A conserved glutamate controls the commitment to acyl-adenylate formation in asparagine synthetase

Article abstract of DOI:10.1021/bi1010688

Inhibitor docking studies have implicated a conserved glutamate residue (Glu-348) as a general base in the synthetase active site of the enzyme asparagine synthetase B from Escherichia coli (AS-B). We now report steady-state kinetic, isotope transfer, and positional isotope exchange experiments for a series of site-directed AS-B mutants in which Glu-348 is substituted by conservative amino acid replacements. We find that formation of the β-aspartyl-AMP intermediate, and therefore the eventual production of asparagine, is dependent on the presence of a carboxylate side chain at this position in the synthetase active site. In addition, Glu-348 may also play a role in mediating the conformational changes needed to (i) coordinate, albeit weakly, the glutaminase and synthetase activities of the enzyme and (ii) establish the structural integrity of the intramolecular tunnel along which ammonia is translocated. The importance of Glu-348 in mediating acyl-adenylate formation contrasts with the functional role of the cognate residues in β-lactam synthetase (BLS) and carbapenam synthetase (CPS) (Tyr-348 and Tyr-345, respectively), which both likely evolved from asparagine synthetase. Given the similarity of the chemistry catalyzed by AS-B, BLS, and CPS, our work highlights the difficulty of predicting the functional outcome of single site mutations on enzymes that catalyze almost identical chemical transformations.

Full text of DOI:10.1021/bi1010688

Biochemistry 2010, 49, 9391–9401 9391
DOI: 10.1021/bi1010688
A Conserved Glutamate Controls the Commitment to Acyl-Adenylate Formation in
Asparagine Synthetase†
Megan E. Meyer,‡ Jemy A. Gutierrez,‡,§ Frank M. Raushel, and Nigel G. J. Richards*,‡
‡Department of Chemistry, University of Florida, P.O. Box 117200, Gainesville, Florida 32611, United States, and
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station, Texas 77843, United States
§Present address: Department of Biochemistry, Albert Einstein College of Medicine, 1300 Morris Park Ave.,
Bronx, NY 10461
Received July 3, 2010; Revised Manuscript Received September 17, 2010
ABSTRACT: Inhibitor docking studies have implicated a conserved glutamate residue (Glu-348) as a general base in
the synthetase active site of the enzyme asparagine synthetase B from Escherichia coli (AS-B). We now report
steady-state kinetic, isotope transfer, and positional isotope exchange experiments for a series of site-directed AS-B
mutants in which Glu-348 is substituted by conservative amino acid replacements. We find that formation of the
β-aspartyl-AMP intermediate, and therefore the eventual production of asparagine, is dependent on the presence
of a carboxylate side chain at this position in the synthetase active site. In addition, Glu-348 may also play a role in
mediating the conformational changes needed to (i) coordinate, albeit weakly, the glutaminase and synthetase
activities of the enzyme and (ii) establish the structural integrity of the intramolecular tunnel along which ammonia
is translocated. The importance of Glu-348 in mediating acyl-adenylate formation contrasts with the functional
role of the cognate residues in β-lactam synthetase (BLS) and carbapenam synthetase (CPS) (Tyr-348 and Tyr-345,
respectively), which both likely evolved from asparagine synthetase. Given the similarity of the chemistry catalyzed
by AS-B, BLS, and CPS, our work highlights the difficulty of predicting the functional outcome of single site
mutations on enzymes that catalyze almost identical chemical transformations.
Human asparagine synthetase (hASNS)1 catalyzes the ATP-
Scheme 1: Overall Reaction Catalyzed by Glutamine-Dependent
ASNS
dependent synthesis of
L-asparagine from L-aspartic acid using
L-glutamine as the nitrogen source (Scheme 1) (1). A number of
observations make hASNS of interest as a drug target. For
example, hASNS has been shown to be a predictive biomarker
for ovarian cancer (2, 3). Moreover, the constitutive expression of
hASNS in transformed MOLT-4 cells permits them to grow in
the presence of asparaginase (4), thereby supporting hypotheses
concerning the role of hASNS in the molecular mechanisms that
cause resistance to current clinical protocols for treating acute
lymphoblastic leukemia (ALL) (5). The physiological importance
of hASNS-mediated asparagine biosynthesis in both ovarian
cancer and ALL is also supported by our recent finding that the
adenylated sulfoximine 1 (Figure 1), which is the first inhibitor of
hASNS with nanomolar potency, suppresses proliferation of an
asparaginase-resistant MOLT-4 cell line (6).
For a variety of technical reasons, in part associated with the
need to employ baculovirus-based expression for preparing large
amounts of recombinant, wild-type hASNS (7), almost all
mechanistic and structural studies of the enzyme have been
performed using the glutamine-dependent asparagine synthetase
present in Escherichia coli (AS-B) (8-11). In an important devel-
opment for structure-based discovery of new hASNS inhibitors
with improved activity in cell-based assays (12), our group was
able to obtain a high-resolution crystal structure for the C1A AS-
B mutant (in which Cys-1 is replaced by an alanine residue)
complexed to L-glutamine and AMP (Figure 1A) (13). The
adenylated sulfoximine 1 also inhibits AS-B (14), presumably
by forming similar intermolecular interactions given that residues
within the synthetase active sites of hASNS and AS-B are com-
pletely conserved. Using homology modeling methods (15, 16),
we have been able to develop a computational model of the
hASNS inhibitor 1 bound within the AS-B synthetase site
(Figure 1B) and are systematically validating this model of
the AS-B/sulfoximine complex.2 For example, recent work
has confirmed that the electrostatic interaction between the
phosphate moiety in 1 and the side chain of a conserved active
†This work was supported by the Chiles Endowment Biomedical
Research Program of the Florida Department of Health (N.G.J.R.) and
the National Institutes of Health (Grant DK30343 to F.M.R.). Funding
for a predoctoral studentship (M.E.M.) from an NIH training grant in
Cancer Biology (CA09126) is also gratefully acknowledged.
*To whom correspondence should be addressed: 352-392-3601 (office);
352-846-2095 (fax); richards@qtp.ufl.edu (e-mail).
1Abbreviations: hASNS, human asparagine synthetase; ALL, acute
lymphoblastic leukemia; AS-B, Escherichia coli asparagine synthetase B;
AMP, adenosine 50-monophosphate; βAspAMP, β-aspartyl-AMP; DNFB,
2,4-dinitrofluorobenzene; DTT, dithiothreitol; EPPS, 3-[4-(2-hydroxyethyl)-
1-piperazinyl]propanesulfonic acid; PPi, inorganic pyrophosphate; LC-ESI,
liquid chromatography-electrospray ionization; WT, wild type; PIX, posi-
tional isotope exchange; BLS, β-lactam synthetase; CPS, carbapenam
synthetase; GMPS, guanosine 50-monophosphate synthetase; ASNS, gluta-
mine-dependent asparagine synthetase.
2Y. Ding, R. N. Humkey, and N. G. J. Richards, to be published.
Coordinate data for the model complex are available on request.
r
2010 American Chemical Society
Published on Web 09/20/2010
pubs.acs.org/Biochemistry

9392 Biochemistry, Vol. 49, No. 43, 2010
Meyer et al.
F
IGURE 1: (A) Cartoon representation of the E. coli AS-B C1A mutant complexed to glutamine and AMP (1CT9) (14). The N-terminal
glutaminase and C-terminal synthetase domains of each monomer are colored gray and blue, respectively. Bound glutamine (top) and AMP
(bottom) are shown as CPK models. (B) Computational model for inhibitor 1 bound within the synthetase active site of AS-B, indicating the
possible proximity of the Glu-348 side chain carboxylate and the electron-deficient methyl group (cyan) in the N-adenylated sulfoximine that
mimics ammonia. Coloring: C, gray; H, white; O, red; N, blue; S, yellow; P, orange. Both images were rendered in PYMOL (DeLano Scientific
Software LLC, Palo Alto, CA).
site lysine residue (Lys-449 in AS-B) is critical for inhibitor
binding (17).
primers were obtained from Integrated DNA Technologies, Inc.
