. 2007 Mar;71(1):36–47. doi:10.1128/MMBR.00032-06

Surprising Arginine Biosynthesis: a Reappraisal of the Enzymology and Evolution of the Pathway in Microorganisms

Ying Xu¹,Bernard Labedan^2,^*,Nicolas Glansdorff³

Marine Sciences Research Center, State University of New York at Stony Brook, Stony Brook, New York 11794-5000,¹ Institut de Génétique et Microbiologie, CNRS UMR 8621, Université Paris Sud, Bâtiment 400, 91405 Orsay Cedex, France,² Microbiology and Genetics, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium³

Corresponding author. Mailing address: Institut de Génétique et Microbiologie, CNRS UMR 8621, Université Paris Sud, Bâtiment 400, 91405 Orsay Cedex, France. Phone: 33 1 69 15 35 60. Fax: 33 1 69 15 72 96. E-mail:bernard.labedan@igmors.u-psud.fr.

PMC Copyright notice

PMCID: PMC1847373 PMID:17347518

Abstract

Summary: Major aspects of the pathway of de novo arginine biosynthesis via acetylated intermediates in microorganisms must be revised in light of recent enzymatic and genomic investigations. The enzymeN-acetylglutamate synthase (NAGS), which used to be considered responsible for the first committed step of the pathway, is present in a limited number of bacterial phyla only and is absent fromArchaea. In manyBacteria, shorter proteins related to the Gcn5-relatedN-acetyltransferase family appear to acetylatel-glutamate; some are clearly similar to the C-terminal, acetyl-coenzyme A (CoA) binding domain of classical NAGS, while others are more distantly related. Short NAGSs can be single gene products, as inMycobacterium spp. andThermus spp., or fused to the enzyme catalyzing the last step of the pathway (argininosuccinase), as in members of theAlteromonas-Vibrio group. How these proteins bind glutamate remains to be determined. In someBacteria, a bifunctional ornithine acetyltransferase (i.e., using both acetylornithine and acetyl-CoA as donors of the acetyl group) accounts for glutamate acetylation. In manyArchaea, the enzyme responsible for glutamate acetylation remains elusive, but possible connections with a novel lysine biosynthetic pathway arose recently from genomic investigations. In someProteobacteria (notablyXanthomonadaceae) andBacteroidetes, the carbamoylation step of the pathway appears to involveN-acetylornithine orN-succinylornithine rather than ornithine. The productN-acetylcitrulline is deacetylated by an enzyme that is also involved in the provision of ornithine from acetylornithine; this is an important metabolic function, as ornithine itself can become essential as a source of other metabolites. This review insists on the biochemical and evolutionary implications of these findings.

INTRODUCTION

Microbial genes and enzymes involved in de novo arginine biosynthesis have enjoyed increasing attention over the last fifty years. Early studies focused on the steps of the pathway, but arginine biosynthesis soon became a paradigm for the analysis of regulatory mechanisms; this aspect has remained a major center of interest ever since (reviewed in references12 and31). However, further enzymatic and genomic exploration of the microbial world recently disclosed surprises that brought the pathway itself back into focus (Fig.1). This is the subject of the present review. In order to keep the number of references within reasonable limits, the reader is referred to Caldovic and Tuchman (11) and Charlier and Glansdorff (12) for reviews of most of the early work.

The first committed step (EC 2.3.1.1) of arginine biosynthesis is acetylation ofl-glutamate at theN-α position (Fig.1). Acetylation of the early precursors of arginine distinguishes them from the analogous intermediates in proline biosynthesis and prevents spontaneous cyclization of the semialdehyde arginine precursor. In the classical arginine pathway, the flow of acetylated precursors runs until acetylornithine. The formation of the subsequent intermediate—ornithine—is catalyzed by either of two enzymes, (i) acetylornithine deacetylase (acetylornithinase [AO]) (EC 3.5.1.16) or (ii) ornithine acetyltransferase (OAT) (EC 2.3.1.35), which recycles the acetyl group on glutamate. This acetyl cycle is of obvious energetic significance and is frequently referred to as “more evolved” (see, however, our concluding remarks, below). Ornithine is converted into arginine via citrulline and argininosuccinate. This textbook picture is now challenged at two levels: (i) the very identity, mechanism of action, and origin of the enzymes responsible for glutamate acetylation in different microorganisms and (ii) the extension of the sequence of acetylated intermediates beyond acetylornithine in a number ofBacteria.

De novo arginine biosynthesis viaN-acetylglutamate is a feature characteristic of many prokaryotes, fungi, and plants (including unicellular algae) but not of animals (2); Caldovic and Tuchman (11) have reviewed the synthesis and metabolic role ofN-acetylglutamate in animals. The present analysis focuses mainly on prokaryotes, where the new discoveries were made. The first comprehensive analysis of genes and enzymes of arginine biosynthesis in plants was published recently (69). In many respects, the situations in plants and in prokaryotes such asPseudomonas sp. are similar (69).

GLUTAMATE ACETYLATION INBACTERIA

N-Acetylglutamate Synthase

Early studies of arginine biosynthesis inEscherichia coli andPseudomonas aeruginosa had disclosed an enzyme catalyzing the formation of acetylglutamate from acetyl-coenzyme A (CoA) and glutamate (EC 2.3.1.1).N-Acetylglutamate synthase (NAGS), referred to further as “classical NAGS,” is the product of a single gene (argA) encoding two domains: (i) the N-terminal domain contains a carbamate kinase fold also present in acetylglutamate kinase (NAGK) (EC 2.7.2.8, ArgB), which catalyzes the next step in the pathway (58,59), and (ii) the C-terminal domain contains an acetyl-CoA binding fold present in enzymes of the vast Gcn5-relatedN-acetyltransferase (GNAT) family that transfer the acetyl group from acetyl-CoA to a variety ofN-terminal amino groups (21,53,75). This genetic structure suggests that the N-terminal domain of NAGS is responsible for efficient glutamate binding. With most GNAT enzymes—includingE. coli ArgA—the reaction proceeds by the formation of a ternary complex between the protein and the substrates (sequential mechanism) and not by the formation of an acetylated enzyme intermediate (ping-pong bi-bi mechanism) (3,75).

NAGS is also found in fungi and in vertebrates; in the latter its only function appears to be the provision of acetylglutamate as a cofactor of carbamoylphosphate synthase (11). Within each group—bacteria, fungi, or vertebrates—NAGS amino acid sequences display obvious similarities. Between each one of these groups, however, the similarities become very weak, suggesting ancient divergence or perhaps even independent origin. Nevertheless the limited similarities found between the DNA segments encoding the C-terminal regions of mammalian andNeurospora NAGS were used to clone the mouse and human NAGS genes (10,51). Moreover, a two-domain structure similar to that of bacterial NAGS was suggested for the mammalian enzyme (51). Of note is that fungal NAGS is active only when associated with NAGK (37,56), whereas the mammalian gene can complement anE. coli argA auxotroph by itself (10). This fact and the lack of similarity between fungal NAGS and NAGK (1) suggest that in fungi it may be NAGK and not NAGS that provides the glutamate binding site. In the plantArabidopsis thaliana, there are two bimodular NAGS genes similar to those found inE. coli andPseudomonas but with an insert of 90 codons in the N-terminal domain (69).

Ornithine Acetyltransferase

Figure1 shows that another option for synthesizing acetylglutamate is OAT (ArgJ). This ornithineN-acetyltransferase (EC 2.3.1.35), discovered inMicrococcus glutamicus (70), was first characterized as an alternative to acetylornithine deacetylase (ArgE, EC 3.5.1.16) because it is a transacetylase that recycles the acetyl group from acetylornithine on glutamate, a reversible reaction (Fig.1). Much later, complementation tests and biochemical experiments with purified enzyme established that some OATs could also synthesize acetylglutamate de novo from acetyl-CoA and glutamate (43,45). Not all OATs possess this dual activity, however: some are bifunctional whereas others do not use acetyl-CoA as the acetyl donor (43,76). In the most studied instance (Geobacillus [formerlyBacillus]stearothermophilus), the acetylation activity is about five times lower with acetyl-CoA as the substrate than with acetylornithine; moreover, in the former case, the reaction is irreversible, whereas the acetyl group exchange between acetylornithine and acetylglutamate is fully reversible (76). Active OAT derives from a preprotein that undergoes self-catalyzed cleavage next to a conserved T residue that becomes the catalytic nucleophile after autoproteolysis (44).

There is no detectable similarity between bifunctional OATs (EC 2.3.1.35/EC 2.3.1.1) and classical NAGS (EC 2.3.1.1). This may not be surprising since, unlike NAGS, OAT proceeds by a ping-pong bi-bi catalytic mechanism requiring the formation of a covalent enzyme-acetyl intermediate involving the above-mentioned invariant T residue. Moreover, in contrast to the substrate recognition pattern observed in GNAT enzymes (21,53,75), the CoA moiety of acetyl-CoA does not appear to enter the catalytic site of OAT (76).

Table1 lists OATs that have been characterized in vitro or in vivo (i.e., by complementation tests) as mono- or bifunctional. It should be stressed that most of the genomic annotations reading “bifunctional OATs” in the public databases are invalid since they have been proposed on the sole basis of sequence similarity withG. stearothermophilus OAT and are not supported by enzymatic or genetic evidence.

