Citation
Chemical and kinetic characterization of glutamine hydrolysis catalyzed by Escherichia coli asparagine synthetase B

Material Information

Title:
Chemical and kinetic characterization of glutamine hydrolysis catalyzed by Escherichia coli asparagine synthetase B
Creator:
Schnizer, Holly G., 1969-
Publication Date:
Language:
English
Physical Description:
xiv, 165 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Amides ( jstor )
Ammonia ( jstor )
Biochemistry ( jstor )
Enzymes ( jstor )
Escherichia coli ( jstor )
Hydrolysis ( jstor )
Kinetics ( jstor )
Nitrogen ( jstor )
pH ( jstor )
Reaction mechanisms ( jstor )
Aspartate-Ammonia Ligase -- chemistry ( mesh )
Aspartate-Ammonia Ligase -- metabolism ( mesh )
Aspartate-Ammonia Ligase -- physiology ( mesh )
Catalysis ( mesh )
Department of Biochemistry and Molecular Biology thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Biochemistry and Molecular Biology -- UF ( mesh )
Escherichia coli -- chemistry ( mesh )
Glutamine -- chemistry ( mesh )
Hydrolysis ( mesh )
Models, Chemical ( mesh )
Papain -- chemistry ( mesh )
Papain -- metabolism ( mesh )
Papain -- physiology ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 152-164).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Holly G. Schnizer.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
0028368640 ( ALEPH )
50397440 ( OCLC )

Downloads

This item has the following downloads:


Full Text









CHEMICAL AND KINETIC CHARACTERIZATION OF GLUTAMINE
HYDROLYSIS CATALYZED BY ESCHERICHIA COLI
ASPARAGINE SYNTHETASE B










BY

HOLLY G. SCHNIZER










A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA 1997























This work is dedicated to
my parents, Flint and Dottye Gray, and to my husband, Richard Schnizer,
with love.














ACKNOWvLEDGMENTS


I wish to thank everyone who helped me accomplish the work presented here. I would like to thank my supervisor, Dr. Sheldon Schuster, for his eagerness to discuss new ideas, his enthusiasm in my work, and for his friendship. My work benefited greatly from the guidance and helpful discussions provided by my supervisory committee: Dr. Tom O'Brien, Dr. Tom Yang, Dr. Richard Moyer, Dr. Harry Nick, and Dr. Don Allison. I would like to thank them especially for their continuing support even when my research took a change in direction. I am also indebted to Dr. Nigel Richards and Dr. Jon Stewart for their additional guidance in weekly discussions.I am especially grateful to my advisor on my first project on the hepatitis viruses, Dr. Johnson Lau, who introduced me to the fascinating world of viruses, and who had patience with me during my first few years in the laboratory.

My thanks goes to my friends in Dr. Schuster's Laboratory for providing the daily support I needed to complete this journey: Dr. Sue Boehlein, Ms. Ellen Walworth, and Mr. Gentry Wilkerson. I am especially thankful for Sue's helpful comments and ideas which helped guide my research.

Most of all I am grateful to my husband, Richard Schnizer, for all his helpful scientific advice and most of all for his love, patience, and support which helped me achieve this goal.


i1i














TABLE OF CONTENTS




PAGE

ACKNOW LEDGM ENTS ....................................................................................... iij

LIST OF TABLES .................................................................................................... vi

LIST OF FIGURES .................................................................................................. viii

ABBREVIATIONS .................................................................................................. xii

ABSTRACT ............................................................................................................... xiii

CHAPTERS

I INTRODUCTION ..................................................................................... 1

2 pH DEPENDENCE OF THE GLUTAMINASE REACTION 18
CATALYZED BY ASPARAGINE, SYNTHETASE B .................

Introduction ................................................................................... 18
M aterials and M ethods ............................................................. 23
Results .............................................................................................. 27
Discussion ....................................................................................... 40

3 EFFECT OF TEMPERATURE ON GLUTAMINE HYDROLYSIS 45
CATALYZED BY ASPARAGINE, SYNTHETASE B .................

Introduction ................................................................................... 45
M aterials and M ethods .............................................................. 46
Results .............................................................................................. 51
Discussion ....................................................................................... 64


iv







4 DETERMINATION OF THE RATE DETERMINING STEP
IN ASPARAGINE SYNTHETASE B-CATALYZED
GLUTAMINE HYDROLYSIS ........................................ 67

Introduction ............................................................ 67
Materials and Methods.............................................. 70
Results ..................................................................7
Discussion ............................................................... 98

5 EVIDENCE OF A THIOESTER INTERMEDIATE FORMED
DURING GLUTAMINE HYDROLYSIS CATALYZED BY
ASPARAGINE SYNTHETASE B...................................... 102


Introduction ............................................................ 102
Materials and Methods.............................................. 104
Results.................................................................... 106
Discussion ............................................................... 109

6 CHARACTERIZATION OF A THIOBSTER INTERMEDIATE ...120

Introduction............................................................ 120
Materials and Methods ............................................. 122
Results.................................................................... 123
Discussion............................................................... 134

7 DISCUSSION AND FUTURE DIRECTIONS ........................... 138

REFERENCES............................................................................. 152

BIOGRAPHICAL SKETCH ............................................................ 165













LIST OF TABLES


Table page

2.1 Oligonucleotides used in construction of sitedirected m utants of AS-B .................................................. 2 5

2.2 Kinetic constants of the AS-B-catalyzed hydrolysis
of L-glutam ine at 37C ......................................................... 3 1

2.3 Kinetic constants of the AS-B catalyzed hydrolysis
of L-glutamic-y-monohydroxamate at 370C ................3 4

2.4 Kinetic constants for glutamine hydrolysis
catalzyed by the AS-B mutant H47N at 37C ..............3 7

2.5 Values of pKa for the kcat/Km-pH profiles for
glutamine and LGH hydrolysis catalyzed by wildtype AS-B and for glutamine hydrolysis catalyzed
by the A S-B m utant H47N .................................................. 3 8

2.6 Kinetic parameters for wild-type AS-B and the
H29A, H80A, H47N, and R30A mutants at three
values of pH .............................................................................. 3 9

3.1 Thermodynamic properties of glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and the
R 30A m utant ........................................................................... 5 6

3.2 Kinetic constants for the synthesis of LGH
catalyzed by wild-type AS-B ............................................. 5 9





vi







4.1 Kinetic constants for AS-B catalyzed hydrolysis of
L-glutamine, L-glutamic acid-7-monohydroxamate,
and L-glutamic acid-7-hydrazide ..................................... 7 8

5.1 Isolation of a covalent glutamyl-enzyme
interm ediate by gel filtration ........................................... 11 5

5.2 Time dependence of y-glutamyl AS-B adduct
degradation in SDS solution ................................................ 11 6

5.3 Trapping the thioester by filter binding ...................... 11 7

5.4 Base lability of the covalent adduct isolated by gel
filtration ..................................................................................... 11 9

6.1 Characterization of the glutamyl-enzyme
interm ediate ............................................................................ 1 2 9

7.1 Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by wild-type AS-B according to schemes
(1) and (2) .................................................................................. 1 4 6

7.2 Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by AS-B according to equation (3) ............. 149

7.3 Comparison of the actual and simulated values
obtained for the kcat, slope, intercept, and [TE]
from Figure 7.1 ........................................................................ 15 0















vii













LIST OF FIGURES

Figures age

1.1 Mechanism of nitrogen transfer via an ammonia
intermediate as proposed by Mei and Zalkin, 1989 1 0

1.2 Mechanism of nitrogen transfer via an imide
intermediate 3 as proposed by Richards and
Schuster, 1992 ....................................................................... 1 4

1.3 Mechanism of nitrogen transfer via direct transfer
of the nitrogen from glutamine to aspartyl-AMP
as proposed by Stoker et al., 1996 ................................. 1 5

2.1 Sequence alignment of part of the GAT-domains
of known asparagine synthetases and a partial set
of class II amidotransferases showing highly
conserved residues .............................................................. 2 2

2.2 The pH dependence of kcat for wild-type AS-B
catalyzed-glutamine hydrolysis ..................................... 2 9

2.3 The pH dependence of kcat/Km for wild-type AS-B
catalyzed glutamine hydrolysis ...................................... 3 0

2.4 The pH dependence of kcat for wild-type catalyzed
LG H hydrolysis ...................................................................... 3 2

2.5 The pH dependence of kct/Km for wild-type AS-B
catalyzed LGH hydrolysis ................................................... 33

2.6 The pH dependence of kcat for glutamine
hydrolysis catalzyed by the AS-B mutant H47N ......3 5



viii







2.7 The pH dependence of kcat/Km for glutamine
hydrolysis catalzyed by the AS-B mutant H47N ......3 6

3.1 Temperature dependence of glutamine hydrolysis
catalyzed by wild-type AS-B .......................................... 5 3

3.2 Temperature dependence of LGH hydrolysis
catalyzed by wild-type AS-B .......................................... 5 4

3.3 Temperature dependence of glutamine hydrolysis
catalyzed by the AS-B mutant R30A ............................ 55

3.4 Effect of hydroxylamine concentration on the
synthesis of LGH catalyzed by wild-type AS-B .........5 7

3.5 Effect of glutamate concentration on the synthesis
of LGH catalyzed by wild-type AS-B ............................5 8

3.6 Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 5oC ............................... 60

3.7 Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 37C ...........................6 1

3.8 Replot of the slope from the double-reciprocal
plots of glutamine hydrolysis at 5C versus
glutam ate concentration .................................................... 6 2

3.9 Replot of the slope from the double-reciprocal
plots of glutamine hydrolysis at 370C versus
glutam ate concentration .................................................... 6 3

4.1 Partitioning of the thioester intermediate to Lglutamic-y- monohydroxamate using
hydroxylamine in a reaction catalyzed by wt AS-B 7 6

4.2 Partitioning of the thioester intermediate to
glutamate hydroxamate using hydroxylamine in
the glutaminase reaction catalyzed by AS-B
R 30A ............................................................................................ 7 7


ix








4.3 DON inactivation of AS-B in the absence of
glutam ine .................................................................................. 8 1

4.4 Replot of DON inactivation of AS-B in the absence
(0) and presence of glutamine (A, 20 tM; 0,50
[IM ) ............................................................................................... 8 2

4.5 DON inactivation of AS-B in the presence of 20 j.M
glutam ine .................................................................................. 8 4

4.6 DON inactivation of AS-B in the presence of 50 jaM
glutam in e .................................................................................. 8 5

4.7 DON inactivation of AS-B catalyzed LGH hydrolysis
in the absence of LGH .......................................................... 8 6

4.8 Replot of DON inactivation of AS-B catalyzed LGH
hydrolysis ................................................................................. 8 7

4.9 The effect of various concentrations of LGH on
DON inactivation of AS-B catalyzed LGH hydrolysis 8 8

4.10 LGH protection against DON inactivation of AS-B
catalyzed LGH hydrolysis ..................................................8 9

4.11 DON inactivation of the R30A mutant of AS-B in
the absence of glutamine .................................................. 90

4.12 Replot of DON inactivation of R30A in the absence
of glutam ine ............................................................................ 9 1

4.13 Glutamine protection from DON inhibition of the
R30A catalyzed glutaminase reaction ........................... 92

4.14 Replot of glutamine protection of DON inactivation
of R30A catalyzed glutaminase ......................................9 3

4.15 Glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time .....................................9 5


x









4.16 Effect of the presence of L-glutamate on the burst
in product formation in AS-B catalyzed glutamine
hydrolysis ................................................................................. 9 6

4.17 Mechanism of glutamine hydrolysis catalyzed by
the class II glutamine amidotransferases as
proposed by Mei and Zalkin (1989) .............................. 97

5.1 Effect of glutamine on the steady-state
concentration of the thioester .......................................... 11 8

6.1 Rate of deacylation of the thioester intermediate
formed during wild-type AS-B catalyzed
glutam ine hydrolysis ........................................................... 1 2 6

6.2 The steady-state concentration of the thioester
formed in the presence of ATP ........................................1 27

6.3 Effect of ATP on the rate of deacylation ......................1 28

6.4 Effect of ammonia on the rate of deacylation ...........1 3 0

6.5 Effect of glutamine on the steady-state
concentration of thioester formed in the
glutaminase reaction catalyzed by the AS-B
m utant, R 30A .......................................................................... 1 3 1

6.6 Effect of glutamine on the steady-state
concentration of the thioester formed during the
glutaminase reaction catalyzed by the AS-B
m utant, N 74A .......................................................................... 1 3 2

6.7 The rate of deacylation of the AS-B mutant, N74A,
during glutamine hydrolysis ............................................ 1 3 3

7.1 Comparison of the simulated to the actual data
representing glutamate formation as a function of
tim e .............................................................................................. 1 4 5


xi













ABBREVIATIONS ALL: Acute Lymphoblastic Leukemia AS: asparagine synthetase AS-A: ammonia dependent asparagine synthetase A from E.coli AS-B: asparagine synthetase B from E. coli AMP: adenosine monophosphate ATP: adenosine 5'-triphosphate CIA: mutant of AS-B with cysteine-I replaced by alanine DON: 6-diazo-5-oxonorleucine E. coli: Escherichia coli
GAT: glutamine amide transfer GFAT: glutamine fructose- 6-phosphate amidotransferase GPA: glutamine 5-phosphoribosyl-l-pyrophosphate
amidotransf erase
H47N: mutant of AS-B with histidine-47 replaced by asparagine KIE: kinetic isotope effect LGH: L-glutamic-7-monohydroxamate N74A: AS-B mutant with asparagine-74 replaced by alanine Ppi: pyrophosphate R30A: AS-B mutant with arginine-30 replaced by alanine wt: wild-type







XH













Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

CHEMICAL AND KINETIC CHARACTERIZATION OF GLUTAMINE HYDROLYSIS CATALYZED BY ESCHERICHIA COLI ASPARAGINE SYNTHETASE B

By

Holly Schnizer

December, 1997

Chairman: Dr. Sheldon Schuster Major Department: Biochemistry and Molecular Biology

A detailed model of glutamine hydrolysis catalyzed by
Escherichia coli asparagine synthetase B, a class II amidotransferase, was constructed using a variety of chemical and kinetic techniques. Glutamine hydrolysis catalyzed by the class II amidotransferase family has been proposed to follow a mechanism similar to that of amide hydrolysis catalyzed by the thiol protease, papain. This study presents a comparison of the catalytic behavior of these two hydrolytic enzymes in order to determine if asparagine synthetase Bcatalyzed glutaminase reaction can be described by a similar mechanism. First, this study revealed differences in acid-base catalysis between these two hydrolytic enzymes. In a continuing investigation of the importance of conserved histidines in the


xiii








glutamine utilizing domain of asparagine synthetase B, the similarity of pH profiles of wild-type and an asparagine synthetase B mutant in which a conserved histidine is replaced with an asparagine, suggests that this histidine is not likely to be involved as part of the cys-his dyad which is essential to papain-catalyzed hydrolysis. Second, this study describes the development of a model for glutamine hydrolysis catalyzed by wild-type asparagine synthetase B from a minimal mechanism. The use of such a model was confirmed by the demonstration that one main feature of papain catalysis, the formation of a thioester, was conserved in asparagine synthetase B catalyzed glutamine hydrolysis.
The values for the dissociation constants for glutamine and the inhibition constants for glutamate were determined. Furthermore, the isolation of a putative thioester intermediate provided the opportunity to measure the rate of a single step in the reaction, deacylation. These constants along with the information gained from the kinetic constants of L-glutamic-y-monohydroxamate synthesis catalyzed by asparagine synthetase B were used in a computer simulation of the burst of glutamate production with time which showed that an additional rate limiting conformational change must be included to describe the observed kinetic behavior of wild-type asparagine synthetase B.









xiv












CHAPTER 1
INTRODUCTION



L-Asparagine and Treatment of Malignant Disease
L-asparagine was one of the first amino acids to be isolated from nature (Vauquelin & Robiquet, 1806). The importance of this non-essential amino acid in such metabolic events as the production of aspartate and ammonia (Broome, 1963a), transamination (Meister, 1952), protein synthesis (Cooney & Handschumacher, 1970), and glycosylation (Spiro, 1969) has been established. Considerable interest in L-asparagine metabolism developed after the 1953 discovery (Kidd, 1953a, 1953b) of the therapeutic effects of guinea pig serum against lymphomas in mice; and the subsequent identification of L-asparaginase, the enzyme which catalyzes asparagine hydrolysis, as the inhibitory component (Broome, 1961, 1963a, 1963b). Today, L-asparaginase is used most frequently in the treatment of acute lymphoblastic leukemia (ALL), the most common malignancy in children. When L-asparaginase is used as the sole chemotherapeutic agent, 40-68% of the patients achieve complete remission (Capizzi, et al., 1971; Uren & Handschumacher, 1977). In combination with other chemotherapeutic agents, L-asparaginase treatment results in almost 95% remission in previously untreated ALL (Sanz, et al., 1986). Despite the therapeutic potency of Lasparaginase, this treatment has several major disadvantages which




2


limit its use. First, L-asparaginase is toxic to organs such as the liver, pancreas, kidneys, blood clotting systems, and brain therefore, therapy is often accompanied by a wide variety of deleterious side effects (Crowther, 1971). These side effects are usually reversible and dose dependent however, in some cases they preclude the use of this drug. A second drawback to the use of L-asparaginase is that some patients who achieve remission suffer relapse with tumors resistant to the L-asparaginase therapy (Kiriyama et al., 1989; Lobel et al., 1979; Terebello et al., 1986). And finally, L-asparaginase treatment is confined to ALL because the enzyme may actually enhance the growth of some resistant tumors (Capizzi et al., 1971; Tallal et al., 1970) and, at present, there is no assay to predict sensitivity in solid tumors.
The therapeutic action of L-asparaginase is related to its ability to deplete the extracellular supply of L-asparagine (Broome, 1961; Uren & Handschumacher, 1977). While normal cells are independent of an external source of asparagine, sensitive tumor cells synthesize asparagine very slowly and are dependent on extracellular supplies of this amino acid to survive. Indeed, it has been shown that the activity of the enzyme which synthesizes asparagine, asparagine synthetase (AS), is depressed in tumor cells which are sensitive to asparaginase (Hutson et al., 1997). In contrast, increased AS activity is often observed in cell lines which exhibit a resistance to Lasparaginase (Gantt & Arfin, 1981; Kiriyama et al., 1989; Patterson & Orr, 1969) and a direct correlation between increased AS mRNA and protein content and resistance to asparaginase has been demonstrated in a human leukemia cell line (Hutson et al., 1997).




3


In view of the above mentioned problems associated with
asparaginase therapy, alternative methods of treating ALL need to be explored. Since the effectiveness of this therapy seems to reside in reducing the circulating concentration of asparagine, inhibitors of asparagine synthetase could provide this much needed alternative. In efforts to achieve this goal, several hundred compounds have been evaluated as AS inhibitors, however none have provided both the strength and specificity required for clinical consideration (Cooney et al, 1976). Therefore, a detailed chemical and kinetic investigation of AS is essential in order to develop more effective inhibitors of this enzyme.

Asparagine Synthetase
Asparagine synthetases catalyze the ATP dependent transfer of nitrogen from either glutamine or ammonia to aspartic acid in the formation of asparagine. The biochemical and mechanistic properties of AS have been examined in a variety of sources including mammals (Hongo et al., 1978; Markin et al., 1979; Markin et al., 1981; Milman & Cooney, 1979; Van Heeke & Schuster, 1989b), fungi (Ramos & Wiame, 1979, 1980), plants (Pike & Beevers, 1982; Rognes, 1970), and bacteria (Burchall et al., 1964; Cedar & Schwartz, 1969a, 1969b; MacPhee et al., 1983) as well as in some tumor cell lines (Gantt & Arfin, 1981). In order to facilitate the recovery of sufficient amounts of AS for study, expression systems have been developed for the overexpression of human AS in yeast (Van Heeke & Schuster, 1990) and AS-B in Escherichia coli (Boehlein et al., 1994a).




4


Two classes of AS have been described which do not possess
any sequence similarity. AsnA, found in many procaryotes, encodes for an AS (AS-A) which only uses ammonia as a source of nitrogen in the synthesis of asparagine (reaction 2). The kinetic and chemical mechanism of AS-A, the product of the asnA gene, has been well characterized (Cedar & Schwartz, 1969a, 1969b). In fact, the existence of an aspartyl-AMP intermediate formed during AS catalyzed asparagine synthesis was demonstrated first in experiments using AS-A (Cedar & Schwartz, 1969b). The other AS, encoded by asnB found in both prokaryotes and eukaryotes, catalyzes the synthesis of asparagine using either ammonia or glutamnine as a source of nitrogen (reactions 1 and 2). These enzymes also catalyze glutamine hydrolysis (reaction 3) in the absence of the other substrates; a reaction which is stimulated by the presence of ATP.


L-Asp + L-Gln + ATP L-Asn + L-Glu +AMP + PP (Reaction 1) L-Asp + NH3 + ATP 4 L-Asn + AMP + PP, (Reaction 2)
L-Gln + H,O L-Glu + NH, (Reaction 3)


Glutamine dependent asparagine synthetases are thought to be organized into two functional domains. The glutamine amide transfer (GAT) domain binds glutamine and catalyzes the hydrolysis of glutamine to glutamate and ammonia. The synthetase domain binds ATP and aspartate and catalyzes the ammonia dependent reaction as well as the transfer of nitrogen to aspartate in the glutamine dependent reaction. Evidence for the location of the two




5


active sites in separate domains was revealed through studies using monoclonal antibodies (Pfeiffer et al., 1986, 1987) and chemical modifiers (Larsen & Schuster, 1992) that could selectively inhibit the glutaminase or asparagine synthesis activities of human or bovine AS.
While the mechanism of nitrogen transfer between the two domains is poorly understood, examination of conserved residues within the individual domains has led to the identification of binding sites and intermediates involved in the reaction. For the C-terminal synthetase domain, it has been confirmed that during asparagine synthesis, a P-aspartyl-AMP intermediate is formed which activates the aspartate carboxylate for attack by the amide nitrogen of glutamine (Luehr & Schuster, 1985). Further studies showed a binding site for aspartate located in a region highly conserved among asparagine synthetases (Boehlein et al., 1997a). And finally, another highly conserved region in the AS synthetase domain was found to contain a P-loop-like motif that is present in other enzymes which catalyze the hydrolysis of ATP to AMP and PP (Bork & Koonin, 1994), and has been proposed to represent a pyrophosphate binding element (Boehlein et al. manuscript in progress).
Experiments using site-directed mutagenesis have identified several conserved amino acids in the N-terminal domain that are crucial to glutamine utilization. Sequence analysis reveals that all AS contain an N-terminal cysteine residue. Milman and Cooney first proposed a critical cysteine residue in the glutamine active site when they found AS from mouse pancreas was strongly susceptible to oxidation (Milman & Cooney, 1979). More evidence for an important




6


cysteine residue was revealed upon the observation that bovine pancreatic AS lost glutamine dependent activity when it was covalently modified with the sulfhydryl modifier and glutamine analog, 6-diazo-5-oxo-L-norleucine (DON) (Mehlhaff & Schuster, 1991). Finally, the essentiality of the N-terminal cysteine residue was demonstrated in human AS and E. coli AS-B in experiments replacing cysteine-1 with either an alanine or serine residue (Boehlein et al., 1994a; Sheng et al., 1993; Van Heeke & Schuster, 1989b).
In addition to the conserved cysteine residue, the role of
another completely conserved residue, asparagine-74, also has been determined. When Asn-74 was replaced with either an alanine or an aspartate, the resulting mutant enzymes (N74A & N74D) exhibited a depressed glutaminase activity (Boehlein et al., 1994b). Subsequent studies using the alternative substrate, L-glutamic acid-'ymonohydroxamate (LGH) and the AS-B mutant N74A led to the suggestion that the interaction of this asparagine residue with substrates causes a destabilization in the ES complex (Boehlein et al., 1996) and, therefore, this residue might play a role similar to that of Gln-19 of papain (Drenth et al., 1976; Menard, 1995). For papain, it is thought that Gln-19 forms an 'oxyanion hole' in the stabilization of a tetrahedral intermediate formed when the active site cysteine attacks the substrate. This 'oxyanion hole' contains a hydrogen bonding network involving the NH peptide backbone of cys-25 and the side chain NH, of Gln-19 which stabilizes the oxide anion bonded to the tetrahedral carbon atom. The proposed role of N74 of AS-B in forming such an oxyanion hole is illustrated below:




7


O
1 ENZ
ASN-74 NH HN


H2N- C- S

R
CYS-1



This proposed relationship between the thiol protease, papain, and AS-B was confirmed in studies which showed that a nitrile could act as a weak competitive inhibitor of wild type AS-B and as a substrate for the AS-B mutant N74D (Boehlein et al., 1997b); results which were consistent with similar studies using wild type and a Q19E mutant of papain (Menard, 1995).


Glutamine Amidotransferase Class II
Based on its primary amino acid sequence, asparagine synthetase is a member of a family of enzymes known as the glutamine-dependent amidotransferases. The members of this family catalyze the transfer of nitrogen from glutamine to a variety of acceptor substrates in the biosynthesis of amino acids, carbohydrates, and nucleotides (for review, Zalkin, 1993). Like AS, all amidotransferases share a modular organization with a glutamine utilizing (GAT) domain in common and a synthetase domain which is unique to each member of the family. According to the amino acid sequence alignment of the GAT domain, glutamine amidotransferases are divided into two types (Zalkin, 1993). Class I enzymes (formerly 'G-type' for the trpG gene product, anthranilate synthase) are




8


characterized by three catalytically important residues thought to function as a cys-his-glu catalytic triad analogous to the thiol proteases. This class of enzymes (see Zalkin, 1993 for review) includes GMP synthetase (Tiedeman et al, 1985), anthranilate synthase, (Nichols et al., 1980) carbamoyl-phosphate synthetase (Piette, 1984), imidazole glycerol-phosphate synthase (Carlomagno, 1988), aminodeoxychorismate synthase, CTP synthetase (Weng, 1986), and FGAM synthetase (Schendel et al., 1989). Asparagine synthetase as well as glutamine 5'-phosphoribosyl-1-pyrophosphate amidotransferase (GPA) (Tso et al., 1982), glutamine fructose-6phosphate amidotransferase (GFAT) (Badet-Denisot et al., 1993), and glutamate synthase (Vanoni et al., 1991) comprise the class II amidotransferases (formerly 'F-type' for the purF gene product, glutamine 5-phosphoribosylpyrophosphate) which are characterized by a conserved N-terminal cysteine required for glutamine dependent activity.


Proposals for the Mechanism of Nitrogen Transfer
For the class II enzymes, the formation of a covalent yglutamyl-enzyme intermediate involving the N-terminal cysteine residue was first suggested in studies of GPA where it was shown that attachment of glutamine affinity analogs to cys-1 (Vollmer et al., 1983) or replacement with phenylalanine (Mantsala & Zalkin, 1984) caused selective elimination of the glutamine-dependent activity. Later studies using site directed mutagenesis of conserved residues in the GAT domain of GPA led to the proposal that cys-1 participated as part of a cys-his catalytic dyad in a mechanism analogous to the




9


thiol proteases (figure 1.1)(Mei and Zalkin, 1989). This model of nitrogen transfer involves a two step process: hydrolysis of glutamine to produce free ammonia, followed by its transfer to the acceptor substrate. In analogy to the thiol protease mechanism, a nucleophilic attack of the conserved cys-1 thiolate anion on the amide of glutamine results in a tetrahedral intermediate which, upon release of ammonia, is converted to an acylthioenzyme intermediate

1. The released ammonia is transferred to the activated aspartate intermediate 2 in the synthetase domain for the synthesis of asparagine while the acylenzyme intermediate undergoes hydrolysis to glutamate and free enzyme via a second tetrahedral intermediate. For AS, this model for nitrogen transfer is supported by the observation that all members of its family can catalyze the synthesis of asparagine using ammonia instead of glutamine, demonstrating the presence of binding site for ammonia. Furthermore, a completely conserved cysteine has been shown to be catalytically essential in GPA (Mei & Zalkin, 1989) and AS-B (Boehlein et al., 1994a) and a residue (Asn-74) thought to play a role in forming an oxyanion hole, much like that formed by Gln-19 in papain catalysis (Menard et al. 1995), has been shown for AS-B.
While there is some evidence supporting an ammonia mediated nitrogen transfer, there are also many inconsistencies. First, was the inability to show that the class II enzymes contain a catalytically important histidine cognate to that of the thiol protease papain. Papain requires an active site histidine which functions to stabilize the cysteine thiolate anion and provide a proton to the leaving group in the conversion of the first tetrahedral intermediate to the





10




( O NH2
H NH2 H HCys H

H O I
HisCYs O O HS102
2 H (N C02- <\ I

Cys1 0'

Asp2 + NH3 NH3


0
S
H3N K N
N' Cys / His12
CO2 N
O H


Asp29


(b) O O
Ad O
O 'O 0 NH
0- H NH2
H,- ),jk + AMP
H3N' C2O H
HO OH H3N "CO2
2

(C) 0 0

H3 H o H
HcCys / Hise2 + H20 H3N' N Hiso2
His10 + H20 4, H <
CO2 N C02 S N'r
O H H' transfer CysH

Asp29O Asp29




Figure 1.1 Mechanism of nitrogen transfer via an ammonia intermediate as proposed by Mei and Zalkin, 1989. (a) Mechanism for the hydrolysis of L-glutamine to yield ammonia and an acylenzyme intermediate 1 by analogy to the purF enzyme, GPA. (b) Synthesis of L-asparagine by reaction of ammonia with activated aspartyl-AMP 2. (c) Hydrolysis of the acylenzyme intermediate to yield L-glutamate. This diagram was taken from Richards and Schuster, 1992.








thioester. Initially, it was thought, from results of site-directed mutagenesis (Mei & Zalkin, 1989), that His-102 was involved in such acid-base catalysis for GPA and covalent modification studies using diethyl pyrocarbonate suggested that GFAT (Badet-Denisot & Badet, 1992) also possessed this important histidine. However, sitedirected mutagenesis of each of the conserved histidines in the GAT domain of AS-B revealed little or no change in the kinetic parameters (Boehlein et al., 1994a). Subsequent structural studies of GPA (Kim et al., 1996; Smith et al., 1994) and the N-terminal domain of GFAT (Isupov et al., 1996) did not reveal a histidine in the appropriate position to participate in acid-base catalysis. Alternatively, these structural studies led to the proposal that the a-amino group of cys-1 serves as the general acid-base catalyst (Smith, 1995). Subsequent structural comparisons of GPA with other hydrolytic enzymes placed the class II enzymes into a larger superfamily called the N-terminal nucleophile (Ntn) hydrolases which includes the 20S proteasome and penicillin acylase (Brannigan et al., 1995; Duggleby et al., 1995; Seemuller et al., 1996). This family is characterized by an N-terminal catalytic nucleophile which is thought to be activated by proton transfer from its side chain to the free N-terminus (Seemuller et al., 1996). However, experiments aimed at establishing the functional role of the N-terminal amine using mutants of GPA and GFAT with an additional N-terminal residue have yielded contradictory results (Isupov et al., 1996; Kim et al., 1996).
Further evidence against an ammonia mediated nitrogen
transfer in AS-B catalyzed asparagine synthesis was revealed in an investigation of the kinetic isotope effects (KIE) (Stoker et al. 1996)




12


for AS-B catalyzed glutamine dependent reactions. The KIEs associated with placing 5N in the primary amide of glutamine were examined for both glutamine dependent reactions and the resulting values were interpreted using 5N KIE determinations for papain catalyzed peptide hydrolysis (O'Leary et al, 1.974). For papain, the KIE value was shown to be consistent with a mechanism where the first irreversible step is C-N bond cleavage to form the thioester intermediate (O'Leary et al., 1974). It was expected that C-N bond cleavage would be rate limiting in the AS-B catalyzed glutaminase reaction as well and thus give the same 15N (V/K) isotope effect as papain. Furthermore, an ammonia mediated nitrogen transfer would require a common pathway for glutamine hydrolysis in the two glutamine dependent reactions. Therefore, if AS-B followed an ammonia mediated mechanism, it was expected that the 15N KIEs determined for both glutamine dependent reactions would be identical. Surprisingly, the glutaminase 15N (V/K) was significantly smaller than that obtained for papain catalyzed peptide hydrolysis, indicating that a step other than C-N bond cleavage may be rate limiting. Moreover, the KIE for glutaminase was smaller than that for the glutamine dependent synthesis reaction. These results were regarded as the first direct evidence against the mediation of nitrogen transfer by enzyme-bound ammonia.
In addition to the above mentioned inconsistencies between
AS-B catalytic behavior and an ammonia mediated nitrogen transfer, questions addressing the difficulties of maintaining the free ammonia in the active site in its unprotonated form led to the proposal of alternative mechanisms for nitrogen transfer (Figure 1.2) (Richards &




13


Schuster, 1992). The first proposal for an alternate mechanism of nitrogen transfer involved the formation of an imide intermediate (Richards & Schuster, 1992). According to this hypothesis, a direct nucleophilic attack of the primary amide of glutamine on the Paspartyl-AMP intermediate 2 to form an imide intermediate 3. Reaction of this intermediate with cys-1 causes the release of asparagine and the formation of a thioacylenzyme intermediate 1 which forms free enzyme and glutamate in a mechanism similar to the ammonia mediated mechanism. An attempt to isolate such an intermediate was made using the AS-B mutant (CIA) which lacks the cysteine required to catalyze C-N bond cleavage of the imide intermediate. However, an imide intermediate was not detected by HPLC using chemically synthesized imide as a standard, when CiA AS-B was incubated with glutamine, ATP, and aspartic acid. It was concluded that participation of an imide intermediate in nitrogen transfer catalyzed by AS-B was unlikely (Boehlein et al., unpublished observations).
In a second alternative hypothesis for the mechanism of

nitrogen transfer (Stoker et al., 1996), Cys-1 reacts with glutamine to form the first tetrahedral intermediate 5 (figure 1.3). In contrast to the ammonia mediated mechanism, however, the tetrahedral intermediate makes a subsequent direct attack on the P-aspartylAMP intermediate to form an intermediate in which glutamine is covalently linked to f3-aspartyl-AMP 6. Cleavage of the C-N bond in this intermediate releases the acylenzyme and forms a second tetrahedral intermediate 8. Asparagine is then formed from a loss of





14











oc oo o o o con"

NH 0, 0 H'transter N N*H
HN NHlj 0 AMP HH
H H3N 0 02 HO OH 3 H H
CO2 sH 2 CO2- HO


Cys, SO 1 O
sCysi Asps,
Asp3y


0 00
1 S HN
H S Asn H3N Cys, H C '
H3N K,, H3N CO"
Cys0 + H20 CO2 1 0
COz O. H'O


Asp3 Aspz
H' transfer
-H'

0


H3N O"

CO2 2SH
cys, 0

Asp33









Figure 1.2 Mechanism of nitrogen transfer via an imide intermediate
3 as proposed by Richards and Schuster, 1992. Attack of the primary amide occurs directly on the activated L-aspartate 2. Cys-1 and Asp-33 are then involved in the hydrolysis of the imide to yield L-glutamate and L-asparagine in subsequent steps.






15
















0 CYG,4 0NMO 0
M~j- H +H 1 H

CH, 0-7 HOO0
S' o*






0' 0' 0* N 0 H, H IH.N i CO;



H





H H




0







H2NN'













Figure 1 .3 Mechanism of nitrogen transfer via direct transfer of the nitrogen from glutamine to aspartyl-AMP as proposed by Stoker et al., 1996).




16


AMP from the tetrahedral intermediate. Interestingly, this mechanism would not necessarily require N-protonation before C-N bond cleavage in the glutamine dependent synthesis reaction and, therefore, may explain the differences in the 15N (V/K) isotope effects for the two glutamine dependent reactions.
Despite the progress that has been made in understanding the functions of the individual domains of AS, the mechanism of nitrogen transfer remains unequivocally defined. One difference between the mechanism involving free ammonia and that of a direct attack is the requirement, in the former, of a common mechanism for glutamine hydrolysis in the glutaminase and glutamine dependent asparagine synthesis reactions. Therefore, it is important to understand the mechanistic details of both the glutamine dependent synthesis and glutamine hydrolysis in order to discriminate between the two mechanisms. The ability of AS-B to catalyze the hydrolysis of glutamine in the absence of the other amino acids provides the opportunity to develop a model of the glutaminase reaction from which to identify the individual steps and intermediates in the reaction as well as examine the functional roles played by the conserved amino acids in the GAT domain. In efforts to begin building such a model, this dissertation focuses on the chemical and kinetic mechanisms of glutamine hydrolysis catalyzed by AS-B. Differences and similarities in the mechanism of substrate hydrolysis catalyzed by AS-B and papain will be examined through pH analysis of AS-B catalyzed glutaminase activity and by isolation of a covalent intermediate in the reaction pathway of glutamine hydrolysis catalyzed by AS-B. A variety of kinetic techniques will be used to




17


determine the individual rate constants of the glutaminase reaction and these will be used to build a model of glutamine hydrolysis catalyzed by AS-B.













CHAPTER 2
pH DEPENDENCE OF THE GLUTAMINASE REACTION CATALYZED BY ASPARAGINE SYNTHETASE B


Introduction
Asparagine Synthetase B (AS-B) catalyzes the ATP dependent synthesis of asparagine using aspartic acid and either glutamine or ammonia as a nitrogen source (reactions 1 and 2). In the absence of aspartate and ATP, this enzyme also is capable of catalyzing glutamine hydrolysis (reaction 3).


L-Gln + ATP + L-Asp -- L-Asn + AMP + PP + L-Glu (reaction 1) NH3 + ATP + L-Asp -- L-Asn + AMP + PPi (reaction 2)
L-Gln + H,20 --> Glu + NH3 (reaction 3)


Amino acid sequence alignment of the N-terminal glutamine utilizing domain of AS-B indicates that this enzyme is a member of the class II glutamine amidotransferase family (Zalkin, 1993) which includes glutamine 5'-phosphoribosylpyrophosphate amidotransferase (GPA) (Tso et al., 1982), glutamine fructose-6-phosphate amidotransferase (GFAT) (Badet-Denisot & Badet, 1993), and glutamate synthase (Vanoni et al., 1991). Like AS-B, members of this family of enzymes catalyze the transfer of the amide nitrogen of glutamine to various acceptor molecules including amino acids, nucleotides, and carbohydrates.


18




19


The mechanism of nitrogen transfer remains controversial and poorly understood. Results from studies using site-directed mutagenesis suggested the catalytic importance of a conserved N-terminal cysteine, a histidine, and an aspartic acid in glutamine utilization in GPA (Mei & Zalkin, 1989). This observation led to the hypothesis that glutamine amidotransferases catalyze glutamine hydrolysis by a mechanism similar to that of amide hydrolysis catalyzed by the thiol proteases, typified by papain (for review see Brocklehurst et al., 1987), as shown below (1).


