Kinetic mechanism of asparagine synthetase from Vibrio cholerae

Article abstract of DOI:10.1016/j.bioorg.2003.10.002

Asparagine synthetase B (AsnB) catalyzes the formation of asparagine in an ATP-dependent reaction using glutamine or ammonia as a nitrogen source. To obtain a better understanding of the catalytic mechanism of this enzyme, we report the cloning, expression, and kinetic analysis of the glutamine- and ammonia-dependent activities of AsnB from Vibrio cholerae. Initial velocity, product inhibition, and dead-end inhibition studies were utilized in the construction of a model for the kinetic mechanism of the ammonia- and glutamine-dependent activities. The reaction sequence begins with the ordered addition of ATP and aspartate. Pyrophosphate is released, followed by the addition of ammonia and the release of asparagine and AMP. Glutamine is simultaneously hydrolyzed at a second site and the ammonia intermediate diffuses through an interdomain protein tunnel from the site of production to the site of utilization. The data were also consistent with the dead-end binding of asparagine to the glutamine binding site and PP_i with free enzyme. The rate of hydrolysis of glutamine is largely independent of the activation of aspartate and thus the reaction rates at the two active sites are essentially uncoupled from one another.

Full text of DOI:10.1016/j.bioorg.2003.10.002

BIOORGANIC
CHEMISTRY
Bioorganic Chemistry 32 (2004) 63–75
www.elsevier.com/locate/bioorg
Kinetic mechanism of asparagine synthetase
from Vibrio cholerae
a
Vicente Fresquet, James B. Thoden, Hazel M. Holden,
b
b,
*
a,
*
and Frank M. Raushel
a
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station, TX 77843-3012,USA
b
Department of Biochemistry, University of Wisconsin, Madison, WI 53706, USA
Received 11 June 2003
Abstract
Asparagine synthetase B (AsnB) catalyzes the formation of asparagine in an ATP-depen-
dent reaction using glutamine or ammonia as a nitrogen source. To obtain a better under-
standing of the catalytic mechanism of this enzyme, we report the cloning, expression, and
kinetic analysis of the glutamine- and ammonia-dependent activities of AsnB from Vibrio
cholerae. Initial velocity, product inhibition, and dead-end inhibition studies were utilized in
the construction of a model for the kinetic mechanism of the ammonia- and glutamine-depen-
dent activities. The reaction sequence begins with the ordered addition of ATP and aspartate.
Pyrophosphate is released, followed by the addition of ammonia and the release of asparagine
and AMP. Glutamine is simultaneously hydrolyzed at a second site and the ammonia interme-
diate diffuses through an interdomain protein tunnel from the site of production to the site of
utilization. The data were also consistent with the dead-end binding of asparagine to the glu-
i
tamine binding site and PP with free enzyme. The rate of hydrolysis of glutamine is largely
independent of the activation of aspartate and thus the reaction rates at the two active sites
are essentially uncoupled from one another.
Ó 2003 Elsevier Inc. All rights reserved.
Keywords: Asparagine synthetase B; Dead-end inhibition; Glutamine amidotransferase; Kinetic
mechanism
*
Corresponding authors. Fax: 1-979-845-9452 (F.M. Raushel), 1-608-262-1319 (H.M. Holden).
E-mail addresses: hazel_holden@biochem.wisc.edu (H.M. Holden), raushel@tamu.edu (F.M.
Raushel).
0
045-2068/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2003.10.002