TABLE 1.

Functional assignments for acetylation of glutamate and deacetylation of acetylornithine amongBacteria andArchaea

Organism^a	Presence of^b:					Strict dependence on:
Organism^a	ArgE	ArgJ	ArgA	Arg(A)	ArgH(A)	Strict dependence on:
Bacteria
E. coli	Yes	No	Yes	No	No
Vibrio cholerae	Yes	No	Yes	No	No
Vibrio fischeri	Yes	No	Yes	No	Yes
Vibrio parahaemolyticus	Yes	No	Yes	Yes?	Yes
Vibrio vulnificus	Yes	No	Yes	No	Yes
Photobacterium profundum	Yes	No	Yes	No	Yes
C. psychrerythraea	Yes	Yes	Yes	No	Yes
Moritella abyssi	Yes	ND	ND	ND	Yes
I. loihiensis	Yes	No	No	No	Yes	ArgH(A)
P. haloplanktis	Yes	No	No	No	Yes	ArgH(A)
P. aeruginosa	Yes	m	Yes	No	No
Neisseria gonorrhoeae	No	b	Yes	No	No
Rhizobium meliloti	No	Yes	No	Yes	No
Geobacter sulfurreducens	No	Yes	No	Yes	No
Desulfotalea psychrophila	No	Yes	No	Yes	No
Desulfovibrio vulgaris	No	Yes	No	Yes	No
Helicobacter heilmannii	No	Yes	No	Yes	No
Wolinella succinogenes	No	Yes	No	Yes	No
Chlorobium tepidum	No	Yes	No	Yes	No
Aquifex aeolicus	No	Yes	No	Yes	No
Rhodopirellula bultica	No	Yes	No	Yes	No
Leptospira interrogans	No	Yes	No	Yes	No
Thermus thermophilus	No	m	No	Yes	No	Arg(A)
D. radiodurans	No	Yes	No	Yes	No
Leifsonia xyli	No	Yes	No	Yes	No
Streptomyces coelicolor	No	m	No	Yes	No	Arg(A)
Streptomyces avidinii	No	Yes	No	Yes	No
Nocardia farcinica	No	Yes	No	Yes	No
Mycobacterium tuberculosis	No	m	No	Yes	No	Arg(A)
Mycobacterium leprae	No	Yes	No	Yes	No
Corynebacterium glutamicum	Yes?	m	No	No	No
Symbiobacterium thermophilum	No	Yes	No	Yes	No
B. subtilis	No	b	No	No	No
Geobacillus stearothermophilus	ND	b	ND	ND	ND
Geobacillus kaustophilus	No	Yes	No	Yes	No
Bacillus cereus	No	Yes	No	Yes	No
Bacillus halodenitrificans	No	Yes	No	Yes	No
T. maritima	No	Yes	No	No	No
T. neapolitana	ND	b	ND	ND	ND
Archaea
H. marismortui	Yes?	No	No	Yes	No
Methanosarcina acetivorans	No	Yes	No	Yes	No
Methanosarcina mazei	No	Yes	No	Yes	No
M. jannaschii	Yes?	m	No	No	No
Archaeoglobus fulgidus	Yes?	Yes	No	No	No

Open in a new tab

Groups of organisms are listed in the following order:Proteobacteria (γ, α, δ/ε),Bacteroidetes/Chlorobium,Aquificales,Planctomycetes,Spirochaetes,Thermus/Deinococcus,Firmicutes (actinobacteria and bacilli),Thermotogales, andArchaea.

ArgE, acetylornithinase (?, positive assignment in the absence of genetic evidence has limited value due to similarities between different deacetylases; see reference12); ArgJ, ornithine acetyltransferase (m, monofunctional; b, bifunctional); ArgA, NAGS (acetylglutamate synthase); Arg(A), S-NAGS; ArgH(A), fusion between the genes for argininosuccinase and S-NAGS; ND, not determined.

Mono- or bifunctional OATs have been found in organisms possessing a classical NAGS (81). In the former case (e.g.,P. aeruginosa), NAGS assumes the anaplerotic function of priming the acetyl cycle with a first molecule of acetylglutamate. In the latter case (e.g.,N. gonorrhoeae), there is functional redundancy, but it should be kept in mind that the acetylornithine-dependent transacetylation may be quantitatively the most important reaction, as inG. stearothermophilus (see above). There is an OAT inArabidopsis (69), but its functional pattern is not known.

The X-ray crystal structure of a monofunctional OAT ofStreptomyces clavuligerus involved in clavulanic acid biosynthesis (19), rather than primary metabolism (23), has been determined, but there is presently no in silico clue allowing us to distinguish monofunctional from bifunctional OATs. The phylogenetic tree (Fig.2) rooted by the paralogous OATs involved in clavulanic acid biosynthesis does not show a clear separation between monofunctional and bifunctional ArgJ enzymes. Indeed, the monofunctional enzyme ofP. aeruginosa and the bifunctional one ofN. gonorrhoeae have a common node. However, it appears puzzling that these bacteria share a gene set (the presence of bothargA andargJ) and thus appear to be exceptions in their respective classes, as most other monofunctional OATs are accompanied by a shortarg(A) gene (see below) and the few other organisms known to use a bifunctional OAT have onlyargJ. This may explain why they are grouped on a common node.

FIG. 2. — Phylogenetic tree of the experimentally studied ArgJ enzymes. The amino acid sequences of the ArgJ proteins that have been experimentally studied and characterized as being either mono- or bifunctional were multiply aligned with the program MUSCLE (22). This alignment was then used to reconstruct a phylogenetic tree with Phyml, a fast and efficient program based on a maximum-likelihood analysis of a distance tree (33). The tree was rooted with the two paralogues of ArgJ that are present inStreptomyces clavuligerus (19). The root is marked by a black oval. The presence/absence of enzymatic activities involved in acetyl group transfer (respective EC numbers) is indicated close to each species name. Arg(A) refers to S-NAGS, as discussed in the text. Mono- and bifunctional ArgJ enzymes are bracketed separately. The enzymatic profiles ofP. aeruginosa andN. gonorrhoeae are framed to underline their similarity. Note that the genome ofS. clavuligerus has not been entirely sequenced.

The existence of bifunctional OATs raises a question of qualitative importance. Can OAT actually replace NAGS? Since the completion of a number of genome sequencing projects, a few instances indeed have emerged where no NAGS, but a bifunctional OAT, appears to be present. This is the case inB. subtilis (reinterpretation of old complementation data [55,81]),G. stearothermophilus, andThermotoga spp., assuming thatT. maritima, whose genome was completely sequenced, is similar toT. neapolitana, which has a bifunctional OAT (26,43). The lack of biochemical information on the catalytic properties of most genomically identified OATs precludes further identification. At our present state of knowledge, it is nevertheless safe to assume that a number of microorganisms, perhaps many, depend on a bifunctional OAT for the synthesis of acetylglutamate. Recent observations, however, have revealed that there are other alternatives to OAT for “life without a classical NAGS,” as discussed in the next section.

Substitutes for Classical NAGS: Short NAGS

Evidence of substitutes for classical NAGS first came from genetic observations ofCampylobacter jejuni, an ɛ-proteobacterium (35), and two marine psychrophilic γ-Proteobacteria,Moritella abyssi andM. profunda (82).

The genome ofC. jejuni does not contain a classical NAGS or OAT gene. TheargO gene ofC. jejuni is part of anargCOBD operon, and it complementsE. coli argA mutants. ThisargO gene codes a 146-amino-acid polypeptide presenting low similarity toE. coli streptothricin acetyltransferase and being broadly related to the GNATN-acetyltransferases family. It was able to catalyze the first step of arginine biosynthesis in this organism, although it does not appear to be homologous to the classicalargA gene of other bacteria (Fig.3).

FIG. 3. — Glutamate acetylation: summary of the various ways of carrying out a single reaction, EC 2.3.1.1. The classical or most likely alternative candidates that encode the EC 2.3.1.1 activity are listed. For each are given the protein name, information on its function, its homology, and its taxonomic distribution. For the various combinations involving the (A) component, the domain structure is also outlined. CK is the carbamate kinase domain homologous to ArgB. NAT is the domain homologous toN-acetyltransferases (in gray).

M. abyssi andM. profunda display an unusual structure for theargH gene that codes for the argininosuccinate lyase catalyzing the last step (EC 4.3.2.1) of arginine biosynthesis (82). TheargH gene is extended by a 170-codon stretch able to complement anargA E. coli mutant. In contrast withC. jejuni argO, the cognate amino acid sequence is clearly similar to the C-terminal domain of NAGS (Fig.3). The gene is present at the end of anargE/CBFGH(A) operon, whereargE—coding for an acetylornithinase (AO)—and the rest of the cluster are expressed divergently, a pattern characteristic of many enteric bacteria. After the original report (82), theargH(A) gene was found in similar genetic contexts in closely related bacteria belonging to theAlteromonas-Vibrio group (81). Two of them—Pseudoalteromonas haloplanktis (46) andIdiomarina loihiensis (36)—do not harbor a genetically identifiable NAGS or OAT (Fig.4). Since both bacteria are Arg⁺, the ArgH(A) protein appears as a new type of arginine biosynthetic enzyme able to catalyze in vivo both the first committed step of the pathway and the last one. Functional redundancy occurs in some organisms of this group: several have a gene for a putative classical NAGS and one of them,Colwellia psychrerythraea, has adjacent genes for an ArgH(A) fusion and a classical NAGS (Fig.4); it also has an OAT (47). More surprisingly,Pseudoalteromonas atlantica displays anargC toargH cluster containingargA next toargH. Therefore, the comparison between the twoPseudoalteromonas species and their close relativesIdiomarina andColwellia allows us to propose scenarios explaining theargH(A) fusion (Fig.4; see concluding remarks for details).