Binding Acylation Deacylation
0 0NH, HHO
R-C R- C HS-Enz -' -R- CO aR- C + HS-Enz (1)
NH, NH2 O. S-Enz O OH
+I
R-C"NH2 R-C-OH
HS-Enz
S-Enz S-Enz
TH I TH II


In analogy to the papain mechanism, a nucleophilic attack of the Nterminal cysteine on the side-chain amide of glutamine forms a tetrahedral intermediate (TH I). Protonation of the amide allows C-N bond cleavage to produce a thioester intermediate and free ammonia. The anunmonia is then transferred to the acceptor molecule and the thioester intermediate is broken down, via another tetrahedral intermediate (TH II), to produce glutamate and free enzyme. As illustrated above, an active site cysteine and an acid-base catalyst (a role played by his-159 in papain) are vital components to such a reaction scheme. The cysteine is involved in the activation of glutamine as described above. In papain, the histidine is thought to




20


play three essential acid-base roles including 1) stabilization of the thiolate; 2) donation of a proton to the amide of the first tetrahedral intermediate to increase its leaving group ability; and finally 3) activation of water for a nucleophilic attack on the thioester intermediate by accepting a proton (Storer & Menard, 1994).
This mechanism of ammonia mediated nitrogen transfer is
supported by the ability of the amidotransferases to use ammonia as a nitrogen source and thus the demonstration of an ammonia binding site. Furthermore, the presence of a completely conserved Nterminal cysteine, which has been shown to be essential for glutamine dependent activity of GPA (Mei & Zalkin, 1989) as well as both human and E. coli asparagine synthetases (Boehlein et al., 1994a; Sheng et al., 1993; Van Heeke & Schuster, 1989b), suggests that this residue may function in a manner similar to that of the active site cysteine of the thiol proteases. On the other hand, the presence of an important histidine in the active site of the class II enzymes has become increasingly doubtful. While the studies of GPA mentioned above (Mei & Zalkin, 1989) and chemical modification studies of GFAT (Badet-Denisot & Badet, 1992) seemed to indicate the presence of an essential histidine in these enzymes, site-directed mutagenesis of residues, known to be conserved among asparagine synthetases, did not reveal any histidines required for activity in ASB (Boehlein, et al., 1994a). Moreover, recent structural studies of GPA (Kim et al., 1996; Smith et al., 1994) and the glutamine utilizing domain of GFAT (Isupov et al., 1996) suggested that there were no histidines in position to act as part of a cys-his dyad. Structural comparisons later revealed that the protein fold of GPA (Brannigan et




21


al., 1995) was homologous to that of a family of N-terminal nucleophile (Ntn) hydrolases which also includes penicillin acylase (Duggleby et al., 1995), the 20S proteasome (Lowe et al., 1995; Seemuller et al., 1996), and aspartylglucosamidase(Fisher et al., 1993; Oinonen et al. 1995). It is thought that members of the Ntn family use the N-terminal Ser, Thr, or Cys as the nucleophilic catalyst and the N-terminal c-amino group as the acid-base catalyst. However, whether or not the class II amidotransferases use the Nterminal amino group in acid-base catalysis remains uncertain as results from experiments which have placed an extra amino acid at the N-terminus of GPA (Kim et al., 1996) have been inconclusive.
Further investigation is necessary in order to understand fully how acid-base catalysis occurs during the glutamine hydrolysis catalyzed by the class II amidotransferases. In fact for AS-B, the possibility of an essential histidine still remains. A computer model of a hypothetical AS-B glutamine binding site, constructed using the crystallographic coordinates for the covalent adduct of DON and E. coli GPA, revealed that histidine-47 of AS-B was in an appropriate position to function as part of a cys-his dyad cognate to that of papain (Richards & Schuster, 1997). Moreover, this histidine is partially conserved in asparagine synthetases and a histidine in a comparable position is found in several other class II amidotransferases (Figure 2.1), thus its importance cannot be overlooked. In an effort to gain a clearer understanding as to how acid-base catalysis occurs during glutamine utilization by AS-B, it is the purpose of this chapter to explore the glutaminase activity of this





22





LtHMAS CGIWAL< >HRGPD< >FGFHRLAV< >NGEIYNHKAL GnHmAS CGIWAI,< >HRGPD< >FGFHRLAV< >NGEIYNEKAL RatAS CGIWAL< >HRGPD< >FGFHRLAV< >NGEIYNEKAL
MurAS CGIWAL< >HRGPD< >FGFHRLAV< >NGE-YNHKAL
HuriAS CGIWAL< >HRGPD< >FGFHRLAV< >NGEIYNHKKM SoyAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNHEEL
LotAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIFNHEEL
FavaAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNHEEL PeaNAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNHEEL AlflAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNHEDL PeaRAS CGILAV< >HRGPE< >LAQQRLAI< >NGEIYNHEDL AspaAS CGILAV< >HRGPD< >LSHQRLAI< >NGEIYNHEEL ArabAS CGILAV< >HRGPD< >LAHQRLP.V< >NGEIY.HEEL BrssAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNHEEL RiceAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNKEEL MaizAS CGILAV< >HRGPD< >LAHQRLAI< >NGEIYNHEEL ScerAS CGIFAA< >HRGPD< -FVHERLAI< >NGEIYNHIQL CeleAS CGVFSI< >HRQPD< >LVHERLAI< >NGEIVNHGEL ECoAS CSIFQV< >HRQPD< >LAHERLSI< >NQEIYNHQAL
1 29 47 74 80
ScerGA CGILGI< >HRGQD< >FTQQRVS.< >NGNLVNTASL HumGS CGIFAY< >YRGYD< >HKQQDMDL< >NGIITNYKDL
ScerGS CGIFGY< >YRGYD< >.TKQNPNR< >NGIITNFREL






Figure 2.1: Sequence alignment of part of the GAT-domains of known asparagine synthetases and a partial set of class II amidotransferases showing highly conserved residues. Numbering is according to the amino acid sequence of AS-B. Sequence. alignments taken from Richards & Schuster, 1997. LtHmAS = Cricetulus longicaudatus AS; GnHmAS = Mesocricetus auratus AS; RatAS = Rattus norvegicus AS; MurAS = Mus musculus AS; HumAS = Homo sapiens AS; Soy AS = Glycine max AS; Lot AS = Lotus japoni cus AS; Fava AS Vicia jaba AS; PeaNAS Pisum sath a AS (root); AspaAS = Asparagus officials AS; ArabAS Arabidopsis thaliana AS; BrssAS = Brassica oleracea AS; RiceAS = Oryza sativa AS; MaizAS = Zea ma vs; Scer AS Saccharomyces cerevisiae AS; CeleAS = elegant AS; EcoAS = Escherichia coli AS, ScerGA Saccharomyces cerevisiae GPA; HumGS Homo sapiens GFAT-, ScerGS Saccharomyces cerevisiae GFAT.




23


enzyme as a function of pH. In order to address the possibility that H47functions in an acid-base capacity, the effects of mutations at this position on the pH-rate profile will be characterized.


Materials and Methods
Chemicals and Reagents
Restriction and modifying enzymes were purchased from Promega (Madison, WI) or New England Biolabs (Beverly, MA). Other chemicals, including L-glutamine and L-glutamic-y-monohydroxamate, were obtained from SIGMA Chemical Company (St. Louis, MO) and were of the highest commercial purity. Oligonucleotide primers were synthesized on an Applied Biosystems 380B DNA synthesizer by the DNA Synthesis Core Facility of the Interdisciplinary Center for Biotechnology Research at the University of Florida. Polymerase chain reactions (PCR) were performed on an Ericomp (San Diego, CA) thermocycler using the GeneAmp DNA Amplification Reagent Kit with AmpliTaq from PerkinElmer. Double stranded DNA sequencing was performed using the U.S. Biochemical Corp. Sequenase 2.0 Sequencing kit. Bacterial Strains and Plasmids
Bacterial strains BL21DE3pLysS (F, ompT, rb, mb-), supplied by Studier and Moffat; and NM522 (supE, thi (lac-proAB), hsd5, (rm)/F' pro AB, lac Iq Z M15), obtained from Stratagene (La Jolla, CA), are both derivatives of Escherchia coli K-12. The plasmid pBluescript was obtained from Stratagene (La Jolla, CA) and the plasmid pETB has been described previously (Boehlein et al., 1994a). E. coli host cells were transformed according to the procedures of Hanahan (Hanahan, 1983).




24


Enzyme Preparation

Construction of recombinant wild-type AS-B and mutants in
which His-29 or His-80 have been replaced by alanine (H29A, H80A) has been described elsewhere (Boehlein et al., 1994a). PCR cassette mutagenesis was used to replace Histidine-47 with alanine (H47A) or asparagine (H47N) using oligonucleotide primer pairs ss35OR and ss65 or ss348R and ss65, respectively (Table 2-1) and pETB as a template. The PCR parameters involved 35 cycles of denaturation at 94C for 1 min; annealing at 550C for 1 min; and extension at 72C for two minutes and the final cycle was followed by a 5 minute extension step at 72C. The plasmid, pETB, and the PCR products were digested with HpaI and XhoI and the 50 bp product was cloned into pETB. The resulting plasmids were sequenced through the insertion and no alterations other than those intended were found. All wild-type and mutant proteins were expressed, and purified using standard procedures (Boehlein et al., 1994a). Protein concentrations were determined by the method of Bradford (1976) using an assay kit supplied by Bio-Rad and mouse immunoglobulin G to construct a standard curve. During the course of these investigations an alternate method of determining the protein concentration revealed that the current use of mouse immunoglobulin G as a standard gives erroneously high protein concentrations for AS-B. Amino acid composition analysis of AS-B, carried out by the Protein Chemistry Core Facility of ICBR at the University of Florida, showed that the actual value for the concentration of this enzyme is 2.7 times less than that obtained using the Bradford assay. Therefore, all of the values of enzyme




25


concentration and therefore all kcat and kcat/Km values given in this chapter have been corrected by dividing the protein concentration by the factor 2.7.


TABLE 2.1
Oligonucleotides used in construction of site-directed mutants of AS-B


Oligo Oligonucleotide Sequence
Number

ss 6 5 5' A GCT TCC CAT ATG TGT TCA ATT TTT GGC GTA
TTC GAT 3'
ss350R 5' CGC CCC CGC GTT' AAC GTC GAC AAT TGA CAG ACG
TTC GGC GGC GAG A 3'
ss349R 5' CGC CCC CGC GTT AAC GTC GAC AAT TGA CAG ACG
TTC GGC GGC GAG A 3'

Glutaminase Assays
Initial rates of L-glutamine and L-glutamic-y-monohydroxamate (LGH) hydrolysis were determined by a modified procedure which uses glutamate dehydrogenase in the presence of NAD+ to measure glutamate concentration (Bernt and Bergmeyer, 1974). Reactions (100 p1) containing 100mM Bis-Tris and Tris-HC1 (pH 6.0-9.0) each (the kinetic constants appear to be independent of the nature and concentrations of these buffers), 8mM MgCl2 and various concentrations of glutamine or LGH were initiated by the addition of wild-type (0.7-2 gg) or mutant (4 pg) AS-B and incubated at 370C. A control containing denatured enzyme was used to monitor any glutamate formed from spontaneous breakdown. The reactions were terminated by boiling and a coupling reagent (300 mM glycine, 250




26


mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2 units of glutamate dehydrogenase) was added and incubated 10 minutes at room temperature. The absorbance at 340 nm was measured and the concentration of glutamate was determined by comparison to a standard curve. Initial velocities were measured at 9 different substrate concentrations and each velocity was an average of 2-4 measurements. Stability of the wild-type and mutant AS-B enzymes at each pH was confirmed by incubation in buffer of the appropriate pH for 20 minutes followed by an assay of activity at pH 8. Initial velocity conditions were confirmed at each pH by linear plots of v versus time and v versus enzyme concentration. Data Analysis
The values of Km and kcat were determined from least-squares fit to vo vs [S] plots using the Prism software package supplied by Graphpad, Inc., (San Diego, CA) according to equation 1:



V max IS]
Km + IS]



Where v is the maximal velocity, [S] is the concentration of st Vmax is the maximal velocity, and Km is the Michaelis constant. The pH dependence of kcat/Km obtained from equation 1 were fit to the bell-shaped curve described by equation 2:



Ylim
Y = (2)
1 + 10(pkl-pH) + 10(PH-PK2)




27


In the above equation, Y is the observed kca/Km at a given pH, Yjj is the maximum value of kcat/Km, and pK1 and pK. are the lower and higher acid dissociation constants, respectively.


Results
Effect of pH on the Steady-state Kinetic Parameters of wild-type ASB Catalyzed Glutaminase

The steady-state kinetic parameters, kcat and kat/Km, for glutamine hydrolysis were measured as a function of pH. As illustrated in Figures 2.2 and 2.3, and Table 2.2, kcat is constant over the pH range 6.0-9.0 whereas the kcat/Kmexhibits a bell shaped curve. Since the pKa of the c-amino group of glutamine is 9.13, the Km and thus k1at/KM were corrected for protonated glutamine using the Henderson-Hasselbach equation in an attempt to show how the ionization of the substrate might affect the kcat/Km-pH profile. Indeed, the correction caused a complete disappearance of the basic limb in the kcat/K.(app) plot (Figure 2.3). Non-linear regression of the experimental data to equation 1 allowed determination of the pKa for the non-corrected data which is shown in Table 2.5.
A comparison of the pH dependencies of the hydrolyses of
glutamine and the alternate substrate, LGH, can provide information about whether these reactions follow the same mechanism. Therefore, a pH profile of the wild-type AS-B catalyzed LGH hydrolysis was also examined (figure 2.4 and 2.5, and Table 2.3). Similar to the glutamine profile, the kcat appeared independent of pH for most of the pH range studied with slight increases at the low and high pH. The kcar/Km,(app) profile for LGH is bell shaped with a peak




28


at neutral pH and an acidic and basic pKa close to that obtained with glutamine (Table 2.5). Unlike the glutamine profile, correction for protonated LGH had little effect on the pH-kcat/Km profile (Figure 2.5). Kinetic Characterization of the Glutaminase Reaction catalyzed by the AS-B mutant H47N
As part of a continuing search for a histidine which may serve in a general acid-base capacity during AS-B catalyzed glutamine hydrolysis, Histidine-47 was examined using site-directed mutagenesis. Mutations were made which replaced the histidine with an alanine (H47A) or an asparagine (H47N) and the resulting enzymes were expressed. H47A was insoluble whereas H47N expressed at high levels and was successfully purified. The glutaminase activity of H47N was approximately 10-fold slower than that of wild-type AS-B (compare Tables 2.2 and 2.4). In addition to these changes in kct, there was a slight increase in the Km value for glutamine. pH Dependence of the Kinetic Constants of wild-type AS-B and its H47N
The pH dependence of the kinetic constants for H47N catalyzed glutamine hydrolysis was examined. The pH dependence of kcac was minimal (figure 2.6) except for a slight decrease at low pH which was not evident in the wild-type. The pH profile of kcat/Km(app) revealed a bell-shaped curve (figure 2.7) similar to that of wild-type catalyzed LGH hydrolysis except the curve was slightly narrower. The decrease in kcat/Km(app) on the basic limb was evident even after the correction for protonated glutamine.




29








3.0

2.5-

2.0
a
1.5


0.5

0.0- I I I I I
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
pH


Figure 2.2 The pH dependence of kct for wild-type AS-B catalyzed glutamine hydrolysis. The kinetic constants were determined as described in materials and methods.





30










3000


2500


2000

1500E

1000


500
A
0I I I I I I
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 pH










Figure 2.3 The pH dependence of keat /Km for wild-type AS-B catalyzed glutamine hydrolysis. The kinetic constants were determined as described in the materials and methods. The solid line (-) represents the actual data. The dashed line (---) represents the data corrected for the protonation of glutamine.




31













Table 2.2
Kinetic constants of the AS-B-catalyzed hydrolysis of L-Glutan-ine at 37C


pH kcat K. kcat/Km
(sec') (mM) (Msec-')

6.00 1.51 0.05 5.78 0.60 261 6.25 1.67 0.10 3.44 0.33 485 6.50 1.89 0.03 2.4 0.1 787
7.00 1.94 0.03 1.01 0.06 1921 7.25 2.05 0.05 1.22 0.08 1680 7.50 1.92 0.03 1.01 0.06 1901 7.75 2.13 0.03 1.32 0.07 1614 8.00 2.21 0.03 1.15 0.05 1922 8.50 2.46 0.05 1.50 0.06 1640 8.75 2.48 0.05 1.92 0.09 1292 9.00 2.56 0.05 2.33 0.12 1099




32












5











0



5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5

pH


Figure 2.4 The pH dependence of k,:a, for wild-type catalyzed LGH hydrolysis. The kinetic constants were determined as described in materials and methods.





33














5000



4000



3000

E
2000



1000


0I I I I I I
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 pH







Figure 2.5 The pH dependence of kcat/Km for wild-type AS-B catalyzed LGH hydrolysis. The kinetic constants were determined as described in materials and methods. The solid line (-) represents the actual data. The dashed line (---) represents the data corrected for the protonation of glutamine.




34












Table 2.3:
Kinetic constants of the AS-B-catalyzed hydrolysis of Lglutamic-y-monohydroxamate at 37C


pH kcat at k nKm
(sec') (mM) (M-'sec-')

6.00 3.46 0.05 6.51 0.29 531 6.25 3.16 0.05 2.39 0.15 1322 6.50 2.86 0.03 1.11 0.05 2577 6.75 2.59 0.03 0.68 0.02 3809 7.00 2.67 0.03 0.66 0.02 4045 7.25 2.65 0.03 0.58 0.03 4569 7.50 2.67 0.03 0.65 0.02 4108 7.75 2.65 0.03 0.63 0.03 4206 8.00 2.38 0.05 0.83 0.05 2867 8.25 2.54 0.03 1.11 0.05 2288 8.50 3.00 0.05 1.53 0.08 1961 8.75 3.10 0.05 2.26 0.11 1372 9.00 4.05 0.19 7.41 0.78 546




35










0.6

0.5
f
,,p.. 0.4%%o 0.3- m

0.2-

0.1

0.0.
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
pH


Figure 2.6 The pH dependence of k,,at for glutamine hydrolysis catalyzed by the AS-B mutant H47N. The kinetic constants were determined as described in materials and methods.




36












200



150



100
5


50



0 I I ,I I
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
pH











Figure 2.7 The pH dependence of kat/Km for glutamine hydrolysis catalyzed by the AS-B mutant H47N. The kinetic constants were determined as described in materials and methods.




37
















Table 2.4:
Kinetic constants for glutamine hydrolysis catalyzed
by the AS-B mutant H47N at 370C

pH kcat Km kcat/Km

(sec-') (mM) (M-'sec-')

6.00 0.19 0.01 18.6 2.8 10
6.25 0.22 0.008 13.1 2.0 17
6.50 0.30 + 0.005 8.3 0.9 36
6.75 0.30 0.008 5.4 0.4 55
7.00 0.30 0.005 3.2 0.2 94
7.25 0.30 0.003 2.4 0.1 125 7.50 0.38 + 0.005 2.5 0.1 152 7.75 0.32 0.003 2.5 0.1 128 8.00 0.32 0.003 3.0 0.1 107 8.25 0.40 0.005 4.1 0.2 98
8.50 0.46 0.008 6.1 0.3 75
8.75 0.38 0.01 7.8 0.7 49
9.00 0.43 0.01 10.5 1.1 41




38
















Table 2.5:
Values of pKa for the kat/Km-pH profiles for glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and for glutamine hydrolysis
catalyzed by the AS-B mutant H47N


Enzyme Substrate km,/Km
pKI pK,
wild type glutamine 6.7 0.1 8.8 0.1
LGH 6.7 0.1 8.0 0.1

H47N glutamine 7.2 0.2 8.1 0.2






39

















cn +I +I





+1 +1 +1 mt +1
I-C c C Ln cl W-) M
Ln C'I 09 t:
ct C-i C,4 Ln








00


+1 +1 +1 +1
CL4 00
C:) cc
E




+1 +1 +1 +1
M M Ln
3u C14 OC 00







00 cn





+1 +1
r- cn C
In +1
cq
Lr



+1 +1 +1 Ln

cz 00 -Z m





N oc




40


Comparison of the Kinetic Parameters of Wild-type, H80A. and H29A, at various pH values
The kinetic constants for the glutaminase activity of the three mutants (H29A, H47N, and H80A) are compared with wild-type in Table 2.6. Substitution of an alanine for the histidines at positions 29 and 80 has little effect on the kcat at pH 6.5 and 7.5 while this substitution decreases kcat/Km significantly. However, unlike wild type the k,,t value of H80A increases with increasing pH to a value at pH 9 which is double that of wild type. The most striking effect caused by the mutations at positions 29 and 80 is seen in the dramatic increase in Km especially at high pH.


Discussion
The importance of a histidine in the mechanism of glutamine hydrolysis catalyzed by AS-B and thus the mechanistic similarity between the AS-B catalyzed glutaminase reaction and the papain catalyzed peptide hydrolysis have been questioned for several years. Papain catalyzes the hydrolysis of peptide, amide, and ester bonds by the mechanism shown in (1). All substrates tested with papain have shown a bell-shaped pH dependence for kcat/Km with pKa's of approximately 4 and 8.5. As shown in the simplified model below, it is considered that these ionizations represent the thiol of cysteine-25 and the imidazole of histidine-159 which are involved in the acylation process (Menard et al., 1990):




41


pK1 4 pK2 8.5
-SH +HIm- -S- +Him- -S- Imactive enzyme
form



In order to understand how acid-base catalysis occurs in the AS-B active site during glutamine hydrolysis, the pH dependence of the kinetic constants has been examined for this enzyme. For both glutamine and LGH, the kt/Km value has a bell-shaped pH dependency. In contrast, kct appears independent of pH in the range of 6-9. The similarity in pH profiles for these two substrates suggests that their hydrolysis occurs by similar mechanisms. Furthermore, the values of kcat/Km for LGH are about 2-fold higher than those for glutamine (compare Tables 2.2 and 2.3). If substrate hydrolysis follows a mechanism like that of papain, then Nprotonation of the leaving group on the first tetrahedral intermediate is required for C-N bond cleavage to occur. Therefore, the faster acylation rate for LGH, reflected in kcat/KM, is consistent with the fact that hydroxylamine (pKa=8.2) is a better leaving group than ammonia (pKa-9.2). The pH independence of kcat may arise from a change in rate determining step between two steps with similar rates. However, further investigation, using a more extensive pH range is necessary for a clearer picture of the ionizations affecting kcat. Unfortunately, the AS is highly unstable when the pH is greater than

9 or less than 6.
The bell-shaped curves displayed in the kcat/Km profiles for ASB hydrolysis of glutamine and LGH (with pKa values of 6.6 and 8.3) are narrower than that reported for papain. The stability of the ion-




42


pair in papain is reflected in the width of the pH-profile for kc t/K. (Menard et al., 1991). Therefore, it is possible that AS-B contains an ion pair like that of papain but lacks some of the factors which play a role in ion pair stabilization. On the other hand, the pH profile of kcat/Km may relate to ionizations in free substrate. In this study, the pKa values of the c-amino groups of the substrates are 9.13 therefore, the possibility that the alkaline pKa values observed in the kCat/Km profiles arise from the ionization of glutamine and LGH cannot be excluded. In fact, the basic limb of the glutamine kcat/Km profile is completely eliminated when Km values are corrected for protonated glutamine. However, the same treatment has little effect on the ktKm profile for LGH.
While the above examination of the kinetic behavior of wildtype AS-B as a function of pH indicates the potentiality of an ion pair similar to that of papain, past efforts have failed to reveal a histidine which functions as part of that pair. In this study, substituting an asparagine for histidine at position 47 resulted in a reduction of both kcat and kcat/Km of approximately 10-fold for the glutaminase activity. While these changes are significant, a much more dramatic effect was expected if this residue were playing such an essential role in catalysis as has been postulated. For example, replacement of histidine-353 with an asparagine in the class I enzyme, CPS, resulted in a drop in glutaminase activity from 1.1 ptmol/homg for wt to <0.006 itmol/hemg for H353N (Miran et al, 1991). Subsequent structural studies which showed that this residue was in position to interact with the active site histidine added further support that His353 was involved in acid-base catalysis (Thoden et al., 1997).




43



Additionally, the pH profiles for glutamine hydrolysis catalyzed by H47N (Figures 2.6 & 2.7) resemble those of wt including similar pKa values of 7.2 and 8.5 (Table 2.5) which suggests that histidine-47 is not responsible for one of these ionizations.

Previous investigations of H29 and H80 failed to reveal that a histidine in either of these positions was essential for AS-B catalyzed asparagine synthesis (Boehlein et al., 1994a). While these studies were performed at pH 8 in the presence of an imidizole buffer, it is possible that the importance of the active site histidine could have been masked. Table 2.6 presents the results of a further investigation using the glutaminase activity at pH 6.5, 7.5 and 9.0. The lack of a significant decrease in the k at value for the glutaminase reaction catalyzed by these mutants confirms the previous conclusions that these histidines probably are not involved as part of the essential cys-his dyad which has been proposed.
Interpretations of this and any pH analysis of an enzymatic reaction rely on a number of assumptions which have been summarized by Knowles (1976). First, as noted by Peller and Alberty (1959), the association of the pKa values of the pH-kca,/Km with ionizations of the free enzyme and free substrate assumes that only one ionization state of the enzyme is directly convertible to product. In fact, the pH dependence of the kcat/Km values observed in papain catalyzed substrate hydrolysis has been found to be more complex than originally thought (Brocklehurst, 1994). Several studies have provided evidence that the development of the full nucleophilic character of the ion pair depends not only on the ionization at pH 4 but on a second ionization at pH 4 and, in some cases, ionization with





44


pKa 5 (Brocklehurst, 1988 a,b). Altered pKa values also may result if the rate constant for substrate dissociation is less than the net rate constant for product formation. This is unlikely to be the case for AS-B glutamine hydrolysis since investigations into the dissociation constant of glutamine have suggested that a rapid equilibrium mechanism is plausible (see chapter 4). Determination of the pKa of a competitive inhibitor or of the backwards reaction will be necessary to confirm this assumption.
With respect to these assumptions it is important to take heed in comparison of the pH profiles of AS-B and papain. On the other hand, with the results presented here, it is difficult to envision a role for histidine-47 in acid-base catalysis when one considers the lack of effect its replacement with asparagine made on the pH profiles. In light of the presence of this histidine in many asparagine synthetases as well as other class II amidotransferases it was important to include this residue among potential acid-base catalysts. Indeed, the 10-fold loss of activity in the H47N mutant may reflect a minor or indirect importance of this residue in glutamine binding and/or utilization. On the other hand, the importance of the histidine to papain activity is underscored in the complete conservation of this amino acid across the entire family of cysteine proteases. The lack of such a completely conserved histidine in the class II family of amidotransferases suggests that an alternate mechanism of acid-base catalysis probably occurs in these enzymes and therefore demonstrates dissimilarities in mechanism for the two families of enzymes.














CHAPTER 3
EFFECT OF TEM~vPERATURE ON GLUTAMINE HYDROLYSIS CATALYZED BY ASPARAGINE SYNTHETASE B


Intro du ct ion

The following chapters present experiments which are aimed at the determination of rate constants and the identification of intermediates in order to develop a descriptive model of the glutaminase reaction. Much of the investigation requires a very slow reaction rate and, therefore, the reactions were examined at 5'C. Variation in temperature can cause alterations in the mechanism of a reaction including differences in rate determining steps. In order to validate the comparison of the results from experiments at 5'C with those performed at the temperature at which the reaction is normally investigated (37C), the steady state kinetics of glutamine hydrolysis will be examined over a range of temperatures and Arrhenius plots will be used to detect any changes in rate limiting step with temperature for this reaction. The temperature dependence of LGII hydrolysis catalyzed by wild-type AS-B and that of glutamine hydrolysis catalyzed by the AS-B mutant R30A will be examined in a similar manner and the activation energies of the three reactions will be compared. Similarity in the pH profiles for glutamine and LGH hiydrolysis presented in the previous chapter suggests that these reactions occur by a common mechanism. Further evidence for a common mechanism could be obtained 45




46



through similarities in the activation parameters which would suggest that the same step is rate limiting in both reactions. In contrast, work in the following chapters suggests that glutamine hydrolysis catalyzed by the AS-B mutant, R30A, has a different rate limiting step than that catalyzed by wt AS-B and these differences might be revealed in different values for the activation parameters.
A complete description of the glutaminase reaction requires 1) the order of product release, 2) the inhibition constants of the products, and 3) knowledge of whether or not the reaction is reversible. Therefore, in order to address these issues, product inhibition studies were employed and the ability of the enzyme to catalyze the synthesis of glutamine and LGH was examined.


Materials and Methods

Chemicals and Reagents
[14C]-glutamine was purchased from Amersham. Scintillation fluid, ScintiVersTM II* was obtained from Fischer (Orlando, FL). DE81 anion exchange chromatography paper was purchased from Whatman (Hillsboro, OR). Other chemicals, including L-glutamine and L-glutamic-y-monohydroxamate, and hydroxylamine were obtained from SIGMA Chemical Company (St. Louis, MO) and were of the highest commercial purity.
Enzyme Preparation
Construction of recombinant wild-type AS-B and the mutant AS-B in which R30 has been replaced by alanine (R30A) have been described elsewhere (Boehlein et al., 1994a,b). Both wild-type and mutant proteins were expressed., and purified using standard




47



procedures (Boehlein et al., 1994a). Protein concentrations were determined by the method of Bradford (1976) using an assay kit supplied by Bio-Rad and mouse immunoglobulin G to construct a standard curve. During the course of these investigations an alternate method of determining the protein concentration revealed that the current use of mouse immunoglobulin G as a standard gives erroneously high protein concentrations for AS-B. Amino acid composition analysis of AS-B, carried out by the Protein Chemistry Core Facility of ICBR at the University of Florida, showed that the actual value for the concentration of this enzyme is 2.7 times smaller than that obtained using the Bradford assay. Therefore, all of the values of enzyme concentration and therefore all kcat and kcat/Km values given in this chapter have been corrected by dividing the protein concentration by the factor 2.7. Glutaminase Assays
Initial rates of L-glutamine and L-glutamic--monohydroxamate (LGH) hydrolysis were determined by a modified procedure which uses glutamate dehydrogenase in the presence of NAD+ to measure glutamate concentration (Bernt & Bergmeyer, 1974). Reactions (100 tl) containing 100mM Bis-Tris and Tris-HC1 (pH 8.0) each, 8mM MgC12 and various concentrations of glutamine or LGH were initiated by the addition of wild-type or mutant R30A AS-B and incubated at various temperatures (4C-40C). A control reaction lacking enzyme was used to monitor any glutamate formed from spontaneous breakdown. The reactions were terminated by the addition of 20 tl of IN acetic acid. A coupling reagent (300 mM glycine, 250 mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2 units of glutamate dehydrogenase) was added




48


and incubated 10 minutes at room temperature. The absorbance at 340 nm was measured and the concentration of glutamate was determined by comparison to a standard curve. Initial velocities were measured at 9 different substrate concentrations and each velocity was an average of 3 measurements. Stability of the wild-type and mutant AS-B enzymes at each temperature was confirmed by incubation in buffer of the appropriate temperature for various amounts of time followed by an assay of activity at pH 8. Initial velocity conditions were confirmed at each pH by linear plots of v versus time and v versus enzyme concentration. The pH of the buffer was adjusted to pH 8 at each temperature studied to account for the pKa change with temperature.
Determination of I- for glutamate and hydroxylamine in the AS-B catalyzed synthesis of LGH
The Michaelis constant, KM, of glutamate in the synthesis of LGH was determined by incubating wild-type AS-B with various concentrations of glutamate and a saturating concentration of hydroxylamine (150 mM) in Tris-HC1, pH 8.0, at 37'C for 15 min. In a similar manner the Km for hydroxylamine was determined by incubating the wild-type enzyme with various concentrations of hydroxylamine and a saturating concentrations of glutamate (150 mM) in Tris-HC1, pH 8.0, at 37C for 20 min. In both cases, the 300 ptl reaction was terminated by the addition of 100 tl of 16 % TCA. LGH formation was determined in a final volume of 500 tl by adding a solution containing 80% TCA, 6N HC1, and 10% FeC13 in 0.02 N HC1 to the remaining reaction, centrifuging the samples in a microcentrifuge to remove particulates, and measuring the absorbance of the




49



hydroxamate-FeC13 complex at 540 nm. A standard curve using a stock solution of authentic LGH was used to quantitate the product formed in these reactions.
The ability of AS-B to catalyze the synthesis of glutamine also was examined. Wild-type AS-B (37 jig) was combined with glutamate and ammonium chloride (100 mM each) in Tris-HC1 (200 mM) in a total volume of 50 jil. The reactions were incubated for 4 hours at 370C followed by the addition of 20 tl of 2N acetic acid. The terminated reactions were filtered through a 2 jim filter and derivatized with phenylisothiocyanate (PITC). The glutamine and glutamate derivatives were separated by HPLC and quantitated spectroscopically. This assay procedure can detect less than 100 pmoles under our standard procedures Product inhibition of the glutaminase reaction
Wild-type AS-B was incubated in a 100 jtl reaction containing various concentrations of [4C] glutamine (SA 880-20,000 dpm/nmol) in 100mM Bis-Tris and Tris-HC1 at 37C for 10 min in the presence of various concentrations of glutamate or NH4Cl. The reactions were terminated by the addition of 10 jil of 1N acetic acid. Glutamate was separated from glutamine on Whatman DE-81 ion-exchange chromatography paper using deionized distilled water as the mobile phase. After 6 hours, elution was stopped, and the paper was dried and cut into two strips to separate glutamine from glutamate. The radioactivity associated with each strip was counted in ScintiVersll* scintillation fluid on a Beckman LS 61C scintillation counter. Control experiments were carried out in an identical manner except that the enzyme was excluded from the reaction mixture. The separation of




50



the amino acids was visualized in a non-radioactive control by spraying the paper with 0.2% ninhydrin in 95% ethanol. Glutamine was found to migrate with the solvent front while glutamate remained at the origin. The amount of "4C-labeled glutamate formed in the reaction was determined by subtracting the percentage of total radioactive counts located at the origin in the control reactions from the percentage in each sample.
In a similar manner, the inhibition of glutamate also was examined at 5'C except that the reaction time was 40 min. Data Analysis

The values of Kmn and kcat were determined from least-squares fit to vo vs [S] plots using the Prism software package supplied by Graphpad, Inc., (San Diego, CA) according to equation 1: V ma x [S]
v = (1)
Km + [S](

Where v is the initial velocity, [S] is the concentration of substrate, Vmax is the maximal velocity, and Km is the Michaelis constant.
As shown in the equation below, the value for -Ea/RT can be derived from the slope of a plot of the log kcat versus l/T (K) where A is a constant which is related to the collision frequency and steric factors.
k = Ae(-Ea/RT) (2)
Values for the energy, enthalpy, and entropy of activation then were calculated using the following equations:
AH* = Ea RT (3)
AG* = RT ln(k3T/h) RTln(k) (4)




51



AG* = AH* TAS* (5)
where AH* is the enthalpy of activation, R is the gas constant, kB is boltzmans constant, h is Planck's constant, k is the forward rate constant; AG* is the free energy of activation, and AS' is the entropy of activation.


Results
Effect of temperature on glutamine and LGH hydrolysis catalyzed by wild-type AS-B and the AS-B mutant R30A

The steady-state kinetic parameters of glutamine and LGH
hydrolysis were determined at a range of temperatures from 5' C to 400 C. The Arrhenius plots (Figures 3.1 & 3.2) were observed to be linear indicating either that there was no change in the ratedetermining step over the temperature range studied or that the activation energies for the limiting steps were very similar.
The temperature dependence of the catalytic rate constants of the R30A-catalyzed hydrolysis of glutamine also displayed linearity in the Arrhenius plot, as shown in Figure 3.3. The activation parameters are given in Table 3.1. LGH synthesis catalyzed by wild-type AS-B
AS-B was found to catalyze the formation of LGH from
glutamate and hydroxylamine. The reaction rate of 0.29 sec' (Table
3.2, Figures 3.4 & 3.5) was about 12% of that of the corresponding forward rate of 2.4 sec1 at pH 8 (see chapter 2, Table 2.3), and was fast enough to obtain the kinetic parameters of LGH synthesis catalyzed by wild-type AS-B (Table 3.2).
The steady-state kinetic parameters of reversible reactions can




52


be used to calculate the equilibrium constant for the overall reaction using the Haldane relationship for a one-substrate/two-product reaction given in equation (6) (Cleland, 1963):


VfKip Vf Kp Kiq
Keq K -(6)
VrKia Vr Ka



where Keq is the equilibrium constant for the conversion of substrate to product, V, and V, represent the forward and reverse reaction rates, respectively, and the indices p, q, and a refer to hydroxylamine, glutamate and LGH, respectively. Using equation (6) and the following parameters, obtained at 37oC: k'r = 2.4 sec-', Km(NH2OH)= 95 mM, Ki(gu,)=182 mM, k'cat= 0.29 sec1, Ki(LGH)= 0.83 mM; a value of Keq=172 was calculated. Using this value for Kcq and the equation, AGo = -RTlnKeq, the difference in free energy between substrates and products, AGO, was determined to be -13.2 kJ/mol K. Product inhibition glutamine hydrolysis catalyzed by wild-type AS-B
Figures 3.6 and 3.7 show the plots for the inhibition of the
glutaminase reaction by glutamate at 5oC and 37oC, respectively. The Ki values for glutamate determined from the replot of the slope versus glutamate concentration (Figure 3.8 & 3.9) at 50C and 37oC were 260 mM and 182 mrM, respectively. An attempt to obtain the Ki for ammonia was made. However, NH4C1 concentrations up to 200 mM seemed to have little effect on the rate of the glutaminase reaction.




53









2


C 1



0 .j
-1



-2
0.00300 0.00325 0.00350 0.00375
1/'1(K)







Figure 3.1: Temperature dependence of glutamine hydrolysis catalyzed by wild-type AS-B. The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and Tris-HCI as described in the materials and methods.




54















1



0
.i
-1




-2-r
0.0032 0.0033 0.0034 0.0035 0.0036 0.0037
1/T (K)


Figure 3.2: Temperature dependence of LGH hydrolysis catalyzed by wild-type AS-B. The initial rates at each temperature were determined at pH8.0 in 100 mM Bis-Tris and Tris-HC1 as described in materials and methods.




55
















0






-2



0.00300 0.00325 0.00350 0.00375
1/T (K)


Figure 3.3: Temperature dependence of glutamine hydrolysis catalyzed by the AS-B mutant R30A. The initial rates at each temperature were determined at pH 8.0 in 100 mM Bis-Tris and Tris-HCI as described in materials and methods.




56














Table 3.1: Thermodynamic properties of glutamine and LGH hydrolysis catalyzed by wild-type AS-B and the R30A mutant.


Substrate/ AG* AH* AS* TAS'
enzyme kJ mol' kJ mol' J mol-'K-' kJ mol'
(370C) (370C) (5-400C) (370C)
gln/wt 167 89 -252 -78
LGH/wt 152 82 -226 -70
gln/R30A 147 77 -226 -70


The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and Tris-HCI at temperatures ranging from 5-400C, under conditions described in the materials and methods.




57








200



S150
E

S100



50


0
0 50 100 150
[NH2OH] mM








Figure 3.4: Effect of hydroxylamine concentration on the synthesis of LGH catalyzed by wild-type AS-B. Wild-type AS-B (18.5 4g) was incubated with various concentrations of hydroxylamine and saturating concentrations of glutamate (150 mM) in Tris-HC1, pH 8.0 at 370C for 20 min. The reaction was terminated by the addition of 16% TCA and LGH formation was measured as described in materials and methods.




58







250


200

E
150

0
E
S100


50



050 7'5 100 125
(Glu] mM








Figure 3.5: Effect of glutamate concentration on the synthesis of LGH catalyzed by wild-type AS-B. Wild-type AS-B (18.5 gg) was incubated with various concentrations of glutamate and a saturating concentration of hydroxylamine (150 mM) in Tris-HC1, pH 8.0 at 370C for 15 min. The reactions were terminated by the addition of 16% TCA and the formation of LGH was measured as described in materials and methods.





59














Table 3.2: Kinetic constants for the synthesis of LGH catalyzed by wild-type AS-B.


Substrate kcat (sec-') K,1 (rnaM) kcat/K (M-1 sec')

glutamate 0.36 0.02 76 5 4.7
hydroxylamine 0.29 0.01 95 7 2.9


The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and Tris-HC1 at 371C, by varying one substrate while keeping the second substrate saturating. The conditions of the reaction and the determination of the LGH formed in the reaction are described in the materials and methods.




60








0.3


E
~0.2
E 0

%m 0.1Ii I I I i
-5 0 5 10 15 20 25

1/[GIn] mM1





Figure 3.6: Glutamate inhibition of the glutaminase reaction catalyzed by wild-type AS-B at 5C. The rate of glutamine hydrolysis was determined as described in the materials and methods in the presence of 0 (i), 10 (A), 50 (V), 100 (*), and 136 (0) mM glutamate.




61












0.00250.00200.00150.00100.0005

II I I
-1 0 1 2 3

1/Gin (mM-1')


Figure 3.7: Glutamate inhibition of the glutaminase reaction catalyzed by wild-type AS-B at 37oC. The rate of glutamine hydrolysis was determined as described in the materials and methods in the presence of 0 (U), 40 (*), 80 (A), and 100 (V) mM glutamate.




62











0.0150 0.01250.0100O 0.00750.00500.00250.0000 ,
0 25 50 75 100 125 150
[Glu] mM


Figure 3.8: Replot of the slope from the double-reciprocal plots of glutamine hydrolysis at 50C versus glutamate concentration.




63













0.00080.0006,, 0.00040.00020.0000 I I I I I
0 25 50 75 100 125
[Glu] mM


Figure 3.9: Replot of the slope from the double-reciprocal plots of glutamine hydrolysis at 37oC versus glutamate concentration.




64



Discussion

Many of the experiments in the following chapters required a reaction rate slower than that observed at 37'C, the standard temperature at which the reaction is examined. One way to obtain a slower rate is to examine the reaction at low temperatures therefore, many of the experiments in the following chapters have been conducted at 50C. In preparing for these experiments it was important to know whether the results obtained at 5'C could be compared to those at the standard temperature. Temperature variations often cause changes in the rate-limiting step of a reaction. Since non-linear Arrhenitis plots are indicative of such changes, the temperature dependence of the glutaminase reaction was examined to confirm that there was not a change in the rate limiting step between 5'C and 37'C. In the case of the glutaminase reaction, it is difficult to conclude that a single step is rate determining across the entire range of temperatures studied. Certainly, the linear Arrhenius plots suggest that a change in the rate limiting step between reactions at 5'C and 37'C is unlikely. However, it should be noted that two steps in the reaction may have similar activation energies, and therefore a change in rate limiting step would remain unnoticed in these experiments. Such an event is a possibility in AS-B catalyzed glutamine hydrolysis as demonstrated by the unexpected similarity in activation constants for reactions catalyzed by wild-type and the mutant R30A (Table 3.1 and Figures

3.1 and 3.3). Partitioning experiments in chapter 4 imply that in reactions catalyzed by R3OA, hydrolysis of the thioester is rate limiting. In contrast, direct measurements of the deacylation step of glutamnine hydrolysis suggest that an additional step contributes to the




65



overall rate of the reaction catalyzed by the wild-type enzyme. The differences in the nature of the rate determining step in the reactions catalyzed by the wild-type and mutant enzyme were expected to be reflected in the activation parameters. The lack of any differences in the parameters suggests that the two slow steps in the wild-type catalyzed reaction have similar activation energies.

The patterns of product inhibition can provide information on the order of product release. However, the studies of product inhibition presented here were inconclusive due to the inability of ammonia to inhibit the glutaminase reaction. On the other hand, the examination of glutamate inhibition provided a needed inhibition constant which will be used in the model described in chapter 7.

The inability of ammonia to inhibit the glutaminase reaction and the lack of detectable glutamine synthesis using ammonia and glutamate as substrates suggests that the acylation step is essentially irreversible under the conditions of the examination. With the ability to detect glutamine concentrations to 0.1 nmol (see materials and methods), an upper limit on the turnover number of kcat< 5xl104 sec-1 can be set for glutamine synthesis at 37'C catalyzed by the wild-type AS-B under the conditions studied. In contrast, the reversibility of LGH hydrolysis catalyzed by wild-type AS-B was easily demonstrated and has a keat which is 12% of the forward reaction. The difference in the ability to synthesize the reverse reaction may be a reflection of the greater nucleophilicity of hydroxylamine. The kinetic parameters, kcat and Kin, obtained from the analysis of LGH synthesis allowed the determination of the equilibrium constant for the overall reaction.




