabor, H., and C. W. Tabor. 1969. Partial separation of two pools of arginine in Escherichia coli: preferential use of exogenous rather than endogenous arginine for the biosynthesis of 1,4-diaminobutane. J. Biol. Chem. 244:6386–6387.
304. Tabor, H., C. W. Tabor, M. S. Cohn, and E. W. Hafner. 1981. Streptomycine resistance (rpsL) produces an absolute requirement for polyamines for growth of an Escherichia coli strain unable to synthesize putrescine and spermidine [Δ(speA-speB)ΔspeC]. J. Bacteriol. 147:702–704.
305. Tabor, H., C. W., Tabor, and E. W. Hafner. 1978. Escherichia coli mutants completely deficient in adenosylmethionine decarboxylase and in spermidine biosynthesis. J. Biol. Chem. 253:3671–3676.
306. Taylor, K., Z. Hradecna, and W. Szybalski. 1967. Asymmetric distribution of transcribing regions on the complementary strands of coliphage lambda DNA. Proc. Natl. Acad. Sci. USA 57:1618–1625.
307. Tian, G., D. Lim, J. Carey, and W. K. Maas. 1992. Binding of the arginine repressor of Escherichia coli K12 to its operator sites. J. Mol. Biol. 226:387–397.
308. Tian, G., D. Lim, J. D. Oppenheim, and W. K. Maas. 1994. Explanation for different types of regulation of arginine biosynthesis in Escherichia coli B and Escherichia coli K12 caused by a difference between their arginine repressors. J. Mol. Biol. 235:221–230.
308a. Tian, G., and W. K. Maas. 1994. Mutational analysis of the arginine repressor of Escherichia coli. Mol. Microbiol. 13:599–608.
309. Trotta, P., M. E. Burt, R. H. Haschemeyer, and A. Meister. 1971. Reversible dissociation of carbamoylphosphate synthase into a regulated synthesis subunit and a subunit required for glutamine utilization. Proc. Natl. Acad. Sci. USA 68:2599–2603.
310. Trudel, M., M. Springer, M. Graffe, G. Fayat, S. Blanquet, and M. Grunberg-Manago. 1984. Regulation of E. coli phenylalanyl-tRNA synthetase operon in vivo. Biochim. Biophys. Acta 782:10–17.
311. Turnbough, C. L., Jr. 1983. Regulation of Escherichia coli aspartate transcarbamylase synthesis by guanosine tetraphosphate and pyrimidine ribonucleoside triphosphate. J. Bacteriol. 153:998–1007.
312. Turnbough, C. L., Jr., and B. R. Bochner. 1985. Toxicity of the pyrimidine biosynthetic pathway intermediate carbamoylaspartate in Salmonella typhimurium. J. Bacteriol. 163:500–505.
313. Turnbough, C. L., Jr., K. L. Hicks, and J. P. Donahue. 1983. Attenuation control of pyrBI operon expression in Escherichia coli K12. Proc. Natl. Acad. Sci. USA 80:368–372.
314. Udaka, S. 1966. Pathway-specific pattern of control of arginine biosynthesis in bacteria. J. Bacteriol. 91:617–621.
315. Udaka, S. 1970. Isolation of the arginine repressor in Escherichia coli. Nature (London) 228:336–338.
316. Urban, C., and R. T. F. Celis. 1990. Purification and properties of a kinase from Escherichia coli K-12 that phosphorylates two periplasmic transport proteins. J. Biol. Chem. 265:1783–1786.
317. Urm, E., H. Y. Lang, G. Zubay, N. Kelker, and W. K. Maas. 1973. In vitro repression of N-α-acetylornithinase in Escherichia coli. Mol. Gen. Genet. 121:1–7.
318. Van Boggelen, R. A., V. Vaughn, and F. C. Neidhart. 1983. Gene for heat-inducible lysyl-tRNA synthetase (lysU) maps near cadA in Escherichia coli. J. Bacteriol. 153:1066–1068.
319. Van De Casteele, M., M. Demarez, C. Legrain, N. Glansdorff, and A. Piérard. 1990. Pathways of arginine biosynthesis in extreme thermophilic archaeo- and eubacteria. J. Gen. Microbiol. 136:1177–1183.