64
V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
1
. Introduction
Asparagine synthetase is a key enzyme in the de novo formation of asparagine in
eukaryotes and has been associated with resistance to acute lymphoblastic leukemia
treatments. Chemotherapy of acute lymphoblastic leukemia involves the administra-
tion of multiple drugs, almost always including the enzyme asparaginase where the
proposed mechanism of action for asparaginase is the catalytic depletion of cellular
asparagine [1]. However, resistant tumors have been reported upon prolonged treat-
ment of asparaginase. Several studies have reported that cellular resistance to aspar-
aginase administration is correlated with elevated expression levels of asparagine
synthetase [2–5].
In bacteria there are two unrelated genes (asnA and asnB) that code for proteins
that can synthesize the de novo formation of asparagine. AsnA is present only in
prokaryotes and catalyzes the ammonia-dependent synthesis of asparagine [6]. AsnB
is present in both eukaryotes and prokaryotes and synthesizes an ATP-dependent
conversion of aspartic acid into asparagine using either glutamine or ammonia as
the nitrogen source [7]. Previous studies have postulated that the glutamine-depen-
dent synthesis of asparagine takes place via three separate reactions as summarized
in Scheme 1. In this mechanism, ammonia is formed through the hydrolysis of glu-
1
tamine. The activation of aspartate occurs by the formation of a b-aspartyl-AMP
intermediate from ATP and then asparagine is formed in the final reaction through
a nucleophilic attack of ammonia on the b-aspartyl-AMP intermediate [8].
A comparison of amino acid sequences shows that AsnB belongs to the glutamine
amidotransferase family of enzymes. AsnB is a member of the Ntn (PurF or Class II)
type of glutamine amidotransferases [9], which also includes glutamine phosphoribo-
sylpyrophosphate amidotransferase [10] and glutamine fructose-6-phosphate amido-
transferase [11]. This type of glutamine amidotransferase is characterized by an
N-terminal cysteine residue that is essential for the hydrolysis of glutamine and for-
mation of ammonia. The three-dimensional X-ray crystal structure of the C1A mu-
tant of AsnB from Escherichia coli has been solved in the presence of glutamine and
AMP [12]. The enzyme is a homodimer with two active sites separated by a hydro-
ꢀ
phobic intramolecular tunnel of ꢀ20 A. The C-terminal synthetase domain binds
ATP and aspartate and is the site for the formation of the b-aspartyl intermediate
and the subsequent reaction with ammonia. The ammonia intermediate has been
proposed to diffuse through the interdomain tunnel from the site of production to
the site of utilization. The synthetase component has a high degree of homology with
GMP synthetase and is included in the category of ATP pyrophosphatases [12].
Several previous attempts have been made to determine the kinetic mechanism of
asparagine synthetase [13–17]. However, very different kinetic mechanisms have been
proposed depending on the enzyme source and some of these are presented in
Scheme 2. For example, a uni-uni-bi-ter ping pong mechanism (Scheme 2A) was
1
0
Abbreviations used: AMP-CH2-PP, a; b-methylene adenosine 5 -triphosphate; AsnA, asparagine
synthetase A; AsnB, asparagine synthetase B; CPS, carbamoyl phosphate synthetase.

V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
65
Scheme 1.
Scheme 2.
described for mouse pancreatic AsnB where glutamine binds first and glutamate is
released prior to the binding of ATP and aspartate [13]. An alternative bi-uni-uni-
ter ping pong mechanism (Scheme 2B) has also been proposed where glutamine
and ATP bind prior to the release of glutamate and the binding of aspartate [14].
The kinetic mechanism of AsnB from E. coli is the only prokaryote enzyme analyzed
to date. Two alternative mechanisms that differ in the order of binding and release of
glutamine and glutamate have been proposed [15]. These kinetic mechanisms are
summarized in Schemes 2C and D. However, the proposed models are unable to ad-
equately explain all of the kinetic data observed with this enzyme. For example, it
has been demonstrated that AsnB can hydrolyze glutamine in the absence of the
other substrates and catalyzes the synthesis of the b-aspartyl intermediate in the ab-
sence of glutamine [15]. In order to obtain a better understanding of the kinetic
mechanism of this key enzyme, the bacterial asparagine synthetase from Vibrio
cholerae was purified and analyzed for the glutamine- and ammonia-dependent syn-
thesis of asparagine.