66


The free energy, AGO, derived from this equilibrium constant was calculated to be -13.2 kJ/mol.












CHAPTER 4
DETERMINATION OF THE RATE DETERMINING STEP IN ASPARAGINE
SYNTHETASE B-CATALYZED GLUTAMINE HYDROLYSIS


Introduction
Asparagine synthetase B (AS-B) is thought to be organized into two functional domains: the glutamine amide transfer (GAT) domain, which catalyzes the hydrolysis of glutamine to glutamate; and the synthetase domain, which catalyzes the formation of a 3-aspartylAMP intermediate and the synthesis of asparagine. While the overall reaction involves three substrates, four products, and possibly twenty rate constants; the ability of the enzyme to catalyze glutamine hydrolysis in the absence of the other substrates provides the opportunity to examine glutamine utilization under a simpler one substrate, two product reaction. A detailed description of this reaction would provide a useful tool in understanding the roles played by conserved amino acid residues of the GAT domain in glutamine utilization. Therefore, this chapter presents a first step towards characterizing the glutaminase reaction catalyzed by AS-B.
Analysis of the amino acid sequence of the GAT domain suggests that AS-B belongs to a family of enzymes known as the glutamine amidotransferases (Zalkin, 1993). Like AS-B, the enzymes of this family catalyze nitrogen transfer using glutamine as the primary source of nitrogen. While all enzymes in the glutamine amidotransferase family contain a cysteine which is essential for

67




68


glutamine dependent activity, the position of this cysteine as well as other sequence differences in the GAT domain further divides the family into two classes. The essential cysteine residue is located within the GAT domain for class I enzymes whereas it resides at the N-terminus in the class II enzymes such as glutamine phosphoribosylpyrophosphate amidotransferase (GPA) (Tso et al., 1982) glucosamine-6-phosphate synthetase (GFAT) (Badet et al., 1987), and asparagine synthetase (Humbert et al., 1980). Initially, site directed mutagenesis studies suggested that the N-terminal cysteine as well as a histidine and aspartate residue were involved in the reaction catalyzed by GPA (Mei & Zalkin, 1989). As this catalytic triad is found in many thiol proteases (Brocklehurst et al., 1987), it was proposed that the glutamine hydrolysis reaction catalyzed by the class II enzymes occurs via a mechanism analogous to that of peptide hydrolysis catalyzed by the thiol proteases, such as papain. This hypothesis was extended to the class I enzymes as well (Mei & Zalkin, 1989). Since then, the importance of a catalytic triad composed of a conserved cysteine, histidine, and glutamate residues has been demonstrated for the class I enzymes through site-directed mutagenesis (Amuro et al., 1985; Miran et al., 1991) and x-ray crystallography (Tesmer et al., 1996; Thoden et al., 1997).
On the other hand, the picture remains more ambiguous for the class II amidotransferases as subsequent studies confirmed only the importance of the conserved N-terminal cysteine residue (Boehlein et al., 1994a). In fact, a more recent alignment of the GAT domain which includes a greater number of amidotransferase sequences reveals that there are no histidines which are conserved across this




69


entire class of enzymes (Boehlein et al., 1994a; Zalkin, 1993). Moreover, studies using site-directed mutagenesis of AS-B (Chapter 2; Boehlein et al., 1994a) and x-ray crystallography of GFAT (Isupov et al., 1996) and GPA (Kim et al., 1995) suggest that the class II amidotransferases function differently than papain with respect to acid-base catalysis. While the lack of a conserved histidine in the class II amidotransferases contrasts with the essentiality of such a residue in thiol proteases, recent evidence for an acyl enzyme intermediate (chapter 5 & 6) formed during glutamine hydrolysis catalyzed by AS-B provides a basis for comparison of the hydrolysis reactions catalyzed by the two enzyme families. A minimal description of amide hydrolysis catalyzed by papain, which occurs via a thioester intermediate (E-TE), is shown in equation 1 below:


k, k
S + ES --2-- E-TE E+P2 (equation 1)
k -1 + P t

In relating this equation to glutamine hydrolysis catalyzed by AS-B, ES is the enzyme-substrate complex, E-TE is the acylenzyme intermediate, and P1 and P2 are ammonia and glutamate, respectively. In an effort to understand the mechanism of glutamine hydrolysis catalyzed by AS-B, the glutaminase reaction will be examined in order to develop a model for this particular aspect of the enzymatic reaction. The minimal model shown above will be used as a starting point in this investigation to provide evidence (in addition to that shown in chapter 5 for the existence of a thioester intermediate, to determine whether steps leading to its formation or




70


those involved in its breakdown are rate limiting, and to compare and contrast catalytic behavior of three groups of hydrolytic enzymes: a class II amidotransferase (AS-B), a class I amidotransferase (CAD), and a thiol protease (papain).


Materials and Methods
Chemicals and Reagents
14C-glutamine was purchased from Amersham. Scintillation fluid, ScintiVersTM II* was obtained from Fischer (Orlando, FL). DE81 anion exchange chromatography paper was purchased from Whatman (Hillsboro, OR). Other chemicals, including L-glutamine and L-glutamic-y-monohydroxamate, hydroxylamine, and 6-diazo-5oxonorleucine (DON) were obtained from SIGMA Chemical Company (St. Louis, MO) and were of the highest commercial purity. Enzymes
The construction of recombinant AS-B and the AS-B mutant,
R30A, has been described previously (Boehlein et al., 1994a, 1994b). The enzymes were obtained by overexpression and purified using standard procedures (Boehlein et al., 1994a). Protein concentration was determined by the method of Bradford (1976) using an assay kit supplied by Biorad and immunoglobulin G as a standard. Recent examination of the protein concentration using amino acid analysis of AS-B, performed by the Protein Core Facility of ICBR at the University of Florida, gave a value for protein concentration which was 2.7-fold lower than that obtained with the previous method. Therefore, in this report the enzyme concentration has been corrected by dividing the enzyme concentrations by the factor 2.7.




71


Steady State Kinetics
The kinetic parameters of AS-B-catalyzed glutamine, LGH, and L-glutamine-hydrazide hydrolysis were obtained in an end-point assay which uses glutamate dehydrogenase in the presence of NAD' to measure the concentration of glutamate formed (Bernt & Bergmeyer, 1974). Reactions (100 pl) containing 100mM Bis-Tris and Tris-HCI (pH 7.5) each, 8mM MgCl2 and various concentrations of glutamine, LGH, or glutamic acid-y-hydrazide were initiated by the addition of 1.5 gg wild-type AS-B and incubated at 370C. After 10 minutes, the reactions were terminated by boiling for 2 minutes. Reactions in which the enzyme was added to the mix during the termination period of boiling were used to monitor any glutamate formed from spontaneous breakdown. A coupling reagent (300 mM glycine, 250 mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2 units of glutamate dehydrogenase) was added to the reaction and the resulting mixture was incubated 10 minutes. The absorbance at 340 nm was measured and the concentration of glutamate was determined by comparison to a standard curve. Initial velocities were measured at 9 different substrate concentrations and each velocity was an average of 2-4 measurements.
The values of Km and kcat were determined from least-squares fit to vo vs [S] plots using the Prism software package supplied by Graphpad, Inc., (San Diego, CA).
Determination of the K. for glutamine and LGH
The rate of inactivation by the glutamine analog, 6-diazo-5oxonorleucine (DON), was determined by incubating 30 gtg wt AS-B in a 300 p1 solution containing 100 mM Bis-Tris and Tris-HC1, pH 8 and




72


varying concentrations of DON at 5C. At various time points 20 ptl aliquots were removed from the incubation mixture and the residual activity was measured in 200 pi buffer containing 12 mM glutamine. Control experiments measuring the activity in the presence of diluted inhibitor concentrations verified that DON did not interfere in the assay of residual activity.
Glutamine protection of DON inhibition was examined as
described above except that wt AS-B was incubated with varying concentrations of DON in the presence of either 20 ptM or 50 tM glutamine. Alternatively, LGH protection of DON inactivation of the wild type enzyme and glutamine protection of R30A inactivation were examined by incubating wild-type or R30A AS-B with a constant concentration of DON (6 4M for wild type and 650 tM for R30A) and varying concentrations of LGH or glutamine, respectively. An analysis of glutamate formation in AS-B catalyzed glutamine hydrolysis as a function of time.
Glutamine hydrolysis catalyzed by wt AS-B was examined at 50C, a temperature at which the reactions were slow enough to mix by hand. The reactions were initiated by adding 0.37 or 0.74 nmol wt AS-B to a 50 pL1 solution containing 100 mM Bis-Tris and Tris-HCl (pH 8) and ['4C]-glutamine (2mM, SA = 9 nCi/tl). After the appropriate amount of time, the reactions were terminated with 20 pI of 2N acetic acid. Glutamate was separated from glutamine on Whatman DE-81 ion-exchange chromatography paper using deionized distilled water as the mobile phase. After 6 hours, elution was stopped, and the paper was dried and cut into two strips to separate glutamine from glutamate. The radioactivity associated




73


with each strip was counted in ScintiVerslI* scintillation fluid on a Beckman LS 61C scintillation counter. Control experiments were carried out in an identical manner except that the enzyme was excluded from the reaction mixture. The separation of the amino acids was visualized in a non-radioactive control by spraying the paper with 0.2% ninhydrin in 95% ethanol. Glutamine was found to migrate with the solvent front while glutamate remained at the origin. The number of nanomoles of 4C-labeled glutamine in each sample was determined by subtracting the percentage of total radioactive counts located at the origin in the control reactions from the percentage in each sample. All values are derived from the average of four separate reactions.
To examine the effect of glutamate on the burst, wild-type AS-B was incubated on ice either alone or with 100 mM L-glutamate, or DGlutamate for 3 hours prior to the reaction. The enzyme/glutamate mixture then was added to the reaction and product formation was determined at various time points as described above. Hydroxylamine Partitioning of the Thioester

A reaction mixture containing either wt AS-B (9.3 kg), or R30A (18.5 gg), and 50 mM glutamine, 100 mM Bis-Tris and Tris-HC1 (pH 8) and a variable concentration of hydroxylamine in a total volume of 300 p.l was incubated at 37C for 45 minutes. The reaction was terminated by the addition of 100 p.l of 16% TCA and a 50 p.l aliquot was removed to determine the amount of glutamate formed using the previously mentioned GDH assay. LGH formation was determined in a final volume of 500 p.1 by adding a solution containing 80% TCA, 6N HCI, and 10% FeCI3 in 0.02 N HC1 to the remaining reaction,




74


and measuring the absorbance of the hydroxamate-FeC13 complex at 540 nm. A standard curve using a stock solution of authentic LGH was used to quantitate the product formed in these reactions.


Results
Partitioning of a thioester intermediate to 7-glutamvl hydroxamate

The minimal mechanism, given in equation 1, requires the formation of a covalent intermediate. Due to the highly reactive nature of hydroxylamine towards acyl groups, this agent has been used to provide evidence for an acylenzyme intermediate and to determine whether its formation or breakdown is rate limiting in hydrolysis reactions catalyzed by serine proteases (Caplow et al., 1964; Epand et al., 1963) as well as by the amidotransferase CAD (Chaparian et al., 1991). Since hydroxylamine competes with H,O in the deacylation step, its presence will cause an increase in the amount of total product formed (hydroxamate plus glutamate) only if this same step is rate limiting (reaction 2). In contrast, if the acylation step is very slow then the presence of hydroxylamine will not have an effect on the amount of total product produced as the slow rate of the acylation step will limit the amount of thioester available for hydrolysis and hydroxaminolysis.
H20 E + Glu
K, k2
E +Gln E.Gln E-TE (Reaction 2)

NH2OH E + LGH




75


In an attempt to show that AS-B catalyzed hydrolysis occurs via a thioester intermediate and to determine which step in hydrolysis is rate limiting, the effect of various concentrations of hydroxylamine on the glutaminase reaction was examined. As shown in Figure 4.1, even very low concentrations of hydroxylamine were sufficient to partition the reaction towards the production of glutamate hydroxamate. Unexpectedly however, an increase in hydroxylamine concentration in the AS-B catalyzed glutaminase reaction caused a decrease in total product formed (figure 4.1).

One potential cause for this decrease in total product formed in the presence of hydroxylamine is a slow release of L-glutamate hydroxamate due to its tight binding to the enzyme. To test this idea, the partitioning experiment was repeated using an AS-B mutant (R30A) which exhibits poor substrate binding. A role in substrate binding has been implied for Arginine-30 in AS-B (Boehlein et al., 1994b) and the cognate arginine in GFAT (Isupov et al., 1996) in studies using site-directed mutagenesis and x-ray crystallography, respectively. Use of R30A in the partitioning experiments resulted in an increase in total product with increasing concentrations of hydroxylamine (figure 4.2).
Comparison of AS-B catalyzed hydrolysis of three substrates
The kinetic constants kcat, Km, and kcat/Km for the AS-B catalyzed hydrolysis of L-glutamine, L-glutamic acid-y-monohydroxamate, and L-glutamic acid-y-hydrazide are given in Table 4.1. As shown in the table, the kcat values for the three substrates were similar while each displayed different kcat/Km values.




76






600
i.

500

) 400E

5g 300. 200100


0 10 20 30 40 50 60 70
[NH2OH] (mM)




Figure 4.1: Partitioning of the thioester intermediate to L-glutamicy-monohydroxamate using hydroxylamine in a reaction catalyzed by wt AS-B. A mixture containing wt AS-B, 50 mM glutamine, 100 mM Bis-Tris and Tris-HCI (pH 8), and the indicated concentrations of hydroxylamine was incubated at 370C for 45 min. The reactions were terminated by the addition of 4% TCA. An aliquot was removed to determine glutamate formed (0) by an end-point assay using glutamate dehydrogenase. LGH formation (A) was measured by adding a FeCl2 solution and measuring the absorbance at 540 nm as described in materials and methods. The nmol of LGH and glutamine formed at each concentration of hydroxylamine were added to give the total product formed (U).




77








2000


1500
E

4-E 10000
L
500


0
0 10 20 30 40
[NH2OH] mM




Figure 4.2: Partitioning of the thioester to glutamate hydroxamate using hydroxylamine in the glutaminase reaction catalyzed by AS-B R30A. A mixture containing wt AS-B, 80 mM glutamine, 100 mM Bis-Tris and Tris-HCI (pH 8), and the indicated concentrations of hydroxylamine was incubated at 370C for 45 min. The reactions were terminated by the addition of 4% TCA. An aliquot was removed to determine glutamate formed (A) by an end-point assay using glutamate dehydrogenase. LGH formation (0) was measured by adding a FeC1, solution and measuring the absorbance at 540 nm as described in materials and methods. The nmol of LGH and glutamine formed at each concentration of hydroxylamine were added to give the total product formed (+).




78













TABLE 4.1:
Kinetic constants for wt AS-B-catalyzed hydrolysis of L-glutamnine, Lglutamic acid-y-monohydroxamate, and L- glutamic acid-,y- hydrazide.




Substrate k~(e) K (M) ki. K sM se c'

L-glutamine 1.9 0.03 1.0 0.06 1900

L-glutamic-y-monohydroxamate 2.5 0.02 0.65 0.02 3846

L-glutamic-y-hydrazide 2.0 0.05 2.3 0.20 869




79


Determination of the Dissociation constants for glutamine and LGH using the covalent modifier. 6-diazo-5-oxonorleucine (DON).
The diazo ketone analogue of glutamine, DON, has been shown to inhibit the glutaminase activity of bovine AS in such a manner that the effects could be eliminated only by the addition of glutamine and not by desalting (Mehlhaff & Schuster, 1991). Furthermore, the x-ray structure of the glutamine active site labeled by DON in another class II amidotransferase, E. coli GPA (Kim et al., 1996), has provided evidence that this analog alkylates the cys-1 residue in a reaction which is comparable to the formation of the glutamylenzyme intermediate. Using the method of Meloche (1967), the affinity analogue, DON, has been used in the determination of the glutamine dissociation constant for the glutaminase reaction catalyzed by anthranilate synthetase (Goto et al., 1976), CAD glutaminase (Chaparian & Evans, 1991) and the amidotransferase PabA/PabB complex (Roux & Walsh, 1992). Likewise, the ability of DON to inactivate wild-type and the R30A mutant of AS-B was examined and the effect of glutamine or LGH on the inactivation was used to determine the dissociation constant of glutamine and LGH. Incubation of wild type AS-B with DON prior to the measurement of the glutaminase activity caused a time-dependent loss of activity over a period of 20 minutes. The semilogrithmic plot of the residual activity versus time shows that DON inactivates AS-B with pseudofirst order kinetics (figure 4.3). This plot also shows that inactivation is irreversible since the residual activity of the enzyme after it had been incubated with DON was determined using 20 mM glutamine, providing a 1400-fold dilution. The inactivation half-times, u, were




80


determined from these plots and were used in a replot to determine the Ki for DON according to the following equation described by Meloche (1967):


(TKi)+ (1)



where u is the time required for the inhibitor to cause a 50% loss of activity (tr= n2/first order rate constant for inactivation); [I] is the inhibitor concentration; T is the time required for one-half inhibition at infinite inhibitor concentration; and Ki is the inhibition constant for the inhibitor. When measuring the residual AS-B glutaminase activity after incubating the enzyme with DON, the replot (tc vs 1/[I]) showed a straight line which extrapolated to a minimum inactivation half-time, T, of 1.22 minutes (figure 4.4). The Ki for DON determined from the slope of this replot was 34 .M.
The dissociation constant for glutamine, K was determined by measuring the competitive binding of glutamine and DON. Therefore, DON inactivation was examined in the presence of 20 jiM and 50 jM glutamine (Figures 4.5 and 4.6). Using the following equation (Meloche, 1967):


TK( 1+ IS) (2)
=[I]




81









2.25
[DON]
>12.00 1 4 iM
A~ 6pM
1.75
03 8 PvM

0 1.50- 0 10OPM
0)
1.25-012p
0 1 4 pM

1.00
0 5 10 15 20 25

time (min)






Figure 4.3: DON inactivation of AS-B in the absence of glutamine. Wild type AS-B was incubated with the concentrations of DON given in the figure. At time points ranging from 0-20 minutes, an aliquot was removed from the mixture and tested for residual activity in a reaction containing 12 mM glutamine. The glutaminase activities of the DON modified enzymes were measured as described in materials and methods.




82












2015


10


5


0
0.00 0.05 0.10 0.15 0.20 0.25
1/DON RM





Figure 4.4: Replot of DON inactivation of AS-B in the absence (0) and presence of glutamine (A. 20 pM; 0, 50 tM).




83


a replot of 'r vs 1/[I] at each concentration of glutamine gave straight lines with a different slope for each glutamine concentration (Figure

4.4). The slope of the line resulting from 20 RLM glutamine is 50.73 and that from 50 LM glutamine is 65.02 corresponding to K, values of 140 iM and 200 jiM respectively.
In a similar manner, DON inactivation and LGH protection against inactivation was examined for wild type AS-B, and DON inactivation and glutamine protection against inactivation was examined for the AS-B mutant R30A (Figures 4.7-4.14). Using equation (1) and the replot of vs 1/[DON] the minimum inactivation half-time, T, for DON inactivation of wild type catalyzed LGH hydrolysis (Figure 4.8) and R30A catalyzed glutamine hydrolysis (Figure 4.12) were 1.4 sec and 3.5 sec, respectively. The K for DON in these two reactions, determined from the slope of the replot was 28 M for LGH hydrolysis and 8.7 mM for inactivation of R30A catalyzed glutamine hydrolysis. The following equation was used to determine the dissociation constant for LGH in the wild type catalyzed glutaminase reaction and for glutamine in the R30A catalyzed reaction:


c=[S]T](Kac) + (T+T Kinact) (3)
KSM



The Ks, derived from the slope of the line in a replot of c vs [substrate] for LGH in the wild type catalyzed hydrolysis reaction (Figure 4.10) was 70 jtM and for glutamine in the R30A catalyzed reaction (Figure 4.14) was 9.5 mM




84










2.25- [DON]

2.00 6 gM
.8gM
1.75 10 gM

Co
0
1.25

1.00
0 5 10 15 20 25

Time (min)




Figure 4.5: DON inactivation of AS-B in the presence of 20 tM glutamine. Wild type AS-B was incubated with the concentrations of DON given in the figure. At time points ranging from 0-20 min, an aliquot was removed from the mixture and tested for residual activity in a reaction containing 12 mM glutamine. The glutaminase activities of the DON modified enzymes were measured as described in materials and methods.




85









[DON]
2.5
4 6pM

A 8 pm
2.0 10pM

V 12pM

0
1.5



60 10 15 20 25

Time (min)





Figure 4.6: DON inactivation of AS-B in the presence of 50 jiM glutamine. Wild type AS-B was incubated with the concentrations of DON given in the figure. At time points ranging from 0-20 min, an aliquot was removed from the mixture and tested for residual activity in a reaction containing 12 mM glutamine. The glutaminase activities of the DON modified enzymes were measured as described in materials and methods.




86










[DON]
2.25

2.00-, 6 M
A 8 M

1.75- 10 RM

-0 1.500
1.25

1.00 ,,,
10 15 20
Time (min)




Figure 4.7: DON inactivation of wild-type catalyzed LGH hydrolysis in the absence of LGH. Wild type AS-B was incubated with the concentrations of DON indicated in the figure. At time points ranging from 0-18 min an aliquot was removed from the mixture and tested for residual activity in a reaction containing 12 mM glutamine. The glutaminase activities of the DON modified enzymes were measured as described in materials and methods.




Full Text
CHEMICAL AND KINETIC CHARACTERIZATION OF GLUTAMINE
HYDROLYSIS CATALYZED BY ESCHERICHIA COLI
ASPARAGINE SYNTHETASE B
BY
HOLLY G. SCHNIZER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1997


103
includes glutamine 5'-phosphoribosylpyrophosphate amido-
transferase (GPA) (Tso et al., 1982), glutamine fructose-6-phosphate
amidotransferase (GFAT) (Badet-Denisot et al., 1993), and glutamate
synthase (Vanoni et al., 1991). These enzymes are characterized by
a highly conserved N-terminal cysteine residue in their mature,
active form. AS-B catalyzes the ATP dependent synthesis of
asparagine utilizing either ammonia (Reaction 1) or glutamine
(Reaction 2) as a nitrogen source. When aspartate is absent, AS-B
catalyzes the hydrolysis of glutamine to glutamate and ammonia
(Reaction 3), in a reaction that is stimulated by ATP (Boehlein et al.,
1994b; Boehlein et al., 1996).
Reaction 1: L-Asp + NH? + ATP > L-Asn + AMP + PPi
Reaction 2: L-Asp + L-Gln + ATP > L-Asn + L-Glu + AMP + PPi
Reaction 3: L-Gln -> L-Glu + NH?
AS-B has been employed in an extensive series of studies aimed at
delineating (i) the molecular mechanism of asparagine synthesis, and
(ii) the functional roles of conserved residues in glutamine binding
and activation (Boehlein et al., 1994a; Boehlein et al., 1994b; Boehlein
et al., 1996; Richards & Schuster, 1997). The high level of
glutaminase activity exhibited by AS-B has also been exploited
(Boehlein et al., 1997b) to show a mechanistic relationship between
Class II amidotransferases and thiol proteases, such as papain (Storer
& Menard, 1994). On the other hand, and despite substantial effort
(Badet, B., personal communication), the thioester intermediate that
must therefore be formed during glutamine breakdown has not yet


105
dpm/nmol) in 100 mM Tris-HCl (pH 8.0) at room temperature (total
reaction volume 100 pi). After 30 s, the reaction was quenched in 1
ml of 8% TCA on a 2.5 cm nitrocellulose filter (0.45 mm porosity) and
100 pi BSA (10 mg/ml) was added. After 2 minutes in the quench
solution the samples were filtered under vacuum on a 2.5 cm
nitrocellulose filter (0.45 pm porosity), and the filter was washed
with 50 ml of 1 N HC1. The filter was transferred to 5 ml of
scintillation fluid (ScintiVers IP) and the 14(y activity was measured
on a Beckman model LS6000IC scintillation counter. This procedure
was repeated in the absence of enzyme as a control for non-specific
binding of L-glutamine to the filter.
Gel Filtration of the Intermediate formed between AS-B and
Glutamine
(A) 1 mM [U-14-C]-L-glutamine (SA 22 000 dpm/nmol) was
incubated with either wt AS-B (0.74 nmol) or the CIA AS-B mutant
(0.74 nmol) in 100 mM Tris-HCl (pH 8.0) at room temperature (total
reaction volume 100 pi). After 30 s, reactions were terminated using
22 pi of 0.1 M NaOAC (pH 4.0) containing 5% SDS, and the protein was
separated from free glutamine by gel filtration using Sephadex G-50
Fine spin column equilibrated with 1% SDS in 0.1 M NaOAc (pH 4.0)
(Penefsky, 1979). The amount of radiolabeled enzyme was then
measured by liquid scintillation counting. (B) In a separate gel
filtration experiment, wt AS-B (2 nmol) was incubated with 2 mM
[U-14C]-L-glutamine (SA 11,000 dpm/nmol) in 100 mM Tris-HCl (pH
8.0) for 30 s at room temperature (total volume 100 pi). After
termination by dilution with 100 pi of 8M guanidinium-HCl in 0.1M
NaOAc (pH 4.0), the protein was isolated by gel filtration on a


66
The free energy, AG, derived from this equilibrium constant was
calculated to be -13.2 kJ/mol.


32
PH
Figure 2.4 The pH dependence of kcat for wild-type catalyzed LGH
hydrolysis. The kinetic constants were determined as described in
materials and methods.


157
Hongo, S, Matsumoto, T, & Sato, T (1978) Purification and properties
of asparagine synthetase from rat liver. Biochim Biophys Acta 552:
258-266
Humbert, R & Simoni, RD (1980) Genetic and biochemical studies
demonstrating a second gene coding for asparagine synthetase in
Escherichia coli. J. Bacteriol. 142: 212-220
Hutson, RG, Kitoh, T, Moraga-Amador, DA, Cosic, S, Schuster, SM,
Kilberg, MS (1997) Amino acid control of asparagine synthetase:
relation to asparaginase resistence in human leukemia cells. Am. J.
Physiol. 272: C1691-C1699
Isupov, MN, Obmolova, G, Butterworth, S, Badet-Denisot, MA, Badet,
B, Polikarpov, I, Littlechild, JA, Teplyakov, A (1996) Substrate
binding is required of the assembly of the active conformation of the
catalytic site in Ntn amidotransferases: Evidence from the 1.8 A
crystal structure of the glutaminase domain of glucosamine 6-
phosphate synthase. Structure 4: 801-810
Kidd, S.T. (1953a) Regression of transplanted lymphomas induced in
vivo by means of normal guinea pig serum. I. course of
transplanted cancers of various kinds in mice and rats given guinea
pig serum, horse serum, or rabbit serum. J. Exp. Med. 98: 565-581
Kidd, S.T. (1953b) Regression of transplanted lymphomas induced in
vivo by means of normal guinea pig serum. II. Studies on the
nature of the active serum constituent. J. Exp. Med. 98: 583-606
Kim, JH, Krahn, JM, Tomchick, DR, Smith, JL, Zalkin, H (1996)
Structure and function of the glutamine
phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 271:
15549-15557
Kiriyama, Y, Kubota, M, Takimoto, T, Kitoh, T, Tanizawa, A, Akiyama,
Y, Mikawa, H (1989) Biochemical characterization of U937 cells
resistant to L-asparaginase: the role of asparagine synthetase.
Leukemia 3: 294-297


24
Enzyme Preparation
Construction of recombinant wild-type AS-B and mutants in
which His-29 or His-80 have been replaced by alanine (H29A, H80A)
has been described elsewhere (Boehlein et ah, 1994a). PCR cassette
mutagenesis was used to replace Histidine-47 with alanine (H47A) or
asparagine (H47N) using oligonucleotide primer pairs ss350R and
ss65 or ss348R and ss65, respectively (Table 2-1) and pETB as a
template. The PCR parameters involved 35 cycles of denaturation at
94C for 1 min; annealing at 55C for 1 min; and extension at 72C for
two minutes and the final cycle was followed by a 5 minute
extension step at 72C. The plasmid, pETB, and the PCR products
were digested with Hpal and Xhol and the 50 bp product was cloned
into pETB. The resulting plasmids were sequenced through the
insertion and no alterations other than those intended were found.
All wild-type and mutant proteins were expressed, and purified
using standard procedures (Boehlein et al., 1994a). Protein
concentrations were determined by the method of Bradford (1976)
using an assay kit supplied by Bio-Rad and mouse immunoglobulin G
to construct a standard curve. During the course of these
investigations an alternate method of determining the protein
concentration revealed that the current use of mouse
immunoglobulin G as a standard gives erroneously high protein
concentrations for AS-B. Amino acid composition analysis of AS-B,
carried out by the Protein Chemistry Core Facility of ICBR at the
University of Florida, showed that the actual value for the
concentration of this enzyme is 2.7 times less than that obtained
using the Bradford assay. Therefore, all of the values of enzyme


3
In view of the above mentioned problems associated with
asparaginase therapy, alternative methods of treating ALL need to
be explored. Since the effectiveness of this therapy seems to reside
in reducing the circulating concentration of asparagine, inhibitors of
asparagine synthetase could provide this much needed alternative.
In efforts to achieve this goal, several hundred compounds have
been evaluated as AS inhibitors, however none have provided both
the strength and specificity required for clinical consideration
(Cooney et al, 1976). Therefore, a detailed chemical and kinetic
investigation of AS is essential in order to develop more effective
inhibitors of this enzyme.
Asparagine Synthetase
Asparagine synthetases catalyze the ATP dependent transfer of
nitrogen from either glutamine or ammonia to aspartic acid in the
formation of asparagine. The biochemical and mechanistic properties
of AS have been examined in a variety of sources including mammals
(Hongo et al., 1978; Markin et al., 1979; Markin et al., 1981; Milman
& Cooney, 1979; Van Heeke & Schuster, 1989b), fungi (Ramos &
Wiame, 1979, 1980), plants (Pike & Beevers, 1982; Rognes, 1970),
and bacteria (Burchall et al., 1964; Cedar & Schwartz, 1969a, 1969b;
MacPhee et al., 1983) as well as in some tumor cell lines (Gantt &
Arfin, 1981). In order to facilitate the recovery of sufficient amounts
of AS for study, expression systems have been developed for the
overexpression of human AS in yeast (Van Heeke & Schuster, 1990)
and AS-B in Escherichia coli (Boehlein et al., 1994a).


107
the previous chapter, then accumulation of this intermediate may
allow an opportunity for its isolation.
Two different methods were used to isolate a covalent adduct
formed between wt AS-B and glutamine. In the initial approach,
labeled protein was obtained by incubating wt AS-B with [44c]-L-
glutamine, denaturing the protein with SDS in acid solution, and
separating the enzyme from free glutamine by gel filtration (Table
1). The radioactivity associated with the enzyme most likely was not
due to non-specific glutamine binding as wt AS-B that had been
covalently modified by 6-diazo-5-oxonorleucine (DON), prior to
incubation with [14c]-L-glutamine under these conditions, did not
give the radiolabeled covalent adduct (Table 5.1). Furthermore,
radiolabeled protein was not formed when [14c]-L-glutamine was
incubated with the CIA AS-B mutant in place of the wild-type
enzyme, despite the observation that this AS-B mutant binds L-
glutamine with high affinity (Boehlein et al., 1994a).
While gel filtration could be used successfully to isolate the
putative y-glutamyl AS-B adduct, quantitative measurements of its
kinetic properties were hampered by a reduction in the amount of
radiolabeled enzyme as a function of the time during which samples
were maintained in the quench solution (Table 5.2). After
approximately fifteen minutes, the observed reduction in
radiolabeled protein stabilized at 59% of the value observed at short
incubation times. Efforts to address this problem by employing 8 M
guanidine-HCl as a denaturant, instead of SDS, did not affect the rate
of loss of radiolabeled enzyme.


20
play three essential acid-base roles including 1) stabilization of the
thiolate; 2) donation of a proton to the amide of the first tetrahedral
intermediate to increase its leaving group ability; and finally 3)
activation of water for a nucleophilic attack on the thioester
intermediate by accepting a proton (Storer & Menard, 1994).
This mechanism of ammonia mediated nitrogen transfer is
supported by the ability of the amidotransferases to use ammonia as
a nitrogen source and thus the demonstration of an ammonia binding
site. Furthermore, the presence of a completely conserved N-
terminal cysteine, which has been shown to be essential for
glutamine dependent activity of GPA (Mei & Zalkin, 1989) as well as
both human and E. coli asparagine synthetases (Boehlein et al.,
1994a; Sheng et al., 1993; Van Heeke & Schuster, 1989b), suggests
that this residue may function in a manner similar to that of the
active site cysteine of the thiol proteases. On the other hand, the
presence of an important histidine in the active site of the class II
enzymes has become increasingly doubtful. While the studies of GPA
mentioned above (Mei & Zalkin, 1989) and chemical modification
studies of GFAT (Badet-Denisot & Badet, 1992) seemed to indicate
the presence of an essential histidine in these enzymes, site-directed
mutagenesis of residues, known to be conserved among asparagine
synthetases, did not reveal any histidines required for activity in AS-
B (Boehlein, et al., 1994a). Moreover, recent structural studies of
GPA (Kim et al., 1996; Smith et al.. 1994) and the glutamine utilizing
domain of GFAT (Isupov et al., 1996) suggested that there were no
histidines in position to act as part of a cys-his dyad. Structural
comparisons later revealed that the protein fold of GPA (Brannigan et


77
[NH2OH] mM
Figure 4.2: Partitioning of the thioester to glutamate hydroxamate
using hydroxylamine in the glutaminase reaction catalyzed by AS-B
R30A. A mixture containing wt AS-B, 80 mM glutamine, 100 mM
Bis-Tris and Tris-HCl (pH 8), and the indicated concentrations of
hydroxylamine was incubated at 37C for 45 min. The reactions
were terminated by the addition of 4% TCA. An aliquot was removed
to determine glutamate formed (A) by an end-point assay using
glutamate dehydrogenase. LGH formation () was measured by
adding a FeCl, solution and measuring the absorbance at 540 nm as
described in materials and methods. The nmol of LGH and glutamine
formed at each concentration of hydroxylamine were added to give
the total product formed (+).


156
Cooney, DA, Jones, MT, Milman, HA, Young, DM, & Jayaram, HN (1980)
Regulators of the metabolism of L-asparagine. A search for
endogenous inhibitors. Int. J. Biochem. 11: 519-539
Crowther, D (1971) L-asparaginase and human malignant disease.
Nature 229: 168-171
Drenth, J, Kalk, KH, & Swen, HM (1976) Binding of chloromethyl
ketone substrate analogues to crystalline papain. Biochemistry 15:
3731-3738
Duggleby, HI, Tolley, SP, Hill, CP, Dodson, EJ, Dodson, G, Moody, PCE
(1995) Penicillin acylase has a single-amino-acid catalytic centre.
Nature 373: 264-268
Epand, RM & Wilson, IB (1963) Evidence for the formation of
hippuryl chymotrypsin during hydrolysis of hippuric acid esters. J.
Biol. Chem. 238: 1718-1723
Fisher, KJ (1993) Post-Translational Processing and Thr-206 are
required for glucosylasparaginase activity. FEES Lett. 323: 271-275
Gantt, IS, & Arfin, SM (1981) Elevated level of asparagine synthetse
activity in physiologically and genetically derepressed Chinese
hamster ovary cells due to increase rates of enzyme synthesis. J.
Biol. Chem. 256: 7311-7315
Goto, Y, Zalkin, H, Keim, PS, Heinrikson, RL (1976) Properties of
anthranilate synthetase component II from Pseudomonas putida.
J. Biol. Chem. 251: 941-949
Hanahan, D (1983) Studies on transformation of Escherichia coli with
plasmids. J. Mol. Biol. 166: 557-580
Hinchman, SK, Henikoff, S, Schuster, SM (1992) A relationship
between Asparagine Synthetase A and Aspartyl tRNA synthetase. J.
Biol. Chem. 267: 144-149


95
Figure 4.15: Glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time. The progress curve shows the nmol
of glutamate formed versus time using 2 mM [14C]-glutamine
(SA=9nCi/pl), and either 0.37 nmol () or 0.74 nmol (A) of wt AS-B
in 100 mM Bis-Tris and Tris-HCl (pH 8). After terminating the
reaction with acetic acid, the glutamine and glutamate were
separated by anion exchange chromatography as described in the
materials and methods and the radioactivity associated with each
amino acid was counted. All values are derived from the average of
four separate determinations.


144
deacylation demonstrated in chapter 6 provides a rate constant for
k3. The observation of a burst in glutamate formation with time
presented in chapter 4 suggests that k3 is less than k2 and therefore
k2 will be arbitrarily set at 1 sec'1. Preliminary studies using rapid
quench techniques have confirmed that this value of k2 is reasonable.
The acylation step will be made irreversible since the backwards
reaction seems unlikely. Recall that ammonia has no effect on the
rate of deacylation (chapter 6, Figure 6.4), it does not seem to act as a
substrate in the synthesis of glutamine (chapter 3), and it does not
inhibit the glutaminase reaction (chapter 3). The value for k3 is
estimated from the observation that the rate of LGH synthesis is 12%
of the rate of its hydrolysis. Partitioning studies suggest that
hydroxylamine readily attacks the thioester, therefore it is probably
not the rate determining step for the reverse reaction. Rather, the
step most likely to govern the rate of the reverse reaction is
formation of the thioester from glutamate (k_3). Thus, an upper limit
can be set to 12% of the forward rate for k3. However, the value of
k_3 was varied in order to attempt to fit the simulation to the actual
data. These values were applied to a simulation of the formation of
glutamate with time. It is apparent from the fit shown in Figure 7.1
that scheme (1) is not sufficient to describe the experimental data.
First, such a scheme results in the lack of a burst in glutamate
formation. Second, the slope of glutamate formation as a function of
time and thus kca[ are too high in the simulated data when compared
to the actual values (see Table 7.4).


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
R. Donald Allison
Associate Scientist of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
\ N,
[ //-> J -V <
7 |
Richard W. Moyer ( j
Professor of Moleculai^Genetics
and Microbiology
This Dissertation was submitted to the Graduate Faculty of the
College of Medicine and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
1 f
a
-i-L*
December, 1997
Dean, College of Medicine
Dean, Graduate School


116
TABLE 5.2
Time-dependence of y-glutamyl AS-B adduct degradation in SDS
solution.
Time
(min )
d p
m
< 1
4802
77
1
4165
3
5
3593
460
1 5
2828
220
29
3036
72
The intermediate was formed using 4 nmol of wild-type AS-B and 1
mM [14C] glutamine (S.A. 22,000 dpm/nmol). The reactions were
terminated as described in Table I and then incubated for the
amount of time indicated before they were added to the sephadex
column.


28
at neutral pH and an acidic and basic pKa close to that obtained with
glutamine (Table 2.5). Unlike the glutamine profile, correction for
protonated LGH had little effect on the pH-kca[/Km profile (Figure 2.5).
Kinetic Characterization of the Glutaminase Reaction catalyzed bv the
AS-B mutant H47N
As part of a continuing search for a histidine which may serve
in a general acid-base capacity during AS-B catalyzed glutamine
hydrolysis, Histidine-47 was examined using site-directed
mutagenesis. Mutations were made which replaced the histidine with
an alanine (H47A) or an asparagine (H47N) and the resulting enzymes
were expressed. H47A was insoluble whereas H47N expressed at
high levels and was successfully purified. The glutaminase activity of
H47N was approximately 10-fold slower than that of wild-type AS-B
(compare Tables 2.2 and 2.4). In addition to these changes in kcat,
there was a slight increase in the Km value for glutamine.
pH Dependence of the Kinetic Constants of wild-tvpe AS-B and its
H47N
The pH dependence of the kinetic constants for H47N catalyzed
glutamine hydrolysis was examined. The pH dependence of kcat was
minimal (figure 2.6) except for a slight decrease at low pH which was
not evident in the wild-type. The pH profile of kcat/Km(app) revealed
a bell-shaped curve (figure 2.7) similar to that of wild-type catalyzed
LGH hydrolysis except the curve was slightly narrower. The
decrease in kcat/Km(app) on the basic limb was evident even after the
correction for protonated glutamine.


110
would then regenerate the enzyme and produce glutamate. Direct
support for this mechanistic hypothesis for Class I amidotransferases
has been provided by isolation of the thioester in studies on CAD
(Chaparian & Evans, 1991), Escherichia coli PabA (Roux & Walsh,
1992) and CPS (Lusty, 1992; Wellner et al., 1973). This
accomplishment has contributed significantly to understanding the
mechanism of glutamine hydrolysis in these enzymes by allowing
use of a minimal model to examine the rate determining step of the
reaction, as well as the effects of the other substrates on this step.
Class II amidotransferases, on the other hand, possess a
conserved N-terminal cysteine that has been demonstrated to be
essential for all glutamine-dependent activities of these enzymes
(Badet et al., 1987; Boehlein et al., 1994a; Mei & Zalkin, 1989; Van
Heeke & Schuster, 1989). Although early experiments suggested the
presence of a catalytically important histidine in GPA (Mei & Zalkin,
1989) site directed mutagenesis of histidines within the GAT domain
of AS-B has almost no effect on the kinetic properties of the resulting
mutant enzymes (Boehlein et al., 1994a) and recent X-ray crystal
structures of Bacillus subtilis GPA (Smith et al., 1994), Escherichia
coli GPA (Kim et al., 1996) and an N-terminal fragment of Escherichia
coli GFAT (Isupov et al., 1996) do not reveal a histidine in the
appropriate position to participate in acid-base catalysis. Class II
amidotransferases have therefore been proposed to be members of
the N-terminal nucleophile (Ntn) hydrolase superfamily (Brannigan
et al., 1995), that includes the 20S proteasome (Seemuller et al.,
1996) and penicillin acylase (Duggleby et al., 1995). Ntn hydrolases
are characterized by an N-terminal catalytic nucleophile (hydroxyl or


15
Figure 1.3 Mechanism of nitrogen transfer via direct transfer of the
nitrogen from glutamine to aspartyl-AMP as proposed by Stoker et
al., 1996).