FIG. 4. — Possible scenarios for the fusion event creating theargH(A) gene. The scenarios derive from a comparison of the genetic contexts ofargA andargH in the closely related speciesPseudoalteromonas sp.,*I. loihiensis*, andC. psychrerythraea. This comparison was made with data extracted from two online resources,*Integrated Microbial Genomes* (http://img.jgi.doe.gov/cgi-bin/pub/main.cgi) andThe SEED (http://theseed.uchicago.edu/FIG/index.cgi), as well as from unpublished tools designed by F. Lemoine, O. Lespinet, and B. Labedan to visualize synteny blocks in many multiply-aligned bacterial genomes. Box 1 shows a simplified species tree based on 16S RNA sequences compared with those ofShewanella andMoritella, whereargA is not closely linked to theargC-to-*argH* cluster.

Screening of complete genomes for homologues to thearg(A) part ofargH(A) by looking for virtualargH/arg(A) fusions brought to light a large array of organisms, including someArchaea, where such a sequence is present though not necessarily fused withargH or linked to it. InThermus thermophilus andDeinococcus spp. (bothD. radiodurans andD. geothermalis), the sequence is adjacent to anargGH operon and the genetic context suggests that it is part of it. Remarkably, most of these organisms do not display a homologue of classicalargA but have anargJ gene. In at least two organisms,T. thermophilus andS. coelicolor, OAT is monofunctional (Fig.2). Therefore, both probably depend on this short version of acetylglutamate synthase for arginine synthesis (81). Interestingly,B. subtilis has a bifunctional ArgJ (likeG. stearothermophilus) but no Arg(A) homologue (a functionally meaningful combination), whereas otherBacilli have an Arg(A) homologue and an OAT of unknown specificity (Table1). The relatedG. stearothermophilus has a bifunctional OAT, but the genome is not yet known in its entirety.

Independently of these observations, Errey and Blanchard (24) characterized inMycobacterium tuberculosis a short version of NAGS which is actually retrieved from the genome ofMycobacterium species when databases are screened forarg(A) homologues, as outlined above (81). In this organism as well, there is an OAT that appears to be monofunctional on the basis of genetic information.

Short NAGSs (S-NAGSs) thus also constitute an option for the synthesis of acetylglutamate. Since they lack the N-terminal domain of classical NAGS, and as the cognateM. tuberculosis protein displays an extremely highK_m for glutamate (>600 mM), it is possible that an efficient glutamate binding site is provided by another protein, such as NAGK, that would associate with NAGS to form an active acetylating complex. This hypothesis is reminiscent of the association between yeast NAGS and NAGK, without which NAGS remains inactive (see above). In the case of ArgH(A), it is possible that association with ArgH somehow stabilizes the protein and/or enhances its activity. Obviously, further studies will be necessary to determine the properties of these proteins and what their molecular context is in vivo.

Retrieval of close homologues of the C-terminal domain of classical NAGS likely identifies only a fraction of bacterial acetylglutamate synthases, as already shown by theargO gene ofCampylobacter (see above). In the absence of functional data, it is not possible to predict which of the many Gcn5-relatedN-acetyltransferases could play this role. In some cases, however, the genetic context indicates promising candidates. For example, inXanthomonas campestris, a gene for a putative acetyltransferase of the GNAT family was identified within a cluster of arginine biosynthetic genes betweenargB andargC (17). However, although this gene has been calledargA, it does not appear to be homologous toarg(A).

Feedback Inhibition of Ornithine Synthesis

The metabolic flow of acetylated ornithine precursors is regulated by feedback inhibition. In microorganisms using an acetylornithinase but no acetyl recycling enzyme, it is NAGS that catalyzes the first committed step of the pathway, and it is inhibited by arginine. When an OAT recycles the acetyl group, NAGK can become flow controlling; it is actually inhibited by arginine in many organisms, whether NAGS is present or not (11,12,59,71; also see below). For example, inCorynebacterium glutamicum andT. maritima, which have no recognizable NAGS or S-NAGS but an OAT (monofunctional in the former and probably bifunctional in the latter, by analogy withT. neapolitana [43]), arginine inhibits NAGK (see references26 and63, correcting an earlier report [72] forT. maritima). InP. aeruginosa, where both NAGS and OAT are present, both NAGS and NAGK are inhibited by arginine (34).

However, OAT itself is also a “logical” target for efficient feedback control. InG. stearothermophilus, arginine does not inhibit NAGK (62), but ornithine strongly inhibits both activities of the bifunctional OAT (61). Arginine inhibits OAT inThermus aquaticus (72).

The group of V. Rubio recently disclosed a structural basis for feedback inhibition of NAGS and NAGK by arginine (59). InT. maritima andP. aeruginosa, the arginine-sensitive NAGK is a ring-like homohexamer where arginine binds to each dimeric subunit at a site close to the interdimeric junction. InE. coli, the arginine-insensitive NAGK is a homodimer, whereas the arginine-sensitive NAGS is a homohexamer, as is NAGK inT. maritima andP. aeruginosa. Since NAGK is homologous to the N-terminal domain of NAGS, Ramon-Maiques et al. (59) were able to localize the molecular signature for arginine inhibition in that domain. However,Mycobacterium S-NAGS, where this domain is absent, is also inhibited by arginine (24). Both the arginine and glutamate binding sites thus remain to be identified in S-NAGS.

In the yeastSaccharomyces cerevisiae, both NAGS and NAGK are inhibited by arginine, but NAGS must be associated with NAGK to remain both functional and sensitive to arginine. Deletion of the NAGS gene decreases the sensitivity of the kinase, while making the kinase insensitive renders NAGS insensitive as well (1,56). The situation appears to be similar inNeurospora crassa (83).

The Lysine Connection

In a seminal paper, Nishida et al. (54) described a novel pathway for lysine biosynthesis inT. thermophilus. As in fungi, this pathway uses aminoadipic acid (AAA) as an intermediate but, instead of proceeding to lysine via saccharopine, it converts AAA to lysine by a series of acetylated intermediates analogous to those leading from glutamate to ornithine (Fig.1, steps 1 to 5) and differing by one CH₂ group only.Thermus was known to have two unlinked genes annotatedargC, one of them being part of an arginine operon (4), as well as two genes annotatedargB, oneargD homologue (aliaslysJ) and oneargE homologue (aliaslysK). The latter two were shown to code for bifunctional enzymes, acting on substrates of both the new AAA pathway and ornithine biosynthesis (48,49). InE. coli,N-acetylornithine transaminase (ArgD) andN-succinyl-l,l-diaminopimelate:alpha-ketoglutarate aminotransferase (DapC), an enzyme of lysine biosynthesis, are one and the same protein (40).

Of particular interest from the point of view of glutamate acetylation is the protein LysX, which was shown by genetic disruption to catalyze the synthesis of the first intermediate in the conversion of AAA to lysine inThermus, plausiblyN-acetyl-α-aminoadipate (54,60). LysX enzymes form a subfamily inside the ATP-dependent carboxylate-amino ligase RimK family. Considering the chemical similarity between the substrates and products of the reactions catalyzed by NAGS or S-NAGS and LysX, respectively (even if the mechanisms of those reactions are different since the LysX-catalyzed reaction involves ATP), the question arises whether a LysX homologue could not be involved in glutamate acetylation in some organisms (Fig.3). The substrate specificity ofThermus LysX remains to be investigated; disruption of the cognate gene does not bring about arginine auxotrophy, but acetylglutamate is probably synthesized by the short version of ArgA, coregulated with anargG-argH operon (81) (and not by ArgJ, as erroneously assumed in reference54, sinceThermus ArgJ is monofunctional [4]). LysX homologues have been detected in several organisms (mostlyArchaea; see the next section), but their exact function remains to be defined. Some of them appear in a genetic context that suggests a possible involvement in arginine biosynthesis, such as inChloroflexus aurantiacus, somewhat related toThermus. InLactobacillus plantarum (which has noargA gene), the presence of a LysX homologue closely linked to anargC argBE cluster, distinct from the bipolarcarA argCJDBF operon (7), is intriguing (see Table2).

TABLE 2.