66
V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
2
. Materials and methods
2
.1. Cloning of the AsnB gene
Genomic DNA from V. cholerae was prepared by standard methods and the AsnB
0
gene was PCR amplified such that the forward primer 5 -GGAATTCATATGTGCT
0
0
CTATCTTTGGCATCTTAG-3 and reverse primer 5 -CCGCTCGAGTTAGTAG
0
GCTTGTTTGTGGACG-3 added NdeI and XhoI cloning sites, respectively. The
gene was PCR amplified with Platinum Pfx DNA polymerase (Invitrogen) according
to the manufacturerÕs instructions and standard cycling conditions. The PCR
product was purified with the QIAquick PCR Purification kit (Qiagen), followed
by A-tailing and ligation into pGEM-T Vector (Promega) and transformation into
E. coli DH5a cells. The asparagine synthetase gene was sequenced in the pGEM-T
vector construct with the ABI Prism Big Dye Primer Cycle Sequencing Kit (Applied
Biosystems) to confirm that no mutations were introduced during PCR amplifica-
tion. The pGEM-T Vector was then digested with NdeI and XhoI and the gene sep-
arated from digestion by-products on a 1.0% agarose gel, excised, and purified with
QIAquick Gel Purification kit (Qiagen). The purified asparagine synthetase gene was
then ligated into the expression vector pET-31 (Novagen) that was previously cut
with the same restriction enzymes. E. coli DH5 a cells were transformed with the li-
gation mixture and then plated onto LB media supplemented with 100 lg/ml ampi-
cillin. Individual colonies were selected, cultured overnight, and plasmid DNA was
extracted with the QIAprep Spin Miniprep Kit (Qiagen). Plasmids were tested for
insertion of the AsnB gene by digesting with NdeI and XhoI.
2
.2. Protein expression
For protein expression, E. coli Rosetta (DE3) cells (Novagen) were transformed
with the pET31-asnB plasmid and plated onto LB agarose plates supplemented with
00 lg/ml ampicillin and 30 lg/ml chloramphenicol. Single colonies from the plates
1
were selected and grown overnight in 500 ml LB media (plus antibiotics) to use for
the inoculation of 12 ꢁ 2 L baffled flasks containing 500 ml M9 minimal media (plus
antibiotics). The cells were grown at 37 °C with aeration to an OD600 of ꢀ0.9, at
which time they were cooled on ice for 5 min and transferred to a 16 °C shaker before
induction with 1.0 mM IPTG. The cells were allowed to grow for an additional 18 h
before harvesting by centrifugation at 6000g for 8 min. The cell paste was frozen in
liquid nitrogen and stored at )80 °C.
2
.3. Protein purification
All protein purification steps were carried out on ice or at 4 °C. The frozen cell
paste was thawed in 3 volumes of 20 mM NaH PO (pH 7.5). Cells were lysed on
2 4
ice by four cycles of sonication (45 s) separated by 5 min of cooling. Cellular debris
was removed by centrifugation at 4 °C for 25 min at 20,000g. The clarified lysate was
loaded onto a 50 ml fast-flow Q-Sepharose column pre-equilibrated with the lysis

V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
67
buffer. After loading, the column was washed with about 200 ml of lysis buffer. The
protein was then recovered by gradient elution with 0–500 mM KCl in the lysis buf-
fer. Protein containing fractions were pooled based on SDS–PAGE. The pooled
fractions were diluted with three volumes of lysis buffer and passed over a 50 ml
Green-19 column to remove minor contaminants. The protein was again pooled
on the basis of SDS–PAGE, concentrated, and dialyzed against 20 mM Hepes (pH
7
.5) containing 100 mM NaCl. The dialyzed protein was concentrated to approxi-
mately 10 mg/ml, based on an extinction coefficient of ꢀ0.79 ml/(mg cm) as calcu-
lated by the program Protean (DNAstar, Madison, WI).
2
.4. Asparagine synthetase activity
The asparagine synthetase activity was measured using a myokinase/pyruvate ki-
nase/lactic dehydrogenase coupled assay by following the loss of NADH spectropho-
tometrically [18]. The assay was carried out at 30 °C using 100 mM Hepes, pH 8.0,
1
00 mM KCl, 1.0 mM dithiothreitol (DTT), 28 lg/ml lactic dehydrogenase, 28 lg/
4
ml pyruvate kinase, 0.5 U myokinase, 10 mM MgSO , 1.0 mM phosphoenolpyr-
uvate, 0.44 mM NADH, and 1–10 lg of asparagine synthetase. When glutamine,
ammonia, ATP or aspartate was varied, the concentration range was: 0.05–
1
5 mM, 1–50 mM, 0.05–10 mM, and 0.1–20 mM, respectively.
In the inhibition studies by AMP the activity was quantified by monitoring the
formation of pyrophosphate with a fructose-6-phosphate pyrophosphate-dependent
kinase (PP -PFK), aldolase, triose phosphate isomerase (TPI), and glycerol phos-
i
phate dehydrogenase (GDH) coupled assay. The activity was measured at 30 °C in
2
MgSO
50 ll containing 100 mM Hepes, pH 8.0, 100 mM KCl, 1.0 mM DTT, 10 mM
4 i
, 5.0 mM D-fructose-6-phosphate, 0.25 U PP -PFK, 1.0 U aldolase, 0.5 U
GDH, 5 U TPI, 0.44 mM NADH, and 1–10 lg of asparagine synthetase.
2
.5. Glutaminase activity
The amount of glutamate produced was determined using glutamate dehydroge-
nase as the coupling enzyme in the presence of 3-acetylpyrimidine adenine nucleoside
APAD). The reaction mixtures (200 ll) contained 100 mM Hepes, pH 8.0, 100 mM
KCl, 1.0 mM DTT, 10 mM MgCl with a variable concentration of glutamine (0.05–
0 mM). Reactions were initiated by adding AsnB (5.4 lg) and incubating for 20 min
(
2
1
at 30 °C and then quenched by the addition of 75 ll of 10% TCA. After neutralizing
by the addition of 15 ll of 3 M Trizma base, a 75 ll portion of the reaction mixture
was added to 400 ll of the coupling reagent (0.12 m Hepes, pH 8.0, 0.12 M KCl, 12 U
of glutamate dehydrogenase, and 1.2 mM APAD). After 2 h of incubation the absor-
bance was measured at 363 nm.
2
.6. Asparaginase activity
The hydrolysis of asparagine was measured spectrophotometrically by coupling
the formation of aspartate to the oxidation of NADH using malate dehydrogenase