(Coralville, IA), and DNA sequencing was performed by the core
facility in the Interdisciplinary Center for Biotechnology Research
at the University of Florida. 2,4-Dinitrofluorobenzene (DNFB)
was obtained from Aldrich (St. Louis, MO). Caution: Extreme care
should be taken when handling solutions of DNFB in organic
solvents because this reagent is a potent allergen and will penetrate
many types of laboratory gloves (21).
Despite this initial success in validating our model for how the
adenylated sulfoximine might bind to the synthetase site, we are
undertaking a series of systematic studies to test other predictions
of the computational structure. For example, simple modeling
suggested that the Glu-348 side chain could be positioned close to
the methyl group of sulfoximine 1, a moiety thought to mimic an
ammonia molecule in the transition state undergoing nucleo-
philic attack on a β-aspartyl-AMP intermediate 2 ( βAspAMP) (6),
which is tightly bound by the enzyme (8, 18, 19). We now report
the results of efforts to ascertain the functional role of Glu-348 in
catalysis, which have sought to confirm the possibility that Glu-
348 functions as a general base to facilitate nucleophilic attack
of enzyme-bound ammonia on βAspAMP (Scheme 2). Some-
what unexpectedly, this work implicates Glu-348 in mediating
βAspAMP formation and suggests that this conserved residue
may also participate in the molecular mechanisms coordinating
catalytic activity in the weakly coupled glutaminase and synthe-
tase sites of the enzyme.
Expression and Purification of C-Terminally Tagged
AS-B and AS-B Mutants. The asnB gene encoding AS-B
was cloned into the pET-21c(þ) plasmid (Novagen, Madison,
WI) and expressed following literature procedures (22). Site-
directed mutagenesis to obtain genes encoding AS-B mutants was
performed using the QuikChange kit (Stratagene, La Jolla, CA)
using standard primers. After expression of the C-terminally
tagged AS-B or AS-B mutant, the cellular pellet was resuspended
in 50 mM EPPS, pH 8, containing 300 mM NaCl, 10 mM
imidazole, 1% Triton X, and 0.5 mM DTT (50 mL) before
sonication. Cellular debris was removed by centrifugation
(10000g, 20 min, 4 °C), and the resulting supernatant was loaded
onto a Ni-NTA column (Qiagen, Valencia, CA). This column
was then washed with 50 mM EPPS, pH 8.0, containing 300 mM
NaCl, 20 mM imidazole, and 0.5 mM DTT (5 ꢀ 3 mL) before the
His6-tagged AS-B or AS-B mutant was eluted with 50 mM EPPS,
pH 8.0, containing 300 mM NaCl, 250 mM imidazole, and
0.5 mM DTT (10 ꢀ 3 mL). Fractions containing enzyme were
identified using SDS-PAGE, pooled, and dialyzed against
50 mM EPPS, pH 8.0, containing 5 mM DTT. Glycerol was
MATERIALS AND METHODS
General. Unless otherwise stated, all chemicals were purchased
from Sigma (St. Louis, MO) or Fischer Scientific (Pittsburgh, PA)
and were of the highest available purity. L-Glutamine was recrys-
tallized prior to use, as previously described (8). Protein concentra-
tions were determined using the Bradford assay (Pierce, Rockford,
IL) (20) and corrected as previously described (18). All DNA

Article
Biochemistry, Vol. 49, No. 43, 2010 9393
Scheme 2: Reactions Catalyzed in the Synthetase Active Site of
Glutamine-Dependent ASNS Showing the Structure of the βAs-
pAMP Intermediate 2 and the Transition State for the Subsequent
Attack of Ammonia To Form Asparagine and AMP
addition of trichloroacetic acid, to a final concentration of 4%,
and the resulting solution was added to 300 mM glycine/250 mM
hydrazine buffer, pH 9.0, containing 1.5 mM NADþ and 1 mM
ADP. L-Glutamate dehydrogenase (2.5 units) (Sigma, St. Louis,
MO) was then added (final volume 500 μL), and the absorbance
of the solution at 340 nm was taken after 30 min incubation at
room temperature. The amount of L-glutamate was then quanti-
fied using a standard curve. Measurements were made at specific
substrate and enzyme concentrations in triplicate, and data were
analyzed to obtain the values of V and V/K using standard
computer-based methods (23).
The ratio of L-asparagine and L-glutamate formed in the
synthetase reaction was determined independently by monitoring
peaks associated with both amino acids in a published HPLC-
based assay (18, 24). Hence, reaction mixtures containing enzyme
(2-4 μg), 5 mM ATP, 10 mM
L-aspartic acid, and either 100 mM
NH4Cl or 20 mM -glutamine dissolved in 100 mM EPPS,
L
pH 8.0, containing 10 mM MgCl2 (1 mL total volume) were
incubated for 20 min at 37 °C. Reactions were initiated by the
addition of AS-B, or the appropriate AS-B mutant, and
quenched using glacial AcOH (4%) to precipitate the enzyme.
The solution was then neutralized using 10 M NaOH, and an
aliquot (40 μL) was mixed with 400 mM Na2CO3 buffer (80 μL),
pH 9, DMSO (20 μL), and a saturated solution of DNFB in
absolute EtOH (60 μL). The resulting solution was heated at
50 °C for 50 min before being cooled and quenched with glacial
AcOH (40 μL). The mixture was then analyzed by reverse-phase
HPLC using a Varian Microsorb C18 column. The derivatized
amino acids were eluted using a step gradient of 40 mM formic
acid buffer, pH 3.6, and CH3CN, such that the initial CH3CN
concentration was 14%. After 30 min the amount of CH3CN was
increased to 20% over a period of 30 s, and elution continued for
an additional 10 min. 2,4-Dinitrobenzene-tagged asparagine and
glutamate were detected by UV (365 nm) and quantified by
comparison to authentic standards (Sigma, St. Louis, MO).
Preparation of [18O]Aspartic Acid and [ β,γ-18O6]ATP.
added (to a final concentration of 20%) to the dialyzed enzyme
(approximately 1 mg/mL), and aliquots were stored at -80 °C.
Steady-State Kinetic Assays. Synthetase activity was as-
sayed by monitoring the production of inorganic pyrophosphate
(PPi) spectrophotometrically using a standard coupled enzyme
assay (Sigma Technical Bulletin BI-100) (10).3 Assay mixtures
[18O]-
L-Aspartic acid was prepared by incubating the unlabeled
consisted of 5 mM ATP, 20 mM
reagent (350 μL) dissolved in 100 mM EPPS, pH 8.0, containing
10 mM MgCl2 and either 20 mM -glutamine or 100 mM NH4Cl
L-aspartic acid, and PPi assay
amino acid in 95% H218O under acidic conditions for 14 days, as
outlined previously (8). The final reaction mixture was lyophi-
lized, [18O]aspartic acid was dissolved in water to a final
concentration of 2.1 M, and the solution pH was adjusted to 8
using 10 M NaOH. The extent of 18O labeling was determined by
isocratic C18 LC-ESI mass spectrometry in negative ion mode
(Thermo-Finnigan LCQ). 18O-Labeled ATP was prepared from
18O-labeled carbamoyl phosphate,4 AMP, a trace of ATP,
adenylate kinase, and carbamate kinase as described pre-
viously (25), with minor modifications to the literature
procedure (26). Integration of the 31P signals associated with
the β- and γ-phosphate atoms in the 18O-labeled ATP
indicated that the average level of isotope incorporation
was 75%.
L
as a nitrogen source (1 mL total volume). Reaction was initiated
by the addition of enzyme (2-4 μg), and PPi formation was
determined over a period of 20 min at 37 °C by monitoring the
change in absorbance at 340 nm. In experiments to determine
steady-state kinetic parameters, the initial concentration of either
ATP (0-3 mM) or -aspartic acid (0-10 mM) was varied while
L
maintaining the other substrates at the levels outlined above.