132
Figure 6.6: Effect of glutamine on the steady-state concentration of
the thioester formed during the glutaminase reaction catalyzed by
the AS-B mutant N74A. A glutamyl-enzyme complex was formed by
incubating 0.74 nmol of N74A with various concentrations of L-
glutamine (SA 22,000 dpm/nmol) in a reaction containing 100 mM
Tris-HCl (pH 8.0) for 30 sec at 5C. A radiolabeled enzyme complex
was captured by quenching the reaction in 8% TCA over a
nitrocellulose filter as described in materials and methods and the
radioactivity associated with the filter was measured by scintillation
counting.


33
Figure 2.5 The pH dependence of kcat/Km for wild-type AS-B
catalyzed LGH hydrolysis. The kinetic constants were determined as
described in materials and methods. The solid line () represents
the actual data. The dashed line (---) represents the data corrected
for the protonation of glutamine.


74
and measuring the absorbance of the hydroxamate-FeCl3 complex at
540 nm. A standard curve using a stock solution of authentic LGH
was used to quantitate the product formed in these reactions.
Results
Partitioning of a thioester intermediate to v-glutamyl hvdroxamate
The minimal mechanism, given in equation 1, requires the
formation of a covalent intermediate. Due to the highly reactive
nature of hydroxylamine towards acyl groups, this agent has been
used to provide evidence for an acylenzyme intermediate and to
determine whether its formation or breakdown is rate limiting in
hydrolysis reactions catalyzed by serine proteases (Caplow et al.,
1964; Epand et al., 1963) as well as by the amidotransferase CAD
(Chaparan et al., 1991). Since hydroxylamine competes with H20 in
the deacylation step, its presence will cause an increase in the amount
of total product formed (hydroxamate plus glutamate) only if this
same step is rate limiting (reaction 2). In contrast, if the acylation
step is very slow then the presence of hydroxylamine will not have
an effect on the amount of total product produced as the slow rate of
the acylation step will limit the amount of thioester available for
hydrolysis and hydroxaminolysis.
H20 E + Glu
K,
E + Gln
(Reaction 2)
E + LGH


73
with each strip was counted in ScintiVersII* scintillation fluid on a
Beckman LS 61C scintillation counter. Control experiments were
carried out in an identical manner except that the enzyme was
excluded from the reaction mixture. The separation of the amino
acids was visualized in a non-radioactive control by spraying the
paper with 0.2% ninhydrin in 95% ethanol. Glutamine was found to
migrate with the solvent front while glutamate remained at the
origin. The number of nanomoles of 14C-labeled glutamine in each
sample was determined by subtracting the percentage of total
radioactive counts located at the origin in the control reactions from
the percentage in each sample. All values are derived from the
average of four separate reactions.
To examine the effect of glutamate on the burst, wild-type AS-B
was incubated on ice either alone or with 100 mM L-glutamate, or D-
Glutamate for 3 hours prior to the reaction. The enzyme/glutamate
mixture then was added to the reaction and product formation was
determined at various time points as described above.
Hvdroxvlamine Partitioning of the Thioester
A reaction mixture containing either wt AS-B (9.3 pg), or R30A
(18.5 pg), and 50 mM glutamine, 100 mM Bis-Tris and Tris-HCl (pH
8) and a variable concentration of hydroxylamine in a total volume
of 300 pi was incubated at 37C for 45 minutes. The reaction was
terminated by the addition of 100 pi of 16% TCA and a 50 pi aliquot
was removed to determine the amount of glutamate formed using
the previously mentioned GDH assay. LGH formation was determined
in a final volume of 500 pi by adding a solution containing 80% TCA,
6N HC1, and 10% FeCl? in 0.02 N HC1 to the remaining reaction,


21
al., 1995) was homologous to that of a family of N-terminal
nucleophile (Ntn) hydrolases which also includes penicillin acylase
(Duggleby et al., 1995), the 20S proteasome (Lowe et al., 1995;
Seemuller et al., 1996), and aspartylglucosamidase(Fisher et al.,
1993; Oinonen et al. 1995). It is thought that members of the Ntn
family use the N-terminal Ser, Thr, or Cys as the nucleophilic catalyst
and the N-terminal a-amino group as the acid-base catalyst.
However, whether or not the class II amidotransferases use the N-
terminal amino group in acid-base catalysis remains uncertain as
results from experiments which have placed an extra amino acid at
the N-terminus of GPA (Kim et al., 1996) have been inconclusive.
Further investigation is necessary in order to understand fully
how acid-base catalysis occurs during the glutamine hydrolysis
catalyzed by the class II amidotransferases. In fact for AS-B, the
possibility of an essential histidine still remains. A computer model
of a hypothetical AS-B glutamine binding site, constructed using the
crystallographic coordinates for the covalent adduct of DON and E.
coli GPA, revealed that histidine-47 of AS-B was in an appropriate
position to function as part of a cys-his dyad cognate to that of
papain (Richards & Schuster, 1997). Moreover, this histidine is
partially conserved in asparagine synthetases and a histidine in a
comparable position is found in several other class II
amidotransferases (Figure 2.1), thus its importance cannot be
overlooked. In an effort to gain a clearer understanding as to how
acid-base catalysis occurs during glutamine utilization by AS-B, it is
the purpose of this chapter to explore the glutaminase activity of this


149
Table 7.2: Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by AS-B according to equation (3).
step
forward (k^)
reverse (kn)
1
lxlO8 IVT'sec"1
1.75xl04 sec"
2
1 sec1
lxlO"6M' sec'
3
l.xlO'1 sec'1
l.xlO'2 sec'1
4
7xl0'2sec'1
1 sec'1
5
1.82xl07 sec1
lxlO8 M'1 sec'


127
[Gin] mM
Figure 6.2: The steady-state concentration of the thioester formed in
the presence of ATP.


142
group via specific acid catalysis, then the 15N V/K KIE might increase
at low pHs where it is easier to protonate the leaving group. The
limitation to such a study is the narrow pH range in which AS-B is
stable.
The finding that there are differences in acid-base catalysis
between papain and AS-B raises the question of whether the AS-B
catalyzed glutaminase reaction follows a pathway which has any
similarities to that of papain. Therefore, the existence of an acyl-
enzyme intermediate in the pathway of glutamine hydrolysis
catalyzed by AS-B was investigated. Chapter 5 presents evidence for
such an intermediate as the combination of [14C] glutamine with the
wild-type AS-B resulted in a radiolabeled enzyme which could be
trapped by precipitation on a nitrocellulose filter. Evidence that the
glutamyl-enzyme adduct is a thioester includes: 1) the inability to
form a glutamyl-enzyme adduct with the AS-B mutant, CIA, which
can bind glutamine but lacks the cysteine thought to make the thiol
linkage (Chapter 5) 2) the inability to form an adduct with wild-
type enzyme that had been incubated with a covalent modifier (DON)
which was shown to be competitive with glutamine and to cause
irreversible inhibition of the glutaminase activity (Chapter 4) 3) the
observation that the amount of intermediate saturates with
increasing glutamine concentration and the half saturation point is
consistent with steady state parameters observed for the AS-B
glutaminase reaction (Chapter 5), and 4) the instability of the
intermediate in a basic environment as would be expected of a thiol
ester intermediate (Chapter 5). Future studies to support the
conclusion that the covalent intermediate is a thioester include the


slope
62
Figure 3.8: Replot of the slope from the double-reciprocal plots of
glutamine hydrolysis at 5C versus glutamate concentration.


145
Time (sec)
Figure 7.1: Comparison of the simulated to the actual data
representing glutamate formation as a function of time. The dashed
line () represents the simulation of equations (1) and (2). The
solid line () that parallels the actual data points (E) represents the
simulation from equation (3).


123
mg/ml) was added. After 2 minutes in the quench solution the
samples were filtered by vacuum through a 2.5 cm nitrocellulose
filter (0.45 mm porosity) and the filter was washed with 50 ml of IN
HC1. The filter was transferred to 5 ml of scintillation fluid
(ScintiVers II*) and the 14C activity was measured.
Rate of deacvlation of the thioester intermediate
A glutamyl-enzyme adduct was formed at 5C by incubating
0.74 nmol of either wt, N74A, or R30A with 1.5 mM L-glutamine (SA
20,000 dpm/nmol) in a 100 pi reaction containing 100 mM Bis-Tris
and Tris (pH 8.0). After incubating the reaction for 1 min, 100 pi of
200 mM non-radioactive glutamine was added to each sample to
dilute the unreacted radioactive glutamine. The samples were
quenched at various time intervals after the dilution as described
above and the amount of radioactivity associated with the protein
was determined by scintillation counting.
Results
Characterization of the deacvlation of wild-tvpe AS-B
Isolation of a thioester intermediate formed between wild-type
AS-B and glutamine and the steady-state concentration of this
intermediate have been described in chapter 5.
A plot of thioester concentration versus time of incubation with
unlabeled glutamine followed a mono-exponential decay and the rate
constant for breakdown of the thioester determined from this plot
was 0.16 sec1 (Figure 6.1). This value is four times higher than the
kcat (Table 6.1).


kcat/Km (NT1 sec-
36
Figure 2.7 The pH dependence of kcat/Km for glutamine hydrolysis
catalyzed by the AS-B mutant H47N. The kinetic constants were
determined as described in materials and methods.


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CHEMICAL AND KINETIC CHARACTERIZATION OF GLUTAMINE
HYDROLYSIS CATALYZED BY ESCHERICHIA COLI ASPARAGINE
SYNTHETASE B
By
Holly Schnizer
December, 1997
Chairman: Dr. Sheldon Schuster
Major Department: Biochemistry and Molecular Biology
A detailed model of glutamine hydrolysis catalyzed by
Escherichia coli asparagine synthetase B, a class II amidotransferase,
was constructed using a variety of chemical and kinetic techniques.
Glutamine hydrolysis catalyzed by the class II amidotransferase
family has been proposed to follow a mechanism similar to that of
amide hydrolysis catalyzed by the thiol protease, papain. This study
presents a comparison of the catalytic behavior of these two
hydrolytic enzymes in order to determine if asparagine synthetase B-
catalyzed glutaminase reaction can be described by a similar
mechanism. First, this study revealed differences in acid-base
catalysis between these two hydrolytic enzymes. In a continuing
investigation of the importance of conserved histidines in the


O d
91
Figure 4.12: Replot of DON inactivation of R30A in the absence of
glutamine. The K¡ for DON, determined from the slope of the line and
equation (1), was 8.7 mM.


41
pKi 4 pK2 8.5
-SH +HIm- -S' +HIm- -S' Im-
active enzyme
form
In order to understand how acid-base catalysis occurs in the
AS-B active site during glutamine hydrolysis, the pH dependence of
the kinetic constants has been examined for this enzyme. For both
glutamine and LGH, the kcat/Km value has a bell-shaped pH
dependency. In contrast, kcat appears independent of pH in the range
of 6-9. The similarity in pH profiles for these two substrates
suggests that their hydrolysis occurs by similar mechanisms.
Furthermore, the values of kca[/Km for LGH are about 2-fold higher
than those for glutamine (compare Tables 2.2 and 2.3). If substrate
hydrolysis follows a mechanism like that of papain, then N-
protonation of the leaving group on the first tetrahedral intermediate
is required for C-N bond cleavage to occur. Therefore, the faster
acylation rate for LGH, reflected in kcat/Km, is consistent with the fact
that hydroxylamine (pKa=8.2) is a better leaving group than ammonia
(pKa~9.2). The pH independence of kcat may arise from a change in
rate determining step between two steps with similar rates.
However, further investigation, using a more extensive pH range is
necessary for a clearer picture of the ionizations affecting kcat.
Unfortunately, the AS is highly unstable when the pH is greater than
9 or less than 6.
The bell-shaped curves displayed in the kcat/Km profiles for AS-
B hydrolysis of glutamine and LGH (with pKa values of 6.6 and 8.3)
are narrower than that reported for papain. The stability of the ion-


83
a replot of x vs 1/[I] at each concentration of glutamine gave straight
lines with a different slope for each glutamine concentration (Figure
4.4). The slope of the line resulting from 20 pM glutamine is 50.73
and that from 50 pM glutamine is 65.02 corresponding to Ks values of
140 |iM and 200 pM respectively.
In a similar manner, DON inactivation and LGH protection
against inactivation was examined for wild type AS-B, and DON
inactivation and glutamine protection against inactivation was
examined for the AS-B mutant R30A (Figures 4.7-4.14). Using
equation (1) and the replot of t vs l/[DON] the minimum inactivation
half-time, T, for DON inactivation of wild type catalyzed LGH
hydrolysis (Figure 4.8) and R30A catalyzed glutamine hydrolysis
(Figure 4.12) were 1.4 sec and 3.5 sec, respectively. The K¡ for DON in
these two reactions, determined from the slope of the replot was 28
pM for LGH hydrolysis and 8.7 mM for inactivation of R30A catalyzed
glutamine hydrolysis. The following equation was used to determine
the dissociation constant for LGH in the wild type catalyzed
glutaminase reaction and for glutamine in the R30A catalyzed
reaction:
(3)
The Ks, derived from the slope of the line in a replot of x vs
[substrate] for LGH in the wild type catalyzed hydrolysis reaction
(Figure 4.10) was 70 pM and for glutamine in the R30A catalyzed
reaction (Figure 4.14) was 9.5 mM


69
entire class of enzymes (Boehlein et al., 1994a; Zalkin, 1993).
Moreover, studies using site-directed mutagenesis of AS-B (Chapter
2; Boehlein et al., 1994a) and x-ray crystallography of GFAT (Isupov
et al., 1996) and GPA (Kim et al., 1995) suggest that the class II
amidotransferases function differently than papain with respect to
acid-base catalysis. While the lack of a conserved histidine in the
class II amidotransferases contrasts with the essentiality of such a
residue in thiol proteases, recent evidence for an acyl enzyme
intermediate (chapter 5 & 6) formed during glutamine hydrolysis
catalyzed by AS-B provides a basis for comparison of the hydrolysis
reactions catalyzed by the two enzyme families. A minimal
description of amide hydrolysis catalyzed by papain, which occurs
via a thioester intermediate (E-TE), is shown in equation 1 below:
(equation 1)
+ P
In relating this equation to glutamine hydrolysis catalyzed by AS-B,
ES is the enzyme-substrate complex, E-TE is the acylenzyme
intermediate, and P, and P, are ammonia and glutamate,
respectively. In an effort to understand the mechanism of glutamine
hydrolysis catalyzed by AS-B, the glutaminase reaction will be
examined in order to develop a model for this particular aspect of
the enzymatic reaction. The minimal model shown above will be
used as a starting point in this investigation to provide evidence (in
addition to that shown in chapter 5 for the existence of a thioester
intermediate, to determine whether steps leading to its formation or


38
Table 2.5:
Values of pKa for the kcal/Km-pH profiles for glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and for glutamine hydrolysis
catalyzed by the AS-B mutant H47N
Enzyme
Substrate
km ,/Km
PK,
pK?
wild type
glutamine
6.7 0.1
8.8 0.1
LGH
6.7 0.1
8.0 0.1
H47N
glutamine
7.2 0.2
8.1 0.2


53
Figure 3.1: Temperature dependence of glutamine hydrolysis
catalyzed by wild-type AS-B. The initial rates were determined at
pH 8.0 in 100 mM Bis-Tris and Tris-HCl as described in the materials
and methods.


27
In the above equation, Y is the observed kcat/Kra at a given pH, Ylim is
the maximum value of kcat/Km, and pK; and pfC, are the lower and
higher acid dissociation constants, respectively.
Results
Effect of pH on the Steady-state Kinetic Parameters of wild-tvpe AS-
B Catalyzed Glutaminase
The steady-state kinetic parameters, kcat and kcat/Km, for
glutamine hydrolysis were measured as a function of pH. As
illustrated in Figures 2.2 and 2.3, and Table 2.2. kcal is constant over
the pH range 6.0-9.0 whereas the kca[/Km exhibits a bell shaped curve.
Since the pKa of the a-amino group of glutamine is 9.13, the Km and
thus kcat/Km were corrected for protonated glutamine using the
Henderson-Hasselbach equation in an attempt to show how the
ionization of the substrate might affect the kcat/Km-pH profile. Indeed,
the correction caused a complete disappearance of the basic limb in
the kcat/Km(app) plot (Figure 2 .3). Non-linear regression of the
experimental data to equation 1 allowed determination of the pKa for
the non-corrected data which is shown in Table 2.5.
A comparison of the pH dependencies of the hydrolyses of
glutamine and the alternate substrate, LGH, can provide information
about whether these reactions follow the same mechanism.
Therefore, a pH profile of the wild-type AS-B catalyzed LGH
hydrolysis was also examined (figure 2.4 and 2.5, and Table 2.3).
Similar to the glutamine profile, the kcat appeared independent of pH
for most of the pH range studied with slight increases at the low and
high pH. The kcar/Km(app) profile for LGH is bell shaped with a peak


121
very similar to Gin-19 of papain, as part of an oxyanion hole which
stabilizes the tetrahedral intermediates thought to be formed during
the acylation and deacylation steps in the hydrolysis reaction
(Boehlein et al., 1994b; 1996; 1997). Therefore, such an involvement
should be reflected in a difference in either the steady-state
concentration of the thioester or the rate of deacylation catalyzed by
the AS-B mutant, N74A, relative to those values measured for the wt
enzyme.
Several roles have been suggested for Arginine-30 such as an
involvement in the stabilization of the glutamine binding site, the
communication between the GAT and synthetase domains, and the
release of nitrogen from glutamine (Boehlein et al., 1994b). These
ideas were supported by structural information obtained for GPA
(Kim et al., 1996) and GFAT (Isupov et al., 1996) which revealed a
network of hydrogen bonding between the cognate arginine residue
in these enzymes and other residues involved in catalysis in the
glutamine site including Cys-1 and Asn-101(Asn-74 in AS-B). While
chapter 4 presented evidence which supports the active participation
of Arg-30 in glutamine binding, additional exploration into its
interactions and functions will be necessary to obtain a more precise
and comprehensive picture of its purpose. For example, partitioning
studies (Chapter 4) suggested that the deacylation step is rate
limiting in glutamine hydrolysis catalyzed by the R30A mutant of
AS-B. This conclusion can now be tested by comparison of the rate of
deacylation (determined by direct measurement) with that of the
overall reaction.


133
Figure 6.7: The rate of deacylation of the AS-B mutant N74A during
glutamine hydrolysis. A glutamyl-enzyme adduct was formed at 5C
by incubating 0.74 nmol N74A with 1 mM glutamine (SA 20,000
dpm/nmol) in 100 mM Bis-Tris and Tris (pH 8.0). After incubating
the reaction for 30 sec, the radioactive glutamine was diluted by the
addition of 200 mM non-radioactive glutamine. The samples were
quenched at various time intervals after the dilution as described in
materials and methods and the amounts of radioactivity associated
with the protein was determined by scintillation counting.


This work is dedicated to
my parents, Flint and Dottye Gray,
and to my husband, Richard Schnizer,
with love.
11


80
determined from these plots and were used in a replot to determine
the K¡ for DON according to the following equation described by
Meloche (1967):
(1)
T= (TK:) + T
[I]
where x is the time required for the inhibitor to cause a 50% loss of
activity (x= ln2/first order rate constant for inactivation); [I] is the
inhibitor concentration; T is the time required for one-half inhibition
at infinite inhibitor concentration; and K¡ is the inhibition constant for
the inhibitor. When measuring the residual AS-B glutaminase
activity after incubating the enzyme with DON, the replot (x vs 1/[I])
showed a straight line which extrapolated to a minimum inactivation
half-time, T, of 1.22 minutes (figure 4.4). The K¡ for DON determined
from the slope of this replot was 34 pM.
The dissociation constant for glutamine, Ks, was determined by
measuring the competitive binding of glutamine and DON. Therefore,
DON inactivation was examined in the presence of 20 pM and 50 pM
glutamine (Figures 4.5 and 4.6). Using the following equation
(Meloche, 1967):
(2)


7
O
II
ENZ
ASN-74 NH
HN
H2N C S'
"1
R
CYS-1
This proposed relationship between the thiol protease, papain, and
AS-B was confirmed in studies which showed that a nitrile could act
as a weak competitive inhibitor of wild type AS-B and as a substrate
for the AS-B mutant N74D (Boehlein et al., 1997b); results which
were consistent with similar studies using wild type and a Q19E
mutant of papain (Menard, 1995).
Glutamine Amidotransferase Class II
Based on its primary amino acid sequence, asparagine
synthetase is a member of a family of enzymes known as the
glutamine-dependent amidotransferases. The members of this
family catalyze the transfer of nitrogen from glutamine to a variety
of acceptor substrates in the biosynthesis of amino acids,
carbohydrates, and nucleotides (for review, Zalkin, 1993). Like AS,
all amidotransferases share a modular organization with a glutamine
utilizing (GAT) domain in common and a synthetase domain which is
unique to each member of the family. According to the amino acid
sequence alignment of the GAT domain, glutamine amidotransferases
are divided into two types (Zalkin, 1993). Class I enzymes (formerly
G-type for the trpG gene product, anthranilate synthase) are


64
Discussion
Many of the experiments in the following chapters required a
reaction rate slower than that observed at 37C, the standard
temperature at which the reaction is examined. One way to obtain a
slower rate is to examine the reaction at low temperatures therefore,
many of the experiments in the following chapters have been
conducted at 5C. In preparing for these experiments it was important
to know whether the results obtained at 5C could be compared to
those at the standard temperature. Temperature variations often
cause changes in the rate-limiting step of a reaction. Since non-linear
Arrhenius plots are indicative of such changes, the temperature
dependence of the glutaminase reaction was examined to confirm that
there was not a change in the rate limiting step between 5C and 37C.
In the case of the glutaminase reaction, it is difficult to conclude that a
single step is rate determining across the entire range of temperatures
studied. Certainly, the linear Arrhenius plots suggest that a change in
the rate limiting step between reactions at 5C and 37C is unlikely.
However, it should be noted that two steps in the reaction may have
similar activation energies, and therefore a change in rate limiting step
would remain unnoticed in these experiments. Such an event is a
possibility in AS-B catalyzed glutamine hydrolysis as demonstrated by
the unexpected similarity in activation constants for reactions
catalyzed by wild-type and the mutant R30A (Table 3.1 and Figures
3.1 and 3.3). Partitioning experiments in chapter 4 imply that in
reactions catalyzed by R30A, hydrolysis of the thioester is rate
limiting. In contrast, direct measurements of the deacylation step of
glutamine hydrolysis suggest that an additional step contributes to the


60
1/[Gln] mM'1
Figure 3.6: Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 5C. The rate of glutamine hydrolysis
was determined as described in the materials and methods in the
presence of 0 (0), 10 (A), 50 (), 100 (), and 136 () mM
glutamate.


44
pKa 5 (Brocklehurst, 1988 a,b). Altered pKn values also may result if
the rate constant for substrate dissociation is less than the net rate
constant for product formation. This is unlikely to be the case for
AS-B glutamine hydrolysis since investigations into the dissociation
constant of glutamine have suggested that a rapid equilibrium
mechanism is plausible (see chapter 4). Determination of the pKa of a
competitive inhibitor or of the backwards reaction will be necessary
to confirm this assumption.
With respect to these assumptions it is important to take heed
in comparison of the pH profiles of AS-B and papain. On the other
hand, with the results presented here, it is difficult to envision a role
for histidine-47 in acid-base catalysis when one considers the lack of
effect its replacement with asparagine made on the pH profiles. In
light of the presence of this histidine in many asparagine synthetases
as well as other class II amidotransferases it was important to
include this residue among potential acid-base catalysts. Indeed, the
10-fold loss of activity in the H47N mutant may reflect a minor or
indirect importance of this residue in glutamine binding and/or
utilization. On the other hand, the importance of the histidine to
papain activity is underscored in the complete conservation of this
amino acid across the entire family of cysteine proteases. The lack of
such a completely conserved histidine in the class II family of
amidotransferases suggests that an alternate mechanism of acid-base
catalysis probably occurs in these enzymes and therefore
demonstrates dissimilarities in mechanism for the two families of
enzymes.


159
Mehlhaff, PM & Schuster, SM (1991) Bovine pancreatic asparagine
synthetase explored with substrate analogs and specific monoclonal
antibodies. Arch. Biochem. Biophys. 284: 143-150
Mei, B, and Zalkin, H (1989) Cysteine-Histidine-Aspartate catalytic
triad is involved in glutamine amide transfer function in purF-type
glutamine amidotransferases. J. Biol. Chem. 264: 16613-16619
Meister, A, Soben, H, Tice, SV, Frasen, PE (1952) Transamination and
associated deamidation of asparagine and glutamine. J. Biol. Chem.
197: 319-330
Meloche, HP (1967) Bromopyruvate inactivation of 2-Keto-3-deoxy-
6-phosphogluconic aldolase. I. Kinetic evidence for active site
specificity. Biochemistry 6: 2273-2280
Menard, R, Khouri, HE, Plouffe, C, Dupras, R, Ripoll, D, Vernet, T,
Tessier, DC, Laliberte, F, Thomas, DY, Storer, AC (1990) A protein
engineering study of the role of aspartate 158 in the catalytic
mechanism of papain. Biochemistry 29: 6706-6713
Menard, R, Khouri, HE, Plouffe, C, Laflamme, P, Dupras, R, Vernet, T,
Tessier, DC, Thomas, DY, Storer, AC (1991) Importance of hydrogen
bonding interactions involving the side chain of Aspl58 in the
catalytic mechanism of papain. Biochemistry 30: 5531-5538
Menard, R., Plouffe, C., Laflamme, P., Vernet, T., Tessier, D.C., Thomas,
D.Y., Storer, A.C. (1995) Modification of the Electorstatic Environment
is tolerated in the oxyanion hole of the cysteine protease papain.
Biochemistry 34: 464-471
Milman, HA & Cooney, DA (1979) Partial purification and properties
of L-asparagine synthetase from mouse pancreas. Biochem. J. 181:
51-59
Miran, SG, Chang, SH, & Raushel, FM (1991) Role of four histidine
residues in the amidotransferase domain of carbamoyl phosphate
synthetase. Biochemistry 30: 7901-7907


81
time (min)
[DON]
a 4 (iM
a 6 (i M
8 uM
o 1 0 (iM
1 2 (iM
1 4 (iM
Figure 4.3: DON inactivation of AS-B in the absence of glutamine.
Wild type AS-B was incubated with the concentrations of DON given
in the figure. At time points ranging from 0-20 minutes, an aliquot
was removed from the mixture and tested for residual activity in a
reaction containing 12 mM glutamine. The glutaminase activities of
the DON modified enzymes were measured as described in materials
and methods.


106
Sephadex G-50 Fine spin column (Penefsky, 1979) equilibrated in the
quench solution. The amount of radiolabeled enzyme was measured
by liquid scintillation counting.
Lability of the intermediate formed between AS-B and 14C1-
Glutamine at high pH
The gel filtration procedure was used to determine the stability
of the intermediate at low and high pH. Wild-type AS-B (4.8 nmol)
was incubated 30 s at room temperature with 1.5 mM [14C] -
Glutamine (25,633 dpm/nmol) in a 100 pi reaction containing 100
mM Bis-Tris and Tris-HCl (pH 8). The reactions were terminated by
the addition of 22 pi 0.5 M sodium acetate (pH 4) in 5% SDS. The
protein was separated from free glutamine by the method of
Penefsky (1979) using a Sephadex G-50 fine spin column
equilibrated in either: 1) 0.1 M sodium phosphate (pH 12) in 1% SDS
or 2) 0.1 sodium phosphate (pH 2) in 1% SDS. The amount of
radiolabeled enzyme in 70 pi of the solution which passed through
the column was measured by liquid scintillation counting, and the
protein concentration was measured using an assay kit provided by
BioRad. Values for the protein concentration and the percent of
enzyme labeled at each pH were derived from seven separate
experiments.
Results
The partitioning studies presented in chapter 4 provided
evidence for a thioester formed on the reaction pathway of
glutamine hydrolysis catalyzed by wt AS-B. If the breakdown of the
thioester intermediate is slower than its formation, as indicated in


Ill
thiol) that is thought to be activated by transfer of its proton to the
free N-terminal amino group (Brannigan et al., 1995). Thus, in the
case of AS-B and other Class II amidotransferases, the a-amino group
of Cys-1 would serve as a general acid-base catalyst in place of
histidine. No unambiguous kinetic evidence, however, has yet been
reported that supports such an assumption. The observation of a
significantly smaller amide 15N (Vmax/Km) kinetic isotope effect for
AS-B catalyzed glutamine hydrolysis compared to that observed for
papain catalyzed amide hydrolysis, however, may reflect this
difference in the nature of protonation of the leaving group in the
tetrahedral intermediate (Stoker et al., 1996). Whereas in papain,
His-159 functions as a general acid catalyst (Brocklehurst, et al.,
1987) it has been suggested that, at least for AS-B, N-protonation
occurs by specific acid catalysis (Stoker et al., 1996). On the other
hand, Class II amidotransferases do exhibit many characteristics of
thiol proteases, including an oxyanion hole (Boehlein et al., 1994a;
Boehlein et al., 1997; Isupov et al., 1996; Kim et al., 1996).
In light of these structural and kinetic differences, it is
important to validate the assumption of a thioester intermediate in
AS-B catalyzed glutamine hydrolysis. Toward this goal, the
demonstration of the formation of LGH upon addition of
hydroxylamine to the glutaminase reaction presented in this report
supports the notion that a thioester is formed on the main pathway
of the hydrolysis reaction catalyzed by AS-B. Since the amide of
glutamine is resonance stabilized, hydroxylamine attack requires its
conversion to a more activated carbonyl derivative. Several
intermediates other than a thioester could serve as this activated


2.7The pH dependence of kcat/Km for glutamine
hydrolysis catalzyed by the AS-B mutant H47N 3 6
3.1 Temperature dependence of glutamine hydrolysis
catalyzed by wild-type AS-B 5 3
3.2 Temperature dependence of LGH hydrolysis
catalyzed by wild-type AS-B 5 4
3.3 Temperature dependence of glutamine hydrolysis
catalyzed by the AS-B mutant R30A 5 5
3.4 Effect of hydroxylamine concentration on the
synthesis of LGH catalyzed by wild-type AS-B 5 7
3.5 Effect of glutamate concentration on the synthesis
of LGH catalyzed by wild-type AS-B 5 8
3.6 Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 5C 6 0
3.7 Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 37C 6 1
3.8 Replot of the slope from the double-reciprocal
plots of glutamine hydrolysis at 5C versus
glutamate concentration 6 2
3.9Replot of the slope from the double-reciprocal
plots of glutamine hydrolysis at 37C versus
glutamate concentration 6 3
4.1 Partitioning of the thioester intermediate to L-
glutamic-y- monohydroxamate using
hydroxylamine in a reaction catalyzed by wt AS-B 7 6
4.2 Partitioning of the thioester intermediate to
glutamate hydroxamate using hydroxylamine in
the glutaminase reaction catalyzed by AS-B
R30A 7 7
IX


70
those involved in its breakdown are rate limiting, and to compare
and contrast catalytic behavior of three groups of hydrolytic
enzymes: a class II amidotransferase (AS-B), a class I
amidotransferase (CAD), and a thiol protease (papain).
Materials and Methods
Chemicals and Reagents
14C-glutamine was purchased from Amersham. Scintillation
fluid, ScintiVers II* was obtained from Fischer (Orlando, FL). DE-
81 anion exchange chromatography paper was purchased from
Whatman (Hillsboro, OR). Other chemicals, including L-glutamine and
L-glutamic-y-monohydroxamate, hydroxylamine, and 6-diazo-5-
oxonorleucine (DON) were obtained from SIGMA Chemical Company
(St. Louis, MO) and were of the highest commercial purity.
Enzymes
The construction of recombinant AS-B and the AS-B mutant,
R30A, has been described previously (Boehlein et al., 1994a, 1994b).
The enzymes were obtained by overexpression and purified using
standard procedures (Boehlein et al., 1994a). Protein concentration
was determined by the method of Bradford (1976) using an assay kit
supplied by Biorad and immunoglobulin G as a standard. Recent
examination of the protein concentration using amino acid analysis of
AS-B, performed by the Protein Core Facility of ICBR at the
University of Florida, gave a value for protein concentration which
was 2.7-fold lower than that obtained with the previous method.
Therefore, in this report the enzyme concentration has been
corrected by dividing the enzyme concentrations by the factor 2.7.


4.1 Kinetic constants for AS-B catalyzed hydrolysis of
L-glutamine, L-glutamic acid-y-monohydroxamate,
and L-glutamic acid-y-hydrazide 7 8
5.1 Isolation of a covalent glutamyl-enzyme
intermediate by gel filtration 115
5.2 Time dependence of y-glutamyl AS-B adduct
degradation in SDS solution 116
5.3 Trapping the thioester by filter binding 117
5.4 Base lability of the covalent adduct isolated by gel
filtration 1 1 9
6.1 Characterization of the glutamyl-enzyme
intermediate 1 2 9
7.1 Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by wild-type AS-B according to schemes
(1) and (2) 146
7.2 Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by AS-B according to equation (3) 149
7.3 Comparison of the actual and simulated values
obtained for the kcat, slope, intercept, and [TE]
from Figure 7.1 150
vii


93
Figure 4.14: Replot of glutamine protection of DON inactivation of
R30A catalyzed glutaminase. The Ks, determined from the slope of
the line and equation (3), was 9.5 mM.


54
Figure 3.2: Temperature dependence of LGH hydrolysis catalyzed by
wild-type AS-B. The initial rates at each temperature were
determined at pH8.0 in 100 mM Bis-Tris and Tris-HCl as described in
materials and methods.


151
between the glutaminase and glutamine dependent asparagine
synthesis reactions. If the mechanism of nitrogen transfer involves a
direct attack of the amine of the tetrahedral intermediate on the
acceptor substrate then, at least at high pH, the glutamine dependent
asparagine synthesis reaction might have a faster acylation rate than
the glutaminase reaction since this reaction would not necessarily
require a protonation step.
Another feature of the reaction that will require experimental
testing is the existence of a rate determining conformational change
which has been predicted by the model. Studies are already
underway to measure the rate of fluorescence change during AS-B
catalyzed conversion of glutamine to glutamate using stop-flow
techniques. If the rate is found to be consistent with a rate
determining conformational change then it would be interesting to
compare this rate with that for the AS-B mutant R30A since the
reaction catalyzed by this mutant seems to have a different rate
limiting step (see partitioning experiments in chapter 4).
The studies summarized here represent the first step towards
completing a detailed chemical and kinetic description of glutamine
hydrolysis catalyzed by AS-B. With further experimental support,
the model developed in this work can be used as a basis for
comparison in examining the effects of mutations on glutamine
hydrolysis in efforts to determine the roles of the individual amino
acid residues in the utilization of glutamine. Ultimately, these results
may be significant in forming a bridge toward understanding the
overall mechanism of nitrogen transfer.


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
Sheldon M. Schuster, Chair
Professor of Biochemistry and
Molecular Biology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
//
/
-1/.
///S'/t
Harry S. Nffk
Professor or Neuroscience
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
Thomas W. OBrien
Professor of Biochemistry and
Molecular Biology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
Thomas P. Yang j
Associate Professor of Biochemistry
and Molecular Biology ¡


4.3 DON inactivation of AS-B in the absence of
glutamine 8 1
4.4 Replot of DON inactivation of AS-B in the absence
() and presence of glutamine (A, 20 pM; ,50
pM) 8 2
4.5 DON inactivation of AS-B in the presence of 20 pM
glutamine 8 4
4.6 DON inactivation of AS-B in the presence of 50 pM
glutamine 8 5
4.7 DON inactivation of AS-B catalyzed LGH hydrolysis
in the absence of LGH 8 6
4.8 Replot of DON inactivation of AS-B catalyzed LGH
hydrolysis 8 7
4.9 The effect of various concentrations of LGH on
DON inactivation of AS-B catalyzed LGH hydrolysis 8 8
4.10 LGH protection against DON inactivation of AS-B
catalyzed LGH hydrolysis 8 9
4.11 DON inactivation of the R30A mutant of AS-B in
the absence of glutamine 9 0
4.12 Replot of DON inactivation of R30A in the absence
of glutamine 9 1
4.13 Glutamine protection from DON inhibition of the
R30A catalyzed glutaminase reaction 9 2
4.14 Replot of glutamine protection of DON inactivation
of R30A catalyzed glutaminase 9 3
4.15 Glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time 9 5
x


161
Pike, DC & Beevers, L (1982) Kinetic studies on asparagine synthetase
from mung beans (vigna radiata (L) wilczek) with a phosphonate
analogue of a proposed reaction intermediate. Biochim. Biophys.
Acta. 708: 203-209
Ramos, F & Wiame, JM (1979) The genetics of asparagine synthesis
activity in yeast. Eur. J. Biochem. 94: 409-417
Ramos, F & Wiame, JM (1980) Two asparagine synthetases in
Saccharomyces cerevisiae. Eur. J. Biochem. 108: 373-377
Richards, NGJ, & Schuster, SM (1992) An alternative mechanism for
the nitrogen transfer reaction in asparagine synthetase. FEBS Lett.
313: 98-102
Richards, NGJ, & Schuster, SM (1997) Mechanistic issues in
Asparagine Synthetase catalysis. Adv. Enymol. Relat. Areas Mol. Biol.
In Press
Rognes, SE (1970) A glutamine-dependent asparagine synthetase
from yellow lupine seedlings. FEBS lett. 10: 62-66
Roux, B & Walsh, CT (1992) Para-aminobenzoate synthesis in
Escherichia coli: Kinetic and mechanistic characterization of the
amidotransferase PabA. Biochemistry 31: 6904-6910
Sanz, GF, Sanz, MA, Rafecas, FJ, Martinez, JA, Martin-Aragon, G,
Marty, ML (1986) Teniposide and cytarabine combination
chemotherapy in the treatment of relapsed adolescent and adult
acute lymphoblastic leukemia. Cancer Treat. Rep. 70: 1321-1324
Schendel, FJ, Mueller, E, Stubbe, J, Shiau, A, Smith, JM (1989)
Formylglucinamide ribonucleotide synthetase from Escherichia coli:
cloning, sequencing, overproduction, isolation, and characterization.
Biochemistry 28: 2459-2471
Scofield, MA, Lewis, WS, Schuster, SM (1990) Nucleotide Sequence of
Escherichia coli asnB and deduced amino acid sequence of asparagine
synthetase B. J. Biol. Chem. 265: 12895-12902


96
Time (sec)
Figure 4.16: Effect of the presence of L-glutamate on the burst in
product formation in AS-B catalyzed glutamine hydrolysis. The
figure shows the nmol of glutamate formed versus time in reactions
where wild-type AS-B was incubated alone (+) or with 100 mM L-
glutamate () or D-glutamate (A) prior to addition into the reaction.


164
Vollmer, SJ, Switzer, RL, Hermodson, MA, Bower, SG, Zalkin, H (1983)
The glutamine-utilizing site of Bacillus subtilis glutamine
phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 258:
10582-10585
Wellner, VP, Anderson, PM, & Meister, A (1973) Interaction of
Escherichia coli carbamoyl phosphate synthetase with glutamine.
Biochemistry 12: 2061-2066
Wells, PR (1963) Linear free energy relationships. Chem. Rev. 63:
171
Weng, M, Makaroff, CA, Zalkin, H (1986) Nucleotide sequence of
Escherichia coli pyrG encoding CTP synthetase. J. Biol. Chem. 261:
5568-5574
Zalkin, H (1993) The Amidotransferases. Adv. Enzymol. Relat. Areas
Mol. Biol. 66: 203-309


136
the glutaminase reactions catalyzed by R30A and N74A. While the
deacylation rate was not obtainable for R30A, the steady state
concentration of the thioester formed with this mutant was found to
be slightly higher than that with the wild type enzyme which may be
diagnostic of a slower deacylation step. Indeed, earlier studies
provided evidence that the deacylation step was rate limiting for this
mutant (partitioning studies in chapter 4). Furthermore, previous
structural studies have indicated that the cognate arginine residue in
GPA (Kim et al., 1996) and GFAT (Isupov et al., 1996), R26, is
involved in positioning the conserved asparagine residue (N74 in AS-
B). Since it is thought that this asparagine residue is important in
forming an oxyanion hole in the stabilization of the tetrahedral
intermediates, a decrease in deacylation step in the R30A catalyzed
glutamine hydrolysis could reflect an indirect effect on the
interactions made by this asparagine residue resulting in a decrease
in stability of the second tetrahedral intermediate which occurs
during deacylation. Further examination is necessary to confirm this
idea. However, replacement of asparagine-74 with alanine (N74A)
causes similar changes the amount of thioester formed and the
deacylation rate (Table 6.1). Glutamine hydrolysis catalyzed by the
N74A mutant, like R30A, exhibited a greater concentration of
thioester than the wild type enzyme (Figure 6.6 & Table 6.1).
Moreover, N74A exhibited a slightly slower rate of deacylation than
the wild-type (Figure 6.7 & Table 6.1). Once again, this slower rate
of deacylation may indicate the inability of an enzyme lacking this
important asparagine to stabilize the second tetrahedral
intermediate. Thus the deacylation step is slowed and with a lesser


117
TABLE 5.3
Trapping the thioester by filter binding
Sample
d p m
- enzyme
384
187
CIA
290
115
wt + DON
185
55
w t
6504
183
The intermediate was formed as described in materials and methods
using 2 nmol of enzyme and 1 mM glutamine (S.A. 22,000
dpm/nmol). To examine the effect of DON on intermediate formation,
ImM DON was incubated with wild-type AS-B for 30 minutes. The
enzyme covalently labeled with DON was then diluted into the
reaction mixture containing [14C] glutamine.