LysX homologues detected inArchaea andBacteria and belonging to various subfamilies of the RimK family

Domain	Species	Enzyme^a	Subfamily	Uniprot accession no.	Genetic context oflysX homologue^b
Archaea	Pyrobaculum aerophilum	LysX	LysX	Q8ZXK1	Noarg genes in the neighborhood
Archaea	Pyrococcus abyssi	LysX	LysX	Q9V1I7	lysX,argC,argB,argD,argE
Archaea	Pyrococcus horikoshii	LysX	LysX	O59396	lysX,argC,argB,argD,argE
Archaea	Pyrococcus furiosus	LysX	LysX	Q8U0B7	lysX,argC,argB,argD,argE
Archaea	Thermococcus kodakaraensis	LysX	LysX	Q5JFW0	lysX,argC,argB,argD,argE
Archaea	Picrophilus torridus	LysX	LysX	Q6KZ00,Q6KYZ8	lysX1,argE,lysX2,argD,argB,argC
Archaea	Natronomonas pharaonis	ArgX	LysX	Q3IMD9	argG,argH,hyp,argX,argC,argB,argD,argE,argF
Archaea	Haloarcula marismortui	LysX	LysX	Q5UZ49	argG,argH,hyp,argX,argC,argB,argD,argE,argF
Archaea	Sulfolobus solfataricus P2	LysX	LysX	Q7SI95	argC,lysX,argD
Archaea	Sulfolobus tokodaii str. 7	LysX	LysX	Q976J9	argE,argD,lysX,lysW,lysM,argB,argC
Archaea	Sulfolobus acidocaldarius	lysX	LysX	Q4JAP9	argE,argD,lysX,lysW,lysM,argB,argC
Archaea	Ferroplasma acidarmanus	LysX	LysX	Q41V82	argC,argB,argD,lysX,argE
Archaea	Aeropyrum pernix	LysX	LysX	Q9YBY7	argB,argC,lysX,argD,argE
Bacteria	Thermus thermophilus HB8	LysX	LysX	Q84BR0	hyp,hyp,lysX,argC,argB
Bacteria	Chloroflexus aurantiacus	LysX	LysX	Q3DYF8	lysW,lysX,argC,hyp
Bacteria	Lactobacillus plantarum	RimK	RimK	Q88Z73	lysX,hyp,hyp,argC,argBE (bifunctional)
Archaea	Methanococcus jannaschii	MptN	MptN	Q58037	Noarg genes in the neighborhood
Archaea	Methanosarcina mazei	MptN	MptN	Q8Q0M5	Noarg genes in the neighborhood
Archaea	Methanosarcina acetivorans	MptN	MptN	Q8TKX5	Noarg genes in the neighborhood
Archaea	Methanococcus maripaludis	MptN	MptN	Q6LZC7	Noarg genes in the neighborhood
Archaea	Methanococcus jannaschii	CofF	CofF	Q58407	Noarg genes in the neighborhood
Archaea	Methanococcus maripaludis	CofF	CofF	Q6M0U8	Noarg genes in the neighborhood

Open in a new tab

MptN, tetrahydromethanopterin:α-l-glutamate ligase; CofF, γ-F420-2:α-l-glutamate ligase.

hyp, encoding a hypothetical protein.

GLUTAMATE ACETYLATION INARCHAEA

Much less information on glutamate acetylation is available forArchaea than forBacteria. No classical NAGS gene was detected in the archaeal genomes sequenced so far, butargJ is present in several euryarchaeota, mainly methanogens andArchaeoglobus fulgidus (Table1). The only archaeal OAT to have been studied enzymatically (inMethanocaldococcus jannaschii [43]) appears to be monofunctional, but screening the genome of this organism does not yield sequences homologous to the S-NAGS discussed above. The protein responsible for acetylglutamate synthesis inM. jannaschii thus remains elusive. On the other hand, several archaeal genomes (methanogens) that harbor a putativeargJ gene also display a homologue ofarg(A) encoding S-NAGS (Table1). In the absence of enzymatic data on the cognate ArgJ proteins, it is not possible to speculate on the metabolic significance of these sequences. One thing is clear, however: the classical NAGS does not appear to belong to the archaeal patrimony.

Nishida et al. (54) opened a new perspective on arginine biosynthesis inArchaea in the paper discussed in the previous section, where they suggested that the novel AAA pathway for lysine biosynthesis discovered in the bacteriumT. thermophilus might operate in the euryarchaeotePyrococcus sp. to synthesize both lysine and ornithine. In fact, not only the pyrococci but also many euryarchaeota and crenarchaeota turned out to contain putative arginine genes (Table2), fromargB toargH (8,14,42). According to Nishida et al. (54), the genes annotated asargB,argC,argD, andargE might code for bifunctional proteins acting on molecules differing by one CH₂ group only. The exact function of these genes is still uncertain, however; at least some of the genes of the diaminopimelate pathway for lysine biosynthesis appear to be present in the genomes of pyrococci (74). InSulfolobus, lysine was found to regulate the synthesis of mRNA from a gene cluster annotatedargC,argB (lysZ),lysM (lrp-like),lysW,lysX,argD (lysJ), andargE (lysK), but involvement of these genes in arginine biosynthesis could not be excluded, and no enzyme assays were carried out (8).

What concerns us here, however, is whether a homologue of LysX, identified inThermus by genetic disruption as the first enzyme of the new AAA pathway, could be involved in glutamate acetylation in someArchaea (Fig.3). LysX homologues have indeed been detected in several archaeal genomes in a genetic context that suggests a possible involvement in arginine biosynthesis (Table2). For example, the complete sequences of the extreme haloalkaliphileNatronomonas pharaonis (25) and the halophileHaloarcula marismortui (5) show a near completearg cluster fromargB toargH. Apparently,argA andargJ homologues are missing, but betweenargH andargC, one finds a short gene encoding a putative protein and alysX homologue calledargX (annotated as a putative regulator for no clear reason in reference25). Often,lysX homologues are adjacent to putativeargC,argB, andargD genes, as in thermococci andSulfolobales. InHaloarcula marismortui, as inN. pharaonis, the operon is almost complete with the structureargF (annotatedarcB, normally reserved for catabolic ornithine carbamoyltransferase [OTC]),argE,argD,argB,argC,lysX/argX, unknown sequence,argH, andargG. Table2 further shows that there are in archaeal methanogens other homologues belonging to the RimK family, such as members of the subfamilies MptN (tetrahydromethanopterin:α-l-glutamate ligase) and CofF (γ-F420-2:α-l-glutamate ligase). However, in both cases, noarg genes are found in the neighborhood of the encoding genes. Note thatM. jannaschii has bothmptN andcofF genes, but it is impossible with the present data to determine whether one of them could be involved in acetylglutamate synthesis.

The pattern of glutamate acetylation inArchaea may thus turn out to be as varied as it is inBacteria: Arg(A)-like polypeptides (S-NAGS), other putative acetyltransferases, bifunctional OATs (yet to be identified in this domain, however), and LysX homologues are candidates for this metabolic function (Fig.3).

THE FATE OF ACETYLATED PRECURSORS BEYOND ACETYLORNITHINE

Our textbook vision of arginine biosynthesis beyond acetylornithine was shattered by the recent discovery of acetylornithine carbamoyltransferase (AOTC) (ArgF′) in the γ-proteobacteriumXanthomonas campestris (50,65). AOTC catalyzes the formation of acetylcitrulline from acetylornithine but not that of citrulline from ornithine (Fig.1, Box 1). The enzyme ArgF′ (EC 2.1.3.9) is a distant homologue of the classical OTC (EC 2.1.3.3) but lacks an OTC-specific motif, the SMG motif, and the substrate-binding mechanism of this enzyme appears to be different from that of both ornithine and aspartate carbamoyltransferases (68).X. campestris AOTC is presently the only carbamoyltransferase known to catalyze this reaction in the pure form, but genes similar toargF′, also lacking the SMG motif, were identified in other species:X. axonopodis,Xylella fastidiosa,Bacteroides thetaiotaomicron,Cytophaga hutchinsonii,Tannerella forsythensis, andPrevotella ruminicola (50). Obviously, if acetylcitrulline is formed instead of citrulline, a downstream acetylated compound has to be deacetylated in order to provide arginine as an end product for protein synthesis; Shi et al. (66) actually purified fromX. campestris a deacetylase active on acetylcitrulline. Morizono et al. (50) showed that this enzyme, encoded by a gene formerly annotatedargE because of its similarity with the cognateE. coli gene, is in fact not a novel deacetylase specific for acetylcitrulline, as postulated by Shi et al. (66), but is active with acetylornithine as well. Moreover, sinceargF′ complements an OTC-deficientE. coli mutant but the cognate protein does not carbamoylate ornithine in vitro (66), it is likely thatE. coli ArgE itself also deacetylates acetylcitrulline (Fig.1, Box 1). Acetylornithinases are known to be active with a number of acetylated amino acids (see reference12).

B. fragilis argF′ was shown to be essential to its host, but despite its apparently close relationship to theX. campestris homologue, no AOTC activity could be detected in vitro (50,64). Very recently, thisB. fragilis ArgF′ protein was crystallized and shown to be a novelN-succinyl-l-ornithine transcarbamylase (SOTC) involved in the arginine biosynthesis of this bacterium (67). Since the disruption of the encoding gene,argF′_Bf, rendersB. fragilis auxotrophic for arginine, it appears that the arginine pathway inBacteroides is different from the canonical pathway of most organisms (67). Interestingly, a single mutation convertsB. fragilis SOTC into an AOTC (67). The occurrence of succinylornithine as an intermediate of arginine catabolism in certainBacteria is well known (41,73), but no carbamoyltransferase using this substrate had been reported until now.