Measurements were made at specific substrate and enzyme con-
centrations in triplicate, and data were analyzed to obtain the
values of V and V/K using standard computer-based methods (23).
The determination of the glutaminase activity exhibited by the
C-terminally tagged AS-B and AS-B mutants was performed
using standard procedures (22). Hence, the enzyme (1.5 μg) was
incubated in the presence, or absence, of 5 mM ATP with
Isotope (18O) Transfer Experiments. Reaction mixtures
were prepared containing 100 mM NH4Cl, 10 mM ATP, 20 mM
MgCl2, and 20 mM [18O]aspartic acid dissolved in 100 mM
EPPS, pH 8.0 (1 mL total volume). Asparagine synthesis was
initiated by addition of AS-B (200 μg), and the reaction mixture
was incubated for 3 h at 37 °C before being quenched with
trichloroacetic acid (4% final concentration). After enzyme
precipitation by centrifugation, the supernatant (700 μL) was
L-glutamine (0.2-50 mM) in 100 mM EPPS, pH 8.0, containing
100 mM NaCl and 8 mM MgCl2 (200 μL total volume) for 20 min
at 37 °C. The glutaminase reaction was then quenched by the
3This assay is composed of the enzymes pyrophosphate-dependent
fructose-6-phosphate kinase, aldolase, triose phosphate isomerase, and
glycerophosphate dehydrogenase. Pyrophosphate production is there-
fore measured by monitoring the oxidation of NADH spectroscopically
at 340 nm.
4The sample of isotopically labeled carbamoyl phosphate contained
four 18O atoms bonded to phosphorus (26).

9394 Biochemistry, Vol. 49, No. 43, 2010
Meyer et al.
Table 1: Steady-State Kinetic Parameters for the Glutaminase Activity of C-Terminally Tagged AS-B and the Site-Specific AS-B Mutantsa
ATP absent
kcat (s-1
5 mM ATP present
kcat (s-1
enzyme
Km (mM)
)
kcat/Km (M-1 s-1
)
Km (mM)
)
kcat/Km (M-1 s-1
)
WT AS-B
E348D
E348A
5.0 ( 1.4 (1.67)b
2.7 ( 0.4
5.8 ( 0.8
6.2 ( 0.2 (3.38)
4.1 ( 0.1
6.37 ( 0.04
4.0 ( 0.5
1190 (2020)
1510
1.7 ( 0.5 (1.30)
1.1 ( 0.2
3.5 ( 0.7
6.6 ( 0.1 (5.91)
10.02 ( 0.07
5.8 ( 0.2
3880 (4540)
9100
1100
1660
E348Q
9 ( 4
450
3.9 ( 0.3
4.45 ( 0.07
1140
aReaction mixtures of enzyme (1.5 μg) and
-glutamine (0.2-50 mM) dissolved in 100 mM EPPS, pH 8.0, containing 100 mM NaCl and 10 mM MgCl2
L
(200 μL total volume) were incubated, in the presence or absence of 5 mM ATP, for 20 min at 37 °C. Glutamate was quantified using glutamate
dehydrogenase (22). bValues shown in parentheses are literature values reported previously for recombinant, untagged WT AS-B (18).
Table 2: Steady-State Kinetic Parameters for the Synthetase Activity of C-Terminally Tagged AS-B and the E348D AS-B Varianta
ATP
kcat (s-1
aspartate
kcat (s-1
enzyme
Km(app) (mM)
)
kcat/Km(app) (M-1 s-1
)
Km(app) (mM)
)
kcat/Km(app) (M-1 s-1
)
Glutamine-Dependent Activity
WT AS-B
E348D
0.10 ( 0.01 (0.26)b
0.013 ( 0.004
0.90 ( 0.01 (2.18)b
0.51 ( 0.02
9000 (8380)b
39200
0.58 ( 0.04 (0.85)b
0.13 ( 0.01
0.67 ( 0.02 (1.56)b
0.45 ( 0.01
1200 (1830)b
3450
Ammonia-Dependent Activity
WT AS-B
E348D
0.11 ( 0.03
0.03 ( 0.01
0.96 ( 0.02
0.43 ( 0.03
8700
1.2 ( 0.1 (0.53)c
0.23 ( 0.07
0.75 ( 0.03 (1.09)c
0.30 ( 0.02
630 (800)c
1300
14300
aAll measurements were performed at pH 8. Assay mixtures consisted of enzyme (2-4 μg), PPi assay reagent (350 μL), and either 20 mM
100 mM NH4Cl dissolved in 100 mM EPPS containing 10 mM MgCl2 (1 mL total volume). Assay mixtures were augmented with 20 mM -aspartic acid and
-aspartic acid, respectively. bValues
L-glutamine or
L
ATP (0-3 mM) or 5 mM ATP and aspartic acid (0-10 mM) when determining steady-state parameters for ATP and
L
shown in parentheses are literature values reported previously for untagged WT AS-B (18). cValues shown in parentheses are literature values reported
previously for untagged WT AS-B (30) although the published values of kcat have been corrected by a factor of 2.73 for reasons discussed elsewhere (8).
transferred to a new tube containing EDTA (0.0287 g), glycine
(0.045 g), and D2O (200 μL). The pH was adjusted to 9.5 using
10 M NaOH, and an aliquot was transferred to an NMR tube
(800 μL). 31P NMR spectra were recorded at 121 MHz with
broad-band proton decoupling on a Mercury 300 spectrometer,
at a sample temperature of 23 °C. Spectra were obtained from
500 scans with an acquisition time of 4.0 s. 31P chemical shifts
were referenced to 85% phosphoric acid.
Positional Isotope Exchange (PIX) Experiments. Posi-
tional isotope exchange was examined by ascertaining the extent
of 18O incorporation from [ β,γ-18O6]ATP as a function of
time (28). Assay mixtures contained 5 mM [ β,γ-18O6]ATP, 10
revealed the presence of a contaminating ATPase activity in
preparations of the E348A AS-B mutant, where Glu-348
is replaced by alanine. A similar contaminating activity was
reported in samples of hASNS obtained from a yeast-based
expression system (29), which was eventually identified as being
associated with the presence of a DNA damage-responsive
protein (30). In an effort to determine whether we had unmasked
an intrinsic ATPase activity associated with the synthetase site of
the E348A AS-B mutant or were merely observing an impurity in
the recombinant enzyme, we expressed and purified an AS-B
variant containing a C-terminal polyhistidine (His6) tag. As
observed for human ASNS (7), steady-state assays showed that
the presence of the purification tag did not greatly perturb the
steady-state kinetic parameters for glutaminase activity com-
pared to that of untagged AS-B (Table 1). In addition, the
presence of 5 mM ATP in the assay gave a 3-fold increase in the
specificity constant kcat/Km for the tagged enzyme (Table 1),
consistent with previous observations on untagged AS-B (22, 31).
An approximately 2-fold decrease in kcat was observed for both
the glutamine- and ammonia-dependent synthetase activities of
C-terminally tagged AS-B (Table 2). These small changes in
activity are also reflected by the approximately 7.2:1 Glu:Asn
ratio observed when the C-terminally tagged enzyme is incubated
with saturating concentrations of all substrates (Table 3) rather
than the 1.8:1 Glu:Asn ratio reported for untagged AS-B under
identical conditions (18). We note that the concentration of
asparagine and glutamate formed under the reaction conditions
was measured directly using an HPLC-based assay (18, 24). In
light of the fact that kcat/Km values of the glutaminase and
synthetase activities were little changed by the presence of the
C-terminal His6 motif, all of the AS-B mutants used to delineate
mM L-Asp, and 10 mM MgCl2 dissolved in 100 mM EPPS, pH
8.0, with the appropriate enzyme (3 μM AS-B, 3 μM E348D,
3 μM E348A) (4 mL total volume). Reactions were incubated at
37 °C and aliquots (500 μL) removed at 30 min intervals, which
were then quenched by the addition of 1.0 M EDTA (125 μL) and
D2O (125 μL) before being transferred to an NMR tube. Control
experiments confirmed that these samples contained sufficient
EDTA to chelate all of the Mg2þ ions required for reaction. 31
P
NMR spectra were recorded at 121 MHz with broad-band
proton decoupling on a Mercury 300 spectrometer, at a sample
temperature of 23 °C. Spectra were obtained from 3000 scans
with an acquisition time of 4.0 s. 31P chemical shifts were
referenced to 85% phosphoric acid.