49
hydroxamate-FeCl3 complex at 540 nm. A standard curve using a
stock solution of authentic LGH was used to quantitate the product
formed in these reactions.
The ability of AS-B to catalyze the synthesis of glutamine also
was examined. Wild-type AS-B (37 pg) was combined with
glutamate and ammonium chloride (100 mM each) in Tris-HCl (200
mM) in a total volume of 50 pi. The reactions were incubated for 4
hours at 37C followed by the addition of 20 pi of 2N acetic acid. The
terminated reactions were filtered through a 2 pm filter and
derivatized with phenylisothiocyanate (PITC). The glutamine and
glutamate derivatives were separated by HPLC and quantitated
spectroscopically. This assay procedure can detect less than 100
pmoles under our standard procedures
Product inhibition of the glutaminase reaction
Wild-type AS-B was incubated in a 100 pi reaction containing
various concentrations of [14C] glutamine (SA 880-20,000 dpm/nmol)
in lOOmM Bis-Tris and Tris-HCl at 37C for 10 min in the presence of
various concentrations of glutamate or NH4C1. The reactions were
terminated by the addition of 10 pi of IN acetic acid. Glutamate was
separated from glutamine on Whatman DE-81 ion-exchange
chromatography paper using deionized distilled water as the mobile
phase. After 6 hours, elution was stopped, and the paper was dried
and cut into two strips to separate glutamine from glutamate. The
radioactivity associated with each strip was counted in ScintiVersII*
scintillation fluid on a Beckman LS 61C scintillation counter. Control
experiments were carried out in an identical manner except that the
enzyme was excluded from the reaction mixture. The separation of


23
enzyme as a function of pH. In order to address the possibility that
H47functions in an acid-base capacity, the effects of mutations at this
position on the pH-rate profile will be characterized.
Materials and Methods
Chemicals and Reagents
Restriction and modifying enzymes were purchased from Promega
(Madison, WI) or New England Biolabs (Beverly, MA). Other chemicals,
including L-glutamine and L-glutamic-y-monohydroxamate, were
obtained from SIGMA Chemical Company (St. Louis, MO) and were of the
highest commercial purity. Oligonucleotide primers were synthesized
on an Applied Biosystems 380B DNA synthesizer by the DNA Synthesis
Core Facility of the Interdisciplinary Center for Biotechnology Research
at the University of Florida. Polymerase chain reactions (PCR) were
performed on an Ericomp (San Diego, CA) thermocycler using the
Gene Amp DNA Amplification Reagent Kit with AmpliTaq from Perkin-
Elmer. Double stranded DNA sequencing was performed using the U.S.
Biochemical Corp. Sequenase 2.0 Sequencing kit.
Bacterial Strains and Plasmids
Bacterial strains BL21DE3pLysS (F\ ompT, rb\ mb"), supplied by
Studier and Moffat; and NM522 (supE, thi (lac-proAB), hsd5, (rm -
)/F' pro AB, lac Iq Z M15), obtained from Stratagene (La Jolla, CA), are
both derivatives of Escherchia coli K-12. The plasmid pBluescript
was obtained from Stratagene (La Jolla, CA) and the plasmid pETB
has been described previously (Boehlein et al., 1994a). E. coli host
cells were transformed according to the procedures of Hanahan
(Hanahan, 1983).


68
glutamine dependent activity, the position of this cysteine as well as
other sequence differences in the GAT domain further divides the
family into two classes. The essential cysteine residue is located
within the GAT domain for class I enzymes whereas it resides at the
N-terminus in the class II enzymes such as glutamine
phosphoribosylpyrophosphate amidotransferase (GPA) (Tso et al.,
1982) glucosamine-6-phosphate synthetase (GFAT) (Badet et al.,
1987), and asparagine synthetase (Humbert et al., 1980). Initially,
site directed mutagenesis studies suggested that the N-terminal
cysteine as well as a histidine and aspartate residue were involved in
the reaction catalyzed by GPA (Mei & Zalkin, 1989). As this catalytic
triad is found in many thiol proteases (Brocklehurst et al., 1987), it
was proposed that the glutamine hydrolysis reaction catalyzed by
the class II enzymes occurs via a mechanism analogous to that of
peptide hydrolysis catalyzed by the thiol proteases, such as papain.
This hypothesis was extended to the class I enzymes as well (Mei &
Zalkin, 1989). Since then, the importance of a catalytic triad
composed of a conserved cysteine, histidine, and glutamate residues
has been demonstrated for the class I enzymes through site-directed
mutagenesis (Amuro et al., 1985; Miran et al., 1991) and x-ray
crystallography (Tesmer et al., 1996; Thoden et al., 1997).
On the other hand, the picture remains more ambiguous for the
class II amidotransferases as subsequent studies confirmed only the
importance of the conserved N-terminal cysteine residue (Boehlein et
al., 1994a). In fact, a more recent alignment of the GAT domain
which includes a greater number of amidotransferase sequences
reveals that there are no histidines which are conserved across this


4
Two classes of AS have been described which do not possess
any sequence similarity. AsnA. found in many procaryotes, encodes
for an AS (AS-A) which only uses ammonia as a source of nitrogen in
the synthesis of asparagine (reaction 2). The kinetic and chemical
mechanism of AS-A, the product of the asnA gene, has been well
characterized (Cedar & Schwartz, 1969a, 1969b). In fact, the
existence of an aspartyl-AMP intermediate formed during AS
catalyzed asparagine synthesis was demonstrated first in
experiments using AS-A (Cedar & Schwartz, 1969b). The other AS,
encoded by asnB found in both prokaryotes and eukaryotes,
catalyzes the synthesis of asparagine using either ammonia or
glutamine as a source of nitrogen (reactions 1 and 2). These enzymes
also catalyze glutamine hydrolysis (reaction 3) in the absence of the
other substrates; a reaction which is stimulated by the presence of
ATP.
L-Asp + L-Gln + ATP - L-Asn + L-Glu +AMP + PP¡ (Reaction 1)
L-Asp + NH3 + ATP > L-Asn + AMP + PP¡ (Reaction 2)
L-Gln + H20 L-Glu + NH3 (Reaction 3)
Glutamine dependent asparagine synthetases are thought to be
organized into two functional domains. The glutamine amide
transfer (GAT) domain binds glutamine and catalyzes the hydrolysis
of glutamine to glutamate and ammonia. The synthetase domain
binds ATP and aspartate and catalyzes the ammonia dependent
reaction as well as the transfer of nitrogen to aspartate in the
glutamine dependent reaction. Evidence for the location of the two


6
cysteine residue was revealed upon the observation that bovine
pancreatic AS lost glutamine dependent activity when it was
covalently modified with the sulfhydryl modifier and glutamine
analog, 6-diazo-5-oxo-L-norleucine (DON) (Mehlhaff & Schuster,
1991). Finally, the essentiality of the N-terminal cysteine residue
was demonstrated in human AS and E. coli AS-B in experiments
replacing cysteine-1 with either an alanine or serine residue
(Boehlein et al., 1994a; Sheng et aL, 1993; Van Heeke & Schuster,
1989b).
In addition to the conserved cysteine residue, the role of
another completely conserved residue, asparagine-74, also has been
determined. When Asn-74 was replaced with either an alanine or an
aspartate, the resulting mutant enzymes (N74A & N74D) exhibited a
depressed glutaminase activity (Boehlein et al., 1994b). Subsequent
studies using the alternative substrate, L-glutamic acid-y-
monohydroxamate (LGH) and the AS-B mutant N74A led to the
suggestion that the interaction of this asparagine residue with
substrates causes a destabilization in the ES complex (Boehlein et al.,
1996) and, therefore, this residue might play a role similar to that of
Gln-19 of papain (Drenth et al., 1976; Menard, 1995). For papain, it
is thought that Gln-19 forms an oxyanion hole' in the stabilization of
a tetrahedral intermediate formed when the active site cysteine
attacks the substrate. This oxyanion hole contains a hydrogen
bonding network involving the NH peptide backbone of cys-25 and
the side chain NH, of Gln-19 which stabilizes the oxide anion bonded
to the tetrahedral carbon atom. The proposed role of N74 of AS-B in
forming such an oxyanion hole is illustrated below:


40
Comparison of the Kinetic Parameters of Wild-tvpe. H80A. and H29A.
at various pH values
The kinetic constants for the glutaminase activity of the three
mutants (H29A, H47N, and H80A) are compared with wild-type in
Table 2.6. Substitution of an alanine for the histidines at positions 29
and 80 has little effect on the kcat at pH 6.5 and 7.5 while this
substitution decreases kcat/Km significantly. However, unlike wild
type the kcat value of H80A increases with increasing pH to a value at
pH 9 which is double that of wild type. The most striking effect
caused by the mutations at positions 29 and 80 is seen in the
dramatic increase in Km especially at high pH.
Discussion
The importance of a histidine in the mechanism of glutamine
hydrolysis catalyzed by AS-B and thus the mechanistic similarity
between the AS-B catalyzed glutaminase reaction and the papain
catalyzed peptide hydrolysis have been questioned for several years.
Papain catalyzes the hydrolysis of peptide, amide, and ester bonds
by the mechanism shown in (1). All substrates tested with papain
have shown a bell-shaped pH dependence for kcat/Km with pKas of
approximately 4 and 8.5. As shown in the simplified model below, it
is considered that these ionizations represent the thiol of cysteine-25
and the imidazole of histidine-159 which are involved in the
acylation process (Menard et ah, 1990):


TABLE OF CONTENTS
PAGE
ACKNOWLEDGMENTS u
LIST OF TABLES v
LIST OF FIGURES v
ABBREVIATIONS x
ABSTRACT x
CHAPTERS
1INTRODUCTION 1
2 pH DEPENDENCE OF THE GLUT AMIN ASE REACTION
CATALYZED BY ASPARAGINE SYNTHETASE B....
Introduction
Materials and Methods
Results
Discussion
3EFFECT OF TEMPERATURE ON GLUTAMINE HYDROLYSIS
CATALYZED BY ASPARAGINE SYNTHETASE B
Introduction
Materials and Methods
Results
Discussion
IV


130
Time (sec)
Figure 6.4: Effect of ammonia on the rate of deacylation. A
glutamyl-enzyme adduct was formed at 5C by incubating 0.74 nmol
wild-type AS-B with 1.5 mM glutamine (SA 20,000 dpm/nmol) in
100 mM Bis-Tris and Tris (pH 8.0) in the presence, B, and absence,
O, of 10 mM NH4C1. After incubating the reaction for 1 min, the
radioactive glutamine was diluted by the addition of 200 mM non
radioactive glutamine. The samples were quenched at various time
intervals after the dilution as described in materials and methods
and the amounts of radioactivity associated with the protein was
determined by scintillation counting.


119
Table 5.4
Base lability of the covalent adduct isolated by gel filtration.
Conditions Protein Concentration Enzyme Labeled
(mg/ml) (%)
pH 2
1.18 0.06
37 2
pH 12
1.6 0.2
10 2
In all reactions, wt AS-B (4.8 nmol) was incubated 30 s at room
temperature with 1.5 inM [I4C]-glutamine (25,633 dpm/nmol) in a
reaction containing 100 inM Bis-Tris and Tris-HCl (pH 8). The
reactions were terminated by the addition of 22 pi 0.5 M sodium
acetate (pH 4) in 5% SDS. To determine the stability of the
intermediate at pH 2 and 12, the enzyme was separated from free
glutamine by the method of Penefsky (1979) using a Sephadex G-50
fine spin column equilibrated in 0.1 M sodium phosphate (either pH
2 or pH 12) in 1% SDS. The amount of radiolabeled enzyme in 70 pi
of the solution which passed through the column was then measured
by scintillation counting, and the protein concentration was
measured using an assay kit supplied by BioRad.


4.16 Effect of the presence of L-glutamate on the burst
in product formation in AS-B catalyzed glutamine
hydrolysis 9 6
4.17 Mechanism of glutamine hydrolysis catalyzed by
the class II glutamine amidotransferases as
proposed by Mei and Zalkin (1989) 9 7
5.1 Effect of glutamine on the steady-state
concentration of the thioester 1 1 8
6.1 Rate of deacylation of the thioester intermediate
formed during wild-type AS-B catalyzed
glutamine hydrolysis 12 6
6.2 The steady-state concentration of the thioester
formed in the presence of ATP 1 2 7
6.3 Effect of ATP on the rate of deacylation 128
6.4 Effect of ammonia on the rate of deacylation 130
6.5 Effect of glutamine on the steady-state
concentration of thioester formed in the
glutaminase reaction catalyzed by the AS-B
mutant, R30A 13 1
6.6 Effect of glutamine on the steady-state
concentration of the thioester formed during the
glutaminase reaction catalyzed by the AS-B
mutant, N74A 1 3 2
6.7 The rate of deacylation of the AS-B mutant, N74A,
during glutamine hydrolysis 1 3 3
7.1 Comparison of the simulated to the actual data
representing glutamate formation as a function of
time 1 4 5
xi


43
Additionally, the pH profiles for glutamine hydrolysis catalyzed
by H47N (Figures 2.6 & 2.7) resemble those of wt including similar
pKa values of 7.2 and 8.5 (Table 2.5) which suggests that histidine-47
is not responsible for one of these ionizations.
Previous investigations of H29 and H80 failed to reveal that a
histidine in either of these positions was essential for AS-B catalyzed
asparagine synthesis (Boehlein et al., 1994a). While these studies
were performed at pH 8 in the presence of an imidizole buffer, it is
possible that the importance of the active site histidine could have
been masked. Table 2.6 presents the results of a further
investigation using the glutaminase activity at pH 6.5, 7.5 and 9.0.
The lack of a significant decrease in the kca[ value for the glutaminase
reaction catalyzed by these mutants confirms the previous
conclusions that these histidines probably are not involved as part of
the essential cys-his dyad which has been proposed.
Interpretations of this and any pH analysis of an enzymatic
reaction rely on a number of assumptions which have been
summarized by Knowles (1976). First, as noted by Peller and
Alberty (1959), the association of the pKa values of the pH-kcat/Km
with ionizations of the free enzyme and free substrate assumes that
only one ionization state of the enzyme is directly convertible to
product. In fact, the pH dependence of the kcat/Km values observed in
papain catalyzed substrate hydrolysis has been found to be more
complex than originally thought (Brocklehurst, 1994). Several studies
have provided evidence that the development of the full nucleophilic
character of the ion pair depends not only on the ionization at pH 4
but on a second ionization at pH 4 and. in some cases, ionization with


115
TABLE 5.1
Isolation of a covalent glutamyl-enzyme intermediate by gel
filtration.
Enzyme
d p in
wild-type (5 nmol)
5016 100
wild-type (4 nmol) + 1 mM DON
152 52
CIA (4 nmol)
65 2
The intermediate was formed by incubating the indicated
concentration of enzyme with 1 mM [14C] glutamine (S.A. 22,000
dpm/nmol) as described in the materials and methods. The reactions
were terminated with 22 |_il of 5% SDS in 0.1 M sodium acetate (pH
4.0) and the enzyme was separated from free glutamine using a
Sephadex G-50-80 spin column equilibrated in 1% SDS in 0.1 M
sodium acetate (pH 4.0) according to the method of Penefsky (1979).
In the second entry, wt AS-B (1.5nmol) was incubated with ImM
DON for 30 min prior to dilution into the reaction mixture containing
[14C] glutamine. The amount of radiolabeled enzyme in the column
effluent was then measured by liquid scintillation.


99
In contrast to the above observations for papain, evidence is
presented in this study for a rate limiting thioester breakdown for
AS-B catalyzed glutamine hydrolysis. The similarity in the kcat
values for three substrates with different leaving groups suggests
that a step following the release of the leaving group and thus after
the appearance of a common intermediate is rate limiting (Table 4.1).
Moreover, a burst in glutamate production with time indicates that
an intermediate containing a glutamate moiety is formed quickly
followed by a slow breakdown to product and free enzyme (Figure
4.15). The presence of the burst in the production of glutamate was
verified by the ability to eliminate it when the enzyme was forced
into an enzyme-product form (Figure 4.16). It is possible that
glutamate could bind the enzyme and form a thioester intermediate,
therefore the disappearance of the burst in the presence of
glutamate does not reveal which step after thioester formation is
rate limiting. On the other hand, under the acidic conditions of the
reaction termination (see materials and methods), it is likely that the
acylenzyme intermediate remained intact and precipitated in the
acidic solution. Therefore, the radioactive glutamate counted after
paper chromatography represented only the glutamate which was
not covalently attached to the enzyme. If this is true, then the burst
suggests that a step occurring after deacylation is rate limiting.
Further work to address this issue will be presented in subsequent
chapters.
While the experiments described above suggest a rate
determining step following the formation of a thioester intermediate,
the results from hydroxylamine partitioning studies were more


16
AMP from the tetrahedral intermediate. Interestingly, this
mechanism would not necessarily require N-protonation before C-N
bond cleavage in the glutamine dependent synthesis reaction and,
therefore, may explain the differences in the I5N (V/K) isotope
effects for the two glutamine dependent reactions.
Despite the progress that has been made in understanding the
functions of the individual domains of AS, the mechanism of nitrogen
transfer remains unequivocally defined. One difference between the
mechanism involving free ammonia and that of a direct attack is the
requirement, in the former, of a common mechanism for glutamine
hydrolysis in the glutaminase and glutamine dependent asparagine
synthesis reactions. Therefore, it is important to understand the
mechanistic details of both the glutamine dependent synthesis and
glutamine hydrolysis in order to discriminate between the two
mechanisms. The ability of AS-B to catalyze the hydrolysis of
glutamine in the absence of the other amino acids provides the
opportunity to develop a model of the glutaminase reaction from
which to identify the individual steps and intermediates in the
reaction as well as examine the functional roles played by the
conserved amino acids in the GAT domain. In efforts to begin
building such a model, this dissertation focuses on the chemical and
kinetic mechanisms of glutamine hydrolysis catalyzed by AS-B.
Differences and similarities in the mechanism of substrate hydrolysis
catalyzed by AS-B and papain will be examined through pH analysis
of AS-B catalyzed glutaminase activity and by isolation of a covalent
intermediate in the reaction pathway of glutamine hydrolysis
catalyzed by AS-B. A variety of kinetic techniques will be used to


BIOGRAPHICAL SKETCH
Holly Schnizer was born on April 11, 1969 to Dr. Flint and Dottye
Gray of Kingsport, Tennesse. In 1991, Holly received her Bachelor of
Science degree in chemistry from Centre College in Danville,
Kentucky. The following fall, Holly began her graduate studies at the
University of Florida in the Department of Biochemistry and
Molecular Biology under the supervision of Dr. Sheldon Schuster. In
the spring of 1996, Holly married Dr. Richard Schnizer whom she met
while working in the laboratory. She expects to receive her Ph.D. in
Biochemistry and Molecular Biology in the fall of 1997.
165


CHAPTER 3
EFFECT OF TEMPERATURE ON GLUTAMINE HYDROLYSIS CATALYZED
BY ASPARAGINE SYNTHETASE B
Introduction
The following chapters present experiments which are aimed at
the determination of rate constants and the identification of
intermediates in order to develop a descriptive model of the
glutaminase reaction. Much of the investigation requires a very slow
reaction rate and, therefore, the reactions were examined at 5C.
Variation in temperature can cause alterations in the mechanism of a
reaction including differences in rate determining steps. In order to
validate the comparison of the results from experiments at 5C with
those performed at the temperature at which the reaction is
normally investigated (37C), the steady state kinetics of glutamine
hydrolysis will be examined over a range of temperatures and
Arrhenius plots will be used to detect any changes in rate limiting
step with temperature for this reaction. The temperature
dependence of LGH hydrolysis catalyzed by wild-type AS-B and that
of glutamine hydrolysis catalyzed by the AS-B mutant R30A will be
examined in a similar manner and the activation energies of the
three reactions will be compared. Similarity in the pH profiles for
glutamine and LGH hydrolysis presented in the previous chapter
suggests that these reactions occur by a common mechanism.
Further evidence for a common mechanism could be obtained
45


REFERENCES
Amuro, N, Paluh, JL, & Zalkin, H (1985) Replacement by site-directed
mutagenesis indicates a role for histidine 170 in the glutamine amide
transfer function of anthranilate synthase. J. Biol. Chem. 260:
14844-14849
Arad, D, Langridge, R & Kollman, PA (1990) A simulation of the
sulfur attack in the catalytic pathway of papain using molecular
mechanics and semiempirical quantum mechanics. J. Am. Chem. Soc.
112: 491-502
Badet, B, Vermoote, P, Haumont, P-Y, Lederer, F, Le Goffic, F (1987)
Glucosamine synthetase from Escherichia coli: purification,
properties, and glutamine-utilizing site location. Biochemistry 26:
1940-1948
Badet-Denisot, MA, Badet, B (1992) Chemical modification of
glucosamine-6-phosphate synthase by diethyl pyrocarbonate.
Evidence of a histidine requirement for enzymatic activity. Arch.
Biochem. Biophys. 292: 475-478
Badet-Denisot, MA, Badet, B (1993) Mechanistic investigations on
glucosamine-6-phosphate synthase, Bull. Soc. Chim. Fr. 130: 249-
255
Barshop, BA, Wrenn, RF, Frieden, C (1983) Analysis of numerical
methods for computer simulation of kinetic processes; development
of KINSIM, a flexible portable system. Anal. Biochem. 130: 134-145
Bernt, E & Bergmeyer, HU (1974) in Methods of Enzymatic Analysis
(Bergmeyer, H.U., ed) pp. 1704-1708, Academic Press, New York
152


42
pair in papain is reflected in the width of the pH-profile for kcat/Km
(Menard et al., 1991). Therefore, it is possible that AS-B contains an
ion pair like that of papain but lacks some of the factors which play a
role in ion pair stabilization. On the other hand, the pH profile of
kcat/Km may relate to ionizations in free substrate. In this study, the
pKa values of the a-amino groups of the substrates are 9.13
therefore, the possibility that the alkaline pKa values observed in the
kca[/Km profiles arise from the ionization of glutamine and LGH cannot
be excluded. In fact, the basic limb of the glutamine kcat/Km profile is
completely eliminated when Km values are corrected for protonated
glutamine. However, the same treatment has little effect on the
kcat/Km profile for LGH.
While the above examination of the kinetic behavior of wild-
type AS-B as a function of pH indicates the potentiality of an ion pair
similar to that of papain, past efforts have failed to reveal a histidine
which functions as part of that pair. In this study, substituting an
asparagine for histidine at position 47 resulted in a reduction of both
kcat and kcat/Km of approximately 10-fold for the glutaminase activity.
While these changes are significant, a much more dramatic effect was
expected if this residue were playing such an essential role in
catalysis as has been postulated. For example, replacement of
histidine-353 with an asparagine in the class I enzyme, CPS, resulted
in a drop in glutaminase activity from 1.1 pmol/h*mg for wt to
<0.006 pmol/hmg for H353N (Miran et al, 1991). Subsequent
structural studies which showed that this residue was in position to
interact with the active site histidine added further support that His-
353 was involved in acid-base catalysis (Thoden et al., 1997).


ACKNOWLEDGMENTS
I wish to thank everyone who helped me accomplish the work
presented here. I would like to thank my supervisor, Dr. Sheldon
Schuster, for his eagerness to discuss new ideas, his enthusiasm in
my work, and for his friendship. My work benefited greatly from
the guidance and helpful discussions provided by my supervisory
committee: Dr. Tom OBrien, Dr. Tom Yang, Dr. Richard Moyer, Dr.
Harry Nick, and Dr. Don Allison. I would like to thank them
especially for their continuing support even when my research took
a change in direction. I am also indebted to Dr. Nigel Richards and
Dr. Jon Stewart for their additional guidance in weekly discussions. I
am especially grateful to my advisor on my first project on the
hepatitis viruses, Dr. Johnson Lau, who introduced me to the
fascinating world of viruses, and who had patience with me during
my first few years in the laboratory.
My thanks goes to my friends in Dr. Schusters Laboratory for
providing the daily support I needed to complete this journey: Dr.
Sue Boehlein, Ms. Ellen Walworth, and Mr. Gentry Wilkerson. I am
especially thankful for Sue's helpful comments and ideas which
helped guide my research.
Most of all I am grateful to my husband, Richard Schnizer, for
all his helpful scientific advice and most of all for his love, patience,
and support which helped me achieve this goal.
iii


131
Figure 6.5: Effect of glutamine on the steady state concentration of
thioester formed in the glutaminase reaction catalyzed by the AS-B
mutant R30A. A glutamyl-enzyme complex was formed by
incubating 1.8 nmol of R30A with various concentrations of L-
glutamine (SA 22,000 dpm/nmol) in a reaction containing 100 mM
Tris-HCl (pH 8.0) for 30 sec at 5C. A radiolabeled enzyme complex
was captured by quenching the reaction in 8% TCA over a
nitrocellulose filter as described in materials and methods and the
radioactivity associated with the filter was measured by scintillation
counting.


84
Time (min)
[DON]
a 4 jllM
a 6 fiM
8 jiM
* 10 fiM
12 [iM
14 jiM
Figure 4.5: DON inactivation of AS-B in the presence of 20 pM
glutamine. Wild type AS-B was incubated with the concentrations of
DON given in the figure. At time points ranging from 0-20 min, an
aliquot was removed from the mixture and tested for residual
activity in a reaction containing 12 mM glutamine. The glutaminase
activities of the DON modified enzymes were measured as described
in materials and methods.


CHAPTER 5
EVIDENCE OF A THIOESTER INTERMEDIATE FORMED DURING
GLUTAMINE HYDROLYSIS CATALYZED BY ASPARAGINE
SYNTHETASE B
Introduction
Numerous clinical and basic studies have correlated the
importance of asparagine biosynthesis with the ability of some
leukemia cells to escape the toxic effects of treatment with L-
asparaginase (Chakrabarti & Schuster, 1997). As a result of this
relationship, a great deal of effort has been directed towards
understanding the chemical and kinetic mechanism of asparagine
synthetase, the enzyme mediating asparagine formation (Richards &
Schuster, 1997). It has been the goal of several research groups to
obtain specific, and potent, inhibitors of AS that might prevent
asparagine synthesis in tumor cells, and therefore act as either
adjuncts to L-asparaginase therapy or replacements for this drug
(Cooney et al., 1976; Cooney et al., 1980). Such efforts have been
hampered, however, by an incomplete understanding of the chemical
mechanism by which nitrogen is removed from the side chain amide
of glutamine and transferred to B-aspartyl-AMP to form asparagine
(Richards & Schuster, 1997).
Escherichia coli asparagine synthetase B (AS-B) (Boehlein et al.,
1994a; Scofield et ah, 1990) is a member of the class II (formerly
purF) amidotransferase superfamily (Smith. 1995; Zalkin, 1993), that
102


140
attack on the thioester (Storer & Menard, 1994). If there is a
histidine playing an equivalent role in AS-B catalysis, then it would
be expected that its removal would not only dramatically affect the
rate of the glutaminase reaction but would also change the pH
profile. However, as shown in chapter 2, replacing histidine-47 with
an asparagine only causes a four-fold decrease in the kcat for
glutamine hydrolysis. Moreover, the pH profile of the glutaminase
reaction catalyzed by this mutant is similar to that of the wild-type.
These results suggest that a role for His-47 in acid-base catalysis is
unlikely. On the other hand, the 10-fold drop in activity should not
be ignored, especially in light of the hypothesis presented here that a
conformation change is rate limiting. One further test for the
participation of His-47 in acid-base capacity would be to examine the
individual steps catalyzed by H47N. The work presented in this
dissertation has provided a method to examine the deacylation step
in the glutaminase reaction. Therefore, a comparison of the pH-
dependence of the deacylation rate catalyzed by wild-type AS-B and
its H47N mutant might provide the information necessary to reveal
whether His-47 plays a role in activating a water molecule for a
nucleophilic attack on the thioester.
While there may be differences in acid-base catalysis between
AS-B and papain, the overall pathway of hydrolysis may be similar.
This means that there must be an alternative for the deprotonation
of Cys-1 and the protonation of the leaving group during hydrolysis
catalyzed by AS-B. In fact, such an alternative has been offered in
regards to the class II amidotransferases (Kim et al., 1996).
Structural studies have placed the class II amidotransferases in a


87
1/[D0N] |iM1
Figure 4.8: Replot of DON inactivation of AS-B catalyzed LGH
hydrolysis. The K¡ for DON, determined from the slope of this plot and
equation (1), was 28 pM.


154
Brocklehurst, K, Kowlessur, D, Patel, G, Templeton, W, Quigley, K,
Thomas, EW, Wharton, CW, Willenbrock, F, Szawelski, RJ (1988a)
Consequences of molecular recognition in the S,-S2 intersubsite
region of papain for catalytic-site chemistry. Biochem J. 250: 761-
772
Brocklehurst, K, Brocklehurst, SM, Kowlessur, D, ODriscoll, D, Patel, G,
Salih, E, Templeton, W, Thomas, E, Topham, CM, Willenbrock, F
(1988b) Supracrystallographic resolution of interactions contributing
to enzyme catalysis by use of natural structural variants and
reactivity-probe kinetics. Biochem. J. 256: 543-558
Brocklehurst, K, Willenbrock, F, & Salih, E (1987) in Hydrolytic
Enzymes (Neuberger, A, Brocklehurst, K, eds) Elsevier, Amsterdam
pp 39-158
Broome, J.D. (1961) Evidence that the L-asparaginase activity of
guinea pig serum is responsible for its antilymphoma effects. Nature
(London) 191: 1114-1115
Broome, J.D. (1963a) Evidence that the L-asparaginase activity of
guinea pig serum is responsible for its antilymphoma effects. I.
Properties of the L-asparaginase of guinea pig serum in relation to
those of the antilymphoma substance. J. Exp. Med. 118: 99-120
Broome, J.D. (1963b) Evidence that the L-asparaginase activity of
guinea pig serum is responsible for its antilymphoma effects. II.
Lymphoma 6C3HED cells cultured in a medium devoid of L-
asparaginase lose their susceptibility to the effects of guinea pig
serum in vivo. J. Exp. Med. 118: 121-147
Burchall, JJ, Reichelt, EC, Wolin, MJ (1964) Purification and properties
of the asparagine synthetase of Streptococcus bovis. J. Biol. Chem.
239: 1794-1798
Capizzi, RL, Bertino, JR, Skeel, RT, Creasey, WA, Zanes, R, Olayon, C,
Petterson, RG, Handschumacher, RE (1971) L-asparaginase: clinical,
biochemical, pharmacological, and immunological studies. Ann.
Intern. Med. 74: 893-900


143
examination of the enzyme structure in the presence of 15N and 14C
double labeled glutamine using solid state nmr.
Computer Simulations of glutamate formation as a function of time:
The evidence for the existence of a thioester intermediate
given in chapters 5 and 6 provides a starting point in building a
model for the AS-B glutaminase reaction. In this dissertation the
model has been built from the simplest case which involves the
formation of a covalent intermediate (shown in scheme (1)), and
additions have been made according to the minimal requirements
necessary to make the best fit to the actual data from the
examination of glutamate formation as a function of time.
(1)
k-2 + p k-3
In this model, k, and k, represent the steps leading to and from the
formation of the Michaelis complex; k2 and k 2 the steps involved in
the acylation process; and k3 and k_3 the steps involved in
deacylation. The experiments in chapters 4 and 6 yielded values for
several of the rate constants in this reaction which now can be used
in a simulation of the glutaminase reaction using the computer
program KINSIM (Barshop et al. 1983). The values for the rate
constants to be used in the simulation of scheme (1) are given in
Table 7.1. The dissociation constant (Ks) for glutamine can be used to
obtain the value of k., using the relationship, Ks= k ,/k, and assuming
the value of k, approaches diffusion. The direct measurement of


158
Knowles, JR (1976) The intrinsic pKa-values of functional groups in
enzymes: improper deductions from the pH-dependence of steady-
state parameters. CRC Crit. Rev. Biochem. 4: 165-173
Larsen, MC & Schuster (1992) The topology of the glutamine and ATP
binding sites of human asparagine synthetase. Arch. Biochem.
Biophys. 299: 15-22
Lobel, L, OBrian, RT, McIntosh, S, Aspnes, GT, Capizzi, RL (1979)
Methotrexate and asparaginase combination chemotherapy in
refractory ALL of childhood. Cancer 43: 1089-1094
Lowe, J, Stock, D, Jap, B, Zwickl, P, Baumeister, W, Huber, R (1995)
Crystal structure of the 20S proteasome from the Archaeon T.
acidophilum at 3.4 resolution. Science 268: 533-539
Luehr, CA & Schuster, SM (1985) Purification and properties of beef
pancreatic asparagine synthetase. Arch. Biochem. Biophys. 237:
335-346
Lusty, CJ (1992) Detection of an enzyme bound g-glutamyl acyl ester
of carbamyl phosphate synthetase of Escherichia coli. FEBS LETT
314: 135-138
MacPhee, K, Nelson, R, & Schuster, SM (1983) Neurospora crassa
mutants deficient in asparagine synthetase. J. Bacteriol. 156: 475-
478
Mantsala, P & Zalkin, H (1984) Glutamine Amidotransferase Function.
J. Biol. Chem. 259: 14230-14236
Markin, RS and Schuster, SM (1979) Multiple forms of rat liver L-
asparagine synthetase. Biochem. Biophys. Res. Commun. 88: 583-
588
Markin, RS, Luehr, CA, & Schuster, SM (1981) The kinetic mechanism
of beef pancreatic L-asparagine synthetase. Biochemistry 20: 7226-
7232


50
the amino acids was visualized in a non-radioactive control by
spraying the paper with 0.2% ninhydrin in 95% ethanol. Glutamine
was found to migrate with the solvent front while glutamate
remained at the origin. The amount of 14C-labeled glutamate formed
in the reaction was determined by subtracting the percentage of total
radioactive counts located at the origin in the control reactions from
the percentage in each sample.
In a similar manner, the inhibition of glutamate also was
examined at 5C except that the reaction time was 40 min.
Data Analysis
The values of Km and kCat were determined from least-squares
fit to Vo vs [S] plots using the Prism software package supplied by
Graphpad, Inc., (San Diego, CA) according to equation 1:
v =
Vmax[S]
Km + [S]
(1)
Where v is the initial velocity, [S] is the concentration
of substrate, Vmax is the maximal velocity, and Km is the Michaelis
constant.
As shown in the equation below, the value for -Ea/RT can be
derived from the slope of a plot of the log kcat versus 1/T (K) where
A is a constant which is related to the collision frequency and steric
factors.
k = Ae(-Ea/RT) (2)
Values for the energy, enthalpy, and entropy of activation then were
calculated using the following equations:
AH* = Ea -RT
AG* = RT ln(kcT/h) RTln(k)
(3)
(4)


CHAPTER 7
DISCUSSION AND FUTURE DIRECTIONS
Understanding the mechanism of nitrogen transfer catalyzed
by asparagine synthetase will require the knowledge of the integral
parts of the process such as glutamine utilization. The past few years
of research on AS-B have seen substantial progress towards this goal
especially with respect to the catalytic roles played by the conserved
residues in the glutamine active site (See Boehlein et al., 1994a,b;
Boehlein et al., 1996; Boehlein et al., 1997a,b; Stoker et al., 1996) .
From the knowledge gained in the investigations of AS-B and other
class II amidotransferases, two schemes have developed which are
used to describe glutamine utilization and nitrogen transfer. First, as
described by Mei and Zalkin (1989) (see Figure 1.1), nitrogen
transfer occurs via a free ammonia released from glutamine by a
mechanism similar to thiol protease catalyzed amide hydrolysis.
Alternatively, the nucleophilic attack of the N-terminal cysteine
residue on the amide carbon of glutamine activates the side chain
nitrogen for a direct attack on the acceptor molecule (Stoker et al.,
1996) (see Figure 1.2). One key difference between these two
schemes may lie in the requirement of a common mechanism for
glutamine dependent asparagine synthesis and glutaminase reactions
in the scheme involving free ammonia and the lack of a common
mechanism for the two reactions in the scheme involving the direct
138


104
been characterized for any member of the Class II amidotransferase
family. We now report experimental conditions for the isolation of a
y-glutamyl AS-B adduct, together with evidence supporting the
hypothesis that this covalent derivative is the putative thioester
intermediate formed during AS-B catalyzed glutamine hydrolysis.
The characterization of this intermediate is likely to allow further
insight into the details of the kinetic and chemical mechanisms of
nitrogen transfer in AS-B, and other Class II amidotransferases.
Materials and Methods
Enzymes and reagents.
Recombinant wild-type AS-B and the CIA AS-B mutant, in
which Cys-1 is substituted by alanine, were constructed, expressed
and purified using published procedures (Boehlein et al., 1994a).
Protein concentration was determined using an assay kit supplied by
BioRad and immunoglobulin G as a standard. Recent examination of
the protein concentration by amino acid analysis of AS-B, performed
by the Protein Core Facility of ICBR at the University of Florida, gave
a value for protein concentration which was 2.7-fold lower than that
obtained with the previous method. Therefore, the enzyme
concentration has been corrected by a factor of 2.7. [U-14C]-L-
Glutamine (277 mCi/mmol) was purchased from Amersham
(Arlington Heights, IL). 6-Diazo-5-oxo-L-norleucine (DON) and
Sephadex G-50 Fine were obtained from Sigma Chemical Company.
Isolation of a Covalent Adduct between AS-B and P4Cl-glutamine
Wild-type AS-B (0.74 nmol) or the CIA AS-B mutant (0.74
nmol) was incubated with 1 mM [U-14C]-F-glutamine (SA 22,000


LIST OF TABLES
Table page
2.1 Oligonucleotides used in construction of site-
directed mutants of AS-B 2 5
2.2 Kinetic constants of the AS-B-catalyzed hydrolysis
of L-glutamine at 37C 3 1
2.3 Kinetic constants of the AS-B catalyzed hydrolysis
of L-glutamic-y-monohydroxamate at 37C 3 4
2.4 Kinetic constants for glutamine hydrolysis
catalzyed by the AS-B mutant H47N at 37C 3 7
2.5 Values of pKa for the kcat/Km-pH profiles for
glutamine and LGH hydrolysis catalyzed by wild-
type AS-B and for glutamine hydrolysis catalyzed
by the AS-B mutant H47N 3 8
2.6 Kinetic parameters for wild-type AS-B and the
H29A, H80A, H47N, and R30A mutants at three
values of pH 3 9
3.1 Thermodynamic properties of glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and the
R30A mutant 5 6
3.2 Kinetic constants for the synthesis of LGH
catalyzed by wild-type AS-B 5 9
vi


30
3000
2500-
o 2000-1

i 1500-
j? 1000-
500-
Figure 2.3 The pH dependence of kcat/Km for wild-type AS-B
catalyzed glutamine hydrolysis. The kinetic constants were
determined as described in the materials and methods. The solid
line () represents the actual data. The dashed line (---) represents
the data corrected for the protonation of glutamine.


characterized by three catalytically important residues thought to
function as a cys-his-glu catalytic triad analogous to the thiol
proteases. This class of enzymes (see Zalkin, 1993 for review)
includes GMP synthetase (Tiedeman et al, 1985), anthranilate
synthase, (Nichols et al., 1980) carbamoyl-phosphate synthetase
(Piette, 1984), imidazole glycerol-phosphate synthase (Carlomagno,
1988), aminodeoxychorismate synthase, CTP synthetase (Weng,
1986), and FGAM synthetase (Schendel et al., 1989). Asparagine
synthetase as well as glutamine 5-phosphoribosyl-l-pyrophosphate
amidotransferase (GPA) (Tso et al., 1982), glutamine fructose-6-
phosphate amidotransferase (GFAT) (Badet-Denisot et al., 1993), and
glutamate synthase (Vanoni et al., 1991) comprise the class II
amidotransferases (formerly F-type for the purF gene product,
glutamine 5-phosphoribosylpyrophosphate) which are characterized
by a conserved N-terminal cysteine required for glutamine
dependent activity.
Proposals for the Mechanism of Nitrogen Transfer
For the class II enzymes, the formation of a covalent y-
glutamyl-enzyme intermediate involving the N-terminal cysteine
residue was first suggested in studies of GPA where it was shown
that attachment of glutamine affinity analogs to cys-1 (Vollmer et al.,
1983) or replacement with phenylalanine (Mantsala & Zalkin, 1984)
caused selective elimination of the glutamine-dependent activity.
Later studies using site directed mutagenesis of conserved residues
in the GAT domain of GPA led to the proposal that cys-1 participated
as part of a cys-his catalytic dyad in a mechanism analogous to the


Boehlein, SK, Richards, NGI, and Schuster, SM (1994a) Glutamine-
dependent Nitrogen Transfer in Escherichia coli Asparagine
Synthetase B. J. Biol. Chem. 269: 7450-7457
153
Boehlein, SK, Richards, NGJ, Walworth, ES, Schuster, SM (1994b)
Arginine 30 and Asparagine 74 Have Functional Roles in the
Glutamine Dependent Activities of Escherichia coli Asparagine
Synthetase B. J. Biol. Chem. 269: 26789-26795
Boehlein, SK, Schuster, SM, Richards, NGJ (1996) Glutamic acid-y-
monohydroxamate and hydroxylamine are alternate substrates for
Escherichia coli asparagine synthetase B. Biochemistry 35: 3031-
3037
Boehlein, SK, Walworth, ES, Richards, NGJ, Schuster, SM (1997a)
Mutagenesis and chemical rescue indicate residues involved in (}-
aspartyl-AMP formation by Escherichia coli asparagine synthetase B.
J. Biol. Chem. 272: 12384-12392
Boehlein, SK, Rosa-Rodriquez, JG, Schuster, SM, Richards, NGJ (1997b)
Catalytic activity of the N-terminal domain of Escherichia coli
asparagine synthetase B can be reengineered by a single mutation. J.
Am. Chem. Soc. 119: In press
Bork, P & Koonin, EV (1994) A P-loop-like motif in a widespread ATP
pyrophosphatase domain: implications for the evolution of sequence
motifs and enzyme activity. Prot. Struct. Funct. Genet. 20: 347-355
Bradford, MM (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
Brannigan, JA, Dodson, G, Duggleby, HJ, Moody, PCE, Smith, JL,
Tomchick, DR, Murzin, AG (1995) A protein framework with an N-
terminal nucleophile is capable of self-activation. Nature 378: 416-
419
Brocklehurst, K (1994) A sound basis for pH-dependent kinetic
studies on enzymes. Prot. Eng. 7: 291-299


78
TABLE 4.1:
Kinetic constants for wt AS-B-catalyzed hydrolysis of L-glutamine, L-
glutamic acid-y-monohydroxamate, and L-glutamic acid-y-hydrazide.
Substrate
k^. (sec'1)
K,an;iM)
KJK (M-1 sec1)
L-glutamine
1.9 0.03
1.0 0.06
1900
L-glutamic-y-monohydroxamate
2.5 0.02
0.65 0.02
3846
L-glutamic-y-hydrazide
2.0 0.05
2.3 0.20
869


98
in a time course experiment in which the enzyme was incubated with
L-glutamate prior to its addition to the reaction. The observation
that D-giutamate has no effect on the burst suggests that salt effects
are not a cause of the disappearance of burst in the presence of L-
glutamate.
Discussion
This report describes the characterization of the AS-B catalyzed
glutaminase reaction by a comparison of its kinetic behavior to that
of papain and CAD. The mechanism previously proposed for
glutamine hydrolysis catalyzed by the class II amidotransferases
(Mei & Zalkin, 1989) (figure 4.17), in analogy to that of papain,
involves a nucleophilic attack of the N-terminal cysteine on the side
chain of glutamine to form a tetrahedral intermediate (1).
Subsequent protonation of the leaving group allows the release of
ammonia and the formation of a thioester intermediate (2). An
attack of a water molecule on the thioester then causes the formation
of a second tetrahedral intermediate (3) and finally the production
of free enzyme and glutamate (4). Using the minimal model in
equation 1, it has been found that for papain the formation of the
thioester (k2) is rate limiting when amide substrates are used. This
observation is probably due to the greater leaving group ability of
the thiol group relative to the amine or the hydroxyl group of the
first and second tetrahedral intermediate, respectively (OLeary et
al., 1974; Wells et ak, 1963). This situation causes the reaction to
favor return from the first tetrahedral intermediate to free enzyme
and substrate whereas it favors forward progression from the second
tetrahedral intermediate to free enzyme and product.