The fact thatX. campestris acetylcitrulline deacetylase is also active on acetylornithine should be emphasized, since a metabolism bypassing ornithine would probably be deficient; indeed, ornithine itself is a pivotal metabolite (Fig.5). It is required for incorporation in hydroxamate siderophores (77) and in lipids found in abundance inBacteria, some of which are related to species utilizing the AOTC pathway (16,28). It was even suggested that these lipids may play an important role in organisms pathogenic for eukaryotes; this is precisely the case forXylella,Xanthomonas,Bacteroides, andTannerella. Moreover, under certain circumstances, ornithine may become the main source of putrescine, a precursor of polyamines, which are vital metabolites (see reference12). It can be noted that, in principle, an OAT presenting the broad substrate specificity pattern observed inG. stearothermophilus (76) could play the same metabolic role as AO in organisms using an AOTC if it proved to be active on acetylcitrulline.

FIG. 5. — Ornithine is a pivotal compound in cell metabolism. Arginine biosynthesis and known arginine catabolic pathways are outlined. The name of each pathway, framed with the type of line used to draw it, is identified by the enzyme initiating the pathway. Only key intermediates are mentioned, including the six main products deriving from ornithine. cp, carbamoylphosphate. For a recent review of arginine metabolism, see reference41.

CONCLUDING REMARKS ON THE EVOLUTION OF ARGININE BIOSYNTHESIS

Recent discoveries have led to a reappraisal of glutamate acetylation in microorganisms. From established functional evidence, at least three options for this metabolic step exist: the classical NAGS, the so-called S-NAGS (fused with ArgH or independent), and the bifunctional OAT (Fig.3). S-NAGSs belong to the vast GNAT family but are not all homologous to the C-terminal domain of classical NAGS (Fig.3). In someBacteria or inArchaea, other, as-yet-undetected acetyltransferases may be responsible for the synthesis of acetylglutamate.

Classical two-domain NAGS does not appear to be the ancestral glutamate acetylating enzyme; it occurs inProteobacteria but is absent from many major divisions ofBacteria, including those proposed to branch off early in different phylogenies (Thermotoga,Aquifex, orPlanctomycetes [Table1; also see below]). NAGS has not been found inArchaea either. Due to its wide distribution, including early branching bacterial divisions, OAT appears to have preceded classical NAGS, but it is not known whether the ancestral OAT was bifunctional, a prerequisite to qualify as putative primeval glutamate acetylase. On the other hand, the short, Arg(A)-like precursor of NAGS found in severalBacteria andArchaea (Table1) is probably related to an ancestral acetyltransferase that was present in the last universal common ancestor (LUCA). It may have preceded OAT in the role of acetylglutamate synthetase. In the bacterial domain, this short acetyltransferase would have given rise to the bimodular, classical NAGS by fusion with NAGK (Fig.3). In members of theAlteromonas-Vibrio group, it would have become fused withargH either directly, as originally proposed (see Fig. 7 in reference81), or, perhaps more likely, as the result of a deletion occurring in an organism carrying anargA gene close toargH, as the comparison ofColwellia and two differentPseudoalteromonas spp. suggests (Fig.4). Indeed, inP. atlantica,argH is followed by a completeargA, but inP. haloplanktis, onlyargH(A) is present; inColwellia, on the other hand,argH(A) is closely followed by a completeargA. It is conceivable that theP. haloplanktis configuration results from a deletion joiningargH to the C-terminal domain of an adjacentargA, whereas theColwellia configuration could result from the insertion in the chromosome of a circularizedargA (resident or acquired horizontally) by a crossover betweenargH(A) andargA at the level of theargA C-terminal domain. Alternatively,argA could have been duplicated in tandem in the immediate neighborhood ofargH before the fusion occurred. At any rate, these events would be ancient, since the two NAT domains present inColwellia are distantly related (they have only 26% identity).

In many other instances, S-NAGS would have remained a single gene product (Fig.3), sometimes encoded in an arginine operon-like gene cluster (as inThermus andDeinococcus), but association of the protein with a glutamate binding site providing NAGK appears likely. These events may have occurred in organisms having lost an ancestral bifunctional OAT or originally devoid of such an enzyme. The origin of NAGS in fungi is not clear, but as acetyl-CoA is a poor substrate for fungal OAT, at least in yeast (15), efficient glutamate acetylation may have necessitated association of NAGS with NAGK, though not as a covalent complex. In vertebrates, the bimodular structure of NAGS (51) is reminiscent of that ofBacteria and may point to a bacterial origin. In plants, the similarity with classical prokaryotic NAGS is even stronger (69).

When attempting to correlate the distribution of known and putative glutamate acetylases with the branching pattern of the bacterial subdivisions, one is confronted at the onset with a problem, namely, that this pattern is very deep and therefore remains controversial: whereas some authors still consider extreme thermophilicBacteria (Aquificales,Thermotogales) to be the earliest, deepest branches of the bacterial tree, others have emphasized the possibility that inBacteria extreme thermophily may have been acquired by convergence (79). In keeping with this view, a reappraisal of bacterial phylogeny placed the essentially nonthermophilicPlanctomyces spp. at the base of the tree, with the additional, intriguing feature that these organisms appear to possess the equivalent of a nucleus (9,27). The controversy over the first lines of divergence in the bacterial domain is ongoing (6,13,18,20,30,39,80), and at present, the distribution of known glutamate-acetylating enzymes is compatible with either scheme. Advocates of the ancestry of extreme thermophilicBacteria might take advantage of the argument of the bifunctionality of OAT inThermotoga, which has no NAGS, to consider it primeval. On the other hand, there is an OAT in nonthermophilicPlanctomyces but its functional pattern is not known;Planctomyces has anarg(A)-like sequence as well and no NAGS, but this is also true of the extreme thermophilic bacteriumAquifex (81).

Finally, it should be emphasized that any attempt at drawing a comprehensive evolutionary history of the early steps in the arginine pathway will have to integrate functional information aboutlysX. Its presence in several organisms (manyArchaea) close to genes annotated asarg determinants (Fig.3 and Table2) may indeed suggest a primeval function for LysX, perhaps already in the LUCA. The LUCA was probably a genetically redundant and promiscuous community (29,38,78,80). Several alternatives for glutamate acetylation may therefore have been already present in the LUCA population.

Another conclusion of this review is that in manyBacteria, OAT (ArgJ) appears to have predated AO (ArgE), since the latter emerges amongProteobacteria; it may have been recruited from a pool of deac(et)ylases, perhaps after the loss of OAT or independently, or it may even have been acquired laterally. The traditional view that OAT is “more evolved” than AO should therefore be qualified from the chronological point of view.

The presence of an AO in organisms endowed with an AOTC (Fig.1) rather than an OTC may, however, have a deeper origin than in other bacterial subdivisions. Indeed, the carbamoyltransferase phylogenetic tree suggests thatX. campestris AOTC and the related sequences form a family that arose very early in the course of evolution (52), perhaps even before OTC, since AOTC-like proteins branch off close to the root of the tree (see Fig.1 in reference52). If primordial ornithine and lysine biosynthesis proceeded from aminoadipate and glutamate by the bifunctional AAA/ornithine pathway postulated by Nishida et al. (54),N-acetyllysine andN-acetylornithine would constitute the forelast intermediates of the pathway; a deacetylase of this ancestral pathway (ArgE) would be required to produce both lysine and ornithine but also to deacetylate the acetylcitrulline that would be produced if the primordial carbamoyltransferase was an AOTC rather than an OTC. In this view, OTC could have evolved from AOTC. Today, AOTC-like sequences occur mostly in plant pathogens (Xanthomonas,Xylella) and in intestinal or mouth commensals (Bacteroides,Tannerella), which, at first sight, could suggest that AOTC evolved from OTC, perhaps to escape the action of OTC-inhibiting substances produced by the host or by the bacterium itself, as in the case ofPseudomonas syringae ArgK, an OTC insensitive to the phaseolotoxin produced by the bacterium (57). However, AOTC may have been reacquired in these pathogens from an ancestral state of arginine biosynthesis surviving in some cell lines. Hopefully, further investigations on the phylogeny of carbamoyltransferases and deacetylases will throw some light on this question. It is worth pointing out that the transacetylation rates ofG. stearothermophilus OAT for both lysine and ornithine are very similar (76). Therefore, in principle at least, the deacetylase required to deliver both lysine and ornithine from a putative AAA/ornithine pathway could be replaced in some organisms by an OAT with broad substrate specificity.

This hypothesis of an ancestral AAA/ornithine pathway could have another virtue: it contrasts with the idea that the formation of acetylated arginine precursors could have arisen by recruiting enzymes from the classical, non-acetylated proline biosynthetic pathway. If this had occurred in the LUCA, we would expect the proline pathway to be present inArchaea; in fact, it is mostly absent, except in a few methanogens and inN. pharaonis (where it might constitute a late acquisition), whereas mostArchaea appear to synthesize proline by cyclization of ornithine (32). The latter may have derived from an ancestral AAA/ornithine pathway; being a source of proline may therefore have been an essential function of ornithine very early in metabolic evolution.

