The
N
-acetylglutamate synthase/
N
-acetylglutamate kinase
metabolon of
Saccharomyces cerevisiae
allows co-ordinated
feedback regulation of the first two steps in arginine biosynthesis
Katia Pauwels, Agnes Abadjieva, Pierre Hilven, Anna Stankiewicz and Marjolaine Crabeel
Department of Genetics and Microbiology of the Vrije Universiteit Brussel, Brussels, Belgium
In Saccharomyces cerevisiae, which uses the nonlinear
pathway of arginine biosynthesis, the first two enzymes,
N-acetylglutamate synthase (NAGS) and N-acetylglutamate
kinase (NAGK), are controlled by feedback inhibition. We
have previously shown that NAGS and NAGK associate in
a complex, essential to synthase activity and protein level
[Abadjieva, A., Pauwels, K., Hilven, P. & Crabeel, M. (2001)
J. Biol. Chem. 276, 42869–42880].
The NAGKs of ascomycetes possess, in addition to the
catalytic domain that is shared by all other NAGKs and
whose structure has been determined, a C-terminal domain
of unknown function and structure. Exploring the role of
these two domains in the synthase/kinase interaction, we
demonstrate that the ascomycete-specific domain is required
to maintain synthase activity and protein level.
Previous results had suggested a participation of the third
enzyme of the pathway, N-acetylglutamylphosphate reduc-
tase, in the metabolon. Here, genetic analyses conducted in
yeast at physiological level, or in a heterologous background,
clearly demonstrate that the reductase is dispensable for

using the cyclic pathway of ornithine synthesis [1,2].
The linear pattern of ornithine synthesis is found in
Escherichia coli and some other bacteria and archea [1–5].
The cyclic pattern is more widespread among the procary-
otes [6–13], and it is observed in all investigated ascomyce-
tes, including Candida utilis [14], Saccharomyces cerevisiae
[15], Neurospora crassa [2], and in Chlamydomonas algae
[16]. In the fungi, ornithine synthesis proceeds entirely in the
mitochondria [17,18].
Control of the metabolic flux through a biosynthetic
pathway usually occurs at the level of the first committed
step and is often mediated by the end product of the
pathway. This classical mechanism operates in organisms
using the linear pathway of arginine synthesis: arginine
exerts feedback inhibition on N-acetylglutamate synthase in
E. coli and Salmonella typhimurium [19–21]. In pathways
where acetylglutamate is regenerated, the second enzyme of
arginine biosynthesis, N-acetylglutamate kinase (NAGK)
(EC 2.7.2.8) becomes the main controlling step. Feedback
inhibition of the kinase by arginine has been demonstrated
in several bacteria [7,22,23]. Yet, metabolic control should
occur on the production of acetylglutamate, regardless of its
origin. Therefore, feedback inhibition on both the synthase
and the kinase is believed to be general for organisms using
Correspondence to M. Crabeel, Department of Genetics and
Microbiology of the Vrije Universiteit Brussel, c/o CERIA-COOVI,
Emile Gryson avenue 1, B-1070 Brussels, Belgium.
Fax: + 32 2 526 72 73, Tel.: + 32 2 526 72 84,
E-mail:
Abbreviations:NAGS,N-acetylglutamate synthase; NAGK,

J. Chung, C. McKinstry, M. Karaman and G. Turner,
University of California, Los Angeles, CA, USA, personal
communication). Similar data in yeast were independently
obtained by our group [32]. An increase in synthase activity,
expected to result from higher copy numbers of its structural
gene ARG2, has only been observed with a parallel increase
in the ARG5,6 gene copy number. The yeast synthase/kinase
interaction was demonstrated by coimmunoprecipitation
methods [32].
The physical participation of reductase, the second mat-
urated gene product of ARG5,6, to the synthase/kinase com-
plex, has not been provenso far. Hence, it is not clear whether
synthase activity and protein level require reductase. How-
ever, the existence of mutations in the reductase-encoding
domain of the N. crassa arg-6 gene, which affect synthase
activity, suggests a possible role for the reductase [28,31].
Moreover, increasing the copy-number of a synthetic gene,
only encoding the kinase domain of S. cerevisiae ARG5,6
gene, is not sufficient to increase the activity of yeast NAGS
when coexpressed with high copy-number of ARG2 [32].
Another remarkable result, concerning the regulation of
the first enzymes of the arginine pathway, has been reported
by the team of R. L. Weiss. A series of ornithine-over-
producing N. crassa mutants [33], were mapped to the
N-terminus of N-acetylglutamate kinase and shown to bear
F81L modifications. The data suggest that this single
amino-acid modification of the kinase might result in the
deregulation of the first two enzyme activities of the arginine
pathway, leading to the hypothesis of a co-ordinated
feedback control (R. L. Weiss, S. K. Chae, J. Chung,