Table 2.2
Kinetic constants of the AS-B-catalyzed hydrolysis
of L-Glutamine at 37C
pH
^cat
K
kcat/Km
(sec
(mM)
(M'hsec-1)
6.00
1.51

0.05
5.78

0.60
261
6.25
1.67

0.10
3.44

0.33
485
6.50
1.89

0.03
2.4

0.1
787
7.00
1.94

0.03
1.01

0.06
1921
7.25
2.05
+
0.05
1.22

0.08
1680
7.50
1.92

0.03
1.01

0.06
1901
7.75
2.13

0.03
1.32

0.07
1614
8.00
2.21

0.03
1.15

0.05
1922
8.50
2.46

0.05
1.50

0.06
1640
8.75
2.48

0.05
1.92

0.09
1292
9.00
2.56

0.05
2.33

0.12
1099


75
In an attempt to show that AS-B catalyzed hydrolysis occurs
via a thioester intermediate and to determine which step in
hydrolysis is rate limiting, the effect of various concentrations of
hydroxylamine on the glutaminase reaction was examined. As
shown in Figure 4.1, even very low concentrations of hydroxylamine
were sufficient to partition the reaction towards the production of
glutamate hydroxamate. Unexpectedly however, an increase in
hydroxylamine concentration in the AS-B catalyzed glutaminase
reaction caused a decrease in total product formed (figure 4.1).
One potential cause for this decrease in total product formed in
the presence of hydroxylamine is a slow release of L-glutamate
hydroxamate due to its tight binding to the enzyme. To test this
idea, the partitioning experiment was repeated using an AS-B mutant
(R30A) which exhibits poor substrate binding. A role in substrate
binding has been implied for Arginine-30 in AS-B (Boehlein et al.,
1994b) and the cognate arginine in GFAT (Isupov et al., 1996) in
studies using site-directed mutagenesis and x-ray crystallography,
respectively. Use of R30A in the partitioning experiments resulted in
an increase in total product with increasing concentrations of
hydroxylamine (figure 4.2).
Comparison of AS-B catalyzed hydrolysis of three substrates
The kinetic constants kcat, Km, and kcat/Km for the AS-B catalyzed
hydrolysis of L-glutamine, L-glutamic acid-y-monohydroxamate, and
L-glutamic acid-y-hydrazide are given in Table 4.1. As shown in the
table, the kcat values for the three substrates were similar while each
displayed different kcat/Km values.


55
1/T (K)
Figure 3.3: Temperature dependence of glutamine hydrolysis
catalyzed by the AS-B mutant R30A. The initial rates at each
temperature were determined at pH 8.0 in 100 mM Bis-Tris and
Tris-HCl as described in materials and methods.


59
Table 3.2: Kinetic constants for the synthesis of LGH catalyzed by
wild-type AS-B.
Substrate
kcat
(sec"1)
Kni (mM)
kcat/Kra (M-1 sec"1)
glutamate
0.36
0.02
76 5
4.7
hydroxylamine
0.29
0.01
95 7
2.9
The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and
Tris-HCl at 37C, by varying one substrate while keeping the second
substrate saturating. The conditions of the reaction and the
determination of the LGH formed in the reaction are described in the
materials and methods.


9
thiol proteases (figure l.l)(Mei and Zalkin, 1989). This model of
nitrogen transfer involves a two step process: hydrolysis of
glutamine to produce free ammonia, followed by its transfer to the
acceptor substrate. In analogy to the thiol protease mechanism, a
nucleophilic attack of the conserved cys-1 thiolate anion on the
amide of glutamine results in a tetrahedral intermediate which, upon
release of ammonia, is converted to an acylthioenzyme intermediate
1. The released ammonia is transferred to the activated aspartate
intermediate 2 in the synthetase domain for the synthesis of
asparagine while the acylenzyme intermediate undergoes hydrolysis
to glutamate and free enzyme via a second tetrahedral intermediate.
For AS, this model for nitrogen transfer is supported by the
observation that all members of its family can catalyze the synthesis
of asparagine using ammonia instead of glutamine, demonstrating
the presence of binding site for ammonia. Furthermore, a completely
conserved cysteine has been shown to be catalytically essential in
GPA (Mei & Zalkin, 1989) and AS-B (Boehlein et al 1994a) and a
residue (Asn-74) thought to play a role in forming an oxyanion hole,
much like that formed by Gin-19 in papain catalysis (Menard et al.
1995), has been shown for AS-B.
While there is some evidence supporting an ammonia mediated
nitrogen transfer, there are also many inconsistencies. First, was the
inability to show that the class II enzymes contain a catalytically
important histidine cognate to that of the thiol protease papain.
Papain requires an active site histidine which functions to stabilize
the cysteine thiolate anion and provide a proton to the leaving group
in the conversion of the first tetrahedral intermediate to the


LIST OF FIGURES
Figures page
1.1 Mechanism of nitrogen transfer via an ammonia
intermediate as proposed by Mei and Zalkin, 1989 1 0
1.2 Mechanism of nitrogen transfer via an imide
intermediate 3 as proposed by Richards and
Schuster, 1992 1 4
1.3 Mechanism of nitrogen transfer via direct transfer
of the nitrogen from glutamine to aspartyl-AMP
as proposed by Stoker et al., 1996 1 5
2.1 Sequence alignment of part of the GAT-domains
of known asparagine synthetases and a partial set
of class II amidotransferases showing highly
conserved residues 2 2
2.2 The pH dependence of kcat for wild-type AS-B
catalyzed-glutamine hydrolysis 2 9
2.3 The pH dependence of kcat/Km for wild-type AS-B
catalyzed glutamine hydrolysis 3 0
2.4 The pH dependence of keat for wild-type catalyzed
LGH hydrolysis 3 2
2.5 The pH dependence of kcat/Km for wild-type AS-B
catalyzed LGH hydrolysis 3 3
2.6 The pH dependence of kcal for glutamine
hydrolysis catalzyed by the AS-B mutant H47N 3 5
viii


29
O
CD
CD
3.0
2.5
2.0
1.5
1.0
0.5
0.0
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
pH
Figure 2.2 The pH dependence of kcat for wild-type AS-B catalyzed
glutamine hydrolysis. The kinetic constants were determined as
described in materials and methods.


97
Figure 4.17: Mechanism of glutamine hydrolysis catalyzed by the
class II glutamine amidotransferases as proposed by Mei and Zalkin
(1989).


22
LtHmAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
GnHmAS
CGIWAI,<
>HRGPD<
>FGFHRLAV<
RatAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
MurAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
HumAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
SoyAS
CGILAV<
>HRGPD<
>LAHQRLAI<
Lot AS
CGILAV<
>HRGPD<
>LAHQRLAI<
FavaAS
CGILAV<
>HRGPD<
>LAHQRLAI<
PeaNAS
CGILAV<
>HRGPD<
>LAHQRLAI<
AlflAS
CGILAV<
>HRGPD<
>LAHQRLAI<
PeaRAS
CGILAV<
>HRGPE<
>LAQQRLAI<
AspaAS
CGILAV<
>HRGPD<
>LSHQRLAI<
ArabAS
CGILAV<
>HRGPD<
>LAHQRLA.V<
BrssAS
CGILAV<
>HRGPD<
>LAHQRLAI<
RiceAS
CGILAV<
>HRGPD<
>LAHQRLAI<
MaizAS
CGILAV<
>HRGPD<
>LAHQRLAI<
ScerAS
CGIFAA<
>HRGPD<
>FVHERLAI<
CeleAS
CGVFSK
>HRQPD<
>LVHERLAI<
EcoAS
CSIFQV<
>HRQPD<
>LAHERLSI<
1
29
47
ScerGA
CGILGK
>HRGQD<
>FTQQRVS.<
HumGS
CGIFAY<
>YRGYD<
>HKQQDMDL<
ScerGS
CGIFGY<
>YRGYD<
>.TKQNPNR<
>NGEIYNHKAL
>NGEIYNHKAL
>NGEIYNHKAL
>NGEIYNHKAL
>NGEIYNHKKM
>NGEIYNHEEL
>NGEIFNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEDL
>NGEIYNHEDL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHIQL
>NGEIVNHGEL
>NQEIYNHQAL
74 80
>NGNLVNTAS L
>NGIITNYKDL
>NGIITNFREL
Figure 2.1: Sequence alignment of part of the GAT-domains of
known asparagine synthetases and a partial set of class II
amidotransferases showing highly conserved residues. Numbering is
according to the amino acid sequence of AS-B. Sequence alignments
taken from Richards & Schuster, 1997. LtHmAS = Cricetulus
longicaudatus AS; GnHmAS = Mesocricetus auratus AS; RatAS = Rattus
norvegicus AS; MurAS = Mus musculus AS; HumAS = Homo sapiens
AS; Soy AS = Glycine max AS; Lot AS = Lotus japonicus AS; Fava AS =
Vicia faba AS; PeaNAS = Pisum sativa AS (root); AspaAS = Asparagus
officinalis AS; ArabAS = Arabidopsis thaliana AS; BrssAS = Bras sica
olercea AS; RiceAS = Oryza sativa AS; MaizAS = Zea mays', Seer AS =
Saccharomyces cerevisiae AS; Cele AS = elegans AS; Eco AS =
Escherichia coli AS; ScerGA = Saccharomyces cerevisiae GPA; HumGS =
Homo sapiens GFAT; ScerGS = Saccharomyces cerevisiae GFAT.


160
Nichols, BP, Miozzari, GF, van Cleemput, M, Bennett, GN, Yanofsky, C
(1980) Nucleotide sequences of the trpG regions of Escherichia coli,
Shigella dysenteriae, Salmonella typhimurium, and Serratia
marcescens. J. Mol. Biol. 142: 503-517
Oinonen, C, Tikkanen, R, Rouvinen, J,Peltonen, L (1995) Three-
dimensional structure of human lysosomal aspartylglucosamidase.
Nature Struct. Biol. 2: 1105-1108
OLeary, MH, Urberg, M, Young, AP (1974) Nitrogen isotope effects on
the papain-catalyzed hydrolysis of N-benzoyl-L argininamide.
Biochemistry 13: 2077-2081
Patterson, MK, & Orr, G (1969) Regeneration, tumor, dietary, and L-
asparaginase effects on asparagine biosynthesis in rat liver. Cancer
Res. 29: 1179-1183
Peller, L, and Alberty, RA (1959) Multiple Intermediates in steady
state enzyme kinetics. I. The mechanism involving a single substrate
and product. J. Am. Chem. Soc. 81: 5907-5914
Penefsky, H.S. (1979) A centrifuged-column procedure for the
mechanism of ligand binding by beef heart F,. Methods Enzymol.
56: 527-530
Pfeiffer, NE, Mehlhaff, PM, Wylie, DE, Schuster, SM (1986) Monoclonal
antibodies specific for bovine pancreatic asparagine synthetase. J.
Biol. Chem. 261: 1914-1919
Pfeiffer, NE, Mehlhaff, PM, Wylie, DW, & Schuster, SM (1987)
Topographical separation of the catalytic sites of asparagine
synthetase explored with monoclonal antibodies. J. Biol. Chem. 262:
11565-11570
Piette, J, Nyunoya, H, Lusty, CJ, Cunin, R, Weyens, G, Crabeel, M,
Charlier, D, Glansdorff, N, Pierard (1984) DNA sequence of the carA
gene and the control region of carB: Tandem promoters, respectively
controlled by arginine and the pyrimidines, regulate the synthesis of
carbamoyl-phosphate synthetase in Escherichia coli K-12. Proc. Natl.
Acad. Sci. U.S.A. 81:4134-4138


163
Tesmer, JJG, Klem, TJ, Deras, ML, Davisson, VJ, Smith, JL (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
Thoden, JB, Holden, HM, Wesenberg, G, Raushel, FM, Rayment, I
(1997) Structure of carbamoyl phosphate synthetase: A journey of
96 from substrate to product. Biochemistry 36: 6305-6316
Tiedeman, AA, Smith, JM, Zalkin, H (1985) Nucleotide sequence of the
guaA gene encoding GMP synthetase of Escherichia coli K12. J. Biol.
Chem. 260: 8676-8679
Tso, J.Y., Zalkin, H., van Cleemput, M, Yanofsky, C, Smith, JM (1982)
Nucleotide sequence of Escherichia coli purF and deduced amino acid
sequence of glutamine phosphoribosylpyrophosphate
amidotransferase. J. Biol. Chem. 257: 3525-3531
Uren, JR, & Handschumacher, RE (1977) Enzyme therapy. Cancer 5:
457-487
Van Heeke, G, & Schuster, SM (1989a) Expression of human
asparagine synthetase in Escherichia coli. J. Biol. Chem. 264: 5503-
5509
Van Heeke, G, & Schuster, SM (1989b) The N-terminal cysteine of
human asparagine synthetase is essential for glutamine dependent
activity. J. Biol. Chem. 264: 19475-19477
Van Heeke, G & Schuster, SM (1990) Expression of human asparagine
synthetase in Saccharomyces cerevisiae. Protein Eng. 3: 739-744
Vanoni, MA, Edmondson, DE, Rescigno, M, Zanetti, G, Curti, B (1991)
Mechanistic studies on Azospirillum hrasil.ense glutamate synthetase.
Biochemistry 30: 11478-11484
Vauquelin, LN, Robiquet, PJ (1806) Ann. Chim. (Paris) 57: 88


2
limit its use. First, L-asparaginase is toxic to organs such as the liver,
pancreas, kidneys, blood clotting systems, and brain therefore,
therapy is often accompanied by a wide variety of deleterious side
effects (Crowther, 1971). These side effects are usually reversible
and dose dependent however, in some cases they preclude the use of
this drug. A second drawback to the use of L-asparaginase is that
some patients who achieve remission suffer relapse with tumors
resistant to the L-asparaginase therapy (Kiriyama et al., 1989; Lobel
et al., 1979; Terebello et al., 1986). And finally, L-asparaginase
treatment is confined to ALL because the enzyme may actually
enhance the growth of some resistant tumors (Capizzi et al., 1971;
Tallal et al., 1970) and, at present, there is no assay to predict
sensitivity in solid tumors.
The therapeutic action of L-asparaginase is related to its ability
to deplete the extracellular supply of L-asparagine (Broome, 1961;
Uren & Handschumacher, 1977). While normal cells are independent
of an external source of asparagine, sensitive tumor cells synthesize
asparagine very slowly and are dependent on extracellular supplies
of this amino acid to survive. Indeed, it has been shown that the
activity of the enzyme which synthesizes asparagine, asparagine
synthetase (AS), is depressed in tumor cells which are sensitive to
asparaginase (Hutson et al., 1997). In contrast, increased AS activity
is often observed in cell lines which exhibit a resistance to L-
asparaginase (Gantt & Arfin, 1981; Kiriyama et al., 1989; Patterson &
Orr, 1969) and a direct correlation between increased AS mRNA and
protein content and resistance to asparaginase has been
demonstrated in a human leukemia cell line (Hutson et al., 1997).


94
A step after the formation of the thioester intermediate is rate
determining in glutamine hydrolysis catalyzed bv AS-B
In order to determine which step is rate limiting in the
minimal mechanism, the production of glutamate was studied as a
function of time. A burst of product formation during the first
turnover of the enzyme is an indication that the slow step follows the
appearance of the thioester intermediate.
The time-course of glutamate production was examined at 5C,
a temperature at which the reaction was slow enough to mix by
hand. A graph of nmol of glutamate formed versus reaction time
(figure 4.15) shows that the rate limiting step occurs after the
formation of the thioester intermediate as extrapolation of the line to
time zero intercepts the y-axis at a point above zero. This point of
intersection should represent the concentration of the dominant
intermediate, possibly the thioester, and therefore a time course
using a higher concentration of enzyme should intercept the y-axis at
a higher value. In fact, a higher y-axis intercept is observed with
double the concentration of enzyme (Figure 4.15). The burst of
product formed in the first turnover of an enzymatic reaction occurs
because the steps leading to product formation are fast relative to
those which follow. In the minimal mechanism proposed for AS-B,
the first appearance of glutamate occurs upon the formation of the
thioester intermediate. Therefore, it should be possible to eliminate
the appearance of the burst by forcing the reaction to begin at an
enzyme form which would require passing through the rate
determining step prior to the first appearance of glutamate. As
shown in Figure 4.16, the burst in glutamate formation was abolished


10
2
Asp
HS102
AS029
HiS-102
Figure 1.1 Mechanism of nitrogen transfer via an ammonia
intermediate as proposed by Mei and Zalkin, 1989. (a) Mechanism
for the hydrolysis of L-glutamine to yield ammonia and an
acylenzyme intermediate 1 by analogy to the purF enzyme, GPA. (b)
Synthesis of L-asparagine by reaction of ammonia with activated
aspartyl-AMP 2. (c) Hydrolysis of the acylenzyme intermediate to
yield L-glutamate. This diagram was taken from Richards and
Schuster, 1992.


85
[DON]
a 4 (iM
a 6 ¡aM
A 8 llM
10 pM
12 llM
14 (iM
Figure 4.6: DON inactivation of AS-B in the presence of 50 pM
glutamine. Wild type AS-B was incubated with the concentrations of
DON given in the figure. At time points ranging from 0-20 min, an
aliquot was removed from the mixture and tested for residual
activity in a reaction containing 12 mM glutamine. The glutaminase
activities of the DON modified enzymes were measured as described
in materials and methods.


141
family of hydrolytic enzymes, the Ntn-hydrolases, which use the N-
terminal residue (Ser, Thr, or Cys) as the nucleophilic catalyst.
Members of this family, including the 20S proteasome (Brannigan et
al., 1995; Seemuller et al., 1996), aspartylglucosaminidase (Oinonen
et al., 1995), and penicillin acylase (Duggleby et al., 1995), are
thought to catalyze general acid-base reactions through their N-
terminal amino group. Whether or not the class II amidotransferases
use such a mechanism remains to be established since previous
studies, employing site-directed mutants of GPA and GFAT which
contain an additional residue N-terminal to the cysteine, have given
ambiguous results (Isupov et al., 1996; Kim et al, 1996). Another
alternative for a histidine could be specific acid-base catalysis.
Indeed, this mechanism would explain the unexpected differences
between the V/K 15N KIEs for the glutaminase reaction catalyzed by
AS-B (Stoker et al., 1996) and that for amide hydrolysis catalyzed by
papain (OLeary et al., 1974). For papain, theoretical studies have
suggested that the transfer of the proton from His-159 to the leaving
group amine is concerted with the nucleophilic attack of Cys-25
(Arad et al., 1990). While this conclusion remains to be confirmed,
the large V/K 15N KIEs implied that C-N bond cleavage was the slow
step in the acylation process. On the other hand, the absence of a
V/K 15N KIE in the glutaminase reaction catalyzed by AS-B suggested
that a step during acylation other than C-N bond cleavage was rate
limiting. This observation may reflect a slow protonation of the
leaving group regulated by a specific acid catalyst. This issue could
be addressed by a study of the pH dependence of the 15N V/K KIE. If
the rate of acylation is dependent on the protonation of the leaving


CHAPTER 1
INTRODUCTION
L-Asparagine and Treatment of Malignant Disease
L-asparagine was one of the first amino acids to be isolated
from nature (Vauquelin & Robiquet, 1806). The importance of this
non-essential amino acid in such metabolic events as the production
of aspartate and ammonia (Broome, 1963a), transamination (Meister,
1952), protein synthesis (Cooney & Handschumacher, 1970), and
glycosymtion (Spiro, 1969) has been established. Considerable
interest in L-asparagine metabolism developed after the 1953
discovery (Kidd, 1953a, 1953b) of the therapeutic effects of guinea
pig serum against lymphomas in mice; and the subsequent
identification of L-asparaginase, the enzyme which catalyzes
asparagine hydrolysis, as the inhibitory component (Broome, 1961,
1963a, 1963b). Today, L-asparaginase is used most frequently in the
treatment of acute lymphoblastic leukemia (ALL), the most common
malignancy in children. When L-asparaginase is used as the sole
chemotherapeutic agent, 40-68% of the patients achieve complete
remission (Capizzi, et al., 1971; Uren & Handschumacher, 1977). In
combination with other chemotherapeutic agents, L-asparaginase
treatment results in almost 95% remission in previously untreated
ALL (Sanz, et al., 1986). Despite the therapeutic potency of L-
asparaginase, this treatment has several major disadvantages which


35
0.6-
O
CD
CO
CO
o.
0.5-
0.4-
0.3-
2
0.1-j
0.0
o a
I 1 I 1 1 1 1
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0
pH
Figure 2.6 The pH dependence of kcat for glutamine hydrolysis
catalyzed by the AS-B mutant H47N. The kinetic constants were
determined as described in materials and methods.
9.5


79
Determination of the Dissociation constants for glutamine and LGH
using the covalent modifier. 6-diazo-5-oxonorIeucine (DON).
The diazo ketone analogue of glutamine, DON, has been shown
to inhibit the glutaminase activity of bovine AS in such a manner
that the effects could be eliminated only by the addition of glutamine
and not by desalting (Mehlhaff & Schuster, 1991). Furthermore, the
x-ray structure of the glutamine active site labeled by DON in
another class II amidotransferase, E. coli GPA (Kim et al., 1996), has
provided evidence that this analog alkylates the cys-1 residue in a
reaction which is comparable to the formation of the glutamyl-
enzyme intermediate. Using the method of Meloche (1967), the
affinity analogue, DON, has been used in the determination of the
glutamine dissociation constant for the glutaminase reaction
catalyzed by anthranilate synthetase (Goto et al., 1976), CAD
glutaminase (Chaparan & Evans, 1991) and the amidotransferase
PabA/PabB complex (Roux & Walsh, 1992). Likewise, the ability of
DON to inactivate wild-type and the R30A mutant of AS-B was
examined and the effect of glutamine or LGH on the inactivation was
used to determine the dissociation constant of glutamine and LGH.
Incubation of wild type AS-B with DON prior to the measurement of
the glutaminase activity caused a time-dependent loss of activity
over a period of 20 minutes. The semilogrithmic plot of the residual
activity versus time shows that DON inactivates AS-B with pseudo-
first order kinetics (figure 4.3). This plot also shows that inactivation
is irreversible since the residual activity of the enzyme after it had
been incubated with DON was determined using 20 mM glutamine,
providing a 1400-fold dilution. The inactivation half-times, x, were


82
Figure 4.4: Replot of DON inactivation of AS-B in the absence () and
presence of glutamine (A, 20 pM; #, 50 pM).


72
varying concentrations of DON at 5C. At various time points 20 pi
aliquots were removed from the incubation mixture and the residual
activity was measured in 200 pi buffer containing 12 mM glutamine.
Control experiments measuring the activity in the presence of diluted
inhibitor concentrations verified that DON did not interfere in the
assay of residual activity.
Glutamine protection of DON inhibition was examined as
described above except that wt AS-B was incubated with varying
concentrations of DON in the presence of either 20 pM or 50 pM
glutamine. Alternatively, LGH protection of DON inactivation of the
wild type enzyme and glutamine protection of R30A inactivation
were examined by incubating wild-type or R30A AS-B with a
constant concentration of DON (6 pM for wild type and 650 pM for
R30A) and varying concentrations of LGH or glutamine, respectively.
An analysis of glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time.
Glutamine hydrolysis catalyzed by wt AS-B was examined at
5C, a temperature at which the reactions were slow enough to mix
by hand. The reactions were initiated by adding 0.37 or 0.74 nmol
wt AS-B to a 50 pi solution containing 100 mM Bis-Tris and Tris-HCl
(pH 8) and [14C]-glutamine (2mM, SA = 9 nCi/pl). After the
appropriate amount of time, the reactions were terminated with 20
pi of 2N acetic acid. Glutamate was separated from glutamine on
Whatman DE-81 ion-exchange chromatography paper using
deionized distilled water as the mobile phase. After 6 hours, elution
was stopped, and the paper was dried and cut into two strips to
separate glutamine from glutamate. The radioactivity associated


65
overall rate of the reaction catalyzed by the wild-type enzyme. The
differences in the nature of the rate determining step in the reactions
catalyzed by the wild-type and mutant enzyme were expected to be
reflected in the activation parameters. The lack of any differences in
the parameters suggests that the two slow steps in the wild-type
catalyzed reaction have similar activation energies.
The patterns of product inhibition can provide information on
the order of product release. However, the studies of product
inhibition presented here were inconclusive due to the inability of
ammonia to inhibit the glutaminase reaction. On the other hand, the
examination of glutamate inhibition provided a needed inhibition
constant which will be used in the model described in chapter 7.
The inability of ammonia to inhibit the glutaminase reaction
and the lack of detectable glutamine synthesis using ammonia and
glutamate as substrates suggests that the acylation step is essentially
irreversible under the conditions of the examination. With the
ability to detect glutamine concentrations to 0.1 nmol (see materials
and methods), an upper limit on the turnover number of kcat< 5x1 O'4
sec'1 can be set for glutamine synthesis at 37C catalyzed by the
wild-type AS-B under the conditions studied. In contrast, the
reversibility of LGH hydrolysis catalyzed by wild-type AS-B was
easily demonstrated and has a kcat which is 12% of the forward
reaction. The difference in the ability to synthesize the reverse
reaction may be a reflection of the greater nucleophilicity of
hydroxylamine. The kinetic parameters, kcat and K^, obtained from
the analysis of LGH synthesis allowed the determination of the
equilibrium constant for the overall reaction.


Table 2.3:
Kinetic constants of the AS-B-catalyzed hydrolysis of L-
glutamic-Y-monohydroxamate at 37C
PH kcaL
(sec-1)
Km
(mM)
kCat/Km
(M-'sec'1)
6.00
3.46

0.05
6.51

0.29
531
6.25
3.16

0.05
2.39

0.15
1322
6.50
2.86

0.03
1.11

0.05
2577
6.75
2.59

0.03
0.68

0.02
3809
7.00
2.67

0.03
0.66

0.02
4045
7.25
2.65

0.03
0.58

0.03
4569
7.50
2.67

0.03
0.65
+
0.02
4108
7.75
2.65
+
0.03
0.63

0.03
4206
8.00
2.38

0.05
0.83

0.05
2867
8.25
2.54

0.03
1.11

0.05
2288
8.50
3.00

0.05
1.53

0.08
1961
8.75
3.10

0.05
2.26
+
0.11
1372
9.00
4.05

0.19
7.41

0.78
546


89
Figure 4.10: LGH protection against DON inactivation of AS-B
catalyzed LGH hydrolysis. The Ks determined from the slope of the
line and equation (3), was 70 pM.


Caplow,M & Jencks, WP (1964) The chymotrypsin-catalyzed
hydrolysis and synthesis of N-acetyl-L-tyrosine hydroxamic acid.
J. Biol. Chem. 239: 1640-1652
155
Carlomagno, MS, Chiariotti, L, Alifano, P, Nappo. AG, Bruni, CB (1988)
Structure and function of the Salmonella typhimurium and
Escherichia coli K-12 histidine operons. J. Mol. Biol. 203: 585-606
Cedar, H. & Schwarz, JH (1969a) The asparagine synthetase of
Escherichia coli. I. Biosynthetic role of the enzyme, purification, and
characterization of the reaction products. J. Biol. Chem. 244: 4122-
4127
Cedar, H. & Schwarz, JH (1969b) The asparagine synthetase of
Escherichia coli. II. Studies on mechanism. J. Biol. Chem. 244: 4122-
4127
Chakrabarti, R, & Schuster, SM (1997) L-Asparaginase: Perspectives
on the mechanism of action and resistance. Internat. J. Ped.
Hem./One. In Press
Chaparian, MG & Evans, DR (1991) The catalytic mechanism of the
amidotransferase domain of the Syrian hamster multifunctional
protein CAD. J. Biol. Chem. 266: 3387-3395
Cleland, WW (1963) The kinetics of enzyme catalyzed reactions with
two or more substrates or products. Biochim. Biophys. Acta 67: 104-
137
Cleland, WW (1982) An analysis of Haldane relationships. Methods
Enzymol. 87: 366-369
Cooney, DA & Handschumacher, RE (1970) L-asparaginase and L-
asparagine metabolism. Ann. Rev. Pharmacol. 10: 421-440
Cooney, DA, Driscoll, JS, Milman, HA, Jayaram, HN, Davis, RD (1976)
Inhibitors of L-asparagine synthetase in vitro. Cane. Treat. Rep. 60:
1493-1557


101
protease, papain, and the class I amidotransferase, CAD. While both
AS-B and CAD exhibit a rate determining step occurring after the
formation of the acyl enzyme, a feature unlike the slow acylation
step in amide hydrolysis catalyzed by papain, AS-B may have an
additional step which contributes to the rate of the reaction and is
unlike the sole rate determining deacylation which characterizes CAD
catalysis (Chaparan & Evans, 1991). Additional effort will be
required to understand the details of the mechanisms of these
reactions which cause the differences seen in the rate determining
steps. One step towards accomplishing this goal is a knowledge of
the rates of the individual steps in the reactions. In this regard, the
Ks for glutamine determined in this work not only reveals that the
glutaminase reaction occurs under rapid equilibrium conditions (k.
,>>k2) but also contributes a rate constant to the overall model of the
reaction which will be presented in Chapter 7. In addition, Chapters
5 and 6 represent the first attempt at the examination of an
individual step in the reaction.


CHAPTER 4
DETERMINATION OF THE RATE DETERMINING STEP IN ASPARAGINE
SYNTHETASE B-CATALYZED GLUTAMINE HYDROLYSIS
Introduction
Asparagine synthetase B (AS-B) is thought to be organized into
two functional domains: the glutamine amide transfer (GAT) domain,
which catalyzes the hydrolysis of glutamine to glutamate; and the
synthetase domain, which catalyzes the formation of a |3-aspartyl-
AMP intermediate and the synthesis of asparagine. While the overall
reaction involves three substrates, four products, and possibly twenty
rate constants; the ability of the enzyme to catalyze glutamine
hydrolysis in the absence of the other substrates provides the
opportunity to examine glutamine utilization under a simpler one
substrate, two product reaction. A detailed description of this
reaction would provide a useful tool in understanding the roles
played by conserved amino acid residues of the GAT domain in
glutamine utilization. Therefore, this chapter presents a first step
towards characterizing the glutaminase reaction catalyzed by AS-B.
Analysis of the amino acid sequence of the GAT domain suggests
that AS-B belongs to a family of enzymes known as the glutamine
amidotransferases (Zalkin, 1993). Like AS-B, the enzymes of this
family catalyze nitrogen transfer using glutamine as the primary
source of nitrogen. While all enzymes in the glutamine
amidotransferase family contain a cysteine which is essential for
67


51
AG* = AH* TAS* (5)
where AH* is the enthalpy of activation, R is the gas constant, kB is
boltzmans constant, h is Plancks constant, k is the forward rate
constant; AG* is the free energy of activation, and AS* is the entropy of
activation.
Results
Effect of temperature on glutamine and LGH hydrolysis catalyzed by
wild-tvpe AS-B and the AS-B mutant R30A
The steady-state kinetic parameters of glutamine and LGH
hydrolysis were determined at a range of temperatures from 5 C to
40 C. The Arrhenius plots (Figures 3.1 & 3.2) were observed to be
linear indicating either that there was no change in the rate
determining step over the temperature range studied or that the
activation energies for the limiting steps were very similar.
The temperature dependence of the catalytic rate constants of
the R30A-catalyzed hydrolysis of glutamine also displayed linearity
in the Arrhenius plot, as shown in Figure 3.3. The activation
parameters are given in Table 3.1.
LGH synthesis catalyzed bv wild-tvpe AS-B
AS-B was found to catalyze the formation of LGH from
glutamate and hydroxylamine. The reaction rate of 0.29 sec"1 (Table
3.2, Figures 3.4 & 3.5) was about 12% of that of the corresponding
forward rate of 2.4 sec'1 at pH 8 (see chapter 2, Table 2.3), and was
fast enough to obtain the kinetic parameters of LGH synthesis
catalyzed by wild-type AS-B (Table 3.2).
The steady-state kinetic parameters of reversible reactions can


17
determine the individual rate constants of the glutaminase reaction
and these will be used to build a model of glutamine hydrolysis
catalyzed by AS-B.


slope
63
Figure 3.9: Replot of the slope from the double-reciprocal plots of
glutamine hydrolysis at 37C versus glutamate concentration.


90
[DON] jiM
+ 350
450
a 500
f 600
O 650
. 750
Time (min)
Figure 4.11: DON inactivation of the R30A mutant of AS-B in the
absence of glutamine. R30A was incubated with the concentrations
of DON given in the figure. At time points ranging from 0-18 min, an
aliquot was removed from the mixture and tested for residual
activity in a reaction containing 12 mM glutamine. The glutaminase
activities of the DON modified enzyme were measured as described
in materials and methods.