68
V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
and aspartate aminotransferase as previously described [19]. The reaction was
conducted in a volume of 250 ll at 30 °C in a reaction mixture containing 100 mM
Hepes, pH 8.0, 100 mM KCl, 1.0 mM DTT, 10 mM MgCl , 3.7 mM a-ketoglutar-
2
ate, 0.4 mM NADH, 0.36 U malate dehydrogenase, and 6 U aspartate aminotrans-
ferase.
2
.7. Data analysis
Initial velocity data with a single substrate were fitted to Eq. (1). Initial velocity
data with two substrates conforming to an intersecting pattern were fitted to Eq.
2) and a parallel pattern to Eq. (3). Competitive, uncompetitive, and noncompeti-
tive inhibition patterns were fitted to Eqs. (4)–(6), respectively.
(
v ¼ ðVmax½AꢂÞ=ðK
v ¼ ðVmax½Aꢂ½BꢂÞ=ðKia
v ¼ ðVmax½Aꢂ½BꢂÞ=ðK ½Bꢂ þ K ½Aꢂ þ ½Aꢂ½BꢂÞ;
a
þ ½AꢂÞ;
ð1Þ
ð2Þ
ð3Þ
ð4Þ
ð5Þ
ð6Þ
K
b
þ K
b
½Aꢂ þ K ½Bꢂ þ ½Aꢂ½BꢂÞ;
a
a
b
v ¼ ðVmax½Aꢂ=ðK
v ¼ ðVmax½Aꢂ=ðK
v ¼ ðVmax½Aꢂ=ðK
a
a
a
ð1 þ ½Iꢂ=KisÞ þ ½AꢂÞ;
ð1 þ ½Iꢂ=KisÞ þ ½Aꢂð1 þ ½Iꢂ=KiiÞÞ;
þ ½Aꢂð1 þ ½Iꢂ=KiiÞÞ:
3
. Results
.1. Initial velocity studies
The kinetic mechanism of asparagine synthetase was probed by measuring the ini-
3
tial velocities of the reaction while simultaneously varying the concentration of two
substrates and holding the third substrate constant. The data were fitted to Eq. (2) or
Eq. (3) and visualized as double reciprocal plots. Both ammonia and glutamine were
used as alternate sources of nitrogen and the results are summarized in Table 1.
When glutamine was used as the nitrogen source, the ATP versus aspartate double
reciprocal plot was intersecting, indicating that there was not an irreversible step be-
tween the binding of these two substrates to the enzyme active site [20,21]. When am-
monia was used as the nitrogen source the double reciprocal plots between ATP and
aspartate versus ammonia were parallel. These results indicate that there was an ir-
reversible step between the addition of free ammonia and the other two substrates to
the enzyme active site. When glutamine was used as the nitrogen source, the double
reciprocal plots for ATP and aspartate versus glutamine were also parallel. The over-
all kcat values at saturating substrate concentrations of all substrates were 4.2 ꢃ 0.1
ꢄ1
and 3.4 ꢃ 0.1 s for the glutamine- and ammonia-dependent synthesis of asparagine,
respectively.