RESULTS AND DISCUSSION
Expression, Purification, and Characterization of C-Ter-
minally His6-Tagged AS-B. Early efforts to characterize the
functional role of Glu-348 in untagged, recombinant WT AS-B

Article
Biochemistry, Vol. 49, No. 43, 2010 9395
Table 3: Product Stoichiometry for the Glutamine-Dependent Synthetase
Activity of C-Terminally Tagged AS-B and the E348D AS-B Mutanta
Table 4: Product Stoichiometry for the Ammonia-Dependent Synthetase
Activity of C-Terminally Tagged AS-B and the E348D AS-B Mutanta
enzyme μmol of Asn min-1 mg-1 μmol of Glu min-1 mg-1 Glu:Asn
enzyme
μmol of Asn min-1 mg1 μmol of PPi min-1 mg-1 PPi:Asn
WT AS-B
E348D
0.64 ( 0.04
0.36 ( 0.04
4.6 ( 0.2
7.2:1
18:1
WT AS-B
E348D
0.63 ( 0.03
0.33 ( 0.01
0.59 ( 0.05
0.33 ( 0.06b
1:1
1:1
6.5 ( 0.7b
aReaction mixtures containing enzyme (2-4 μg), 5 mM ATP, 10 mM
-aspartic acid, and 20 mM -glutamine dissolved in 100 mM EPPS, pH 8.0,
aReaction mixtures containing enzyme (2-4 μg), 5 mM ATP, 10 mM
-aspartic acid, and 100 mM NH4Cl dissolved in 100 mM EPPS, pH 8.0,
L
L
L
containing 10 mM MgCl2 (1 mL total volume) were incubated for 20 min at
37 °C. Amino acids were derivatized and quantitated by HPLC using a
published procedure (18). bAlthough the E348A and E348Q AS-B mutants
did not synthesize asparagine under these conditions, they did exhibit
glutaminase activity and formed glutamate at rates of 2.8 ( 0.3 μmol of
Glu min-1 mg-1 and 4.1 ( 0.4 μmol of Glu min-1 mg-1, respectively.
containing 10 mM MgCl2 (1 mL total volume) were incubated for 20 min at
37 °C. Amino acids were derivatized and quantitated by HPLC using a
published procedure (18). bAlthough the E348A and E348Q AS-B mutants
did not synthesize asparagine under these conditions, they did exhibit a low
level of ATPase activity and formed inorganic pyrophosphate at rates of
0.04 ( 0.01 μmol of PPi min-1 mg-1 and 0.07 ( 0.01 μmol of PPi min-1
mg-1, respectively.
the functional role of Glu-348 were expressed so that the
C-terminal His6 tag was present and purified using a Ni-NTA
column.
Steady-State Kinetic Characterization of the C-Termin-
ally His6-Tagged E348A, E348D, and E348Q AS-B
Mutants. Although all of the three AS-B mutants exhibited
similar glutaminase activity to the WT enzyme in the absence of
ATP (Table 1), only the E348D AS-B mutant was capable of
catalyzing asparagine formation when glutamine was employed
as a nitrogen source (Table 2). Thus, no synthetase activity could
be observed for the E348A and E348Q mutants even using a
sensitive HPLC-based assay. Given that substitution of Glu-348
might have affected the structure of the protein, we evaluated the
effect of ATP on the glutaminase activities of the three AS-B
mutants. Perhaps surprisingly, the glutaminase activity of the
E348D AS-B mutant was stimulated more than 6-fold relative to
WT AS-B by the presence of 5 mM ATP (Table 1), suggesting
that any intramolecular interactions involved in coordinating
catalysis at the two active sites might have been affected by the
altered size of the side chain. In contrast, ATP-dependent
stimulation of glutaminase activity in the E348A and E348Q
AS-B mutants was less than that of WT enzyme, with only a
2-fold increase seen for the E348Q AS-B mutant (Table 1). Taken
overall, these results supported the hypothesis that Glu-348 was
functionally important, consistent with the predictions of our
model for the AS-B/sulfoximine complex.
Using direct measurements of glutamate, asparagine, and PPi
formation, we next examined whether replacing Glu-348 by
aspartate might give rise to “uncoupled” product stoichiometries.
Both WT AS-B and the E348D AS-B mutant formed pyrophos-
phate and asparagine in a 1:1 ratio when all substrates were
present (Table 4). In an unexpected observation, the E348A and
E348Q AS-B mutants were also able to form PPi under these
reaction conditions (Figure 2A) despite being unable to catalyze
asparagine formation, at least at the detection limits of our
sensitive HPLC-based assay (18). Hence, these two AS-B mu-
tants either possessed an intrinsic ATPase activity or were unable
to catalyze the addition of ammonia to the βAspAMP inter-
mediate. Similar findings were made when the nitrogen source
F
IGURE 2: Specific activities of PPi formation in the presence and
absence of aspartate for the (A) glutamine-dependent and (B) ammonia-
dependent reactions of WT AS-B and the three E348 mutants.
purification methodology, we attributed this activity to AS-B
itself rather than any contaminant, a hypothesis that is supported
by the fact that the E348D AS-B mutant converted ATP to AMP
and PPi in the absence of aspartate at a rate that was approxi-
mately 60% of that seen for asparagine production. Given that
the Asn:PPi ratio was 1:1 (at least within the limits of our assays)
for both WT AS-B and the E348D AS-B mutant when all
substrates are present in the reaction mixture, it seems that
aspartate acts to suppress this intrinsic ATPase activity, perhaps
through a conformational change that prevents water from
entering the synthetase active site or by ensuring that ATP is
bound within the enzyme so as to preclude its hydrolysis. Finally,
the presence of aspartate did not affect the low rate of ATP
breakdown catalyzed by the E348A and E348Q AS-B mutants to
any significant extent under these reaction conditions, which was
taken as evidence for the idea that these two mutants were
incapable of catalyzing the formation of βAspAMP.
was ammonia rather than L-glutamine (Figure 2B). The rate of
pyrophosphate production for WT AS-B and the three AS-B
mutants was therefore remeasured in assay mixtures from which
aspartate was absent, i.e., in the presence of only MgCl2, glu-
tamine, and ATP. These experiments revealed that AS-B did
seem to exhibit an intrinsic ability to catalyze the conversion of
ATP into AMP and PPi in the presence of a nitrogen source that
was 10% (or less) of the synthetase activity when aspartate was
omitted from the reaction mixture (Figure 2). Given the new

9396 Biochemistry, Vol. 49, No. 43, 2010
Meyer et al.
F
IGURE 3: (A) 31P NMR spectrum of the reaction mixture containing ATP, NH4Cl, MgCl2, 18O-labeled aspartate, and the E348D AS-B mutant
in EPPS, pH 8 (full details are provided in Materials and Methods). Peaks for 18O- and 16O-labeled AMP are separated by 0.02 ppm (inset). No
18O-labeled PPi was observed. (B) 31P NMR spectrum of the reaction mixture containing ATP, NH4Cl, MgCl2, 18O-labeled aspartate, and the
E348A AS-B mutant in EPPS, pH 8 (full details are provided inMaterials and Methods). No NMR detectable amounts of either AMP orPPi were
produced under these conditions. 31P chemical shifts are referenced to 85% phosphoric acid.