OUTLOOK

The novel information discussed in this review raises a number of questions that further genomic research and enzyme characterization could resolve. Whether S-NAGS associates in vivo with other glutamate-binding proteins to utilize glutamate efficiently is one of them. How the activity of S-NAGS is feedback inhibited by arginine is another. The enzymology and structure of the ArgH(A) enzyme is a new topic which is presently being explored; the presence of this protein in human pathogens (V. parahaemolyticus andV. vulnificus) and perhaps in pathogens of fishes or other marine organisms may lead to medical and ecological applications. The functional annotation of most OATs requires substantiation; as many organisms possessing an OAT also feature an S-NAGS gene, identifying the protein actually responsible for glutamate acetylation in vivo will require complementation tests withE. coli argA andargE mutants as well as biochemical analyses. OAT structural studies could also identify the possible “signature” of bifunctional OATs. Still other questions are as follows. Which enzymes are responsible for glutamate acetylation in microorganisms where no short or classical NAGS could be identified and where OAT, when present, does not use acetyl-CoA as a substrate? How ancient is the AOTC pathway, and was it already present in the LUCA? The perspectives opened by comparing arginine biosynthesis with the new lysine biosynthetic pathway discovered inThermus are fascinating. The new phylogeny proposed forBacteria, withPlanctomyces as the earliest possible branch, close to LUCA, enhances the interest of these investigations.

Acknowledgments

N.G. is grateful to G. T. Taylor for his hospitality at the Marine Sciences Research Center of Stony Brook University during the preparation of the manuscript.

B.L. thanks the CNRS and ANR for financial support.