V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
69
Table 1
Kinetic parameters from the initial velocity patterns for combinations of substrates
a
Substrate pair Gln
Ammonia ATP Aspartate
Pattern
K
d
(mM)
K
c
(mM)
K
a
ðlM)
KiaðlM)
K
b
(mM)
Kib (mM)
c
Gln–ATP
Gln–Asp
0.5 ꢃ 0.02
0.8 ꢃ 0.07
95 ꢃ 5
94 ꢃ 6
P
0.9 ꢃ 0.04
0.8 ꢃ 0.03
P
P
P
NH
NH
3
–ATP
–Asp
11 ꢃ 0.9
13 ꢃ 0.6
3
b
d
ATP–Asp
122 ꢃ 7
17 ꢃ 4
1.0 ꢃ 0.06 0.14 ꢃ 0.05
I
a
The activity was determined monitoring the rate of AMP formation.
Glutamine was used as the nitrogen source.
A fit of the data to Eq. (3).
b
c
d
A fit of the data to Eq. (2).
3
.2. Inhibition by substrate analogs
The initial velocity studies were unable to distinguish whether the binding of ATP
and aspartate to the enzyme was sequential or random. Chemically unreactive mim-
ics of ATP and aspartate were therefore used as dead-end inhibitors to help deter-
mine the order of substrate binding to asparagine synthetase. a; b-Methylene
0
adenosine 5 -triphosphate (AMP-CH
2
-PP) was utilized as a mimic for ATP. This
compound was competitive versus ATP and noncompetitive versus aspartate. Un-
competitive patterns were obtained versus glutamine and ammonia. The first aspar-
tate analog to be tested was b-methyl aspartate as a mixture of four diastereomers
(Sigma). Although (2S,3R) b-methyl aspartate has been reported to inhibit aspara-
gine synthetase from E. coli [22], no inhibition was detected for the unfractionated
mixture of stereoisomers with AsnB from V. cholerae up to a concentration of
2
0 mM. However, L-cysteine sulfinic acid was found to be a good competitive inhib-
itor with respect to aspartate and uncompetitive versus ATP. Uncompetitive inhibi-
tion was observed versus ammonia. No inhibition of the partial glutaminase reaction
was detected using L-cysteine sulfinic acid. When this compound was tested as an al-
ternate substrate for aspartate, a very slow production of AMP was detected that
was less than 0.5% of the overall rate observed with aspartate. This latter result is
in agreement with previous work carried out with the E. coli enzyme [18]. The
2
dead-end inhibition patterns by AMP-CH -PP and L-cysteine sulfinic acid are con-
sistent with the ordered addition of ATP to free enzyme followed by aspartate. The
c-methyl ester of glutamic acid was utilized as a mimic of glutamine. This compound
was a weak competitive inhibitor versus glutamine and ammonia and uncompetitive
versus ATP and aspartate. All of the dead-end inhibition data are summarized in
Table 2.
3
.3. Product inhibition
Product inhibition studies were carried out in an attempt to obtain more informa-
tion on the order of product release from the active site. Glutamate and AMP were

70
V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
Table 2
Inhibition by product and substrate analogs
Inhibitor Substrate
PP
K
is
K
ii
Pattern
i
ATP
Asp
0.021 ꢃ 0.02
0.85 ꢃ 0.15
C
0.65 ꢃ 0.05
0.85 ꢃ 0.09
0.85 ꢃ 0.06
NC
UC
UC
C
NH
Gln
Gln
3
Asn
AMP-CH
0.023 ꢃ 0.003
1.1 ꢃ 0.2
2
-PP
ATP
Asp
Gln
C
2.6 ꢃ 0.7
2.6 ꢃ 0.4
3.2 ꢃ 0.2
3.7 ꢃ 0.3
2.5 ꢃ 0.1
NC
UC
UC
UC
C
NH
3
L
L
-Cysteine sulfinic acid
ATP
Asp
Gln
1.1 ꢃ 0.1
7.2 ꢃ 1.2
6.4 ꢃ 0.4
2.2 ꢃ 0.1
29 ꢃ 3
NC
UC
UC
UC
C
NH
3
-glutamic acid c-methyl ester
ATP
Asp
Gln
27 ꢃ 3
8.9 ꢃ 0.9
6.5 ꢃ 0.7
When ATP or Asp was varied, ammonia was used as nitrogen source. The data of competitive (C),
NH
3
C
uncompetitive (UC) or noncompetitive (NC) patterns were fitted to Eq. (4)–(6), respectively.
unable to inhibit AsnB up to concentrations of 80 and 12 mM, respectively. How-
ever, pyrophosphate and asparagine were good inhibitors of the enzyme. Pyrophos-
phate was a competitive, noncompetitive, and uncompetitive inhibitor versus ATP,
aspartate, and ammonia, respectively. This set of inhibition patterns for PP
these three substrates is identical to the collection of inhibition patterns obtained
for the ATP analog, AMP-CH -PP. This result is consistent with the conclusion that
PP is acting as a general dead-end inhibitor that binds to free enzyme rather than as
a typical product inhibitor.
i
versus
2
i
Asparagine inhibited the ammonia- and glutamine-dependent activities. However,
the inhibition was different in each case and is illustrated in Fig. 1. Asparagine inhib-
ited completely the glutamine-dependent activity. However, the inhibition of the am-
monia-dependent activity was only partial. This behavior suggests that asparagine
binds predominantly to the site for glutamine. When the asparagine inhibition of
the glutaminase partial reaction was analyzed, a competitive inhibition pattern ver-
sus glutamine was obtained, with a K value of 60 lM. This result demonstrates that
i
the inhibition of the glutamine-dependent reaction by asparagine is predominantly a
consequence of binding directly to the glutamine binding site. The kinetic parameters
for the inhibition by products are presented in Table 2.
The observation of asparagine binding to the glutamine site, suggested that AsnB
may have an asparaginase activity. In order to determine if the enzyme was able to
hydrolyze asparagine, this activity was measured in absence of the other substrates.
A small production of aspartate was detected with a K
0.07 ꢃ 0.007 s . However, this activity was not inhibited by the addition of saturat-
ing concentrations of the other substrates or substrate analogs. Therefore, it cannot
m
of 0.9 ꢃ 0.2 mM and a kcat of
ꢄ1