pombe, N. crassa,andCandida albicans, have two specific
features: (a) they are encoded together with NAGPR as a
bi-functional precursor protein that is processed into two
distinct enzymes in the mitochondria, and (b) they possess
an extra region of about 200 amino acids at their
C-terminus, that we call the ascomycete-specific domain
(ASD) [29,30]. It is tempting to speculate that the
ascomycete-specific domain (ASD) of the kinase might
play a role in formation of the synthase/kinase protein
complex.
This work investigates three important unsolved ques-
tions related to the structure and function of the yeast
NAGS/NAGK metabolon. We analyse (a) the role of the
reductase in the activity and protein level of the synthase, (b)
the role of the ASD of the kinase in its interaction with the
synthase, and (c) the significance of the yeast NAGS/
NAGK metabolon in terms of its co-ordinated feedback
regulation by arginine.
Experimental procedures
Strains and growth conditions
S. cerevisiae. The wild-type strain of this laboratory is
S1278b (Mat a). MG471 (Mat a, ura3–471) was directly
derived from S1278b by M. Grenson, Universiteit
Brussel, Belgium. The strains YeBR5 (Mat a, ura3–471,
Darg5::gen
R
), YeBR6 (Mat a, ura3–471, Darg6::gen
R
,
arg5

procedure allowed scarless removal of the NAGK encoding
ARG6 region from the chromosomal ARG5,6 gene (deletion
from amino acid 84–493 in the ORF encoding the kinase/
reductase precursor). The resulting ura3
–
, Darg6 mutant
strain can be restored to prototrophy by plasmid pYB7,
expressing ARG6 from a GAL promoter. This confirms
that, as expected, SA2 expresses active NAGPR from the
remaining ARG5 region of the ARG5,6 gene.
All yeast strains were grown at 30 °ConM.ammedium,
a minimal medium containing 0.02
M
(NH
4
)
2
SO
4
,3%
glucose, vitamins, and trace minerals [41]. Where required,
uracil,
L
-histidine or
L
-arginine was added to a concentra-
tion of 25 lgÆmL
)1
. Genes which are transcriptionally
controlled by the GAL promoter were induced by growing

of cells at D
600
of 0.250 grown on rich medium plus
ampicilline, were harvested by centrifugation, washed and
resuspended in minimal medium to a concentration of
10
10
cellsÆmL
)1
. Drops of 10 lL of 10-fold serial dilutions
(from 10
10
cellsÆmL
)1
to 10
5
cellsÆmL
)1
) were spotted on
minimal medium with or without arginine (100 lgÆmL
)1
),
and with or without IPTG (1 m
M
). Sets of four plates were
incubated at 37 °C, 30 °Cor25°C.
Oligonucleotides
BY4, BY5: [32], HP72 ¼ GTCTCACAACAACAATTGG
CTGTGATCAAGGTG. HP73 ¼ CACCTTGATCACA
GCTAATTGTTGTTGTGAGAC. HP79 ¼ CACACG