319a. Van de Casteele, M., C. Legrain, L. Desmares, P. G. Chen, A. Piérard, and N. Glansdorff. Molecular physiology of carbamoylation under extreme conditions: what can we learn from extreme thermophylic microorganisms? Comp. Physiol. Biochem., in press.
320. Vander Wauven, C., and V. Stalon. 1984. Enzymes of arginine degradation in Pseudomonas cepacia and Klebsiella aerogenes. Arch. Int. Physiol. Biochim. 92:B67.
321. Van Huffel, C., E. Dubois, and F. Messenguy. 1992. Cloning and sequencing of arg3 and arg11 genes of Schizosaccharomyces pombe on a 10-kb DNA fragment. Eur. J. Biochem. 205:33–43.
322. Van Vliet, F., A. Boyen, and N. Glansdorff. 1988. On interspecies gene transfer: the case of the argF gene of Escherichia coli. Ann. Inst. Pasteur/Microbiol. (Paris) 139:493–496.
323. Van Vliet, F., M. Crabeel, A. Boyen, C. Tricot, V. Stalon, P. Falmagne, Y. Nakamura, S. Baumberg, and N. Glansdorff. 1990. Sequences of the genes encoding argininosuccinate synthase in Escherichia coli and Saccharomyces cerevisiae: comparison with methanogenic archaebacteria and mammals. Gene 95:99–104.
324. Van Vliet, F., R. Cunin, A. Jacobs, J. Piette, D. Gigot, M. Lauwereys, A. Piérard, and N. Glansdorff. 1984. Evolutionary divergence of genes for ornithine and aspartate carbamoyltransferases—complete sequence and mode of regulation of the Escherichia coli argF gene: comparison with argI and pyrB. Nucleic Acids Res. 12:6277–6289.
325. Verkamp, E., V. Backman, J. Björnsson, D. Söl, and G. Eggertson. 1993. The periplasmic dipeptide permease system transports 5-aminolevulinic acid in Escherichia coli. J. Bacteriol. 175:1452–1456.
326. Vissers, S., Y. Dessaux, C. Legrain, and J. M. Wiame. 1981. Feedback inhibition by arginine on ornithine carbamoyltransferase of Agrobacterium tumefaciens. Arch. Int. Physiol. Biochim. 89:B83-B84.
327. Vogel, H. J. 1957. Repression and induction as control mechanisms of enzyme biogenesis: the "adaptive" formation of acetylornithinase, p. 276–289. In W. D. McElroy and B. Glass (ed.), The Chemical Basis of Heredity. Johns Hopkins Press, Baltimore.
328. Vogel, H. J. 1960. Repression of an acetylornithine permeation system. Proc. Natl. Acad. Sci. USA 46:488–494.
329. Vogel, H. J. 1961. Aspects of repression in the regulation of enzyme synthesis: pathway-wide control and enzyme-specific response. Cold Spring Harbor Symp. Quant. Biol. 26:163–171.
330. Vogel, H. J. 1970. Arginine biosynthetic system in Escherichia coli. Methods Enzymol. 17A:249–251.
331. Vogel, H. J., and E. E. Jones. 1970. Acetylornithine-δ-aminotransferase (Escherichia coli). Methods Enzymol. 17A:260–264.
332. Vogel, H. J., and W. L. McLellan. 1970. N-Acetyl-γ-glutaminokinase (Escherichia coli). Methods Enzymol. 17A:251–255.
333. Vogel, H. J., and W. L. McLellan. 1970. N-Acetylglutamic-γ-semialdehyde dehydrogenase (Escherichia coli). Methods Enzymol. 17A:255–260.
334. Vogel, H. J., and W. L. McLellan. 1970. Acetylornithinase (Escherichia coli). Methods Enzymol. 17A:265–269.
335. Vogel, H. J., and R. H. Vogel. 1974. Enzymes of arginine biosynthesis and their respective controls. Adv. Enzymol. 40:65–90.
336. Vyas, S., and W. K. Maas. 1963. Feedback inhibition of acetylglutamate synthase by arginine in Escherichia coli. Arch. Biochem. Biophys. 100:542–546.
337. Wandinger-Ness, A. U., and R. L. Weiss. 1987. A single precursor protein for two separable mitochondrial enzymes in Neurospora crassa. J. Biol. Chem. 262:5823–5830.