REFERENCES

1.Abadjieva, A., K. Pauwels, P. Hilven, and M. Crabeel. 2001. A new yeast metabolon involving at least the two first enzymes of arginine biosynthesis. J. Biol. Chem. 276:42869-42880. [DOI] [PubMed] [Google Scholar]
2.Alonso, E., and V. Rubio. 1989. Participation of ornithine aminotransferase in the synthesis and catabolism of ornithine in mice. Studies using gabaculine and arginine deprivation. Biochem. J. 259:131-138. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Bachmann, C., S. Krahenbuhl, and J. P. Colombo. 1982. Purification and properties of acetyl-CoA: l-glutamateN-acetyltransferase from human liver. Biochem. J. 205:123-127. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Baetens, M., C. Legrain, A. Boyen, and N. Glansdorff. 1998. Genes and enzymes of the acetyl cycle of arginine biosynthesis in the extreme thermophilic bacteriumThermus thermophilus. Microbiology 144:479-492. [DOI] [PubMed] [Google Scholar]
5.Baliga, N. S., R. Bonneau, M. T. Facciotti, M. Pan, G. Glusman, E. W. Deutsch, P. Shannon, Y. Chiu, R. S. Weng, R. R. Gan, P. Hung, S. V. Date, E. Marcotte, L. Hood, and W. V. Ng. 2004. Genome sequence ofHaloarcula marismortui: a halophilic archaeon from the Dead Sea. Genome Res. 14:2221-2234. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Bocchetta, M., S. Gribaldo, A. Sanagelantoni, and P. Cammarano. 2000. Phylogenetic depth of the bacterial generaAquifex andThermotoga inferred from analysis of ribosomal protein, elongation factor, and RNA polymerase subunit sequences. J. Mol. Evol. 50:366-380. [DOI] [PubMed] [Google Scholar]
7.Bringel, F., L. Frey, S. Boivin, and J. C. Hubert. 1997. Arginine biosynthesis and regulation inLactobacillus plantarum: thecarA gene and theargCJBDF cluster are divergently transcribed. J. Bacteriol. 179:2697-2706. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Brinkman, A. B., S. D. Bell, R. J. Lebbink, W. M. de Vos, and J. van der Oost. 2002. TheSulfolobus solfataricus Lrp-like protein LysM regulates lysine biosynthesis in response to lysine availability. J. Biol. Chem. 277:29537-29549. [DOI] [PubMed] [Google Scholar]
9.Brochier, C., and H. Philippe. 2002. Phylogeny: a non-hyperthermophilic ancestor for bacteria. Nature 417:244. [DOI] [PubMed] [Google Scholar]
10.Caldovic, L., H. Morizono, X. Yu, M. Thompson, D. Shi, R. Gallegos, N. A. Allewell, M. Malamy, and M. Tuchman. 2002. Identification, cloning and expression of the mouseN-acetylglutamate synthase gene. Biochem. J. 364:825-831. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Caldovic, L., and M. Tuchman. 2003.N-Acetylglutamate and its changing role through evolution. Biochem. J. 372:279-290. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Charlier, D., and N. Glansdorff. September 2004, posting date. Chapter 3.6.1.10, Biosynthesis of arginine and polyamines.In R. Curtiss III et al. (ed.), EcoSal—Escherichia coli andSalmonella: cellular and molecular biology. ASM Press, Washington, DC.http://www.ecosal.org. [DOI] [PubMed]
13.Cicarelli, F. D., T. Doerks, C. von Mering, C. J. Creevey, B. Snel, and P. Bork. 2006. Toward automatic reconstruction of a highly resolved tree of life. Science 311:1283-1287. [DOI] [PubMed] [Google Scholar]
14.Cohen, G., V. Barbe, D. Flament, M. Galperin, R. Heilig, O. Lecompte, O. Poch, D. Prieur, J. Querellou, R. Ripp, J. C. Thierry, J. van der Oost, J. Weissenbach, I. Zivanovic, and P. Forterre. 2003. An integrated analysis of the genome of the hyperthermophilic archaeonPyrococcus abyssi. Mol. Microbiol. 47:1495-1512. [DOI] [PubMed] [Google Scholar]
15.Crabeel, M., A. Abadjieva, P. Hilven, J. Desimpelaere, and O. Soetens. 1997. Characterization of theSaccharomyces cerevisiae ARG7 gene encoding ornithine acetyltransferase, an enzyme with acetylglutamate synthase activity. Eur. J. Biochem. 250:232-241. [DOI] [PubMed] [Google Scholar]
16.Cullinane, M., C. Baysse, J. P. Morissey, and F. O'Gara. 2005. Identification of two lysophosphatidic acid acyltransferase genes with overlapping function inPseudomonas fluorescens. Microbiology 151:3071-3080. [DOI] [PubMed] [Google Scholar]
17.da Silva, A. C., J. A. Ferro, F. C. Reinach, C. S. Farah, L. R. Furlan, R. B. Quaggio, C. B. Monteiro-Vitorello, M. A. Van Sluys, N. F. Almeida, L. M. Alves, A. M. do Amaral, M. C. Bertolini, L. E. Camargo, G. Camarotte, F. Cannavan, J. Cardozo, F. Chambergo, L. P. Ciapina, R. M. Cicarelli, L. L. Coutinho, J. R. Cursino-Santos, H. El-Dorry, J. B. Faria, A. J. Ferreira, R. C. Ferreira, M. I. Ferro, E. F. Formighieri, M. C. Franco, C. C. Greggio, A. Gruber, A. M. Katsuyama, L. T. Kishi, R. P. Leite, E. G. Lemos, M. V. Lemos, E. C. Locali, M. A. Machado, A. M. Madeira, N. M. Martinez-Rossi, E. C. Martins, J. Meidanis, C. F. Menck, C. Y. Miyaki, D. H. Moon, L. M. Moreira, M. T. Novo, V. K. Okura, M. C. Oliveira, V. R. Oliveira, H. A. Pereira, A. Rossi, J. A. Sena, C. Silva, R. F. de Souza, L. A. Spinola, M. A. Takita, R. E. Tamura, E. C. Teixeira, R. I. Tezza, M. Trindade dos Santos, D. Truffi, S. M. Tsai, F. F. White, J. C. Setubal, and J. P. Kitajima. 2002. Comparison of the genomes of twoXanthomonas pathogens with differing host specificities. Nature 417:459-463. [DOI] [PubMed] [Google Scholar]
18.Daubin, V., M. Gouy, and G. Perriere. 2002. A phylogenomic approach to bacterial phylogeny: evidence of a core of genes sharing a common history. Genome Res. 12:1080-1090. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.de la Fuente, A., J. F. Martin, A. Rodriguez-Garcia, and P. Liras. 2004. Two proteins with ornithine acetyltransferase activity show different functions inStreptomyces clavuligerus: Oat2 modulates clavulanic acid biosynthesis in response to arginine. J. Bacteriol. 186:6501-6507. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Di Giulio, M. 2003. The ancestor of the Bacteria domain was a hyperthermophile. J. Theor. Biol. 224:277-283. [DOI] [PubMed] [Google Scholar]
21.Dyda, F., D. C. Klein, and A. B. Hickman. 2000. GCN5-related N-acetyltransferases: a structural overview. Annu. Rev. Biophys. Biomol. Struct. 29:81-103. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Edgar, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32:1792-1797. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Elkins, J. M., N. J. Kershaw, and C. J. Schofield. 2005. X-ray crystal structure of ornithine acetyltransferase clavulanic acid biosynthesis gene cluster. Biochem. J. 385:565-573. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Errey, J. C., and J. S. Blanchard. 2005. Functional characterization of a novel ArgA fromMycobacterium tuberculosis. J. Bacteriol. 187:3039-3044. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Falb, M., F. Pfeiffer, P. Palm, K. Rodewald, V. Hickmann, J. Tittor, and D. Oesterhelt. 2005. Living with two extremes: conclusions from the genome sequence ofNatronomonas pharaonis. Genome Res. 15:1336-1343. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Fernandez-Murga, M. L., F. Gil-Ortiz, J. Llacer, and V. Rubio. 2004. Arginine biosynthesis inThermotoga maritima: characterization of the arginine-sensitiveN-acetyl-l-glutamate kinase. J. Bacteriol. 186:6142-6149. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Fuerst, J. A. 2005. Intracellular compartmentation in planctomycetes. Annu. Rev. Microbiol. 59:299-328. [DOI] [PubMed] [Google Scholar]
28.Gao, J. L., B. Weissenmayer, A. M. Taylor, J. Thomas-Oates, I. M. Lopez-Lara, and O. Geiger. 2004. Identification of a gene required for the formation of lyso-ornithine lipid, an intermediate in the biosynthesis of ornithine-containing lipids. Mol. Microbiol. 53:1757-1770. [DOI] [PubMed] [Google Scholar]
29.Glansdorff, N. 2000. About the last common ancestor, the universal life-tree and lateral gene transfer: a reappraisal. Mol. Microbiol. 38:177-185. [DOI] [PubMed] [Google Scholar]
30.Glansdorff, N., and Y. Xu. 2004. Phylogeny of extremophiles.In C. Gerday and N. Glansdorff (ed.), Extremophiles. Encyclopedia of life support systems. Eolss Publishers, Oxford, United Kingdom.
31.Glansdorff, N., and Y. Xu. Microbial arginine biosynthesis: pathway, regulation and industrial production.In V. F. Wendisch (ed.), Amino acid biosynthesis: pathways, regulation and metabolic engineering. Microbiology monographs, vol. 5, in press. Springer, Berlin, Germany.
32.Graupner, M., and R. H. White. 2001.Methanococcus jannaschii generates l-proline by cyclization of l-ornithine. J. Bacteriol. 183:5203-5205. [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Guindon, S., and O. Gascuel. 2003. A simple, fast and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst. Biol. 52:696-704. [DOI] [PubMed] [Google Scholar]
34.Haas, D., V. Kurer, and T. Leisinger. 1972.N-Acetylglutamate synthetase ofPseudomonas aeruginosa. An assay in vitro and feedback inhibition by arginine. Eur. J. Biochem. 31:290-295. [DOI] [PubMed] [Google Scholar]
35.Hani, E. K., D. Ng, and V.-L. Chan. 1999. Arginine biosynthesis inCampylobacter jejuni TGH9011: determination of theargCOBD cluster. Can. J. Microbiol. 45:959-969. [PubMed] [Google Scholar]
36.Hou, S., J. H. Saw, K. S. Lee, T. A. Freitas, C. Belisle, Y. Kawarabayasi, S. P. Donachie, A. Pikina, M. Y. Galperin, E. V. Koonin, K. S. Makarova, M. V. Omelchenko, A. Sorokin, Y. I. Wolf, Q. X. Li, Y. S. Keum, S. Campbell, J. Denery, S. Aizawa, S. Shibata, A. Malahoff, and M. Alam. 2004. Genome sequence of the deep-sea gamma-proteobacteriumIdiomarina loihiensis reveals amino acid fermentation as a source of carbon and energy. Proc. Natl. Acad. Sci. USA 101:18036-18041. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Kim, J. H., and R. L. Weiss. 1995. Genetic analysis of interactions between arg-14 and arg-6 products inNeurospora crassa. Mol. Cell 5:461-466. [Google Scholar]
38.Labedan, B., A. Boyen, M. Baetens, D. Charlier, C. Pingguo, R. Cunin, V. Durbecq, N. Glansdorff, G. Herve, C. Legrain, Z. Liang, C. Purcarea, M. Roovers, R. Sanchez, T. L. Toong, M. Van de Casteele, F. Van Vliet, Y. Xu, and Y. F. Zhang. 1999. The evolutionary history of carbamoyltransferases: a complex set of paralogous genes was already present in the last universal common ancestor. J. Mol. Evol. 49:461-473. [DOI] [PubMed] [Google Scholar]
39.Lakshminarayan, M. I., E. V. Koonin, and L. Aravind. 2004. Evolution of bacterial RNA polymerase: implications for large-scale bacterial phylogeny, domain accretion, and horizontal gene transfer. Gene 335:73-88. [DOI] [PubMed] [Google Scholar]
40.Ledwidge, R., and J. S. Blanchard. 1999. The dual biosynthetic capability of N-acetylornithine aminotransferase in arginine and lysine biosynthesis. Biochemistry 38:3019-3024. [DOI] [PubMed] [Google Scholar]
41.Lu, C.-D. 2006. Pathways and regulation of bacterial arginine metabolism and perspectives for obtaining arginine overproducing strains. Appl. Microbiol. Biotechnol. 70:261-272. [DOI] [PubMed] [Google Scholar]
42.Makarova, K. S., L. Aravind, M. Y. Galperin, N. V. Grishin, R. L. Tatusov, Y. I. Wolf, and E. V. Koonin. 1999. Comparative genomics of Archaea (Euryarchaeota): evolution of conserved protein families, the stable core, and the variable shell. Genome Res. 9:608-628. [PubMed] [Google Scholar]