restriction sites for classical cloning in the pYX223 vector
(from R&D systems). The latter is a 2 micron-based yeast–
E. coli shuttle vector, bearing HIS3 as selection marker, and
in which the expression of the inserted genes is put under the
control of a GAL promoter. The BY4/HP73 and HP72/
BY5 primer pairs were used to construct pHP17, BY4/
HP79, and HP80/BY5 for pHP21 and BY4/HP81 and
HP82/BY5 for pHP22.
Plasmids of the pYK series were all derived from the
E. coli expression vector pTrc99a (Pharmacia) and contain
different insertions, all obtained by PCR amplification. The
inserted fragments allow the expression of the ORF under
the transcriptional control of the IPTG-inducible strong
bacterial trp-lac promoter and under the translational
control of an appropriate Shine–Dalgarno sequence.
Plasmids pYK1 expresses the ARG6 ORF, cloned as an
NcoI–HindIII fragment amplified using K1 and K2 as
primers and plasmid pYB3 as a template. Plasmid pYK7
expresses the ARG2 ORF, cloned as a NcoI–PstI fragment
(primers AA29 and K8 and pYB2 as a template). With
primer AA29, a tag of six histidine codons is fused in frame
to the C-terminus of the ARG2 ORF for immunodetection
of the enzyme. Plasmid pYK8 was obtained by inserting
the ARG6 ORF and its trp-lac promoter (from position
)115) as a PstI–HindIII fragment (primers K4/K2, tem-
plate pYK1) into plasmid pYK7. The artificial operon of
plasmid pYK11 expresses a bi-cistronic ARG5/ARG6
mRNA under the control of the trp-lac promoter and
was obtained by inserting a HindIII fragment (primers
K10/K12, template pYK3), containing the reductase

Table 1. Main features of the plasmids used in this work.
Plasmids Cloning vector
Origin of
insert Nature of insert Expressed protein
pYB2 pYX213 (2l, URA3) S1278b PromoterGAL ò ARG2 ORF-HAtag (32) WT NAGS-HA
pYB3 pYX223 (2l, HIS3) S1278b PromoterGAL ò ARG5,6 ORF (32) WT NAGK + WT NAGPR
(amino acids 1–863)
pYB7 pYX223 (2l, HIS3) S1278b PromoterGAL ò ARG6 (32) WT NAGK (amino acids 1–537)
pYB8 pYX223 (2l, HIS3) S1278b PromoterGAL ò ARG5 (32) WT NAGPR (amino acids 1–38 +
amino acids 494–863)
pHP17 pYX223 (2l, HIS3) S1278b PromoterGAL ò ARG5,6 ORF (F99L) FB
R
NAGK + WT NAGPR
pHP21 pYX223 (2l, HIS3) S1278b PromoterGAL ò ARG5,6 ORF
(Damino acids 355–493)
NAGK (DASD) + WT NAGPR
pHP22 pYX223 (2l, HIS3) S1278b PromoterGAL ò ARG5,6 ORF
(Damino acids 85–347)
NAGK (DCD) + WT NAGPR
p238 YCp50 (ARS-CEN, S288c GCN4 (4 uORFs untranslated)
a
Constitutive expression of Gcn4p
URA3)
pYK1 pTrc99a S1278b PromoterTrc ò ARG6 WT NAGK (amino acids 58 to 51)
pYK7 pTrc99a S1278b PromoterTrc ò ARG2 ORF-HIS6tag WT NAGS-HIS6
pYK8 pTrc99a S1278b PromoterTrc ò ARG2 ORF-HIS6tag +
PromoterTrc ò ARG6
WT NAGS-HIS6 + WT NAGK
(amino acids 58–513)
pYK11 pTrc99a S1278b PromoterTrc ò ARG2 ORF-HIS6tag +

Standard deviations generally did not exceed 15%.
Western blots
A standard chemiluminescence Western blotting protocol
(Roche) was used to analyse the yeast NAGS expressed in
E. coli from plasmids pYK7, pYK8, and pYK11. Equal
amounts of total proteins of the different crude extracts were
separated by SDS/PAGE on 12% gels, and then blotted on
an ECL Hybond nitrocellulose membrane (Amersham
Pharmacia Biotech) in transfer buffer [25 m
M
Tris,
192 m
M
glycine, 20% (v/v) methanol] using a Mini PRO-
TEAN 3 blotting cell (Bio-Rad). Specific primary mouse
anti-HIS Ig (Santa Cruz Biotechnology) (0.1 ngÆmL
)1
)and
40 UÆmL
)1
peroxidase-labelled secondary antibody (Roche)
were used to detect the tagged synthase protein. Chemilu-
minescence was monitored by autoradiography. Detection
of Haemaglutinin (HA)-tagged NAGS, expressed by the
pYB2 plasmid in yeast cells, was as described previously [32].
Results
At physiological levels, the presence
of N-acetylglutamyl phosphate reductase
is dispensable to synthase activity
In order to determine the influence of N-acetylglutamyl