V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
71
1
1
20
00
8
6
4
2
0
0
0
0
0
0
2
4
6
8
10
12
[
Asn] mM
Fig. 1. Inhibition of the glutamine- ðdÞ and ammonia- ðsÞ dependent activities by asparagine. Additional
details are provided in the text.
Table 3
Kinetic parameters for the hydrolysis of glutamine
ab
ꢄ1
Substrate combinations
K
d
(mM)
k
cat (s
)
Glutamine
Glutamine + ATP
Glutamine + ATP + Asp
0.63 ꢃ 0.04
0.97 ꢃ 0.06
1.6 ꢃ 0.3
3.9 ꢃ 0.1
5.6 ꢃ 0.1
5.8 ꢃ 0.2
a
The activity was monitored by the following the production of glutamate in the presence or absence
of 5.0 mM ATP and 5.0 mM aspartate.
b
The data were fitted using Eq. (1).
be determined from these data whether or not the very low level of asparagine
hydrolysis is the result of a contamination by an asparaginase.
It has been reported that AsnB from E. coli is different from other glutamine
amidotransferases in that the rate of glutamine hydrolysis is not fully coupled with
the synthesis of asparagine [23]. In the E. coli enzyme, a stoichiometric relationship
of 1.3–1.5 glutamate/PP
[
i
is produced at saturating concentrations of the substrates
15]. In order to determine whether the AsnB from V. cholerae has this characteristic
stoichiometry, the partial glutaminase activity was measured in the presence and ab-
sence of the other two substrates. As shown in the Table 3, the kinetic parameters for
the glutaminase activity do not change significantly in the presence of ATP and as-
partate. The relative rate of the glutaminase activity is higher than unity, indicating
that the reactions are partially uncoupled from one another.
4
. Discussion
We report in this paper the cloning, expression, and purification of asparagine
synthetase from V. cholerae. This enzyme catalyzes a reaction that requires three