Another interesting finding in these studies of product stoi-
chiometry was that the Glu:Asn ratio for the E348D AS-B
mutant increased to a value of 18:1; i.e., the glutaminase and
synthetase activities had become more uncoupled than those of
the wild-type enzyme (7.2:1 ratio) (Table 3). This observation is
consistent with the hypothesis that Glu-348 plays a role in
facilitating the attack of ammonia on the βAspAMP intermedi-
ate, perhaps by acting as a general base. On the other hand, the
inability of the E348A and E348Q AS-B mutants to produce
asparagine even though their glutaminase activities were similar
to that of the wild-type enzyme suggested that the side chain of
Glu-348 might have another function in catalysis.
Steady-state kinetic parameters were determined for the syn-
thetase reaction catalyzed by the E348D mutant in an effort to
delineate the functional role of Glu-348 (Table 2). A 5-fold
decrease in the Km(app) value for aspartate was observed for both
the glutamine- and ammonia-dependent synthetase reactions,
together with a kcat that was half of that observed for the wild-
type enzyme, as measured by PPi production (Table 2). The latter
value is consistent with the relative rates of asparagine production
measured by HPLC for the two enzymes when all substrates are
present at saturating concentrations (Table 3). As a result of these
altered steady-state parameters, kcat/Km(app) for aspartate was
increased almost 3-fold for the E348D AS-B mutant when
glutamine was the nitrogen source. When ammonia was used in
these assays, a 2-fold increase in kcat/Km(app) for aspartate was
observed. The steady-state parameters for ATP also exhibited
similar changes for the E348D AS-B mutant independent of the
nitrogen source. Given that both kcat and Km(app) were decreased
for both substrates in the glutamine-dependent synthetase reac-
tion, it was possible that the mutation was slowing the formation
rate of βAspAMP relative to the rate of breakdown, which would
not be anticipated if ammonia addition to the intermediate was
being impaired by the shorter side chain in the E348D AS-B
mutant. In turn, this implied that the inability of the E348A and
E348Q AS-B mutants to form asparagine might be associated with
problems in forming the βAspAMP intermediate rather than with
enzyme-facilitated ammonia attack to form asparagine and AMP.

Article
Biochemistry, Vol. 49, No. 43, 2010 9397
Scheme 3: Schematic Representation of the Possible Outcomes of the 18O PIX Experiment If Acyl-Adenylation Is Reversible and the
PPi Formed Can Undergo Rotation or “Flipping” within the Synthetase Active Site (25)
18O Isotope Transfer Experiments. 31P NMR-based assays
were therefore undertaken to resolve the question of whether the
βAspAMP intermediate was being formed by the E348A AS-B
mutant. Thus, when His6-tagged WT AS-B was incubated with
18O-labeled aspartic acid, ATP, MgCl2, and NH4Cl in EPPS
buffer, pH 8, the 31P NMR spectrum showed signals from
unreacted ATP, PPi (δ 15.11 ppm), and two peaks for the 31P
nucleus in AMP separated by 0.02 ppm, providing clear evidence
for the incorporation of 18O into AMP and hence confirming
formation of the βAspAMP intermediate (8, 26, 27, 32). The
addition of unlabeled AMP into the sample confirmed the
assignment of the AMP peak. Similarly, incubation of the
E348D AS-B mutant with 18O-labeled aspartic acid under
identical conditions showed clear evidence for 18O incorporation
into AMP, again showing that this mutant was capable of
catalyzing the formation of βAspAMP (Figure 3A). We note
that the amounts of AMP and PPi relative to unreacted ATP were
lower, consistent with the reduced synthetase activity of the
E348D AS-B mutant. In contrast, incubation of the E348A AS-B
mutant under identical conditions gave a mixture for which
the 31P NMR spectrum only contained peaks from nuclei in
unreacted ATP (Figure 3B). Hence, no detectable levels of AMP
and PPi were produced by these mutant enzymes, providing
further support for the idea that Glu-348 plays an important role
in catalyzing βAspAMP formation. Given that replacement
of Glu-348 by alanine or glutamine might have altered the
“on-enzyme” equilibrium constant for βAspAMP formation
but that this intermediate was still being produced in a reversible
addition reaction, positional isotope exchange experiments were
undertaken to investigate this possibility (28).
assumed to be irreversible, despite conflicting reports concerning
the absence of ATP/PPi exchange reported in early work on
ASNS from various sources (8, 33). Furthermore, when incu-
bated with ATP and aspartic acid in the absence of a nitrogen
source, AS-B exhibits burst kinetics on the basis of either AMP or
PPi production, ruling out the possibility that all substrates must
be present before βAspAMP can be formed on the enzyme (8, 12).
Our PIX investigation afforded an opportunity to evaluate the
reversibility of βAspAMP intermediate formation in WT AS-B
because acyl-adenylate formation is accompanied by the release
of 18O-labeled PPi when the enzyme is incubated with
[ β,γ-18O6]ATP 3 (Scheme 3). As a result, the 18O labels in this
enzyme-bound reaction product can scramble their positions,
either by rotation or by “flipping” within the active site. In the
event that the adenylation reaction is reversible, there is a 67%
chance that an 18O label is incorporated into the oxygen bridging
the R- and β-phosphorus atoms in the regenerated ATP
(Scheme 3). Similarly, if PPi can “flip” within, or dissociate and
rebind to, the synthetase active site, then an 18O label will be
transferred to the R,β bridging position with concomitant
transfer of a 16O isotope to the γ-phosphorus. Clearly, a
combination of rotation and flipping events is possible, leading
to a mixture of differentially 18O-labeled ATP isotopomers.
These changes can be followed by monitoring the chemical shift
changes in the 31P NMR of the nuclei in the 18O-labeled ATP
present in the assay mixture (26).
His6-tagged WT AS-B was incubated with aspartate and
[ β,γ-18O6]ATP 3 (25) in the absence of either glutamine or
ammonia. These conditions were chosen to minimize any for-
ward commitment of the βAspAMP intermediate arising from
reaction with ammonia in the active site, which would diminish
the extent of PIX. However, no changes in the 31P signals for the
R- and γ-phosphorus peaks of ATP were observed under these
Positional Isotope Exchange (PIX) Experiments. In ear-
lier efforts to resolve the complicated kinetic mechanism of AS-B
(18) βAspAMP formation in the synthetase active site was

9398 Biochemistry, Vol. 49, No. 43, 2010
Meyer et al.
conditions over a period of 3.5 h (Figure 4A), setting an upper
limit of 0.003 s-1 for PIX. In addition, 31P resonances associated
with AMP and PPi were not observed in our NMR spectra
(Figure S1; see Supporting Information), which was somewhat
surprising given the low intrinsic ATPase activity observed for
WT AS-B in the absence of aspartate (Figure 2).
The upper limit of 0.003 s-1 for PIX in C-terminally tagged
WT AS-B is consistent with prior observations on argininosucci-
nate synthetase (34), another enzyme that forms an adenylated
intermediate in the absence of aspartate (35), and in which
the catalytic domain has the same three-dimensional fold as
the C-terminal domain of AS-B (Figure S2; see Supporting
Information) (36). Indeed, AS-B and argininosuccinate synthe-
tase are only two members of a large group of enzymes that form
adenylated intermediates and possess this domain fold (identified
by an SGGDXS loop motif, where X is a hydrophobic residue
(Figure 5) (37)), including β-lactam synthetase (BLS) (38), car-
bapenam synthetase (CPS) (39), ATP sulfurylase (40), ammonia-
dependent NADþ synthetase (41), 4-thiouridine synthetase (42),
and guanosine-50-monophosphate synthetase (GMPS) (43).