Gcn4p transcriptional transactivator (Table 2).
Synthase activity was assayed in crude extracts of arginine
starved SS1, YeBR5, SA2 and YeBR6, with and without
the plasmid p238 (Table 2). No synthase activity was
detectable in absence of the kinase (SA2 and YeBR6 vs.
SS1). In contrast, the absence of reductase did not affect
considerably the synthase activity, though a small decrease
was observed (YeBR5 vs. SS1). These data demonstrate
that, at physiological level, the synthase activity requires the
presence of the kinase, and that the additional presence of
the reductase is dispensable.
Activity and protein level of the yeast synthase
expressed in
E. coli
, require the coexpression
of the yeast kinase but not of the yeast reductase
The E. coli strain XA4 (argA
–
), which is defective in
N-acetylglutamate synthase, cannot be restored to arginine
prototrophy by a trp-lac-promoter-driven expression of the
Table 2. Physiological levels of the N-acetylglutamate synthase in strains bearing different deletions in the ARG5,6 gene.
Strain Relevant genotype
Status of
NAGS activity (nmolÆmin
)1
Æmg
)1
protein)NAGK NAGPR
SS1 ARG5,6, Darg3 + + 2.2

constructions, including the empty vector pTrc99a, are
transformed in strain XA4. SDS/PAGE/Coomassie Blue
analysis and kinase activity assays confirmed that XA4
(pYK8) and XA4(pYK11) are over-expressing functional
kinase protein (data not shown). Beside the kinase protein,
XA4 (pYK11) expresses the reductase protein, however, in
lower amounts. The functionality of the reductase, encoded
by pYK11, was verified by complementation of the
reductase deficient E. coli strain XC33 (argC
–
) (data not
shown).
First, all four plasmids were tested for their efficiency to
complement the argA
–
deficiency of the XA4 strain, using
spot tests of serial dilutions incubated at 37 °C. Under
noninducing conditions (Fig. 2A), pYK8 and pYK11 (both
expressing the kinase protein) allow growth of the arginine-
deficient mutant in the absence of arginine. On the other
hand, plasmids pYK7 and the empty vector pTrc99a (both
lacking the yeast kinase ORF) were completely unable to
complement the mutation. These data demonstrate that the
presence of the kinase is essential to yeast synthase activity
while the additional presence of the reductase (pYK11 vs.
pYK8) does not improve complementation. The observa-
tion that complementation is even slightly lower in the
presence of the reductase, could be due to a lower copy
number of pYK11, which is larger than pYK8. Unexpect-
edly, expression of pYK8 and pYK11 under induced

the reductase, showed a slight decrease in synthase activity,
which can presumably be ascribed to a lower plasmid copy
number.
To test whether the absence of synthase activity is the
result of low levels of NAGS protein, an immunoWestern
Table 3. Yeast N-acetylglutamate synthase specific activity in the XA4
(argA
–
) E. coli background.
Plasmid
Yeast enzymes
expressed
NAGS activity
after IPTG induction
(nmolÆmin
)1
Æmg
)1
protein)
pTrc99A none <0.2
a
pYK7 NAGS <0.2
a
pYK8 NAGS + NAGK 44
pYK11 NAGS + NAGK + NAGPR 30
a
Below detection.
Fig. 2. Spot growth tests of the E. coli strain XA4(argA
–
) transformed with various plasmids as indicated. In each row, from left to right, 10 lLof

The ascomycete-specific domain of
N
-acetylglutamate
kinase is required to maintain
N
-acetylglutamate
synthase activity and protein level
Yeast N-acetylglutamate kinase consists of two distinguish-
able domains. The N-terminal domain is conserved in both
eucaryotes and procaryotes and is therefore inferred to be
the catalytic active domain (CD). The C-terminal domain is
specific to ascomycetes (ASD). It extends from about amino
acid348toaresiduelocatedbetweenaminoacid510and
540, the region in which the kinase/reductase precursor is
maturated [29]. We addressed the question whether the two
kinase domains are needed to observe synthase activity and
stability. By inference, this would indicate a role for each
domainintheassociationoftheNAGS/NAGKina
complex.
For this experiment, new high copy number plasmids
were derived from pYB3, each lacking one of the kinase
domains. Plasmid pYB3 encodes the full length ARG5,6
gene, plasmid pHP21 is truncated for the ascomycete
specific domain of the kinase (Daa355–493) and plasmid
pHP22 is truncated for the catalytic domain of the kinase
(Daa85–347). The functionality of the kinase protein
encoded by those plasmids was assessed by transforming
the plasmids in the Darg5,6 genetic background of strain
KA42 and measuring kinase activity. As expected
KA42(pHP22) lacks any kinase activity while