5
active sites in separate domains was revealed through studies using
monoclonal antibodies (Pfeiffer et al., 1986, 1987) and chemical
modifiers (Larsen & Schuster, 1992) that could selectively inhibit the
glutaminase or asparagine synthesis activities of human or bovine
AS.
While the mechanism of nitrogen transfer between the two
domains is poorly understood, examination of conserved residues
within the individual domains has led to the identification of binding
sites and intermediates involved in the reaction. For the C-terminal
synthetase domain, it has been confirmed that during asparagine
synthesis, a (3-aspartyl-AMP intermediate is formed which activates
the aspartate carboxylate for attack by the amide nitrogen of
glutamine (Luehr & Schuster, 1985). Further studies showed a
binding site for aspartate located in a region highly conserved among
asparagine synthetases (Boehlein et al., 1997a). And finally, another
highly conserved region in the AS synthetase domain was found to
contain a P-loop-like motif that is present in other enzymes which
catalyze the hydrolysis of ATP to AMP and PP¡ (Bork & Koonin,
1994), and has been proposed to represent a pyrophosphate binding
element (Boehlein et al. manuscript in progress).
Experiments using site-directed mutagenesis have identified
several conserved amino acids in the N-terminal domain that are
crucial to glutamine utilization. Sequence analysis reveals that all AS
contain an N-terminal cysteine residue. Milman and Cooney first
proposed a critical cysteine residue in the glutamine active site when
they found AS from mouse pancreas was strongly susceptible to
oxidation (Milman & Cooney, 1979). More evidence for an important


CHEMICAL AND KINETIC CHARACTERIZATION OF GLUTAMINE
HYDROLYSIS CATALYZED BY ESCHERICHIA COLI
ASPARAGINE SYNTHETASE B
BY
HOLLY G. SCHNIZER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1997

This work is dedicated to
my parents, Flint and Dottye Gray,
and to my husband, Richard Schnizer,
with love.
11

ACKNOWLEDGMENTS
I wish to thank everyone who helped me accomplish the work
presented here. I would like to thank my supervisor, Dr. Sheldon
Schuster, for his eagerness to discuss new ideas, his enthusiasm in
my work, and for his friendship. My work benefited greatly from
the guidance and helpful discussions provided by my supervisory
committee: Dr. Tom OBrien, Dr. Tom Yang, Dr. Richard Moyer, Dr.
Harry Nick, and Dr. Don Allison. I would like to thank them
especially for their continuing support even when my research took
a change in direction. I am also indebted to Dr. Nigel Richards and
Dr. Jon Stewart for their additional guidance in weekly discussions. I
am especially grateful to my advisor on my first project on the
hepatitis viruses, Dr. Johnson Lau, who introduced me to the
fascinating world of viruses, and who had patience with me during
my first few years in the laboratory.
My thanks goes to my friends in Dr. Schusters Laboratory for
providing the daily support I needed to complete this journey: Dr.
Sue Boehlein, Ms. Ellen Walworth, and Mr. Gentry Wilkerson. I am
especially thankful for Sue's helpful comments and ideas which
helped guide my research.
Most of all I am grateful to my husband, Richard Schnizer, for
all his helpful scientific advice and most of all for his love, patience,
and support which helped me achieve this goal.
iii

TABLE OF CONTENTS
PAGE
ACKNOWLEDGMENTS u
LIST OF TABLES v
LIST OF FIGURES v
ABBREVIATIONS x
ABSTRACT x
CHAPTERS
1INTRODUCTION 1
2 pH DEPENDENCE OF THE GLUT AMIN ASE REACTION
CATALYZED BY ASPARAGINE SYNTHETASE B....
Introduction
Materials and Methods
Results
Discussion
3EFFECT OF TEMPERATURE ON GLUTAMINE HYDROLYSIS
CATALYZED BY ASPARAGINE SYNTHETASE B
Introduction
Materials and Methods
Results
Discussion
IV

4 DETERMINATION OF THE RATE DETERMINING STEP
IN ASPARAGINE SYNTHETASE B-CATALYZED
GLUTAMINE HYDROLYSIS 67
Introduction 67
Materials and Methods 70
Results 74
Discussion 98
5 EVIDENCE OF A THIOESTER INTERMEDIATE FORMED
DURING GLUTAMINE HYDROLYSIS CATALYZED BY
ASPARAGINE SYNTHETASE B 102
Introduction 102
Materials and Methods 104
Results 106
Discussion 109
6 CHARACTERIZATION OF A THIOESTER INTERMEDIATE I20
Introduction 120
Materials and Methods i22
Results 123
Discussion I34
7 DISCUSSION AND FUTURE DIRECTIONS I38
REFERENCES I52
BIOGRAPHICAL SKETCH
165

LIST OF TABLES
Table page
2.1 Oligonucleotides used in construction of site-
directed mutants of AS-B 2 5
2.2 Kinetic constants of the AS-B-catalyzed hydrolysis
of L-glutamine at 37C 3 1
2.3 Kinetic constants of the AS-B catalyzed hydrolysis
of L-glutamic-y-monohydroxamate at 37C 3 4
2.4 Kinetic constants for glutamine hydrolysis
catalzyed by the AS-B mutant H47N at 37C 3 7
2.5 Values of pKa for the kcat/Km-pH profiles for
glutamine and LGH hydrolysis catalyzed by wild-
type AS-B and for glutamine hydrolysis catalyzed
by the AS-B mutant H47N 3 8
2.6 Kinetic parameters for wild-type AS-B and the
H29A, H80A, H47N, and R30A mutants at three
values of pH 3 9
3.1 Thermodynamic properties of glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and the
R30A mutant 5 6
3.2 Kinetic constants for the synthesis of LGH
catalyzed by wild-type AS-B 5 9
vi

4.1 Kinetic constants for AS-B catalyzed hydrolysis of
L-glutamine, L-glutamic acid-y-monohydroxamate,
and L-glutamic acid-y-hydrazide 7 8
5.1 Isolation of a covalent glutamyl-enzyme
intermediate by gel filtration 115
5.2 Time dependence of y-glutamyl AS-B adduct
degradation in SDS solution 116
5.3 Trapping the thioester by filter binding 117
5.4 Base lability of the covalent adduct isolated by gel
filtration 1 1 9
6.1 Characterization of the glutamyl-enzyme
intermediate 1 2 9
7.1 Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by wild-type AS-B according to schemes
(1) and (2) 146
7.2 Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by AS-B according to equation (3) 149
7.3 Comparison of the actual and simulated values
obtained for the kcat, slope, intercept, and [TE]
from Figure 7.1 150
vii

LIST OF FIGURES
Figures page
1.1 Mechanism of nitrogen transfer via an ammonia
intermediate as proposed by Mei and Zalkin, 1989 1 0
1.2 Mechanism of nitrogen transfer via an imide
intermediate 3 as proposed by Richards and
Schuster, 1992 1 4
1.3 Mechanism of nitrogen transfer via direct transfer
of the nitrogen from glutamine to aspartyl-AMP
as proposed by Stoker et al., 1996 1 5
2.1 Sequence alignment of part of the GAT-domains
of known asparagine synthetases and a partial set
of class II amidotransferases showing highly
conserved residues 2 2
2.2 The pH dependence of kcat for wild-type AS-B
catalyzed-glutamine hydrolysis 2 9
2.3 The pH dependence of kcat/Km for wild-type AS-B
catalyzed glutamine hydrolysis 3 0
2.4 The pH dependence of keat for wild-type catalyzed
LGH hydrolysis 3 2
2.5 The pH dependence of kcat/Km for wild-type AS-B
catalyzed LGH hydrolysis 3 3
2.6 The pH dependence of kcal for glutamine
hydrolysis catalzyed by the AS-B mutant H47N 3 5
viii

2.7The pH dependence of kcat/Km for glutamine
hydrolysis catalzyed by the AS-B mutant H47N 3 6
3.1 Temperature dependence of glutamine hydrolysis
catalyzed by wild-type AS-B 5 3
3.2 Temperature dependence of LGH hydrolysis
catalyzed by wild-type AS-B 5 4
3.3 Temperature dependence of glutamine hydrolysis
catalyzed by the AS-B mutant R30A 5 5
3.4 Effect of hydroxylamine concentration on the
synthesis of LGH catalyzed by wild-type AS-B 5 7
3.5 Effect of glutamate concentration on the synthesis
of LGH catalyzed by wild-type AS-B 5 8
3.6 Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 5C 6 0
3.7 Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 37C 6 1
3.8 Replot of the slope from the double-reciprocal
plots of glutamine hydrolysis at 5C versus
glutamate concentration 6 2
3.9Replot of the slope from the double-reciprocal
plots of glutamine hydrolysis at 37C versus
glutamate concentration 6 3
4.1 Partitioning of the thioester intermediate to L-
glutamic-y- monohydroxamate using
hydroxylamine in a reaction catalyzed by wt AS-B 7 6
4.2 Partitioning of the thioester intermediate to
glutamate hydroxamate using hydroxylamine in
the glutaminase reaction catalyzed by AS-B
R30A 7 7
IX

4.3 DON inactivation of AS-B in the absence of
glutamine 8 1
4.4 Replot of DON inactivation of AS-B in the absence
() and presence of glutamine (A, 20 pM; ,50
pM) 8 2
4.5 DON inactivation of AS-B in the presence of 20 pM
glutamine 8 4
4.6 DON inactivation of AS-B in the presence of 50 pM
glutamine 8 5
4.7 DON inactivation of AS-B catalyzed LGH hydrolysis
in the absence of LGH 8 6
4.8 Replot of DON inactivation of AS-B catalyzed LGH
hydrolysis 8 7
4.9 The effect of various concentrations of LGH on
DON inactivation of AS-B catalyzed LGH hydrolysis 8 8
4.10 LGH protection against DON inactivation of AS-B
catalyzed LGH hydrolysis 8 9
4.11 DON inactivation of the R30A mutant of AS-B in
the absence of glutamine 9 0
4.12 Replot of DON inactivation of R30A in the absence
of glutamine 9 1
4.13 Glutamine protection from DON inhibition of the
R30A catalyzed glutaminase reaction 9 2
4.14 Replot of glutamine protection of DON inactivation
of R30A catalyzed glutaminase 9 3
4.15 Glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time 9 5
x

4.16 Effect of the presence of L-glutamate on the burst
in product formation in AS-B catalyzed glutamine
hydrolysis 9 6
4.17 Mechanism of glutamine hydrolysis catalyzed by
the class II glutamine amidotransferases as
proposed by Mei and Zalkin (1989) 9 7
5.1 Effect of glutamine on the steady-state
concentration of the thioester 1 1 8
6.1 Rate of deacylation of the thioester intermediate
formed during wild-type AS-B catalyzed
glutamine hydrolysis 12 6
6.2 The steady-state concentration of the thioester
formed in the presence of ATP 1 2 7
6.3 Effect of ATP on the rate of deacylation 128
6.4 Effect of ammonia on the rate of deacylation 130
6.5 Effect of glutamine on the steady-state
concentration of thioester formed in the
glutaminase reaction catalyzed by the AS-B
mutant, R30A 13 1
6.6 Effect of glutamine on the steady-state
concentration of the thioester formed during the
glutaminase reaction catalyzed by the AS-B
mutant, N74A 1 3 2
6.7 The rate of deacylation of the AS-B mutant, N74A,
during glutamine hydrolysis 1 3 3
7.1 Comparison of the simulated to the actual data
representing glutamate formation as a function of
time 1 4 5
xi

ABBREVIATIONS
ALL: Acute Lymphoblastic Leukemia
AS: asparagine synthetase
AS-A: ammonia dependent asparagine synthetase A from E.coli
AS-B: asparagine synthetase B from E. coli
AMP: adenosine monophosphate
ATP: adenosine 5'-triphosphate
CIA: mutant of AS-B with cysteine-1 replaced by alanine
DON: 6-diazo-5-oxonorleucine
E. coli: Escherichia coli
GAT: glutamine amide transfer
GFAT: glutamine fructose-6-phosphate amidotransferase
GPA: glutamine 5'-phosphoribosyl-l-pyrophosphate
amidotransferase
H47N: mutant of AS-B with histidine-47 replaced by asparagine
KIE: kinetic isotope effect
LGH: L-glutamic-y-monohydroxamate
N74A: AS-B mutant with asparagine-74 replaced by alanine
PP¡: pyrophosphate
R30A: AS-B mutant with arginine-30 replaced by alanine
wt: wild-type
xii

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CHEMICAL AND KINETIC CHARACTERIZATION OF GLUTAMINE
HYDROLYSIS CATALYZED BY ESCHERICHIA COLI ASPARAGINE
SYNTHETASE B
By
Holly Schnizer
December, 1997
Chairman: Dr. Sheldon Schuster
Major Department: Biochemistry and Molecular Biology
A detailed model of glutamine hydrolysis catalyzed by
Escherichia coli asparagine synthetase B, a class II amidotransferase,
was constructed using a variety of chemical and kinetic techniques.
Glutamine hydrolysis catalyzed by the class II amidotransferase
family has been proposed to follow a mechanism similar to that of
amide hydrolysis catalyzed by the thiol protease, papain. This study
presents a comparison of the catalytic behavior of these two
hydrolytic enzymes in order to determine if asparagine synthetase B-
catalyzed glutaminase reaction can be described by a similar
mechanism. First, this study revealed differences in acid-base
catalysis between these two hydrolytic enzymes. In a continuing
investigation of the importance of conserved histidines in the

glutamine utilizing domain of asparagine synthetase B, the similarity
of pH profiles of wild-type and an asparagine synthetase B mutant in
which a conserved histidine is replaced with an asparagine, suggests
that this histidine is not likely to be involved as part of the cys-his
dyad which is essential to papain-catalyzed hydrolysis. Second, this
study describes the development of a model for glutamine hydrolysis
catalyzed by wild-type asparagine synthetase B from a minimal
mechanism. The use of such a model was confirmed by the
demonstration that one main feature of papain catalysis, the
formation of a thioester, was conserved in asparagine synthetase B
catalyzed glutamine hydrolysis.
The values for the dissociation constants for glutamine and the
inhibition constants for glutamate were determined. Furthermore,
the isolation of a putative thioester intermediate provided the
opportunity to measure the rate of a single step in the reaction,
deacylation. These constants along with the information gained from
the kinetic constants of L-glutamic-y-monohydroxamate synthesis
catalyzed by asparagine synthetase B were used in a computer
simulation of the burst of glutamate production with time which
showed that an additional rate limiting conformational change must
be included to describe the observed kinetic behavior of wild-type
asparagine synthetase B.
xiv

CHAPTER 1
INTRODUCTION
L-Asparagine and Treatment of Malignant Disease
L-asparagine was one of the first amino acids to be isolated
from nature (Vauquelin & Robiquet, 1806). The importance of this
non-essential amino acid in such metabolic events as the production
of aspartate and ammonia (Broome, 1963a), transamination (Meister,
1952), protein synthesis (Cooney & Handschumacher, 1970), and
glycosymtion (Spiro, 1969) has been established. Considerable
interest in L-asparagine metabolism developed after the 1953
discovery (Kidd, 1953a, 1953b) of the therapeutic effects of guinea
pig serum against lymphomas in mice; and the subsequent
identification of L-asparaginase, the enzyme which catalyzes
asparagine hydrolysis, as the inhibitory component (Broome, 1961,
1963a, 1963b). Today, L-asparaginase is used most frequently in the
treatment of acute lymphoblastic leukemia (ALL), the most common
malignancy in children. When L-asparaginase is used as the sole
chemotherapeutic agent, 40-68% of the patients achieve complete
remission (Capizzi, et al., 1971; Uren & Handschumacher, 1977). In
combination with other chemotherapeutic agents, L-asparaginase
treatment results in almost 95% remission in previously untreated
ALL (Sanz, et al., 1986). Despite the therapeutic potency of L-
asparaginase, this treatment has several major disadvantages which

2
limit its use. First, L-asparaginase is toxic to organs such as the liver,
pancreas, kidneys, blood clotting systems, and brain therefore,
therapy is often accompanied by a wide variety of deleterious side
effects (Crowther, 1971). These side effects are usually reversible
and dose dependent however, in some cases they preclude the use of
this drug. A second drawback to the use of L-asparaginase is that
some patients who achieve remission suffer relapse with tumors
resistant to the L-asparaginase therapy (Kiriyama et al., 1989; Lobel
et al., 1979; Terebello et al., 1986). And finally, L-asparaginase
treatment is confined to ALL because the enzyme may actually
enhance the growth of some resistant tumors (Capizzi et al., 1971;
Tallal et al., 1970) and, at present, there is no assay to predict
sensitivity in solid tumors.
The therapeutic action of L-asparaginase is related to its ability
to deplete the extracellular supply of L-asparagine (Broome, 1961;
Uren & Handschumacher, 1977). While normal cells are independent
of an external source of asparagine, sensitive tumor cells synthesize
asparagine very slowly and are dependent on extracellular supplies
of this amino acid to survive. Indeed, it has been shown that the
activity of the enzyme which synthesizes asparagine, asparagine
synthetase (AS), is depressed in tumor cells which are sensitive to
asparaginase (Hutson et al., 1997). In contrast, increased AS activity
is often observed in cell lines which exhibit a resistance to L-
asparaginase (Gantt & Arfin, 1981; Kiriyama et al., 1989; Patterson &
Orr, 1969) and a direct correlation between increased AS mRNA and
protein content and resistance to asparaginase has been
demonstrated in a human leukemia cell line (Hutson et al., 1997).

3
In view of the above mentioned problems associated with
asparaginase therapy, alternative methods of treating ALL need to
be explored. Since the effectiveness of this therapy seems to reside
in reducing the circulating concentration of asparagine, inhibitors of
asparagine synthetase could provide this much needed alternative.
In efforts to achieve this goal, several hundred compounds have
been evaluated as AS inhibitors, however none have provided both
the strength and specificity required for clinical consideration
(Cooney et al, 1976). Therefore, a detailed chemical and kinetic
investigation of AS is essential in order to develop more effective
inhibitors of this enzyme.
Asparagine Synthetase
Asparagine synthetases catalyze the ATP dependent transfer of
nitrogen from either glutamine or ammonia to aspartic acid in the
formation of asparagine. The biochemical and mechanistic properties
of AS have been examined in a variety of sources including mammals
(Hongo et al., 1978; Markin et al., 1979; Markin et al., 1981; Milman
& Cooney, 1979; Van Heeke & Schuster, 1989b), fungi (Ramos &
Wiame, 1979, 1980), plants (Pike & Beevers, 1982; Rognes, 1970),
and bacteria (Burchall et al., 1964; Cedar & Schwartz, 1969a, 1969b;
MacPhee et al., 1983) as well as in some tumor cell lines (Gantt &
Arfin, 1981). In order to facilitate the recovery of sufficient amounts
of AS for study, expression systems have been developed for the
overexpression of human AS in yeast (Van Heeke & Schuster, 1990)
and AS-B in Escherichia coli (Boehlein et al., 1994a).

4
Two classes of AS have been described which do not possess
any sequence similarity. AsnA. found in many procaryotes, encodes
for an AS (AS-A) which only uses ammonia as a source of nitrogen in
the synthesis of asparagine (reaction 2). The kinetic and chemical
mechanism of AS-A, the product of the asnA gene, has been well
characterized (Cedar & Schwartz, 1969a, 1969b). In fact, the
existence of an aspartyl-AMP intermediate formed during AS
catalyzed asparagine synthesis was demonstrated first in
experiments using AS-A (Cedar & Schwartz, 1969b). The other AS,
encoded by asnB found in both prokaryotes and eukaryotes,
catalyzes the synthesis of asparagine using either ammonia or
glutamine as a source of nitrogen (reactions 1 and 2). These enzymes
also catalyze glutamine hydrolysis (reaction 3) in the absence of the
other substrates; a reaction which is stimulated by the presence of
ATP.
L-Asp + L-Gln + ATP - L-Asn + L-Glu +AMP + PP¡ (Reaction 1)
L-Asp + NH3 + ATP > L-Asn + AMP + PP¡ (Reaction 2)
L-Gln + H20 L-Glu + NH3 (Reaction 3)
Glutamine dependent asparagine synthetases are thought to be
organized into two functional domains. The glutamine amide
transfer (GAT) domain binds glutamine and catalyzes the hydrolysis
of glutamine to glutamate and ammonia. The synthetase domain
binds ATP and aspartate and catalyzes the ammonia dependent
reaction as well as the transfer of nitrogen to aspartate in the
glutamine dependent reaction. Evidence for the location of the two

5
active sites in separate domains was revealed through studies using
monoclonal antibodies (Pfeiffer et al., 1986, 1987) and chemical
modifiers (Larsen & Schuster, 1992) that could selectively inhibit the
glutaminase or asparagine synthesis activities of human or bovine
AS.
While the mechanism of nitrogen transfer between the two
domains is poorly understood, examination of conserved residues
within the individual domains has led to the identification of binding
sites and intermediates involved in the reaction. For the C-terminal
synthetase domain, it has been confirmed that during asparagine
synthesis, a (3-aspartyl-AMP intermediate is formed which activates
the aspartate carboxylate for attack by the amide nitrogen of
glutamine (Luehr & Schuster, 1985). Further studies showed a
binding site for aspartate located in a region highly conserved among
asparagine synthetases (Boehlein et al., 1997a). And finally, another
highly conserved region in the AS synthetase domain was found to
contain a P-loop-like motif that is present in other enzymes which
catalyze the hydrolysis of ATP to AMP and PP¡ (Bork & Koonin,
1994), and has been proposed to represent a pyrophosphate binding
element (Boehlein et al. manuscript in progress).
Experiments using site-directed mutagenesis have identified
several conserved amino acids in the N-terminal domain that are
crucial to glutamine utilization. Sequence analysis reveals that all AS
contain an N-terminal cysteine residue. Milman and Cooney first
proposed a critical cysteine residue in the glutamine active site when
they found AS from mouse pancreas was strongly susceptible to
oxidation (Milman & Cooney, 1979). More evidence for an important

6
cysteine residue was revealed upon the observation that bovine
pancreatic AS lost glutamine dependent activity when it was
covalently modified with the sulfhydryl modifier and glutamine
analog, 6-diazo-5-oxo-L-norleucine (DON) (Mehlhaff & Schuster,
1991). Finally, the essentiality of the N-terminal cysteine residue
was demonstrated in human AS and E. coli AS-B in experiments
replacing cysteine-1 with either an alanine or serine residue
(Boehlein et al., 1994a; Sheng et aL, 1993; Van Heeke & Schuster,
1989b).
In addition to the conserved cysteine residue, the role of
another completely conserved residue, asparagine-74, also has been
determined. When Asn-74 was replaced with either an alanine or an
aspartate, the resulting mutant enzymes (N74A & N74D) exhibited a
depressed glutaminase activity (Boehlein et al., 1994b). Subsequent
studies using the alternative substrate, L-glutamic acid-y-
monohydroxamate (LGH) and the AS-B mutant N74A led to the
suggestion that the interaction of this asparagine residue with
substrates causes a destabilization in the ES complex (Boehlein et al.,
1996) and, therefore, this residue might play a role similar to that of
Gln-19 of papain (Drenth et al., 1976; Menard, 1995). For papain, it
is thought that Gln-19 forms an oxyanion hole' in the stabilization of
a tetrahedral intermediate formed when the active site cysteine
attacks the substrate. This oxyanion hole contains a hydrogen
bonding network involving the NH peptide backbone of cys-25 and
the side chain NH, of Gln-19 which stabilizes the oxide anion bonded
to the tetrahedral carbon atom. The proposed role of N74 of AS-B in
forming such an oxyanion hole is illustrated below:

7
O
II
ENZ
ASN-74 NH
HN
H2N C S'
"1
R
CYS-1
This proposed relationship between the thiol protease, papain, and
AS-B was confirmed in studies which showed that a nitrile could act
as a weak competitive inhibitor of wild type AS-B and as a substrate
for the AS-B mutant N74D (Boehlein et al., 1997b); results which
were consistent with similar studies using wild type and a Q19E
mutant of papain (Menard, 1995).
Glutamine Amidotransferase Class II
Based on its primary amino acid sequence, asparagine
synthetase is a member of a family of enzymes known as the
glutamine-dependent amidotransferases. The members of this
family catalyze the transfer of nitrogen from glutamine to a variety
of acceptor substrates in the biosynthesis of amino acids,
carbohydrates, and nucleotides (for review, Zalkin, 1993). Like AS,
all amidotransferases share a modular organization with a glutamine
utilizing (GAT) domain in common and a synthetase domain which is
unique to each member of the family. According to the amino acid
sequence alignment of the GAT domain, glutamine amidotransferases
are divided into two types (Zalkin, 1993). Class I enzymes (formerly
G-type for the trpG gene product, anthranilate synthase) are

characterized by three catalytically important residues thought to
function as a cys-his-glu catalytic triad analogous to the thiol
proteases. This class of enzymes (see Zalkin, 1993 for review)
includes GMP synthetase (Tiedeman et al, 1985), anthranilate
synthase, (Nichols et al., 1980) carbamoyl-phosphate synthetase
(Piette, 1984), imidazole glycerol-phosphate synthase (Carlomagno,
1988), aminodeoxychorismate synthase, CTP synthetase (Weng,
1986), and FGAM synthetase (Schendel et al., 1989). Asparagine
synthetase as well as glutamine 5-phosphoribosyl-l-pyrophosphate
amidotransferase (GPA) (Tso et al., 1982), glutamine fructose-6-
phosphate amidotransferase (GFAT) (Badet-Denisot et al., 1993), and
glutamate synthase (Vanoni et al., 1991) comprise the class II
amidotransferases (formerly F-type for the purF gene product,
glutamine 5-phosphoribosylpyrophosphate) which are characterized
by a conserved N-terminal cysteine required for glutamine
dependent activity.
Proposals for the Mechanism of Nitrogen Transfer
For the class II enzymes, the formation of a covalent y-
glutamyl-enzyme intermediate involving the N-terminal cysteine
residue was first suggested in studies of GPA where it was shown
that attachment of glutamine affinity analogs to cys-1 (Vollmer et al.,
1983) or replacement with phenylalanine (Mantsala & Zalkin, 1984)
caused selective elimination of the glutamine-dependent activity.
Later studies using site directed mutagenesis of conserved residues
in the GAT domain of GPA led to the proposal that cys-1 participated
as part of a cys-his catalytic dyad in a mechanism analogous to the

9
thiol proteases (figure l.l)(Mei and Zalkin, 1989). This model of
nitrogen transfer involves a two step process: hydrolysis of
glutamine to produce free ammonia, followed by its transfer to the
acceptor substrate. In analogy to the thiol protease mechanism, a
nucleophilic attack of the conserved cys-1 thiolate anion on the
amide of glutamine results in a tetrahedral intermediate which, upon
release of ammonia, is converted to an acylthioenzyme intermediate
1. The released ammonia is transferred to the activated aspartate
intermediate 2 in the synthetase domain for the synthesis of
asparagine while the acylenzyme intermediate undergoes hydrolysis
to glutamate and free enzyme via a second tetrahedral intermediate.
For AS, this model for nitrogen transfer is supported by the
observation that all members of its family can catalyze the synthesis
of asparagine using ammonia instead of glutamine, demonstrating
the presence of binding site for ammonia. Furthermore, a completely
conserved cysteine has been shown to be catalytically essential in
GPA (Mei & Zalkin, 1989) and AS-B (Boehlein et al 1994a) and a
residue (Asn-74) thought to play a role in forming an oxyanion hole,
much like that formed by Gin-19 in papain catalysis (Menard et al.
1995), has been shown for AS-B.
While there is some evidence supporting an ammonia mediated
nitrogen transfer, there are also many inconsistencies. First, was the
inability to show that the class II enzymes contain a catalytically
important histidine cognate to that of the thiol protease papain.
Papain requires an active site histidine which functions to stabilize
the cysteine thiolate anion and provide a proton to the leaving group
in the conversion of the first tetrahedral intermediate to the

10
2
Asp
HS102
AS029
HiS-102
Figure 1.1 Mechanism of nitrogen transfer via an ammonia
intermediate as proposed by Mei and Zalkin, 1989. (a) Mechanism
for the hydrolysis of L-glutamine to yield ammonia and an
acylenzyme intermediate 1 by analogy to the purF enzyme, GPA. (b)
Synthesis of L-asparagine by reaction of ammonia with activated
aspartyl-AMP 2. (c) Hydrolysis of the acylenzyme intermediate to
yield L-glutamate. This diagram was taken from Richards and
Schuster, 1992.

11
thioester. Initially, it was thought, from results of site-directed
mutagenesis (Mei & Zalkin, 1989), that His-102 was involved in such
acid-base catalysis for GPA and covalent modification studies using
diethyl pyrocarbonate suggested that GFAT (Badet-Denisot & Badet,
1992) also possessed this important histidine. However, site-
directed mutagenesis of each of the conserved histidines in the GAT
domain of AS-B revealed little or no change in the kinetic parameters
(Boehlein et al., 1994a). Subsequent structural studies of GPA (Kim
et ah, 1996; Smith et ah, 1994) and the N-terminal domain of GFAT
(Isupov et ah, 1996) did not reveal a histidine in the appropriate
position to participate in acid-base catalysis. Alternatively, these
structural studies led to the proposal that the a-amino group of cys-1
serves as the general acid-base catalyst (Smith, 1995). Subsequent
structural comparisons of GPA with other hydrolytic enzymes placed
the class II enzymes into a larger superfamily called the N-terminal
nucleophile (Ntn) hydrolases which includes the 20S proteasome and
penicillin acylase (Brannigan et ah, 1995; Duggleby et ah, 1995;
Seemuller et ah, 1996). This family is characterized by an N-terminal
catalytic nucleophile which is thought to be activated by proton
transfer from its side chain to the free N-terminus (Seemuller et ah,
1996). However, experiments aimed at establishing the functional
role of the N-terminal amine using mutants of GPA and GFAT with an
additional N-terminal residue have yielded contradictory results
(Isupov et ah, 1996; Kim et ah, 1996).
Further evidence against an ammonia mediated nitrogen
transfer in AS-B catalyzed asparagine synthesis was revealed in an
investigation of the kinetic isotope effects (KIE) (Stoker et ah 1996)

12
for AS-B catalyzed glutamine dependent reactions. The KIEs
associated with placing 15N in the primary amide of glutamine were
examined for both glutamine dependent reactions and the resulting
values were interpreted using 15N KIE determinations for papain
catalyzed peptide hydrolysis (OLeary et al, 1974). For papain, the
KIE value was shown to be consistent with a mechanism where the
first irreversible step is C-N bond cleavage to form the thioester
intermediate (OLeary et al., 1974). It was expected that C-N bond
cleavage would be rate limiting in the AS-B catalyzed glutaminase
reaction as well and thus give the same 15N (V/K) isotope effect as
papain. Furthermore, an ammonia mediated nitrogen transfer would
require a common pathway for glutamine hydrolysis in the two
glutamine dependent reactions. Therefore, if AS-B followed an
ammonia mediated mechanism, it was expected that the 15N KIEs
determined for both glutamine dependent reactions would be
identical. Surprisingly, the glutaminase 15N (V/K) was significantly
smaller than that obtained for papain catalyzed peptide hydrolysis,
indicating that a step other than C-N bond cleavage may be rate
limiting. Moreover, the KIE for glutaminase was smaller than that
for the glutamine dependent synthesis reaction. These results were
regarded as the first direct evidence against the mediation of
nitrogen transfer by enzyme-bound ammonia.
In addition to the above mentioned inconsistencies between
AS-B catalytic behavior and an ammonia mediated nitrogen transfer,
questions addressing the difficulties of maintaining the free ammonia
in the active site in its unprotonated form led to the proposal of
alternative mechanisms for nitrogen transfer (Figure 1.2) (Richards &

13
Schuster, 1992). The first proposal for an alternate mechanism of
nitrogen transfer involved the formation of an imide intermediate
(Richards & Schuster, 1992). According to this hypothesis, a direct
nucleophilic attack of the primary amide of glutamine on the p-
aspartyl-AMP intermediate 2 to form an imide intermediate 3.
Reaction of this intermediate with cys-1 causes the release of
asparagine and the formation of a thioacylenzyme intermediate 1
which forms free enzyme and glutamate in a mechanism similar to
the ammonia mediated mechanism. An attempt to isolate such an
intermediate was made using the AS-B mutant (CIA) which lacks the
cysteine required to catalyze C-N bond cleavage of the imide
intermediate. However, an imide intermediate was not detected by
HPLC using chemically synthesized imide as a standard, when CIA
AS-B was incubated with glutamine, ATP, and aspartic acid. It was
concluded that participation of an imide intermediate in nitrogen
transfer catalyzed by AS-B was unlikely (Boehlein et al., unpublished
observations).
In a second alternative hypothesis for the mechanism of
nitrogen transfer (Stoker et al., 1996), Cys-1 reacts with glutamine to
form the first tetrahedral intermediate 5 (figure 1.3). In contrast to
the ammonia mediated mechanism, however, the tetrahedral
intermediate makes a subsequent direct attack on the p-aspartyl-
AMP intermediate to form an intermediate in which glutamine is
covalently linked to p-aspartyl-AMP 6. Cleavage of the C-N bond in
this intermediate releases the acylenzyme and forms a second
tetrahedral intermediate 8. Asparagine is then formed from a loss of

14
ASPm
Figure 1.2 Mechanism of nitrogen transfer via an imide intermediate
3 as proposed by Richards and Schuster, 1992. Attack of the
primary amide occurs directly on the activated L-aspartate 2. Cys-1
and Asp-33 are then involved in the hydrolysis of the imide to yield
L-glutamate and L-asparagine in subsequent steps.

15
Figure 1.3 Mechanism of nitrogen transfer via direct transfer of the
nitrogen from glutamine to aspartyl-AMP as proposed by Stoker et
al., 1996).

16
AMP from the tetrahedral intermediate. Interestingly, this
mechanism would not necessarily require N-protonation before C-N
bond cleavage in the glutamine dependent synthesis reaction and,
therefore, may explain the differences in the I5N (V/K) isotope
effects for the two glutamine dependent reactions.
Despite the progress that has been made in understanding the
functions of the individual domains of AS, the mechanism of nitrogen
transfer remains unequivocally defined. One difference between the
mechanism involving free ammonia and that of a direct attack is the
requirement, in the former, of a common mechanism for glutamine
hydrolysis in the glutaminase and glutamine dependent asparagine
synthesis reactions. Therefore, it is important to understand the
mechanistic details of both the glutamine dependent synthesis and
glutamine hydrolysis in order to discriminate between the two
mechanisms. The ability of AS-B to catalyze the hydrolysis of
glutamine in the absence of the other amino acids provides the
opportunity to develop a model of the glutaminase reaction from
which to identify the individual steps and intermediates in the
reaction as well as examine the functional roles played by the
conserved amino acids in the GAT domain. In efforts to begin
building such a model, this dissertation focuses on the chemical and
kinetic mechanisms of glutamine hydrolysis catalyzed by AS-B.
Differences and similarities in the mechanism of substrate hydrolysis
catalyzed by AS-B and papain will be examined through pH analysis
of AS-B catalyzed glutaminase activity and by isolation of a covalent
intermediate in the reaction pathway of glutamine hydrolysis
catalyzed by AS-B. A variety of kinetic techniques will be used to

17
determine the individual rate constants of the glutaminase reaction
and these will be used to build a model of glutamine hydrolysis
catalyzed by AS-B.

CHAPTER 2
pH DEPENDENCE OF THE GLUT AMIN ASE REACTION CATALYZED BY
ASPARAGINE SYNTHETASE B
Introduction
Asparagine Synthetase B (AS-B) catalyzes the ATP dependent
synthesis of asparagine using aspartic acid and either glutamine or
ammonia as a nitrogen source (reactions 1 and 2). In the absence of
aspartate and ATP, this enzyme also is capable of catalyzing
glutamine hydrolysis (reaction 3).
L-Gln + ATP + L-Asp L-Asn + AMP + PPj + L-Glu (reaction 1)
NH3 + ATP + L-Asp -> L-Asn + AMP + PPj (reaction 2)
L-Gln + H20 -4 Glu + NH? (reaction 3)
Amino acid sequence alignment of the N-terminal glutamine utilizing
domain of AS-B indicates that this enzyme is a member of the class II
glutamine amidotransferase family (Zalkin, 1993) which includes
glutamine 5-phosphoribosylpyrophosphate amidotransferase (GPA)
(Tso et al., 1982), glutamine fructose-6-phosphate amidotransferase
(GFAT) (Badet-Denisot & Badet, 1993), and glutamate synthase (Vanoni
et al., 1991). Like AS-B, members of this family of enzymes catalyze
the transfer of the amide nitrogen of glutamine to various acceptor
molecules including amino acids, nucleotides, and carbohydrates.

19
The mechanism of nitrogen transfer remains controversial and poorly
understood. Results from studies using site-directed mutagenesis
suggested the catalytic importance of a conserved N-terminal cysteine,
a histidine, and an aspartic acid in glutamine utilization in GPA (Mei &
Zalkin, 1989). This observation led to the hypothesis that glutamine
amidotransferases catalyze glutamine hydrolysis by a mechanism
similar to that of amide hydrolysis catalyzed by the thiol proteases,
typified by papain (for review see Brocklehurst et al., 1987), as shown
below (1).
Binding Acylation Deacylation
TH I TH II
In analogy to the papain mechanism, a nucleophilic attack of the N-
terminal cysteine on the side-chain amide of glutamine forms a
tetrahedral intermediate (TH I). Protonation of the amide allows C-N
bond cleavage to produce a thioester intermediate and free ammonia.
The ammonia is then transferred to the acceptor molecule and the
thioester intermediate is broken down, via another tetrahedral
intermediate (TH II), to produce glutamate and free enzyme. As
illustrated above, an active site cysteine and an acid-base catalyst (a
role played by his-159 in papain) are vital components to such a
reaction scheme. The cysteine is involved in the activation of
glutamine as described above. In papain, the histidine is thought to

20
play three essential acid-base roles including 1) stabilization of the
thiolate; 2) donation of a proton to the amide of the first tetrahedral
intermediate to increase its leaving group ability; and finally 3)
activation of water for a nucleophilic attack on the thioester
intermediate by accepting a proton (Storer & Menard, 1994).
This mechanism of ammonia mediated nitrogen transfer is
supported by the ability of the amidotransferases to use ammonia as
a nitrogen source and thus the demonstration of an ammonia binding
site. Furthermore, the presence of a completely conserved N-
terminal cysteine, which has been shown to be essential for
glutamine dependent activity of GPA (Mei & Zalkin, 1989) as well as
both human and E. coli asparagine synthetases (Boehlein et al.,
1994a; Sheng et al., 1993; Van Heeke & Schuster, 1989b), suggests
that this residue may function in a manner similar to that of the
active site cysteine of the thiol proteases. On the other hand, the
presence of an important histidine in the active site of the class II
enzymes has become increasingly doubtful. While the studies of GPA
mentioned above (Mei & Zalkin, 1989) and chemical modification
studies of GFAT (Badet-Denisot & Badet, 1992) seemed to indicate
the presence of an essential histidine in these enzymes, site-directed
mutagenesis of residues, known to be conserved among asparagine
synthetases, did not reveal any histidines required for activity in AS-
B (Boehlein, et al., 1994a). Moreover, recent structural studies of
GPA (Kim et al., 1996; Smith et al.. 1994) and the glutamine utilizing
domain of GFAT (Isupov et al., 1996) suggested that there were no
histidines in position to act as part of a cys-his dyad. Structural
comparisons later revealed that the protein fold of GPA (Brannigan et

21
al., 1995) was homologous to that of a family of N-terminal
nucleophile (Ntn) hydrolases which also includes penicillin acylase
(Duggleby et al., 1995), the 20S proteasome (Lowe et al., 1995;
Seemuller et al., 1996), and aspartylglucosamidase(Fisher et al.,
1993; Oinonen et al. 1995). It is thought that members of the Ntn
family use the N-terminal Ser, Thr, or Cys as the nucleophilic catalyst
and the N-terminal a-amino group as the acid-base catalyst.
However, whether or not the class II amidotransferases use the N-
terminal amino group in acid-base catalysis remains uncertain as
results from experiments which have placed an extra amino acid at
the N-terminus of GPA (Kim et al., 1996) have been inconclusive.
Further investigation is necessary in order to understand fully
how acid-base catalysis occurs during the glutamine hydrolysis
catalyzed by the class II amidotransferases. In fact for AS-B, the
possibility of an essential histidine still remains. A computer model
of a hypothetical AS-B glutamine binding site, constructed using the
crystallographic coordinates for the covalent adduct of DON and E.
coli GPA, revealed that histidine-47 of AS-B was in an appropriate
position to function as part of a cys-his dyad cognate to that of
papain (Richards & Schuster, 1997). Moreover, this histidine is
partially conserved in asparagine synthetases and a histidine in a
comparable position is found in several other class II
amidotransferases (Figure 2.1), thus its importance cannot be
overlooked. In an effort to gain a clearer understanding as to how
acid-base catalysis occurs during glutamine utilization by AS-B, it is
the purpose of this chapter to explore the glutaminase activity of this

22
LtHmAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
GnHmAS
CGIWAI,<
>HRGPD<
>FGFHRLAV<
RatAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
MurAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
HumAS
CGIWAL<
>HRGPD<
>FGFHRLAV<
SoyAS
CGILAV<
>HRGPD<
>LAHQRLAI<
Lot AS
CGILAV<
>HRGPD<
>LAHQRLAI<
FavaAS
CGILAV<
>HRGPD<
>LAHQRLAI<
PeaNAS
CGILAV<
>HRGPD<
>LAHQRLAI<
AlflAS
CGILAV<
>HRGPD<
>LAHQRLAI<
PeaRAS
CGILAV<
>HRGPE<
>LAQQRLAI<
AspaAS
CGILAV<
>HRGPD<
>LSHQRLAI<
ArabAS
CGILAV<
>HRGPD<
>LAHQRLA.V<
BrssAS
CGILAV<
>HRGPD<
>LAHQRLAI<
RiceAS
CGILAV<
>HRGPD<
>LAHQRLAI<
MaizAS
CGILAV<
>HRGPD<
>LAHQRLAI<
ScerAS
CGIFAA<
>HRGPD<
>FVHERLAI<
CeleAS
CGVFSK
>HRQPD<
>LVHERLAI<
EcoAS
CSIFQV<
>HRQPD<
>LAHERLSI<
1
29
47
ScerGA
CGILGK
>HRGQD<
>FTQQRVS.<
HumGS
CGIFAY<
>YRGYD<
>HKQQDMDL<
ScerGS
CGIFGY<
>YRGYD<
>.TKQNPNR<
>NGEIYNHKAL
>NGEIYNHKAL
>NGEIYNHKAL
>NGEIYNHKAL
>NGEIYNHKKM
>NGEIYNHEEL
>NGEIFNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEDL
>NGEIYNHEDL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHEEL
>NGEIYNHIQL
>NGEIVNHGEL
>NQEIYNHQAL
74 80
>NGNLVNTAS L
>NGIITNYKDL
>NGIITNFREL
Figure 2.1: Sequence alignment of part of the GAT-domains of
known asparagine synthetases and a partial set of class II
amidotransferases showing highly conserved residues. Numbering is
according to the amino acid sequence of AS-B. Sequence alignments
taken from Richards & Schuster, 1997. LtHmAS = Cricetulus
longicaudatus AS; GnHmAS = Mesocricetus auratus AS; RatAS = Rattus
norvegicus AS; MurAS = Mus musculus AS; HumAS = Homo sapiens
AS; Soy AS = Glycine max AS; Lot AS = Lotus japonicus AS; Fava AS =
Vicia faba AS; PeaNAS = Pisum sativa AS (root); AspaAS = Asparagus
officinalis AS; ArabAS = Arabidopsis thaliana AS; BrssAS = Bras sica
olercea AS; RiceAS = Oryza sativa AS; MaizAS = Zea mays', Seer AS =
Saccharomyces cerevisiae AS; Cele AS = elegans AS; Eco AS =
Escherichia coli AS; ScerGA = Saccharomyces cerevisiae GPA; HumGS =
Homo sapiens GFAT; ScerGS = Saccharomyces cerevisiae GFAT.

23
enzyme as a function of pH. In order to address the possibility that
H47functions in an acid-base capacity, the effects of mutations at this
position on the pH-rate profile will be characterized.
Materials and Methods
Chemicals and Reagents
Restriction and modifying enzymes were purchased from Promega
(Madison, WI) or New England Biolabs (Beverly, MA). Other chemicals,
including L-glutamine and L-glutamic-y-monohydroxamate, were
obtained from SIGMA Chemical Company (St. Louis, MO) and were of the
highest commercial purity. Oligonucleotide primers were synthesized
on an Applied Biosystems 380B DNA synthesizer by the DNA Synthesis
Core Facility of the Interdisciplinary Center for Biotechnology Research
at the University of Florida. Polymerase chain reactions (PCR) were
performed on an Ericomp (San Diego, CA) thermocycler using the
Gene Amp DNA Amplification Reagent Kit with AmpliTaq from Perkin-
Elmer. Double stranded DNA sequencing was performed using the U.S.
Biochemical Corp. Sequenase 2.0 Sequencing kit.
Bacterial Strains and Plasmids
Bacterial strains BL21DE3pLysS (F\ ompT, rb\ mb"), supplied by
Studier and Moffat; and NM522 (supE, thi (lac-proAB), hsd5, (rm -
)/F' pro AB, lac Iq Z M15), obtained from Stratagene (La Jolla, CA), are
both derivatives of Escherchia coli K-12. The plasmid pBluescript
was obtained from Stratagene (La Jolla, CA) and the plasmid pETB
has been described previously (Boehlein et al., 1994a). E. coli host
cells were transformed according to the procedures of Hanahan
(Hanahan, 1983).