43.Marc, F., P. Weigel, C. Legrain, Y. Almeras, M. Santrot, and V. Sakanyan. 2000. Characterization and kinetic mechanism of mono- and bifunctional ornithine acetyltransferases from thermophilic microorganisms. Eur. J. Biochem. 267:5217-5226. [DOI] [PubMed] [Google Scholar]
44.Marc, F., P. Weigel, C. Legrain, N. Glansdorff, and V. Sakamyan. 2001. An invariant threonine is involved in self-catalyzed cleavage of the precursor protein for ornithine acetyltransferase. J. Biol. Chem. 276:25404-25410. [DOI] [PubMed] [Google Scholar]
45.Martin, P. R., and M. H. Mulks. 1992. Sequence analysis and complementation studies of theargJ gene encoding ornithine acetyltransferase fromNeisseria gonorrhoeae. J. Bacteriol. 174:2694-2701. [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Medigue, C., E. Krin, G. Pascal, V. Barbe, A. Bernsel, P. N. Bertin, F. Cheung, S. Cruveiller, S. D'Amico, A. Duilio, G. Fang, G. Feller, C. Ho, S. Mangenot, G. Marino, J. Nilsson, E. Parrilli, E. P. Rocha, Z. Rouy, A. Sekowska, M. L. Tutino, D. Vallenet, G. von Heijne, and A. Danchin. 2005. Coping with cold: the genome of the versatile marine Antarctica bacteriumPseudoalteromonas haloplanktis TAC125. Genome Res. 15:1325-1335. [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Methe, B. A., K. E. Nelson, J. W. Deming, B. Momen, E. Melamud, X. Zhang, J. Moult, R. Madupu, W. C. Nelson, R. J. Dodson, L. M Brinkac, S. C. Daugherty, A. S. Durkin, R. T. DeBoy, J. F. Kolonay, S. A. Sullivan, L. Zhou, T. M. Davidsen, M. Wu, A. L. Huston, M. Lewis, B. Weaver, J. F. Weidman, H. Khouri, T. R. Utterback, T. V Feldblyum, and C. M. Fraser. 2005. The psychrophilic lifestyle as revealed by the genome sequence ofColwellia psychrerythraea 34H through genomic and proteomic analyses. Proc. Natl. Acad. Sci. USA 102:10913-10918. [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Miyasaki, J., N. Kobashi, T. Fujii, M. Nishiyama, and H. Yamane. 2002. Characterization oflysK gene as anargE homolog inThermus thermophilus HB27. FEBS Lett. 512:269-274. [DOI] [PubMed] [Google Scholar]
49.Miyasaki, J., N. Kobashi, M. Nishiyama, and H. Yamane. 2001. Functional and evolutionary relationship between arginine biosynthesis and prokaryotic lysine biosynthesis through α-aminoadipate. J. Bacteriol. 183:5067-5073. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Morizono, H., J. Cabrera-Luque, D. Shi, R. Gallegos, S. Yamaguchi, X. Yu, N. Allewell, M. Malamy, and M. Tuchman. 2006. Acetylornithine transcarbamylase: a novel enzyme in arginine biosynthesis. J. Bacteriol. 188:2974-2982. [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Morizono, H., L. Caldovic, D. Shi, and M. Tuchman. 2004. MammalianN-acetylglutamate synthase. Mol. Gen. Metab. 81(Suppl. 1):S4-S11. [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Naumoff, D., Y. Xu, N. Glansdorff, and B. Labedan. 2004. Retrieving sequences of enzymes experimentally characterized but erroneously annotated: the case of the putrescine carbamoyltransferase. BMC Genomics 5:52. [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Neuwald, A. F., and D. Landsman. 1997. GCN5-related histoneN-acetyltransferases belong to a diverse superfamily that includes the yeast SPT10 protein. Trends Biochem. Sci. 22:154-155. [DOI] [PubMed] [Google Scholar]
54.Nishida, H., M. Nishiyama, N. Kobashi, T. Kosuge, T. Hoshino, and H. Yamane. 1999. A prokaryotic gene cluster involved in synthesis of lysine through the amino adipate pathway: a key to the evolution of amino acid biosynthesis. Genome Res. 9:1175-1183. [DOI] [PubMed] [Google Scholar]
55.O'Reilly, M., and K. M Devine. 1994. Sequence analysis of the citrulline biosynthetic operonargC-F from Bacillus subtilis. Microbiology 140:1023-1025. [DOI] [PubMed] [Google Scholar]
56.Pauwels, K., A. Abadjieva, P. Hilven, A. Stankiewicz, and M. Crabeel. 2003. The N-acetylglutamate synthase/N-acetylglutamate kinase metabolon ofSaccharomyces cerevisiae allows co-ordinated feedback regulation of the first two steps in arginine biosynthesis. Eur. J. Biochem. 270:1014-1024. [DOI] [PubMed] [Google Scholar]
57.Peet, R. C., and N. J. Panopoulos. 1987. Ornithine carbamoyltransferase genes and phaseolotoxin immunity inP. syringae. EMBO J. 1:3585-3591. [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Ramon-Maiques, S., A. Marina, F. Gil-Ortiz, I. Fita, and V. Rubio. 2002. Structure of acetylglutamate kinase, a key enzyme for arginine biosynthesis and a prototype for the amino acid kinase enzyme family, during catalysis. Structure 10:329-342. [DOI] [PubMed] [Google Scholar]
59.Ramon-Maiques, S., M. L. Fernandez-Murga, F. Gil-Ortiz, A. Vagin, I. Fita, and V. Rubio. 2006. Structural bases of feed-back control of arginine biosynthesis, revealed by the structures of two hexameric N-acetylglutamate kinases, fromThermotoga maritima andPseudomonas aeruginosa. J. Mol. Biol. 356:695-713. [DOI] [PubMed] [Google Scholar]
60.Sakai, H., M. N. Vassylyeva, T. Matsuura, S. Sekine, K. Gotoh, M. Nishiyama, T. Terada, M. Shirouzu, S. Kuramitsu, D. G. Vassylyev, and S. Yokoyama. 2003. Crystal structure of a lysine biosynthetic enzyme, LysX, fromThermus thermophilus HB8. J. Mol. Biol. 332:729-740. [DOI] [PubMed] [Google Scholar]
61.Sakanyan, V., D. Charlier, C. Legrain, A. Kochikyan, I. Mett, A. Pierard, and N. Glansdorff. 1993. Primary structure, partial purification and regulation of key enzymes of the acetyl cycle of arginine biosynthesis inBacillus stearothermophilus: dual function of ornithine acetyltransferase. J. Gen. Microbiol. 139:393-402. [DOI] [PubMed] [Google Scholar]
62.Sakanyan, V., C. Legrain, D. Charlier, and A. Kochikyan. 1993. N-acetylglutamate-5-phosphotransferase of the thermophilic bacteriumBacillus stearothermophilus: nucleotide sequence of the gene and enzyme characterization. Genetika 29:556-570. (In Russian.) [PubMed] [Google Scholar]
63.Sakanyan, V., P. Petrosyan, M. Lecocq, A. Boyen, C. Legrain, M. Demarez, J. N. Hallet, and N. Glansdorff. 1996. Genes and enzymes of the acetyl cycle of arginine biosynthesis inCorynebacterium glutamicum: enzyme evolution in the early steps of the arginine pathway. Microbiology 142:99-108. [DOI] [PubMed] [Google Scholar]
64.Shi, D., R. Gallegos, J. Deponte III, H. Morizono, X. Yu, N. M. Allewell, M. Malamy, and M. Tuchman. 2002. Crystal structure of a transcarbamylase-like protein from the anaerobic bacteriumBacteroides fragilis at 2.0 Å resolution. J. Mol. Biol. 320:899-908. [DOI] [PubMed] [Google Scholar]
65.Shi, D., H. Morizono, Y. Xiaolin, L. Roth, L. Caldovic, N. M. Allewell, M. H. Malamy, and M. Tuchman. 2005. Crystal structure of aN-acetylornithine transcarbamylase fromXanthomonas campestris: a novel enzyme in a new arginine biosynthetic pathway found in several eubacteria. J. Biol. Chem. 280:14366-14369. [DOI] [PubMed] [Google Scholar]
66.Shi, D., Y. Xiaolin, L. Roth, H. Morizono, Y. Hathout, N. M. Allewell, and M. Tuchman. 2005. Expression, purification, crystallization and preliminary X-ray crystallographic studies of a novel acetylcitrulline deacetylase fromXanthomonas campestris. Acta Crystallogr. F 61:676-679. [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Shi, D., H. Morizono, J. Cabrera-Luque, Y. Xiaolin, L. Roth, M. H. Malamy, N. M. Allewell, and M. Tuchman. 2006. Structure and catalytic mechanism of a novel N-succinyl-l-ornithine transcarbamylase in arginine biosynthesis ofBacteroides fragilis. J. Biol. Chem. 281:20623-20631. [DOI] [PubMed] [Google Scholar]
68.Shi, D., Y. Xiaolin, L. Roth, H. Morizono, M. Tuchman, and N. M. Allewell. 2006. Structures of N-acetylornithine transcarbamoylase fromXanthomonas campestris complexed with substrates and substrate analogs imply mechanisms for substrate biding and catalysis. Proteins 64:532-542. [DOI] [PubMed] [Google Scholar]
69.Slocum, R. D. 2005. Genes, enzymes and regulation of arginine biosynthesis in plants. Plant Physiol. Biochem. 43:729-745. [DOI] [PubMed] [Google Scholar]
70.Udaka, S., and S. Kinoshita. 1958. Studies on l-ornithine fermentation. I. The biosynthetic pathway of l-ornithine inMicrococcus glutamicus. J. Gen. Appl. Microbiol. 4:283-288. [Google Scholar]
71.Udaka, S. 1966. Pathway-specific pattern of control of arginine biosynthesis in bacteria. J. Bacteriol. 91:617-621. [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Van de Casteele, M., M. Demarez, C. Legrain, N. Glansdorff, and A. Pierard. 1990. Pathways of arginine biosynthesis in extreme thermophilic archaeaobacteria and eubacteria. J. Gen. Microbiol. 136:1177-1183. [Google Scholar]
73.Vander Wauven, C., and V. Stalon. 1985. Occurrence of succinyl derivatives in the catabolism of arginine inPseudomonas cepacia. J. Bacteriol. 164:882-886. [DOI] [PMC free article] [PubMed] [Google Scholar]
74.Velasco, A. M., J. I. Leguina, and A. Lazcano. 2002. Molecular evolution of the lysine biosynthetic pathways. J. Mol. Evol. 55:445-459. [DOI] [PubMed] [Google Scholar]
75.Vetting, M. W., L. P. S. de Carvalho, M. Yu, S. S. Hedge, S. Magnet, S. L. Roderick, and J. S. Blanchard. 2005. Structure and functions of the GNAT superfamily of acetyltransferases. Arch. Biochem. Biophys. 433:212-226. [DOI] [PubMed] [Google Scholar]
76.Weigel, P., F. Marc, S. Simon, and V. Sakanyan. 2002. OrnithineN-acetyltransferase and arginine biosynthesis in thermophilic bacteria. Recent Res. Dev. Microbiol. 6:95-106. [Google Scholar]
77.Winkelmann, G. 2001. Microbial siderophore-mediated transport. Biochem. Soc. Trans. 30:691-696. [DOI] [PubMed] [Google Scholar]
78.Woese, C. R. 1998. The universal ancestor. Proc. Natl. Acad. Sci. USA 95:6854-6859. [DOI] [PMC free article] [PubMed] [Google Scholar]
79.Xu, Y., and N. Glansdorff. 2002. Was our ancestor a hyperthermophilic prokaryote? Comp. Biochem. Physiol. A 133:677-688. [DOI] [PubMed] [Google Scholar]
80.Xu, Y., and N. Glansdorff. 2007. Lessons from extremophiles: conjectures on early life evolution and border conditions of life, p. 409-421.In C. Gerday and N. Glansdorff (ed.), Physiology and biochemistry of extremophiles. ASM Press, Washington, DC.
81.Xu, Y., N. Glansdorff, and B. Labedan. 2006. Bioinformatic analysis of an unusual gene-enzyme relationship in the arginine biosynthetic pathway among marine gamma proteobacteria: implications concerning the formation ofN-acetylated intermediates in prokaryotes. BMC Genomics 7:4. [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Xu, Y., Z. Liang, C. Legrain, H. J. Ruger, and N. Glansdorff. 2000. Evolution of arginine biosynthesis in the bacterial domain: novel gene-enzyme relationship from psychrophilicMoritella strains (Vibrionaceae) and the evolutionary significance ofN-α-acetyl ornithinase. J. Bacteriol. 182:1609-1615. [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Yu, Y. G., G. E. Turner, and R. L. Weiss. 1996. Acetylglutamate synthase fromNeurospora crassa: structure and regulation of expression. Mol. Microbiol. 22:545-554. [DOI] [PubMed] [Google Scholar]

Movatterモバイル変換

PERMALINK

Surprising Arginine Biosynthesis: a Reappraisal of the Enzymology and Evolution of the Pathway in Microorganisms

Ying Xu

Bernard Labedan

Nicolas Glansdorff

Abstract

INTRODUCTION

FIG. 1.