used as a positive control. Only when the gels were
deliberately overloaded, did synthase become detectable in
the extracts from strains bearing pHP21 and pHP22, yet in
amounts comparable to the basal level produced in the
negative control (Fig. 4B).
The results demonstrate that the ascomycete-specific
domain of the kinase is required for accumulation of the
synthase. However, if this domain is assumed to be
Fig. 3. NAGS detection by immunoWestern blot analysis of total pro-
tein extracts of E. coli strain transformed with plasmid pYK7 expressing
His
6
-tagged yeast N-acetylglutamate synthase (NAGS), pYK8 expres-
sing His
6
-tagged NAGS and N-acetylglutamate kinase (NAGK) or
pYK11 expressing His
6
-tagged NAGS, NAGK and N-acetylglutamyl
phosphate reductase. Plasmid pTrc99a is the corresponding empty
cloning vector. MM, molecular mass markers. The arrow indicates the
protein band corresponding to NAGS.
Table 4. N-acetylglutamate synthase specific activity in strains coex-
pressing promoter GAL-driven ARG2 and ARG5,6 genes: effect of
domain deletions in the N-acetylglutamate kinase.
Strain
Status
of NAGK
NAGS activity
(nmolÆmin

corresponds to the phenylalanine 99 in S. cerevisiae.We
constructed the yeast kinase ARG5,6 F99L mutant in a
vector with a GAL promoter, yielding plasmid pHP17.
Plasmid pYB2, encoding the yeast synthase, was cotrans-
formed with pHP17 in the strain YeBR6. YeBR6
(pYB2 + pYB3), over-expressing both the wild-type kinase
and synthase, was used as a reference strain. The trans-
formants were grown on galactose medium and N-acetyl-
glutamate kinase activity in cell extracts was assayed in the
presence of increasing arginine concentrations. Figure 5A
compares the arginine inhibition curves of the wild-type and
F99L mutant kinases. The arginine concentration required
to inhibit 50% of the activity of the wild-type kinase (I
0.5
)is
0.1 m
M
, a value that is comparable to an I
0.5
of 0.05 m
M
Fig. 4. ImmunoWestern blot detection of N-acetylglutamate synthase in
total protein extracts of yeast strain 14S31b bearing plasmid pairs as
indicated above the lanes. pYB2 expresses a haemaglutinin-tagged
NAGS, pYB3 expresses the wild-type NAGK/NAGPR, pHP21 and
pHP22 are derived from pYB3 and, respectively, lack the ascomycete-
specific domain and the catalytic domain of the kinase encoding
region, pHP17, also derived from pYB3, bears the F99L modification
in NAGK. pYX213 and pYX223 are the empty cloning vectors. (A)
Equal amounts of total protein were loaded in lanes 1–4, and double

of 10 m
M
).
Synthase activity and its arginine sensitivity were also
assayed using the same extracts (Fig. 5B). In the presence of
the wild-type kinase, 0.015 m
M
arginine is required to reach
I
0.5
of the synthase. This value corresponds with the
I
0.5
-value of 0.02 m
M
published by Wipf and Leisinger
[25]. It is noticeable that the yeast synthase is 10 times more
sensitive to feedback inhibition by arginine than the
N. crassa synthase (50% inhibition at 0.16 m
M
[28]). When
coexpressed with the F99L mutant kinase, the synthase
behaves quite differently than when coexpressed with the
wild-type kinase. The synthase specific activity is reduced
fivefold and, like the mutant kinase, the enzyme becomes
much less sensitive to arginine feedback inhibition (I
0.5
of
0.75 m
M