72
V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
different substrates with the net production of four products and two reactive inter-
mediates. The chemistry takes place within two different active sites that are sepa-
ꢀ
rated in three-dimensional space by approximately 20 A. In order to more fully
understand this complex reaction mechanism we have analyzed both the glutamine-
and ammonia-dependent activities. The analysis of the ammonia-dependent synthe-
sis of asparagine is somewhat easier to manage since all of the chemistry is occurring
within a single active site. When glutamine is used as a nitrogen source, different
chemical reactions can occur simultaneously at the individual active sites connected
by the interdomain molecular tunnel.
Two sets of parallel double reciprocal plots were obtained when either ATP or as-
partate was varied versus ammonia. This result is consistent with the irreversible re-
lease of a product between the binding of ammonia and the other two substrates.
Given the likely chemical mechanism for this transformation, pyrophosphate is there-
fore proposed to be the product formed immediately after the binding of ATP and as-
partate. The binding of ATP and aspartate could either be ordered or random. In
order to decide between these two possibilities, two different dead-end substrate
mimics of ATP and aspartate were utilized. The non-hydrolyzable ATP analog,
2
AMP-CH -PP, was found to be competitive versus ATP and noncompetitive versus
aspartate. In contrast, cysteine sulfinic acid was competitive versus aspartate and
uncompetitive versus ATP. This set of inhibition patterns is only consistent with the
ordered binding of ATP prior to the addition of aspartate to the enzyme. The uncom-
petitive inhibition patterns by the dead-end inhibitors AMP-CH -PP and cysteine sul-
2
finic acid versus ammonia are also fully consistent with the release of pyrophosphate
prior to the binding of ammonia [21]. Once ATP and aspartate have bound to the
enzyme, PP and the b-aspartyl-AMP intermediate are formed following nucleophilic
i
attack by the side chain carboxylate of aspartic acid on the a-phosphoryl group of
ATP. Pyrophosphate is released and ammonia binds to the synthetase active site
where it reacts with the b-aspartyl-AMP intermediate to form asparagine and AMP.
The determination of the order of asparagine and AMP release from the active
site was attempted by using these compounds as product inhibitors. However, these
two compounds did not behave as typical product inhibitors of the forward reaction.
With AMP no inhibition was observed up to a concentration of 12 mM. In contrast,
asparagine was found to be a very good inhibitor of the reaction when glutamine was
used as the nitrogen source. The inhibition versus glutamine was competitive and
thus the primary mode for the inhibition exhibited by asparagine is via binding di-
rectly to the glutamine site within the glutaminase domain and not through a direct
interaction with the synthetase domain. Moreover, the inhibition of the overall reac-
tion by high concentrations of asparagine when ammonia is used as the nitrogen
source is only partial. From these results it can be concluded that asparagine cannot
be used as a product inhibitor since it apparently does not bind to the synthetase do-
main at reasonable concentrations. This situation is probably exacerbated by the
rather weak binding of AMP to free enzyme. Nevertheless, we propose that aspara-
gine is released prior to the dissociation of AMP. This conclusion is based entirely on
the symmetry of the reaction whereby ATP has been shown to bind prior to the
addition of aspartate.

V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
73
The reaction mechanism when glutamine is used as the nitrogen-source is similar.
Parallel double reciprocal patterns were obtained when glutamine was varied versus
ATP or aspartate. This is the expected result for a two-site ping pong mechanism
when glutamine can bind to the glutaminase domain independently from the binding
of the other two substrates within the synthetase domain. However, we have ob-
tained no data that enables us to determine the timing for the release of glutamate.
A similar situation is observed in the steady state kinetic analysis of the reaction cat-
alyzed by carbamoyl phosphate synthetase (CPS) where the double reciprocal plots
for glutamine versus either ATP or bicarbonate are parallel [24]. However, there are
some significant differences observed for the interplay between the glutaminase do-
main and the synthetase domain for asparagine synthetase and that previously ob-
served for comparable domains within carbamoyl phosphate synthetase. With
CPS a conformational change that is triggered by the binding and reaction of
ATP and bicarbonate significantly enhances the rate of ammonia production from
the hydrolysis of glutamine. With asparagine synthetase the rate of glutamine hydro-
lysis is apparently independent of the binding of ATP and/or aspartate to the synthe-
tase domain. With CPS, the triggering mechanism serves as a mechanical device for
ensuring that the proper stoichiometry is maintained throughout the reaction cycle.
Thus, ammonia is not injected into the interdomain tunnel until the reactive carboxy
phosphate intermediate is waiting at the other end. With asparagine synthetase the
rate of glutamine hydrolysis does not seem to be coupled to events within the synthe-
tase domain and thus the stoichiometry must be maintained by the inherent catalytic
efficiency of the two sites. The overall kinetic mechanism for asparagine synthetase
from V. cholerae, including the formation of two dead-end complexes, is summarized
in Scheme 3.
The kinetic model proposed in this paper is in general agreement with previous
experiments conducted with other asparagine synthetases and is able to explain some
peculiar catalytic features of this enzyme. Isotope trapping studies demonstrated that
AsnB from E. coli is able to synthesize the b-aspartyl intermediate in the absence of
Gln
Glu
E’
(E’D
E’CS)
E’S
E’
Asn
E’Q
ATP
Asp
PP
i
NH
3
Asn
AMP
E
EA
PP
(EAB
EIP) EI (EIC
EQR) ER
E
i
EP
Scheme 3.