With one exception (see below), all enzymes in this family that
have been investigated (34, 38, 44), including AS-B (8), do not
catalyze the exchange of PPi into ATP perhaps because PPi is
bound deeply within their synthetase sites when it forms strong,
noncovalent interactions with the SGGDXS loop of the common
adenylating domain (Figure 6B), which precludes (i) PPi dis-
sociation until late in the kinetic mechanism and (ii) facile ATP/
PPi exchange. Moreover, on the basis of crystallography and
molecular modeling of AS-B (Ding et al., unpublished results)
PPi is likely coordinated to a Mg2þ ion that is bound by the
conserved aspartate (Asp-238 in AS-B) in the loop motif. We
hypothesize that the presence of the metal, and numerous PPi/
loop noncovalent interactions, results in restricted rotation about
the P-O bonds, thereby lowering the rate constant associated
with PIX to a value that is too low for observation using 31P
NMR. In contrast, both ATP/PPi exchange and a detectable PIX
rate have been reported for GMPS (44), perhaps reflecting
differences in the conformational properties of GMPS when
compared with other enzymes that possess this evolutionarily
related adenylation domain.
PIX experiments employing the E348D AS-B mutant under
identical conditions to those employed with WT AS-B also failed
to lead to detectable PIX (Figure 4B). Because 18O isotope
transfer shows that this mutant can form the βAspAMP inter-
mediate, this result meant that we could not rule out the
possibility that the impaired catalytic turnover of this AS-B
mutant relative to WT AS-B (Table 2) resulted from reversible
formation of βAspAMP rather than from impaired ammonia
attack on the intermediate due to incorrect positioning of the
carboxylate side chain as implied by our computational model
(Figure 1B). As for WT AS-B, 31P resonances associated with
AMP and PPi were not observed in the NMR spectrum of the
reaction mixture (Figure S3; see Supporting Information). Given
the intrinsic ATPase activity observed for the E348D AS-B
mutant in the absence of aspartate (Figure 2), it appears that
this requires the presence of a nitrogen source, which may be
necessary to induce the correct conformation for the conversion
of ATP into AMP and PPi.
F
IGURE 4: 31P NMR spectra of signals associated with the R-P (left)
and γ-P (right) in ATP isolated from PIX reaction mixtures contain-
ing (A) no enzyme, (B) WT AS-B, (C) the E348D AS-B mutant, and
(D) the E348A AS-B mutant after incubation for 3.5 h at 37 °C. No
NMR detectableamounts of eitherAMP orPPi were producedunder
these conditions. 31P chemical shifts are referenced to 85% phospho-
ric acid.
with the ATP present in these reaction mixtures after a period
of 3.5 h, a result consistent with the notion that this AS-B
mutant cannot catalyze production of the βAspAMP inter-
mediate. Unfortunately, the absence of PIX also means that
βAspAMP may be formed reversibly although the simplest
hypothesis remains that the E348A AS-B mutant cannot
In the final set of PIX studies, the His6-tagged E348A AS-B
mutant was incubated with aspartate and [ β,γ-18O6]ATP 3 (25),
again in the absence of either glutamine or ammonia. As before,
no PIX could be observed from the 31P resonances associated

Article
Biochemistry, Vol. 49, No. 43, 2010 9399
F
IGURE 7: Structural overlay of the “catalytic dyad” residues that
are conserved in BLS (Tyr-348 and Glu-382, blue) and CPS (Tyr-345
and Glu-380, green) with the corresponding residues (Glu-348 and
Asp-384) in the AS-B synthetase site of AS-B. The image was
rendered in PYMOL.
F
IGURE 5: Structural overlay of the “P-loop” in the adenylating
domains of AS-B (green), argininosuccinate synthetase (yellow), and
BLS (blue) together with the βAspAMP intermediate, MgPPi, and the
functionally important residues Asp-238 and Glu-348 in a computa-
tional model of the AS-B/glutamine/βAspAMP/MgPPi complex (Ding
and Richards, unpublished results). βAspAMP, PPi, Asp-238, and Glu-
348 are shown in stick representations, and the Mg(II) ion is rendered as
a magenta sphere. Atom coloring: C, gray; H, white; O, red; N, blue; P,
orange. Image rendered in PYMOL.
Scheme 4: Kinetic Model for the Glutaminase and Synthetase
Reactions Catalyzed by AS-Ba
aReprinted from ref 18 with permission from Elsevier.
CONCLUSIONS
Taken overall, these experiments clearly demonstrate that Glu-
348 has an important functional role in catalysis. Although it is
possible that this residue does participate as a general base during
the reaction of ammonia with βAspAMP, our experiments show
that the negatively charged side chain is somehow essential for
reaction of ATP with bound aspartate to form the adenylated
intermediate. Substantial proof for this assertion is provided by
the 18O-transfer experiments on the E348A and E348D AS-B
mutants. Such a hypothesis is also consistent with the altered Glu:
Asn stoichiometry exhibited by WT AS-B and the E348D AS-B
mutant. This can be seen by considering the kinetic scheme for
the substrate binding steps in AS-B (Scheme 4) (18). In this
proposal, glutamine hydrolysis takes place even in the presence of
saturating concentrations of ATP and aspartate because the rate
constants of glutamine binding to the E Asp ATP complex (k6)
F
IGURE 6: View of the intramolecular tunnel (black dots) in the
E. coli AS-B C1A/glutamine/AMP complex (1CT9) showing the
location of the Glu-348 (E348) side chain relative to the tunnel
connecting the two active sites. The cartoon representations of the
glutaminase and synthase domains of the C1A AS-B mutant are
colored blue and cyan, respectively. Glutamine and AMP are ren-
dered as CPK models while Glu-348 is shown in a stick representa-
tion. Coloring: C, gray; O, red; N, blue; P, orange. The image was
rendered in PYMOL.
3
3
catalyze formation of this intermediate, thereby precluding
asparagine production. More direct 31P NMR-based methods
of detecting enzyme-bound βAspAMP appear to be ruled out
by the size of the enzyme, which is active in solution as a dimer
(Li et al., unpublished observations). As in the other two PIX
studies, no 31P resonances associated with AMP and PPi were
observed (Figure S4; see Supporting Information), although
the very low intrinsic ATPase activity (Figure 2) might have
produced insufficient amounts of these products for detection
in our spectra. It remains possible, however, that ATP break-
down cannot take place in the absence of a nitrogen source.
and subsequent hydrolysis (k11) are similar to the rate of
βAspAMP formation (k3 or k8). In the case of the E348D AS-
B mutant, the observed rate of asparagine production is halved,
presumably because of a decrease in the rate constant for
βAspAMP formation (k3 or k8) that is similar in magnitude.
Thus, more glutamine is hydrolyzed per asparagine in the E348D
AS-B mutant because the steady-state concentration of the
E ATP Asp Gln complex is increased for the mutant enzyme
3
3
3
relative to WT AS-B under identical conditions. Of course, this
might also be the case if attack of ammonia on the intermediate
were significantly impaired by the loss of the Glu-348 side chain

9400 Biochemistry, Vol. 49, No. 43, 2010
Meyer et al.
Scheme 5: Reactions Catalyzed by (A) BLS and (B) CPS Showing the Similarity of Chemical Mechanisms of These Enzymes to That
of Glutamine-Dependent ASNS
acting as a general base, leading to reversible βAspAMP forma-
tion and leakage of ammonia from the enzyme. In this case, the
very low rate of PIX (<0.003 s-1) might then result from a lack
ability to control the proton transfer steps that result in
irreversible β-lactam ring formation. In light of this result, it
remains possible that Glu-348 can participate as a general base to
facilitate the attack of ammonia on βAspAMP. If this is the case,
then the decreased rate of βAspAMP formation in the E348D
AS-B, and the inability of the E348A and E348Q mutants to
adenylate aspartate, may arise from the lack of a large forward
commitment for asparagine production. The differences in the
ability of the E348A AS-B and Y345A CPS mutants to form the
adenylate derivatives of their respective substrates demonstrate
the difficulty of predicting subtle differences in active site proper-
ties for enzymes that share a common ancestor and catalyze
essentially identical chemical reactions.
of pyrophosphate rotation due to strong complexation by Mg2þ
.