vity in extracts was assayed in the presence of increasing
arginine concentrations. The results are presented in Fig. 6.
In extracts of KA44(pYB2 + pYB3), the wild-type
kinase proved to be sensitive to arginine inhibition. The
inhibition curve displays a normal hyperbolic shape and the
I
0.5
-value of 0.26 m
M
arginine in the illustrated experiment
(Fig. 6) is comparable to the I
0.5
of 0,1 m
M
measured with
YeBR6(pYB2 + pYB3) extracts (Fig. 5A). In fact, three
similar, less detailed experiments (data not shown), display
an I
0.5
-value closer to 0.1 m
M
. Interestingly, the apparent
affinity of the kinase for the feedback inhibitor is markedly
lower when the synthase is absent (I
0.5
-value of 1.5 m
M
). In
addition, the inhibition curve becomes reproducibly sigmo-
idal. These data show that the kinase requires an interaction

kinase is inefficiently targeted to the yeast mitochondria in
the absence of reductase.
The present data in E. coli further show that no synthase
protein is detectable in the absence of kinase, a situation
similar to the one observed previously in yeast. We attribute
this drastic reduction in steady state concentration of the
protein to an instability of the yeast synthase when not
associated to the kinase. Because it is also observed in the
heterologous E. coli background, this apparent instability is
likely to be an intrinsic feature of the protein, rather than to
result from of a yeast specific degradation process. Alter-
natively, the uncomplexed synthase might present structural
features rendering it susceptible to proteolytic degradation
Fig. 6. Feedback inhibition by arginine of yeast N-acetylglutamate
kinaseactivityinextractsofDarg2 strain KA44(pYB2 + pYB3) (d)
and KA44(pYX213 + pYB3) (s), after growth on galactose medium
supplemented with arginine.
1022 K. Pauwels et al. (Eur. J. Biochem. 270) Ó FEBS 2003
in general. In any case, the lack of synthase protein in the
absence of kinase has been observed with expression
systems using totally different promoters and translation
initiation signals. Therefore, the hypothesis that it results
from an effect on transcription or translation can be
reasonably excluded.
The data in hand today do not allow to tell if the lack of
synthase activity in the absence of kinase fully correlates
with the physical disappearance of the enzyme, or if inactive
free synthase can subsist. However, as discussed below, the
kinetic properties of the synthase are likely to be modulated
by its association with the kinase.

results show that the yeast mutant kinase is feedback
resistant as well. In comparison to the wild-type yeast
kinase, 100 times more arginine is required to reach half-
inhibition of the F99L yeast mutant kinase. Our results
illustrate further that feedback regulation of the wild-type
yeast synthase is strongly dependent upon the presence of a
normally regulated kinase. In the presence of the wild-type
kinase, the synthase is fully inhibited by 0.1 m
M
arginine,
while 10 m
M
arginine is required to inhibit completely the
synthase activity when the partner is a feedback-resistant
mutant kinase. Moreover, the kinetic properties of the
synthase appear dependent upon its association with the
kinase. Indeed, in the context of the mutated kinase,
the synthase specific activity was reduced by 80% while the
amount of enzyme remained unchanged.
Contrasting with the strict requirement of NAGK for
NAGS activity, the absence of NAGS increased the activity
of NAGK by approximately twofold, possibly reflecting
inhibition of NAGK in the NAGS/NAGK complex. The
presence of NAGS had also the effect of rendering
hyperbolic the inhibition of the kinase by arginine, whereas
in absence of NAGS the inhibition was sigmoidal and
exhibited an increased I
0.5
-value, strongly suggesting that
more than one site for arginine has to be occupied to