74
V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
nitrogen source [15]. These same studies are also consistent with the ordered binding
of ATP and aspartate in the absence of any nitrogen source. The inhibition of the
ammonia-dependent activity is consistent with the binding of asparagine to the glu-
tamine site. Mutagenesis of the cysteine 1 to alanine produces a protein that is unable
to synthesize asparagine from glutamine but with a normal activity when ammonia is
used as substrate. The binding of glutamine to this asparagine synthetase mutant
inhibits the ammonia-dependent activity [25,26].
Acknowledgments
We thank Dr. John Lindquist (Department of Bacteriology, University of Wis-
consin-Madison) for supplying V. cholerae cell stock. This work was supported in
part by the NIH (DK 30343) and (GM 55513) and the Robert A. Welch Foundation
(
A-840).
References
[
[
1] R. Chakrabarti, S.M. Schuster, Int. J. Pediatric Hematol. Oncol. 4 (1997) 597–611.
2] R. Pieters, E. Klumper, G.J.L. Kaspers, A.J.P. Veerman, ASP Crit. Rev. Oncol. Hematol. 25 (1997)
11–26.
[
[
3] G.J.L. Kaspers, R. Pieters, C.H. Van Zantwijk, E.R. Van Wering, A. Van Der Does-Van Der Berg,
A.J.P. Veerman, Blood 92 (1998) 259–266.
4] R.G. Hutson, T. Kitoh, D.A.M. Amador, S. Cosic, S.M. Schuster, M.S. Kilberg, Am. J. Physiol. 272
(1997) C1691–C1699.
[
[
[
[
[
5] A.M. Aslanian, B.S. Fletcher, M.S. Kilberg, Biochem. J. 357 (2001) 321–328.
6] H. Cedar, J.H. Schwartz, J. Biol. Chem. 244 (1969) 4112–4121.
7] H.A. Milman, D.A. Cooney, Biochem. J. 181 (1979) 51–59.
8] N.G. Richards, S.M. Schuster, Adv. Enzymol. 72 (1998) 145–198.
9] H. Zalkin, J.L. Smith, Avd. Enzymol. Relat. Areas Mol. Biol. 72 (1998) 87–144.
[
[
10] J.Y. Tso, M.A. Hermodson, H. Zalkin, J. Biol. Chem. 257 (1982) 3532–3536.
11] K.O. Broschat, C. Gorka, J.D. Page, C.L. Martin-Berger, M.S. Davies, H. Huang, E.A. Gulve, W.J.
Salsgiver, T.P. Kasten, J. Biol. Chem. 277 (2002) 14764–14770.
[
12] T.M. Larsen, S.K. Boehlein, S.M. Schuster, N.G.J. Richards, J.B. Thoden, H.M. Holden, I.
Rayment, Biochemistry 38 (1999) 16146–16157.
[
[
[
13] H.A. Milman, D.A. Cooney, C.Y. Huang, J. Biol. Chem. 255 (1980) 1862–1866.
14] R.S. Markin, C.A. Luehr, S.M. Schuster, Biochemistry 20 (1981) 7226–7232.
15] S.K. Boehlein, J.D. Stewart, E.S. Walworth, R. Thirumoorthy, N.G.J. Richards, S.M. Schuster,
Biochemistry 37 (1998) 13230–13238.
[
[
[
16] P.M. Mehlhaff, C.A. Luehr, S.M. Schuster, Biochemistry 24 (1985) 1104–1110.
17] S. Hongo, T. Sato, Arch. Biochem. Biophys. 238 (1985) 410–417.
18] S.K. Boehlein, T. Nakatsu, J. Hiratake, R. Thiromoorthy, J.D. Stewart, N.G.J. Richards, S.M.
Schuster, Biochemistry 40 (2001) 11168–11175.
[
[
[
[
19] J.B. Thoden, R. Marti-Arbona, F.M. Raushel, H.M. Holden, Biochemistry 42 (2003) 4874–4882.
20] W.W. Cleland, Biochim. Biophys. Acta 67 (1967) 104–137.
21] W.W. Cleland, in: P.D. Boyer (Ed.), The Enzymes, Academic Press, 1970, pp. 1–70.
22] I.B. Parr, S.K. Boehlein, A.B. Dribben, S.M. Schuster, N.G.J. Richards, J. Med. Chem. 39 (1996)
2367–2378.
[
23] A.R. Tesson, T.S. Soper, M. Ciustea, N.G. Richards, Arch. Biochem. Biophys. 413 (2003) 23–31.

V. Fresquet et al. / Bioorganic Chemistry 32 (2004) 63–75
75
[
[
[
24] F.M. Raushel, P.M. Anderson, J.J. Villafranca, Biochemistry 17 (1978) 5587–5591.
25] S.K. Boehlein, N.G.J. Richards, S.M. Schuster, J. Biol. Chem. 269 (1994) 7450–7457.
26] S. Sheng, D.A. Moraga-Amador, G. van Heeke, R.D. Allison, N.G. Richards, S.M. Schuster, J. Biol.
Chem. 268 (1993) 16771–16780.