338. Wang, H., D. Charlier, and N. Glansdorff. 1994. Probing Escherichia coli K-12 arginine repressor binding to a symmetric consensus operator sequence. Arch. Int. Physiol. Biochim. Biophys. 102:B43.
339. Wertheimer, S. J., and Z. Leifer. 1983. Putrescine and spemidine sensitivity of lysine decarboxylase in Escherichia coli: evidence for a constitutive enzyme and its mode of regulation. Biochem. Biophys. Res. Commun. 114:882–888.
340. Weyens, G., D. Charlier, M. Roovers, A. Piérard, and N. Glansdorff. 1988. On the role of the Shine-Dalgarno sequence in determining the efficiency of translation initiation at a weak start codon in the car operon of Escherichia coli K12. J. Mol. Biol. 204:1045–1048.
341. Weyens, G., K. Rose, P. Falmagne, N. Glansdorff, and A. Piérard. 1985. Synthesis of Escherichia coli carbamoylphosphate synthase initiates at a UUG codon. Eur. J. Biochem. 150:111–115.
342. Wickner, R. B., C. W. Tabor, and H. Tabor. 1970. Purification of adenosylmethionine decarboxylase from Escherichia coli W: evidence for covalently bound pyruvate. J. Biol. Chem. 245:2132–2139.
343. Wild, J. R., and M. E. Wales. 1990. Molecular evolution and genetic engineering of proteins domains involving aspartate transcarbamoylase. Annu. Rev. Microbiol. 44:193–218.
344. Wilson, O. H., and J. T. Holden. 1969. Arginine transport and metabolism in osmotically shocked and unshocked cells of Escherichia coli W. J. Biol. Chem. 244:2737–2742.
345. Wilson, O. H., and J. T. Holden. 1969. Stimulation of arginine transport in osmotically shocked Escherichia coli W cells by purified arginine-binding protein fractions. J. Biol. Chem. 244:2743–2749.
346. Wissenbach, U., B. Keck, and G. Unden. 1993. Physical map location of the new artPIQMJ genes of Escherichia coli, encoding a periplasmic arginine transport system. J. Bacteriol. 175:3687–3688.
347. Wright, J. M., and S. M. Boyle. 1982. Negative control of ornithine decarboxylase and arginine decarboxylase by adenosine-3',5'-cyclic monophosphate in Escherichia coli. Mol. Gen. Genet. 186:482–487.
348. Wu, W. H., and D. R. Morris. 1973. Biosynthetic arginine decarboxylase from Escherichia coli. Purification and properties. J. Biol. Chem. 248:1687–1695.
349. Xie, Q. W., C. W. Tabor, and H. Tabor. 1989. Spermidine biosynthesis in Escherichia coli: promoter and termination regions of the speED operon. J. Bacteriol. 171:4457–4465.
350. Xie, Q. W., C. W. Tabor, and H. Tabor. 1993. Deletion mutations in the speED operon: spermidine is not essential for the growth of Escherichia coli. Gene 126:115–117.
351. Yang, H. L., G. Zubay, E. Urm, G. Relness, and M. Cashel. 1974. Effect of guanosine tetraphosphate, guanoside pentaphosphate and β-r-methylguanosine pentaphosphate on gene expression of Escherichia coli in vitro. Proc. Natl. Acad. Sci. USA 71:63–67.
352. Yanofsky, C., and R. Kolter. 1982. Attenuation in amino acid biosynthetic operons. Annu. Rev. Genet. 16:113–134.
353. Ycas, M. 1974. On earlier states of the biochemical systems. J. Theor. Biol. 44:145–160.
354. York, M. K., and M. Stodolsky. 1981. Characterization of P1 arg derivatives from Escherichia coli K-12 transduction. I. IS1 elements flank the argF gene segment. Mol. Gen. Genet. 181:230–240.
355. Zidwick, M. J., G. Keller, and P. Rogers. 1984. Regulation and coupling of argECBH mRNA and enzyme synthesis in cell extracts of Escherichia coli. J. Bacteriol. 159:640–646.
356. Zidwick, M. J., J. Korshus, and P. Rogers. 1984. Positive control of expression of the argECBH gene cluster in vitro by guanosine 5'-diphosphate 3'-diphosphate. J. Bacteriol. 159:647–651.
Chapter25