A third explanation for the increased decoupling of the glutamin-
ase and synthetase activities in the E348D AS-B mutant is that
changing the length of the negatively charged side chain impairs
the adoption of a conformation that is necessary for the efficient
translocation of ammonia between the two active sites. In this
case, it can be imagined that the tunnel is more solvent exposed in
the E348D AS-B mutant leading to loss of ammonia because of
an impaired rate of βAspAMP formation.
Whichever of these arguments is correct, Glu-348 also seems to
participate in the molecular mechanisms that couple the gluta-
minase and synthetase activities, albeit to a smaller extent than
observed for other glutamine-dependent amidotransferases (45).
Thus, when ATP is present, the kcat/Km for glutamine in the
glutaminase reaction increases 6-fold for the E348D AS-B
mutant compared to only a 2-fold increase for the wild-type
enzyme. Removal of the side chain effectively eliminates the ATP
dependence of glutaminase activity (Table 1). On this point, we
note that the negatively charged side chain of Glu-348 is located
at the C-terminal end of the ammonia tunnel linking the active
sites (Figure 6) and may be involved in interactions that facilitate
ammonia transfer, perhaps in a manner similar to those observed
in computational studies on glutamine fructose-6-phosphate
amidotransferase (46, 47).
Our observations of the functional properties of Glu-348 are
given added interest by the observation that the tyrosine residue
located in the cognate position in the synthetase sites of BLS
(Tyr-348) and CPS (Tyr-345) has been shown to participate in
acid/base catalysis leading to formation of a β-lactam from an
acyl-AMP intermediate (Scheme 4) (48). As might have been
expected from the chemical similarity of the reactions catalyzed
by all three enzymes, evidence from crystallography (39, 49, 50)
and sequence alignment studies (51) is consistent with the
hypothesis that BLS and CPS, which are involved in secondary
metabolite biosynthesis, both evolved from ASNS. Structural
overlay clearly shows that Tyr-348, Tyr-345, and Glu-348 all
occupy identical locations in the three synthetase sites (Figure 7).
In contrast to our observation that mutation of Glu-348 impairs
acyl-adenylate formation, replacement of Tyr-345 in CPS by
alanine yields a mutant in which β-lactam ring formation is
impaired, leading to reversible formation of acyl-AMP 4 (Scheme 5)
and the appearance of ATP/PPi exchange (48). Thus, the Y345A
CPS mutant can catalyze the adenylation reaction but lacks the
ACKNOWLEDGMENT
We thank Dr. LaKenya Williams for synthesizing the sample
of [ β,γ-18O6]ATP that was used in the PIX experiments.
SUPPORTING INFORMATION AVAILABLE
Figures S1-S4 as described in the text. This material is
available free of charge via the Internet at http://pubs.acs.org.
REFERENCES
1. Richards, N. G. J., and Schuster, S. M. (1998) Mechanistic issues in
asparagine synthetase catalysis. Adv. Enzymol. Relat. Areas Mol. Biol.
72, 145–198.
2. Lorenzi, P. L., Llamas, J., Gunsior, M., Ozbun, L., Reinhold, W. C.,
Varma, S., Ji, H., Kim, H., Hutchinson, A. A., Kohn, E. C., Goldsmith,
P. K., Birrer, M. J., and Weinstein, J. N. (2008) Asparagine synthetase is
a predictive biomarker of L-asparaginase activity in ovarian cancer cell
lines. Mol. Cancer Ther. 7, 3123–3128.
3. Lorenzi, P. L, Reinhold, W. C., Rudelius, M., Gunsior, M., Shankavaram,
U., Bussey, K. J., Scherf, U., Eichlerl, G. S., Martin, S. E., Chin, K., Gray,
J. W., Kohn, E. C., Horak, I. D., Von Hoff., D. D., Raffeld, M.,
Goldsmith, P. K., Caplen, N. J., and Weinstein, J. N. (2006) Asparagine
synthetase as a causal, predictive biomarker for L-asparaginase activity in
ovarian cancer cells. Mol. Cancer Ther. 5, 2613–2623.
4. Aslanian, A. M., Fletcher, B. S., and Kilberg, M. S. (2001) Asparagine
synthetase expression alone is sufficient to induce L-asparaginase
resistance in MOLT-4 human leukaemia cells. Biochem. J. 357,
321–328.
5. Richards, N. G. J., and Kilberg, M. S. (2006) Asparagine synthetase
chemotherapy. Annu. Rev. Biochem. 75, 629–654.
6. Gutierrez, J. A., Pan, Y.-X., Koroniak, L., Hiratake, J., Kilberg,
M. S., and Richards, N. G. J. (2006) An inhibitor of human
asparagine synthetase suppresses proliferation of an L-asparaginase-
resistant leukemia cell line. Chem. Biol. 13, 1339–1347.
7. Ciustea, M., Gutierrez, J. A., Abbatiello, S. E., Eyler, J. R., and
Richards, N. G. J. (2005) Efficient expression, purification, and
characterization of C-terminally tagged, recombinant human aspar-
agine synthetase. Arch. Biochem. Biophys. 440, 18–27.

Article
Biochemistry, Vol. 49, No. 43, 2010 9401
8. Boehlein, S. K., Stewart, J. D., Walworth, E. S., Thirumoorthy, R.,
Richards, N. G. J., and Schuster, S. M. (1998) Kinetic mechanism of
Escherichia coli asparagine synthetase B. Biochemistry 37, 13230–
13238.
9. Boehlein, S. K., Walworth, E. S., Richards, N. G., and Schuster, S. M.
(1997) Mutagenesis and chemical rescue indicate residues involved in
β-aspartyl-AMP formation by Escherichia coli asparagine synthetase
B. J. Biol. Chem. 272, 12384–12392.
10. Boehlein, S. K., Richards, N. G., Walworth, E. S., and Schuster, S. M.
(1994) Arginine 30 and asparagine 74 have functional roles in the
glutamine dependent activities of Escherichia coli asparagine synthe-
tase B. J. Biol. Chem. 269, 26789–26795.
involved in β-aspartyl-AMP formation by Escherichia coli asparagine
synthetase B. J. Biol. Chem. 272, 12384–12392.
32. Villafranca, J. J. (1989) Positional isotope exchange using phos-
phorus-31 nuclear magnetic resonance. Methods Enzymol. 177, 390–
403.
33. Rognes, S. E. (1975) Glutamine-dependent asparagine synthetase
from Lupinus luteus. Phytochemistry 14, 1975–1982.
34. Hilscher, L. W., Hanson, C. D., Russell, D. H., and Raushel, F. M.
(1985) Measurement of positional isotope exchange rates in enzyme-
catalyzed reactions by fast atom bombardment mass spectrometry:
application to argininosuccinate synthetase. Biochemistry 24, 5888–
5893.
11. Boehlein, S. K., Walworth, E. S., and Schuster, S. M. (1997) Identi-
fication of cysteine-523 in the aspartate binding site of Escherichia coli
asparagine synthetase B. Biochemistry 36, 10168–10177.
12. Blundell, T. L., Jhoti, H., and Abell, C. (2002) High-throughput
crystallography for lead discovery in drug design. Nat. Rev. Drug
Discov. 1, 45–54.
13. Boehlein, S. K., Nakatsu, T., Hiratake, J., Thirumoorthy, R., Stewart,
J. D., Richards, N. G. J., and Schuster, S. M. (2001) Characterization
of inhibitors acting at the synthetase site of Escherichia coli asparagine
synthetase B. Biochemistry 40, 11168–11175.
14. Larsen, T. M., Boehlein, S. K., Schuster, S. M., Richards, N. G.,
Thoden, J. B., Holden, H. M., and Rayment, I. (1999) Three-dimen-
sional structure of Escherichia coli asparagine synthetase B: a short
journey from substrate to product. Biochemistry 38, 16146–16157.
15. Petrey, D., and Honig, B. (2005) Protein structure prediction: inroads
into biology. Mol. Cell 20, 811–819.
16. Arnold, K., Bordoli, L., Kopp, J., and Schwede, T. (2006) The
SWISS-MODEL workspace: a web-based environment for protein
structure homology modelling. Bioinformatics 22, 195–201.