of the Myxococcus xanthus argE gene. J. Bacteriol. 180, 6412–
6414.
4. Van de Casteele, M., Demarez, M., Legrain, C., Glansdorff, N. &
Pie
´
rard, A. (1990) Pathways of arginine biosynthesis in extreme
thermophilic archaeo- and eubacteria. J. Gen. Microbiol. 136,
1177–1183.
5. Xu, Y., Liang, Z., Legrain, C., Ruger, H. & Glansdorff, N. (2000)
Evolution of arginine biosynthesis in the bacterial domain: novel
gene-enzyme relationships from psychrophilic Moritella strains
(Vibrionaceae) and evolutionary significance of N-alpha-acetyl
ornithinase. J. Bacteriol. 182, 1609–1615.
6. Udaka, S. & Kinoshita, S. (1958) Studies on 1-ornithine fermen-
tation. I. The biosynthetic pathway of 1-ornithine in Micrococcus
glutamicus. J. General Appl. Microbiol. 4, 272–282.
7. Hoare, D. & Hoare, S. (1966) Feedback regulation of arginine
biosynthesis in blue-green algae. J. Bacteriol. 92, 375–379.
8. Haas, D., Kurer, V. & Leisinger, T. (1972) N-acetylglutamate
synthetase of Pseudomonas aeruginosa. An assay in vitro and
feedback inhibition by arginine. Eur. J. Biochem. 31, 290–295.
9. Shinners, E. & Catlin, B. (1978) Arginine biosynthesis in Neisseria
gonorrhoeae: enzymes catalyzing the formation of ornithine and
citrulline. J. Bacteriol. 136, 131–135.
10. Meile, L. & Leisinger, T. (1984) Enzymes of arginine biosynthesis
in methanogenic bacteria. Experientia 40, 899–900.
11. Sakanyan, V., Kochikyan, A., Mett, I., Legrain, C., Charlier, D.,
Pie
´
rard, A. & Glansdorff, N. (1992) A re-examination of the

Escherichia coli. Regulation of synthesis and activity by arginine.
J. Biol. Chem. 250, 1690–1693.
21. Abdelal, A. & Nainan, O. (1979) Regulation of N-acetylglutamate
synthesis in Salmonella typhimurium. J. Bacteriol. 137, 1040–1042.
22. Udaka, S. (1966) Pathway-specific pattern of control of arginine
biosynthesis in bacteria. J. Bacteriol. 91, 617–621.
23. Haas, D. & Leisinger, T. (1975) N-acetylglutamate 5-phospho-
transferase of Pseudomonas aeruginosa. Catalytic and regulatory
properties. Eur. J. Biochem. 52, 377–383.
24. Haas, D., Kurer, V. & Leisinger, T. (1972) N-acetylglutamate
synthetase of Pseudomonas aeruginosa. An assay in vitro and
feedback inhibition by arginine. Eur. J. Biochem. 31, 290–295.
25. Wipf, B. & Leisinger, T. (1979) Regulation of activity and
synthesis of N-acetylglutamate synthase from Saccharomyces
cerevisiae. J. Bacteriol. 140, 874–880.
26. Hilger, F., Culot, M., Minet, M., Pie
´
rard, A., Grenson, M. &
Wiame, J.M. (1973) Studies on the kinetics of the enzyme sequence
mediating arginine synthesis in Saccharomyces cerevisiae. J. Gen.
Microbiol. 75, 33–41.
27. Wolf, E. & Weiss, R. (1980) Acetylglutamate kinase. A
mitochondrial feedback-sensitive enzyme of arginine biosynthesis
in Neurospora crassa. J. Biol. Chem. 255, 9189–9195.
28. Hinde, R.W., Jacobson, J.A., Weiss, R.L. & Davis, R.H. (1986)
N-acetyl-L-glutamate synthase of Neurospora crassa.Character-
istics, localization, regulation, and genetic control. J. Biol. Chem.
261, 5848–5852.
29. Boonchird, C., Messenguy, F. & Dubois, E. (1991) Determination
of amino acid sequences involved in the processing of the ARG5/

enzyme for arginine biosynthesis and a prototype for the amino
acid kinase enzyme family, during catalysis. Structure (Camb) 10,
329–342.
38. Crabeel, M., Seneca, S., Devos, K. & Glansdorff, N. (1988)
Arginine repression of the Saccharomyces cerevisiae ARG1 gene.
Comparison of the ARG1 and ARG3 control regions. Curr. Genet
13, 113–124.
39. Wach, A. (1996) PCR-synthesis of marker cassettes with long
flanking homology regions for gene disruptions in S. cerevisiae.
Yeast 12, 259–265.
40. Storici, F., Lewis, K. & Resnick, M. (2001) In vivo site-directed
mutagenesis using oligonucleotides. Nat. Biotechnol. 19, 773–776.
41. Messenguy, F. (1976) Regulation of arginine biosynthesis in
S. cerevisiae: isolation of a cis-dominant constitutive mutant for
ornithine carbamoyltransferase synthesis. J. Bacteriol. 128, 49–55.
1024 K. Pauwels et al. (Eur. J. Biochem. 270) Ó FEBS 2003