17. Ikeuchi, H., Meyer, M. E., Ding, Y., Hiratake, J., and Richards,
N. G. J. (2009) A critical electrostatic interaction mediates inhibitor
recognition by human asparagine synthetase. Bioorg. Med. Chem. 17,
6641–6650.
18. Tesson, A. R., Soper, T. S., Ciustea, M., and Richards, N. G. J. (2003)
Revisiting the steady state kinetic mechanism of glutamine-dependent
asparagine synthetase from Escherichia coli. Arch. Biochem. Biophys.
413, 23–31.
19. Koroniak, L., Ciustea, M., Gutierrez, J. A., and Richards, N. G. J.
(2003) Synthesis and characterization of an N-acylsulfonamide in-
hibitor of human asparagine synthetase. Org. Lett. 5, 2033–2036.
20. Bradford, M. M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 72, 248–254.
35. Ghose, C., and Raushel, F. M. (1985) Determination of the mecha-
nism of the argininosuccinate synthetase reaction by static and
dynamic quench experiments. Biochemistry 24, 5894–5898.
36. Goto, M., Omi, R., Miyahara, I., Sugahara, M., and Hirotsu, K.
(2003) Structures of argininosuccinate synthetase in enzyme-ATP
substrates and enzyme-AMP product forms. J. Biol. Chem. 278,
22964–22971.
37. Bork, P., and Koonin, E. V. (1994) A P-loop-like motif in a wide-
spread ATP pyrophosphatase domain: implications for the evolution
of sequence motifs and enzyme activity. Proteins: Struct., Funct.,
Genet. 20, 347–355.
38. Bachmann, B. O., and Townsend, C. A. (2000) Kinetic mechanism of
the β-lactam synthetase of Streptomyces clavuligerous. Biochemistry
39, 11187–11193.
39. Miller, M. T., Gerratana, B., Stapon, A., Townsend, C. A., and
Rosenzweig, A. C. (2003) Crystal structure of carbapenam synthetase
(CarA). J. Biol. Chem. 278, 40996–41002.
40. Mougous, J. D., Lee, D. H., Hubbard, S. C., Schelle, M. W., Vocadlo,
D. J., Berger, J. M., and Bertozzi, C. R. (2006) Molecular basis for G
protein control of the prokaryotic ATP sulfurylase. Mol. Cell 21, 109–
122.
41. Rizzi, M., Bolognesi, M., and Coda, A. (1998) A novel deamido-
NAD(þ)-binding site revealed by the trapped NAD-adenylate inter-
mediate in the NAD(þ) synthetase structure. Structure 6, 1129–
1140.
42. Waterman, D. G., Ortiz-Lombardi, M., Fogg, M. J., Koonin, E. V.,
and Antson, A. A. (2006) Crystal structure of Bacillus anthracis ThiI, a
tRNA-modifying enzyme containing the predicted RNA-binding
THUMP domain. J. Mol. Biol. 356, 97–110.
43. Tesmer, J. G., Klem, T. J., Deras, M. L., Davisson, V. J., and Smith,
J. L. (1996) The crystal structure of GMP synthetase reveals a novel
catalytic triad and is a structural paradigm for two enzyme families.
Nat. Struct. Biol. 3, 74–86.
21. Thompson, J. S., and Edmonds, O. P. (1980) Safety aspects of
handling the potent allergen FDNB. Ann. Occup. Hyg. 23, 27–33.
22. Boehlein, S., Richards, N., and Schuster, S. (1994) Glutamine-depen-
dent nitrogen transfer in Escherichia coli asparagine synthetase B.
J. Biol. Chem. 269, 7450–7457.
44. Von der Saal, W., Crysler, C. S., and Villafranca, J. J. (1985)
Positional isotope exchange and kinetic experiments with Escherichia
coli guanosine 50-monophosphate synthetase. Biochemistry 24, 5343–
5350.
45. Richards, N. G. J., Humkey, R. N., Li., K., Meyer, M. E., and
Sintjago de Cordova, T. C. (2010) Tunnel and intermediates in the
glutamine-dependent amidotransferases, in Comprehensive Natural
Products II: Chemistry & Biology (Mander, L., and Liu, H.-W., Eds.)
Vol. 8, pp 161-230, Elsevier, Oxford.
46. Floquet, N., Durand, P., Maigret, B., Badet, B., Badet-Denisot,
M.-A., and Perahia, D. (2009) Collective motions in glucosamine-
6-phosphate synthase: influence of ligand binding and role in ammo-
nia channeling and opening of the fructose-6-phosphate binding site.
J. Mol. Biol. 385, 653–664.
23. Cleland, W. W. (1979) Statistical analysis of enzyme kinetic data.
Methods Enzymol. 62, 151–160.
24. Li, K. K., Beeson, W. T., Ghiviriga, I., and Richards, N. G. J. (2007) A
convenient gHMQC-based NMR assay for investigating ammonia
channeling in glutamine-dependent amidotransferases: studies of
Escherichia coli asparagine synthetase B. Biochemistry 46, 4840–4849.
25. Williams, L., Fan, F., Blanchard, J. S., and Raushel, F. M. (2008)
Positional isotope exchange analysis of the Mycobacterium smegmatis
cysteine ligase (MshC). Biochemistry 47, 4843–4850.
26. Cohn, M., and Hu, A. (1980) Isotopic oxygen-18 shifts in phosphorus-
31 NMR of adenine nucleotides synthesized with oxygen-18 in various
positions. J. Am. Chem. Soc. 102, 913–916.
27. Luehr, C. A., and Schuster, S. M. (1985) Purification and character-
ization of beef pancreatic asparagine synthetase. Arch. Biochem.
Biophys. 237, 335–346.
47. Mouilleron, S., Badet-Denisot, M.-A., and Golinelli-Pimpaneau, B.
(2008) Ordering of C-terminal loop and glutaminase domains of
glucosamine-6-phosphate synthase promotes sugar ring opening
and formation of the ammonia channel. J. Mol. Biol. 377, 1174–1185.
48. Raber, M. L., Arnett, S. O., and Townsend, C. A. (2009) A conserved
tyrosyl-glutamyl catalytic dyad in evolutionarily linked enzymes:
carbapenem synthetase and β-lactam synthetase. Biochemistry 48,
4959–4971.
28. Raushel, F. M., and Villafranca, J. J. (1988) Positional isotope
exchange. CRC Crit. Rev. Biochem. 23, 1–26.
29. Sheng, S., Moraga-Amador, D. A., van Heeke, G., Allison, R. D.,
Richards, N. G. J., and Schuster, S. M. (1993) Glutamine inhibits the
ammonia-dependent activities of two Cys-1 mutants of human aspar-
agine synthetase through the formation of an abortive complex.
J. Biol. Chem. 268, 16771–16780.
30. Sheng, S., and Schuster, S. M. (1993) Purification and characteriza-
tion of Saccharomyces cerevisiae DNA damage-responsive protein 48
(DDRP 48). J. Biol. Chem. 268, 4752–4758.
49. Miller, M. T., Bachmann, B. O., Townsend, C. A., and Rosenzweig,
A. C. (2002) The catalytic cycle of β-lactam synthetase observed by
x-ray crystallographic snapshots. Proc. Natl. Acad. Sci. U.S.A. 99,
14752–14757.
50. Miller, M. T., Bachmann, B. O., Townsend, C. A., and Rosenzweig,
A. C. (2001) Structure of β-lactam synthetase shows how to synthesize
antibiotics instead of asparagine. Nat. Struct. Biol. 8, 684–689.
51. Bachmann, B. O., Li, R., and Townsend, C. A. (1998) β-Lactam
synthetase: a new biosynthetic enzyme. Proc. Natl. Acad. Sci. U.S.A.
95, 9082–9086.
31. Boehlein, S. K., Walworth, E. S., Richards, N. G. J., and Schuster,
S. M. (1997) Mutagenesis and chemical rescue indicate residues