24
Enzyme Preparation
Construction of recombinant wild-type AS-B and mutants in
which His-29 or His-80 have been replaced by alanine (H29A, H80A)
has been described elsewhere (Boehlein et ah, 1994a). PCR cassette
mutagenesis was used to replace Histidine-47 with alanine (H47A) or
asparagine (H47N) using oligonucleotide primer pairs ss350R and
ss65 or ss348R and ss65, respectively (Table 2-1) and pETB as a
template. The PCR parameters involved 35 cycles of denaturation at
94C for 1 min; annealing at 55C for 1 min; and extension at 72C for
two minutes and the final cycle was followed by a 5 minute
extension step at 72C. The plasmid, pETB, and the PCR products
were digested with Hpal and Xhol and the 50 bp product was cloned
into pETB. The resulting plasmids were sequenced through the
insertion and no alterations other than those intended were found.
All wild-type and mutant proteins were expressed, and purified
using standard procedures (Boehlein et al., 1994a). Protein
concentrations were determined by the method of Bradford (1976)
using an assay kit supplied by Bio-Rad and mouse immunoglobulin G
to construct a standard curve. During the course of these
investigations an alternate method of determining the protein
concentration revealed that the current use of mouse
immunoglobulin G as a standard gives erroneously high protein
concentrations for AS-B. Amino acid composition analysis of AS-B,
carried out by the Protein Chemistry Core Facility of ICBR at the
University of Florida, showed that the actual value for the
concentration of this enzyme is 2.7 times less than that obtained
using the Bradford assay. Therefore, all of the values of enzyme

concentration and therefore all kcat and kca[/Km values given in this
chapter have been corrected by dividing the protein concentration
by the factor 2.7.
25
TABLE 2.1
Oligonucleotides used in construction of site-directed mutants of
AS-B
Oligo
Number
Oligonucleotide Sequence
ss65
5 A GCT TCC CAT ATG TGT TCA ATT TTT GGC GTA
TTC GAT 3
ss350R
5 CGC CCC CGC GTT AAC GTC GAC AAT TGA CAG ACG
TTC GGC GGC GAG A 3
ss349R
5 CGC CCC CGC GTT AAC GTC GAC AAT TGA CAG ACG
TTC GGC GGC GAG A 3
Glutaminase Assays
Initial rates of L-glutamine and L-glutamic-y-monohydroxamate
(LGH) hydrolysis were determined by a modified procedure which
uses glutamate dehydrogenase in the presence of NAD+ to measure
glutamate concentration (Bernt and Bergmeyer, 1974). Reactions
(100 pi) containing lOOmM Bis-Tris and Tris-HCl (pH 6.0-9.0) each
(the kinetic constants appear to be independent of the nature and
concentrations of these buffers), 8mM MgCl2 and various
concentrations of glutamine or LGH were initiated by the addition of
wild-type (0.7-2 pg) or mutant (4 pg) AS-B and incubated at 37C. A
control containing denatured enzyme was used to monitor any
glutamate formed from spontaneous breakdown. The reactions were
terminated by boiling and a coupling reagent (300 mM glycine, 250

26
mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2 units of
glutamate dehydrogenase) was added and incubated 10 minutes at
room temperature. The absorbance at 340 nm was measured and
the concentration of glutamate was determined by comparison to a
standard curve. Initial velocities were measured at 9 different
substrate concentrations and each velocity was an average of 2-4
measurements. Stability of the wild-type and mutant AS-B enzymes
at each pH was confirmed by incubation in buffer of the appropriate
pH for 20 minutes followed by an assay of activity at pH 8. Initial
velocity conditions were confirmed at each pH by linear plots of v
versus time and v versus enzyme concentration.
Data Analysis
The values of Km and kcat were determined from least-squares
fit to v0 vs [S] plots using the Prism software package supplied by
Graphpad, Inc., (San Diego, CA) according to equation 1:
Vmax [S]
v = (1)
Km + [S]
Where v is the maximal velocity, [S] is the concentration of st
Vmax is the maximal velocity, and Km is the Michaelis constant. The
pH dependence of kcat/Km obtained from equation 1 were fit to the
bell-shaped curve described by equation 2:
Yiim
Y= (2)
l + !0(Pkl-pH) + 10(pH-pK2)

27
In the above equation, Y is the observed kcat/Kra at a given pH, Ylim is
the maximum value of kcat/Km, and pK; and pfC, are the lower and
higher acid dissociation constants, respectively.
Results
Effect of pH on the Steady-state Kinetic Parameters of wild-tvpe AS-
B Catalyzed Glutaminase
The steady-state kinetic parameters, kcat and kcat/Km, for
glutamine hydrolysis were measured as a function of pH. As
illustrated in Figures 2.2 and 2.3, and Table 2.2. kcal is constant over
the pH range 6.0-9.0 whereas the kca[/Km exhibits a bell shaped curve.
Since the pKa of the a-amino group of glutamine is 9.13, the Km and
thus kcat/Km were corrected for protonated glutamine using the
Henderson-Hasselbach equation in an attempt to show how the
ionization of the substrate might affect the kcat/Km-pH profile. Indeed,
the correction caused a complete disappearance of the basic limb in
the kcat/Km(app) plot (Figure 2 .3). Non-linear regression of the
experimental data to equation 1 allowed determination of the pKa for
the non-corrected data which is shown in Table 2.5.
A comparison of the pH dependencies of the hydrolyses of
glutamine and the alternate substrate, LGH, can provide information
about whether these reactions follow the same mechanism.
Therefore, a pH profile of the wild-type AS-B catalyzed LGH
hydrolysis was also examined (figure 2.4 and 2.5, and Table 2.3).
Similar to the glutamine profile, the kcat appeared independent of pH
for most of the pH range studied with slight increases at the low and
high pH. The kcar/Km(app) profile for LGH is bell shaped with a peak

28
at neutral pH and an acidic and basic pKa close to that obtained with
glutamine (Table 2.5). Unlike the glutamine profile, correction for
protonated LGH had little effect on the pH-kca[/Km profile (Figure 2.5).
Kinetic Characterization of the Glutaminase Reaction catalyzed bv the
AS-B mutant H47N
As part of a continuing search for a histidine which may serve
in a general acid-base capacity during AS-B catalyzed glutamine
hydrolysis, Histidine-47 was examined using site-directed
mutagenesis. Mutations were made which replaced the histidine with
an alanine (H47A) or an asparagine (H47N) and the resulting enzymes
were expressed. H47A was insoluble whereas H47N expressed at
high levels and was successfully purified. The glutaminase activity of
H47N was approximately 10-fold slower than that of wild-type AS-B
(compare Tables 2.2 and 2.4). In addition to these changes in kcat,
there was a slight increase in the Km value for glutamine.
pH Dependence of the Kinetic Constants of wild-tvpe AS-B and its
H47N
The pH dependence of the kinetic constants for H47N catalyzed
glutamine hydrolysis was examined. The pH dependence of kcat was
minimal (figure 2.6) except for a slight decrease at low pH which was
not evident in the wild-type. The pH profile of kcat/Km(app) revealed
a bell-shaped curve (figure 2.7) similar to that of wild-type catalyzed
LGH hydrolysis except the curve was slightly narrower. The
decrease in kcat/Km(app) on the basic limb was evident even after the
correction for protonated glutamine.

29
O
CD
CD
3.0
2.5
2.0
1.5
1.0
0.5
0.0
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
pH
Figure 2.2 The pH dependence of kcat for wild-type AS-B catalyzed
glutamine hydrolysis. The kinetic constants were determined as
described in materials and methods.

30
3000
2500-
o 2000-1

i 1500-
j? 1000-
500-
Figure 2.3 The pH dependence of kcat/Km for wild-type AS-B
catalyzed glutamine hydrolysis. The kinetic constants were
determined as described in the materials and methods. The solid
line () represents the actual data. The dashed line (---) represents
the data corrected for the protonation of glutamine.

Table 2.2
Kinetic constants of the AS-B-catalyzed hydrolysis
of L-Glutamine at 37C
pH
^cat
K
kcat/Km
(sec
(mM)
(M'hsec-1)
6.00
1.51

0.05
5.78

0.60
261
6.25
1.67

0.10
3.44

0.33
485
6.50
1.89

0.03
2.4

0.1
787
7.00
1.94

0.03
1.01

0.06
1921
7.25
2.05
+
0.05
1.22

0.08
1680
7.50
1.92

0.03
1.01

0.06
1901
7.75
2.13

0.03
1.32

0.07
1614
8.00
2.21

0.03
1.15

0.05
1922
8.50
2.46

0.05
1.50

0.06
1640
8.75
2.48

0.05
1.92

0.09
1292
9.00
2.56

0.05
2.33

0.12
1099

32
PH
Figure 2.4 The pH dependence of kcat for wild-type catalyzed LGH
hydrolysis. The kinetic constants were determined as described in
materials and methods.

33
Figure 2.5 The pH dependence of kcat/Km for wild-type AS-B
catalyzed LGH hydrolysis. The kinetic constants were determined as
described in materials and methods. The solid line () represents
the actual data. The dashed line (---) represents the data corrected
for the protonation of glutamine.

Table 2.3:
Kinetic constants of the AS-B-catalyzed hydrolysis of L-
glutamic-Y-monohydroxamate at 37C
PH kcaL
(sec-1)
Km
(mM)
kCat/Km
(M-'sec'1)
6.00
3.46

0.05
6.51

0.29
531
6.25
3.16

0.05
2.39

0.15
1322
6.50
2.86

0.03
1.11

0.05
2577
6.75
2.59

0.03
0.68

0.02
3809
7.00
2.67

0.03
0.66

0.02
4045
7.25
2.65

0.03
0.58

0.03
4569
7.50
2.67

0.03
0.65
+
0.02
4108
7.75
2.65
+
0.03
0.63

0.03
4206
8.00
2.38

0.05
0.83

0.05
2867
8.25
2.54

0.03
1.11

0.05
2288
8.50
3.00

0.05
1.53

0.08
1961
8.75
3.10

0.05
2.26
+
0.11
1372
9.00
4.05

0.19
7.41

0.78
546

35
0.6-
O
CD
CO
CO
o.
0.5-
0.4-
0.3-
2
0.1-j
0.0
o a
I 1 I 1 1 1 1
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0
pH
Figure 2.6 The pH dependence of kcat for glutamine hydrolysis
catalyzed by the AS-B mutant H47N. The kinetic constants were
determined as described in materials and methods.
9.5

kcat/Km (NT1 sec-
36
Figure 2.7 The pH dependence of kcat/Km for glutamine hydrolysis
catalyzed by the AS-B mutant H47N. The kinetic constants were
determined as described in materials and methods.

37
Table 2.4:
Kinetic constants for glutamine hydrolysis catalyzed
by the AS-B mutant H47N at 37C
PH
kcat
K.
kcat/Km
(sec'1)
(mM)
(M^sec-1)
6.00
0.19
0.01
18.6 2.8
1 0
6.25
0.22
0.008
13.1 2.0
1 7
6.50
0.30
0.005
8.3 0.9
36
6.75
0.30
0.008
5.4 0.4
55
7.00
0.30
0.005
3.2 0.2
9 4
7.25
0.30
0.003
2.4 0.1
125
7.50
0.38
0.005
2.5 0.1
152
7.75
0.32
0.003
2.5 0.1
128
8.00
0.32
0.003
3.0 0.1
107
8.25
0.40
0.005
4.1 0.2
98
8.50
0.46
0.008
6.1 0.3
75
8.75
0.38
0.01
7.8 0.7
49
9.00
0.43
0.01
10.5 1.1
4 1

38
Table 2.5:
Values of pKa for the kcal/Km-pH profiles for glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and for glutamine hydrolysis
catalyzed by the AS-B mutant H47N
Enzyme
Substrate
km ,/Km
PK,
pK?
wild type
glutamine
6.7 0.1
8.8 0.1
LGH
6.7 0.1
8.0 0.1
H47N
glutamine
7.2 0.2
8.1 0.2

TABLE 2.6:
Kinetic parameters for wild-type AS-B and the H29A, H80A, H47N, and R30A mutants at three
values of pH
kcat]
(sec )
pH 6.5
(mM)
VK,
(M'rsec )
kcat
(sec )
pH 7.5
K,,,
(mM)
k,/K
(M sec )
kca.
(sec )
pH 9.0
K,
(mM)
kCj,t/Km
(M sec )
wt
1.89
0.03
2.40
0.10
787
1.92
0.03
1.01
0.06
1901
2.56
0.05
2.33
0.12
1099
H29A
0.94
0.10
34.47
5.23
27
1.08
0.13
10.36
1.84
104
2.29
0.57
115 28
20
H80A
1.57
0.11
30.92
2.59
51
4.54
0.65
5.02
0.78
904
5.89
0.54
58 6
176
H47N
0.30
0.005
8.3 0.9
36
0.38
0.005
2.46
0.11
154
0.43
0.01
10.47
1.15
41

40
Comparison of the Kinetic Parameters of Wild-tvpe. H80A. and H29A.
at various pH values
The kinetic constants for the glutaminase activity of the three
mutants (H29A, H47N, and H80A) are compared with wild-type in
Table 2.6. Substitution of an alanine for the histidines at positions 29
and 80 has little effect on the kcat at pH 6.5 and 7.5 while this
substitution decreases kcat/Km significantly. However, unlike wild
type the kcat value of H80A increases with increasing pH to a value at
pH 9 which is double that of wild type. The most striking effect
caused by the mutations at positions 29 and 80 is seen in the
dramatic increase in Km especially at high pH.
Discussion
The importance of a histidine in the mechanism of glutamine
hydrolysis catalyzed by AS-B and thus the mechanistic similarity
between the AS-B catalyzed glutaminase reaction and the papain
catalyzed peptide hydrolysis have been questioned for several years.
Papain catalyzes the hydrolysis of peptide, amide, and ester bonds
by the mechanism shown in (1). All substrates tested with papain
have shown a bell-shaped pH dependence for kcat/Km with pKas of
approximately 4 and 8.5. As shown in the simplified model below, it
is considered that these ionizations represent the thiol of cysteine-25
and the imidazole of histidine-159 which are involved in the
acylation process (Menard et ah, 1990):

41
pKi 4 pK2 8.5
-SH +HIm- -S' +HIm- -S' Im-
active enzyme
form
In order to understand how acid-base catalysis occurs in the
AS-B active site during glutamine hydrolysis, the pH dependence of
the kinetic constants has been examined for this enzyme. For both
glutamine and LGH, the kcat/Km value has a bell-shaped pH
dependency. In contrast, kcat appears independent of pH in the range
of 6-9. The similarity in pH profiles for these two substrates
suggests that their hydrolysis occurs by similar mechanisms.
Furthermore, the values of kca[/Km for LGH are about 2-fold higher
than those for glutamine (compare Tables 2.2 and 2.3). If substrate
hydrolysis follows a mechanism like that of papain, then N-
protonation of the leaving group on the first tetrahedral intermediate
is required for C-N bond cleavage to occur. Therefore, the faster
acylation rate for LGH, reflected in kcat/Km, is consistent with the fact
that hydroxylamine (pKa=8.2) is a better leaving group than ammonia
(pKa~9.2). The pH independence of kcat may arise from a change in
rate determining step between two steps with similar rates.
However, further investigation, using a more extensive pH range is
necessary for a clearer picture of the ionizations affecting kcat.
Unfortunately, the AS is highly unstable when the pH is greater than
9 or less than 6.
The bell-shaped curves displayed in the kcat/Km profiles for AS-
B hydrolysis of glutamine and LGH (with pKa values of 6.6 and 8.3)
are narrower than that reported for papain. The stability of the ion-

42
pair in papain is reflected in the width of the pH-profile for kcat/Km
(Menard et al., 1991). Therefore, it is possible that AS-B contains an
ion pair like that of papain but lacks some of the factors which play a
role in ion pair stabilization. On the other hand, the pH profile of
kcat/Km may relate to ionizations in free substrate. In this study, the
pKa values of the a-amino groups of the substrates are 9.13
therefore, the possibility that the alkaline pKa values observed in the
kca[/Km profiles arise from the ionization of glutamine and LGH cannot
be excluded. In fact, the basic limb of the glutamine kcat/Km profile is
completely eliminated when Km values are corrected for protonated
glutamine. However, the same treatment has little effect on the
kcat/Km profile for LGH.
While the above examination of the kinetic behavior of wild-
type AS-B as a function of pH indicates the potentiality of an ion pair
similar to that of papain, past efforts have failed to reveal a histidine
which functions as part of that pair. In this study, substituting an
asparagine for histidine at position 47 resulted in a reduction of both
kcat and kcat/Km of approximately 10-fold for the glutaminase activity.
While these changes are significant, a much more dramatic effect was
expected if this residue were playing such an essential role in
catalysis as has been postulated. For example, replacement of
histidine-353 with an asparagine in the class I enzyme, CPS, resulted
in a drop in glutaminase activity from 1.1 pmol/h*mg for wt to
<0.006 pmol/hmg for H353N (Miran et al, 1991). Subsequent
structural studies which showed that this residue was in position to
interact with the active site histidine added further support that His-
353 was involved in acid-base catalysis (Thoden et al., 1997).

43
Additionally, the pH profiles for glutamine hydrolysis catalyzed
by H47N (Figures 2.6 & 2.7) resemble those of wt including similar
pKa values of 7.2 and 8.5 (Table 2.5) which suggests that histidine-47
is not responsible for one of these ionizations.
Previous investigations of H29 and H80 failed to reveal that a
histidine in either of these positions was essential for AS-B catalyzed
asparagine synthesis (Boehlein et al., 1994a). While these studies
were performed at pH 8 in the presence of an imidizole buffer, it is
possible that the importance of the active site histidine could have
been masked. Table 2.6 presents the results of a further
investigation using the glutaminase activity at pH 6.5, 7.5 and 9.0.
The lack of a significant decrease in the kca[ value for the glutaminase
reaction catalyzed by these mutants confirms the previous
conclusions that these histidines probably are not involved as part of
the essential cys-his dyad which has been proposed.
Interpretations of this and any pH analysis of an enzymatic
reaction rely on a number of assumptions which have been
summarized by Knowles (1976). First, as noted by Peller and
Alberty (1959), the association of the pKa values of the pH-kcat/Km
with ionizations of the free enzyme and free substrate assumes that
only one ionization state of the enzyme is directly convertible to
product. In fact, the pH dependence of the kcat/Km values observed in
papain catalyzed substrate hydrolysis has been found to be more
complex than originally thought (Brocklehurst, 1994). Several studies
have provided evidence that the development of the full nucleophilic
character of the ion pair depends not only on the ionization at pH 4
but on a second ionization at pH 4 and. in some cases, ionization with

44
pKa 5 (Brocklehurst, 1988 a,b). Altered pKn values also may result if
the rate constant for substrate dissociation is less than the net rate
constant for product formation. This is unlikely to be the case for
AS-B glutamine hydrolysis since investigations into the dissociation
constant of glutamine have suggested that a rapid equilibrium
mechanism is plausible (see chapter 4). Determination of the pKa of a
competitive inhibitor or of the backwards reaction will be necessary
to confirm this assumption.
With respect to these assumptions it is important to take heed
in comparison of the pH profiles of AS-B and papain. On the other
hand, with the results presented here, it is difficult to envision a role
for histidine-47 in acid-base catalysis when one considers the lack of
effect its replacement with asparagine made on the pH profiles. In
light of the presence of this histidine in many asparagine synthetases
as well as other class II amidotransferases it was important to
include this residue among potential acid-base catalysts. Indeed, the
10-fold loss of activity in the H47N mutant may reflect a minor or
indirect importance of this residue in glutamine binding and/or
utilization. On the other hand, the importance of the histidine to
papain activity is underscored in the complete conservation of this
amino acid across the entire family of cysteine proteases. The lack of
such a completely conserved histidine in the class II family of
amidotransferases suggests that an alternate mechanism of acid-base
catalysis probably occurs in these enzymes and therefore
demonstrates dissimilarities in mechanism for the two families of
enzymes.

CHAPTER 3
EFFECT OF TEMPERATURE ON GLUTAMINE HYDROLYSIS CATALYZED
BY ASPARAGINE SYNTHETASE B
Introduction
The following chapters present experiments which are aimed at
the determination of rate constants and the identification of
intermediates in order to develop a descriptive model of the
glutaminase reaction. Much of the investigation requires a very slow
reaction rate and, therefore, the reactions were examined at 5C.
Variation in temperature can cause alterations in the mechanism of a
reaction including differences in rate determining steps. In order to
validate the comparison of the results from experiments at 5C with
those performed at the temperature at which the reaction is
normally investigated (37C), the steady state kinetics of glutamine
hydrolysis will be examined over a range of temperatures and
Arrhenius plots will be used to detect any changes in rate limiting
step with temperature for this reaction. The temperature
dependence of LGH hydrolysis catalyzed by wild-type AS-B and that
of glutamine hydrolysis catalyzed by the AS-B mutant R30A will be
examined in a similar manner and the activation energies of the
three reactions will be compared. Similarity in the pH profiles for
glutamine and LGH hydrolysis presented in the previous chapter
suggests that these reactions occur by a common mechanism.
Further evidence for a common mechanism could be obtained
45

46
through similarities in the activation parameters which would
suggest that the same step is rate limiting in both reactions. In
contrast, work in the following chapters suggests that glutamine
hydrolysis catalyzed by the AS-B mutant, R30A, has a different rate
limiting step than that catalyzed by wt AS-B and these differences
might be revealed in different values for the activation parameters.
A complete description of the glutaminase reaction requires 1)
the order of product release, 2) the inhibition constants of the
products, and 3) knowledge of whether or not the reaction is
reversible. Therefore, in order to address these issues, product
inhibition studies were employed and the ability of the enzyme to
catalyze the synthesis of glutamine and LGH was examined.
Materials and Methods
Chemicals and Reagents
[14C]-glutamine was purchased from Amersham. Scintillation
fluid, ScintiVers II* was obtained from Fischer (Orlando, FL). DE-
81 anion exchange chromatography paper was purchased from
Whatman (Hillsboro, OR). Other chemicals, including L-glutamine and
L-glutamic-y-monohydroxamate, and hydroxylamine were obtained
from SIGMA Chemical Company (St. Louis, MO) and were of the
highest commercial purity.
Enzyme Preparation
Construction of recombinant wild-type AS-B and the mutant
AS-B in which R30 has been replaced by alanine (R30A) have been
described elsewhere (Boehlein et al., 1994a,b). Both wild-type and
mutant proteins were expressed, and purified using standard

47
procedures (Boehlein et al., 1994a). Protein concentrations were
determined by the method of Bradford (1976) using an assay kit
supplied by Bio-Rad and mouse immunoglobulin G to construct a
standard curve. During the course of these investigations an
alternate method of determining the protein concentration revealed
that the current use of mouse immunoglobulin G as a standard gives
erroneously high protein concentrations for AS-B. Amino acid
composition analysis of AS-B, carried out by the Protein Chemistry
Core Facility of ICBR at the University of Florida, showed that the
actual value for the concentration of this enzyme is 2.7 times smaller
than that obtained using the Bradford assay. Therefore, all of the
values of enzyme concentration and therefore all kcat and kcat/Km
values given in this chapter have been corrected by dividing the
protein concentration by the factor 2.7.
Glutaminase Assays
Initial rates of L-glutamine and L-glutamic-y-monohydroxamate
(LGH) hydrolysis were determined by a modified procedure which
uses glutamate dehydrogenase in the presence of NAD+ to measure
glutamate concentration (Bernt & Bergmeyer, 1974). Reactions (100
pi) containing lOOmM Bis-Tris and Tris-HCl (pH 8.0) each, 8mM MgCl2
and various concentrations of glutamine or LGH were initiated by the
addition of wild-type or mutant R30A AS-B and incubated at various
temperatures (4C-40C). A control reaction lacking enzyme was used
to monitor any glutamate formed from spontaneous breakdown. The
reactions were terminated by the addition of 20 pi of IN acetic acid. A
coupling reagent (300 mM glycine, 250 mM hydrazine, pH 9, 1 mM
ADP, 1.6 mM NAD+ 2.2 units of glutamate dehydrogenase) was added

48
and incubated 10 minutes at room temperature. The absorbance at
340 nm was measured and the concentration of glutamate was
determined by comparison to a standard curve, initial velocities were
measured at 9 different substrate concentrations and each velocity
was an average of 3 measurements. Stability of the wild-type and
mutant AS-B enzymes at each temperature was confirmed by
incubation in buffer of the appropriate temperature for various
amounts of time followed by an assay of activity at pH 8. Initial
velocity conditions were confirmed at each pH by linear plots of v
versus time and v versus enzyme concentration. The pH of the buffer
was adjusted to pH 8 at each temperature studied to account for the
pKa change with temperature.
Determination of for glutamate and hvdroxylamine in the AS-B
catalyzed synthesis of LGH
The Michaelis constant, Km, of glutamate in the synthesis of LGH
was determined by incubating wild-type AS-B with various
concentrations of glutamate and a saturating concentration of
hydroxylamine (150 mM) in Tris-HCl, pH 8.0, at 37C for 15 min. In
a similar manner the Km for hydroxylamine was determined by
incubating the wild-type enzyme with various concentrations of
hydroxylamine and a saturating concentrations of glutamate (150
mM) in Tris-HCl, pH 8.0, at 37C for 20 min. In both cases, the 300 pi
reaction was terminated by the addition of 100 pi of 16 % TCA. LGH
formation was determined in a final volume of 500 pi by adding a
solution containing 80% TCA, 6N HC1. and 10% FeCl3 in 0.02 N HC1 to
the remaining reaction, centrifuging the samples in a microcentrifuge
to remove particulates, and measuring the absorbance of the

49
hydroxamate-FeCl3 complex at 540 nm. A standard curve using a
stock solution of authentic LGH was used to quantitate the product
formed in these reactions.
The ability of AS-B to catalyze the synthesis of glutamine also
was examined. Wild-type AS-B (37 pg) was combined with
glutamate and ammonium chloride (100 mM each) in Tris-HCl (200
mM) in a total volume of 50 pi. The reactions were incubated for 4
hours at 37C followed by the addition of 20 pi of 2N acetic acid. The
terminated reactions were filtered through a 2 pm filter and
derivatized with phenylisothiocyanate (PITC). The glutamine and
glutamate derivatives were separated by HPLC and quantitated
spectroscopically. This assay procedure can detect less than 100
pmoles under our standard procedures
Product inhibition of the glutaminase reaction
Wild-type AS-B was incubated in a 100 pi reaction containing
various concentrations of [14C] glutamine (SA 880-20,000 dpm/nmol)
in lOOmM Bis-Tris and Tris-HCl at 37C for 10 min in the presence of
various concentrations of glutamate or NH4C1. The reactions were
terminated by the addition of 10 pi of IN acetic acid. Glutamate was
separated from glutamine on Whatman DE-81 ion-exchange
chromatography paper using deionized distilled water as the mobile
phase. After 6 hours, elution was stopped, and the paper was dried
and cut into two strips to separate glutamine from glutamate. The
radioactivity associated with each strip was counted in ScintiVersII*
scintillation fluid on a Beckman LS 61C scintillation counter. Control
experiments were carried out in an identical manner except that the
enzyme was excluded from the reaction mixture. The separation of

50
the amino acids was visualized in a non-radioactive control by
spraying the paper with 0.2% ninhydrin in 95% ethanol. Glutamine
was found to migrate with the solvent front while glutamate
remained at the origin. The amount of 14C-labeled glutamate formed
in the reaction was determined by subtracting the percentage of total
radioactive counts located at the origin in the control reactions from
the percentage in each sample.
In a similar manner, the inhibition of glutamate also was
examined at 5C except that the reaction time was 40 min.
Data Analysis
The values of Km and kCat were determined from least-squares
fit to Vo vs [S] plots using the Prism software package supplied by
Graphpad, Inc., (San Diego, CA) according to equation 1:
v =
Vmax[S]
Km + [S]
(1)
Where v is the initial velocity, [S] is the concentration
of substrate, Vmax is the maximal velocity, and Km is the Michaelis
constant.
As shown in the equation below, the value for -Ea/RT can be
derived from the slope of a plot of the log kcat versus 1/T (K) where
A is a constant which is related to the collision frequency and steric
factors.
k = Ae(-Ea/RT) (2)
Values for the energy, enthalpy, and entropy of activation then were
calculated using the following equations:
AH* = Ea -RT
AG* = RT ln(kcT/h) RTln(k)
(3)
(4)

51
AG* = AH* TAS* (5)
where AH* is the enthalpy of activation, R is the gas constant, kB is
boltzmans constant, h is Plancks constant, k is the forward rate
constant; AG* is the free energy of activation, and AS* is the entropy of
activation.
Results
Effect of temperature on glutamine and LGH hydrolysis catalyzed by
wild-tvpe AS-B and the AS-B mutant R30A
The steady-state kinetic parameters of glutamine and LGH
hydrolysis were determined at a range of temperatures from 5 C to
40 C. The Arrhenius plots (Figures 3.1 & 3.2) were observed to be
linear indicating either that there was no change in the rate
determining step over the temperature range studied or that the
activation energies for the limiting steps were very similar.
The temperature dependence of the catalytic rate constants of
the R30A-catalyzed hydrolysis of glutamine also displayed linearity
in the Arrhenius plot, as shown in Figure 3.3. The activation
parameters are given in Table 3.1.
LGH synthesis catalyzed bv wild-tvpe AS-B
AS-B was found to catalyze the formation of LGH from
glutamate and hydroxylamine. The reaction rate of 0.29 sec"1 (Table
3.2, Figures 3.4 & 3.5) was about 12% of that of the corresponding
forward rate of 2.4 sec'1 at pH 8 (see chapter 2, Table 2.3), and was
fast enough to obtain the kinetic parameters of LGH synthesis
catalyzed by wild-type AS-B (Table 3.2).
The steady-state kinetic parameters of reversible reactions can

52
be used to calculate the equilibrium constant for the overall reaction
using the Haldane relationship for a one-substrate/two-product
reaction given in equation (6) (Cleland, 1963):
(6)
where Keq is the equilibrium constant for the conversion of substrate
to product, Vf and Vr represent the forward and reverse reaction
rates, respectively, and the indices p, q, and a refer to hydroxylamine,
glutamate and LGH, respectively. Using equation (6) and the
following parameters, obtained at 37C: krcat = 2.4 sec'1, Km(NH20H)= 95
mM, Ki(glu)=182 mM, krcat= 0.29 sec'1, K,
0.83 mM; a value of
m(LGH)-
Keq=172 was calculated. Using this value for Kcq and the equation, AG
= -RTlnKeq, the difference in free energy between substrates and
products, AG, was determined to be -13.2 kJ/mol K.
Product inhibition glutamine hydrolysis catalyzed bv wild-tvpe AS-B
Figures 3.6 and 3.7 show the plots for the inhibition of the
glutaminase reaction by glutamate at 5C and 37C, respectively. The
K¡ values for glutamate determined from the replot of the slope
versus glutamate concentration (Figure 3.8 & 3.9) at 5C and 37C
were 260 mM and 182 mM, respectively. An attempt to obtain the K¡
for ammonia was made. However, NH4C1 concentrations up to 200
mM seemed to have little effect on the rate of the glutaminase
reaction.

53
Figure 3.1: Temperature dependence of glutamine hydrolysis
catalyzed by wild-type AS-B. The initial rates were determined at
pH 8.0 in 100 mM Bis-Tris and Tris-HCl as described in the materials
and methods.

54
Figure 3.2: Temperature dependence of LGH hydrolysis catalyzed by
wild-type AS-B. The initial rates at each temperature were
determined at pH8.0 in 100 mM Bis-Tris and Tris-HCl as described in
materials and methods.

55
1/T (K)
Figure 3.3: Temperature dependence of glutamine hydrolysis
catalyzed by the AS-B mutant R30A. The initial rates at each
temperature were determined at pH 8.0 in 100 mM Bis-Tris and
Tris-HCl as described in materials and methods.

56
Table 3.1: Thermodynamic properties of glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and the R30A mutant.
Substrate/
AG*
AH*
AS*
TAS*
enzyme
kJ mol1
kJ mol 1
J mol 'K1
kJ mol"1
(37C)
(37C)
(5-40C)
(37C)
gln/wt
1 67
89
-252
-78
LGH/wt
152
82
-226
-70
gln/R30A
147
77
-226
-70
The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and
Tris-HCl at temperatures ranging from 5-40C, under conditions
described in the materials and methods.

57
Figure 3.4: Effect of hydroxylamine concentration on the synthesis of
LGH catalyzed by wild-type AS-B. Wild-type AS-B (18.5 pg) was
incubated with various concentrations of hydroxylamine and
saturating concentrations of glutamate (150 mM) in Tris-HCl, pH 8.0
at 37C for 20 min. The reaction was terminated by the addition of
16% TCA and LGH formation was measured as described in materials
and methods.

58
Figure 3.5: Effect of glutamate concentration on the synthesis of LGH
catalyzed by wild-type AS-B. Wild-type AS-B (18.5 pg) was
incubated with various concentrations of glutamate and a saturating
concentration of hydroxylamine (150 mM) in Tris-HCl, pH 8.0 at 37C
for 15 min. The reactions were terminated by the addition of 16%
TCA and the formation of LGH was measured as described in
materials and methods.

59
Table 3.2: Kinetic constants for the synthesis of LGH catalyzed by
wild-type AS-B.
Substrate
kcat
(sec"1)
Kni (mM)
kcat/Kra (M-1 sec"1)
glutamate
0.36
0.02
76 5
4.7
hydroxylamine
0.29
0.01
95 7
2.9
The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and
Tris-HCl at 37C, by varying one substrate while keeping the second
substrate saturating. The conditions of the reaction and the
determination of the LGH formed in the reaction are described in the
materials and methods.

60
1/[Gln] mM'1
Figure 3.6: Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 5C. The rate of glutamine hydrolysis
was determined as described in the materials and methods in the
presence of 0 (0), 10 (A), 50 (), 100 (), and 136 () mM
glutamate.

61
Figure 3.7: Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 37C. The rate of glutamine
hydrolysis was determined as described in the materials and
methods in the presence of 0 (), 40 (), 80 (A), and 100 () mM
glutamate.

slope
62
Figure 3.8: Replot of the slope from the double-reciprocal plots of
glutamine hydrolysis at 5C versus glutamate concentration.

slope
63
Figure 3.9: Replot of the slope from the double-reciprocal plots of
glutamine hydrolysis at 37C versus glutamate concentration.

64
Discussion
Many of the experiments in the following chapters required a
reaction rate slower than that observed at 37C, the standard
temperature at which the reaction is examined. One way to obtain a
slower rate is to examine the reaction at low temperatures therefore,
many of the experiments in the following chapters have been
conducted at 5C. In preparing for these experiments it was important
to know whether the results obtained at 5C could be compared to
those at the standard temperature. Temperature variations often
cause changes in the rate-limiting step of a reaction. Since non-linear
Arrhenius plots are indicative of such changes, the temperature
dependence of the glutaminase reaction was examined to confirm that
there was not a change in the rate limiting step between 5C and 37C.
In the case of the glutaminase reaction, it is difficult to conclude that a
single step is rate determining across the entire range of temperatures
studied. Certainly, the linear Arrhenius plots suggest that a change in
the rate limiting step between reactions at 5C and 37C is unlikely.
However, it should be noted that two steps in the reaction may have
similar activation energies, and therefore a change in rate limiting step
would remain unnoticed in these experiments. Such an event is a
possibility in AS-B catalyzed glutamine hydrolysis as demonstrated by
the unexpected similarity in activation constants for reactions
catalyzed by wild-type and the mutant R30A (Table 3.1 and Figures
3.1 and 3.3). Partitioning experiments in chapter 4 imply that in
reactions catalyzed by R30A, hydrolysis of the thioester is rate
limiting. In contrast, direct measurements of the deacylation step of
glutamine hydrolysis suggest that an additional step contributes to the

65
overall rate of the reaction catalyzed by the wild-type enzyme. The
differences in the nature of the rate determining step in the reactions
catalyzed by the wild-type and mutant enzyme were expected to be
reflected in the activation parameters. The lack of any differences in
the parameters suggests that the two slow steps in the wild-type
catalyzed reaction have similar activation energies.
The patterns of product inhibition can provide information on
the order of product release. However, the studies of product
inhibition presented here were inconclusive due to the inability of
ammonia to inhibit the glutaminase reaction. On the other hand, the
examination of glutamate inhibition provided a needed inhibition
constant which will be used in the model described in chapter 7.
The inability of ammonia to inhibit the glutaminase reaction
and the lack of detectable glutamine synthesis using ammonia and
glutamate as substrates suggests that the acylation step is essentially
irreversible under the conditions of the examination. With the
ability to detect glutamine concentrations to 0.1 nmol (see materials
and methods), an upper limit on the turnover number of kcat< 5x1 O'4
sec'1 can be set for glutamine synthesis at 37C catalyzed by the
wild-type AS-B under the conditions studied. In contrast, the
reversibility of LGH hydrolysis catalyzed by wild-type AS-B was
easily demonstrated and has a kcat which is 12% of the forward
reaction. The difference in the ability to synthesize the reverse
reaction may be a reflection of the greater nucleophilicity of
hydroxylamine. The kinetic parameters, kcat and K^, obtained from
the analysis of LGH synthesis allowed the determination of the
equilibrium constant for the overall reaction.

66
The free energy, AG, derived from this equilibrium constant was
calculated to be -13.2 kJ/mol.

CHAPTER 4
DETERMINATION OF THE RATE DETERMINING STEP IN ASPARAGINE
SYNTHETASE B-CATALYZED GLUTAMINE HYDROLYSIS
Introduction
Asparagine synthetase B (AS-B) is thought to be organized into
two functional domains: the glutamine amide transfer (GAT) domain,
which catalyzes the hydrolysis of glutamine to glutamate; and the
synthetase domain, which catalyzes the formation of a |3-aspartyl-
AMP intermediate and the synthesis of asparagine. While the overall
reaction involves three substrates, four products, and possibly twenty
rate constants; the ability of the enzyme to catalyze glutamine
hydrolysis in the absence of the other substrates provides the
opportunity to examine glutamine utilization under a simpler one
substrate, two product reaction. A detailed description of this
reaction would provide a useful tool in understanding the roles
played by conserved amino acid residues of the GAT domain in
glutamine utilization. Therefore, this chapter presents a first step
towards characterizing the glutaminase reaction catalyzed by AS-B.
Analysis of the amino acid sequence of the GAT domain suggests
that AS-B belongs to a family of enzymes known as the glutamine
amidotransferases (Zalkin, 1993). Like AS-B, the enzymes of this
family catalyze nitrogen transfer using glutamine as the primary
source of nitrogen. While all enzymes in the glutamine
amidotransferase family contain a cysteine which is essential for
67

68
glutamine dependent activity, the position of this cysteine as well as
other sequence differences in the GAT domain further divides the
family into two classes. The essential cysteine residue is located
within the GAT domain for class I enzymes whereas it resides at the
N-terminus in the class II enzymes such as glutamine
phosphoribosylpyrophosphate amidotransferase (GPA) (Tso et al.,
1982) glucosamine-6-phosphate synthetase (GFAT) (Badet et al.,
1987), and asparagine synthetase (Humbert et al., 1980). Initially,
site directed mutagenesis studies suggested that the N-terminal
cysteine as well as a histidine and aspartate residue were involved in
the reaction catalyzed by GPA (Mei & Zalkin, 1989). As this catalytic
triad is found in many thiol proteases (Brocklehurst et al., 1987), it
was proposed that the glutamine hydrolysis reaction catalyzed by
the class II enzymes occurs via a mechanism analogous to that of
peptide hydrolysis catalyzed by the thiol proteases, such as papain.
This hypothesis was extended to the class I enzymes as well (Mei &
Zalkin, 1989). Since then, the importance of a catalytic triad
composed of a conserved cysteine, histidine, and glutamate residues
has been demonstrated for the class I enzymes through site-directed
mutagenesis (Amuro et al., 1985; Miran et al., 1991) and x-ray
crystallography (Tesmer et al., 1996; Thoden et al., 1997).
On the other hand, the picture remains more ambiguous for the
class II amidotransferases as subsequent studies confirmed only the
importance of the conserved N-terminal cysteine residue (Boehlein et
al., 1994a). In fact, a more recent alignment of the GAT domain
which includes a greater number of amidotransferase sequences
reveals that there are no histidines which are conserved across this

69
entire class of enzymes (Boehlein et al., 1994a; Zalkin, 1993).
Moreover, studies using site-directed mutagenesis of AS-B (Chapter
2; Boehlein et al., 1994a) and x-ray crystallography of GFAT (Isupov
et al., 1996) and GPA (Kim et al., 1995) suggest that the class II
amidotransferases function differently than papain with respect to
acid-base catalysis. While the lack of a conserved histidine in the
class II amidotransferases contrasts with the essentiality of such a
residue in thiol proteases, recent evidence for an acyl enzyme
intermediate (chapter 5 & 6) formed during glutamine hydrolysis
catalyzed by AS-B provides a basis for comparison of the hydrolysis
reactions catalyzed by the two enzyme families. A minimal
description of amide hydrolysis catalyzed by papain, which occurs
via a thioester intermediate (E-TE), is shown in equation 1 below:
(equation 1)
+ P
In relating this equation to glutamine hydrolysis catalyzed by AS-B,
ES is the enzyme-substrate complex, E-TE is the acylenzyme
intermediate, and P, and P, are ammonia and glutamate,
respectively. In an effort to understand the mechanism of glutamine
hydrolysis catalyzed by AS-B, the glutaminase reaction will be
examined in order to develop a model for this particular aspect of
the enzymatic reaction. The minimal model shown above will be
used as a starting point in this investigation to provide evidence (in
addition to that shown in chapter 5 for the existence of a thioester
intermediate, to determine whether steps leading to its formation or

70
those involved in its breakdown are rate limiting, and to compare
and contrast catalytic behavior of three groups of hydrolytic
enzymes: a class II amidotransferase (AS-B), a class I
amidotransferase (CAD), and a thiol protease (papain).
Materials and Methods
Chemicals and Reagents
14C-glutamine was purchased from Amersham. Scintillation
fluid, ScintiVers II* was obtained from Fischer (Orlando, FL). DE-
81 anion exchange chromatography paper was purchased from
Whatman (Hillsboro, OR). Other chemicals, including L-glutamine and
L-glutamic-y-monohydroxamate, hydroxylamine, and 6-diazo-5-
oxonorleucine (DON) were obtained from SIGMA Chemical Company
(St. Louis, MO) and were of the highest commercial purity.
Enzymes
The construction of recombinant AS-B and the AS-B mutant,
R30A, has been described previously (Boehlein et al., 1994a, 1994b).
The enzymes were obtained by overexpression and purified using
standard procedures (Boehlein et al., 1994a). Protein concentration
was determined by the method of Bradford (1976) using an assay kit
supplied by Biorad and immunoglobulin G as a standard. Recent
examination of the protein concentration using amino acid analysis of
AS-B, performed by the Protein Core Facility of ICBR at the
University of Florida, gave a value for protein concentration which
was 2.7-fold lower than that obtained with the previous method.
Therefore, in this report the enzyme concentration has been
corrected by dividing the enzyme concentrations by the factor 2.7.

71
Steady State Kinetics
The kinetic parameters of AS-B-catalyzed glutamine, LGH, and
L-glutamine-hydrazide hydrolysis were obtained in an end-point
assay which uses glutamate dehydrogenase in the presence of NAD+
to measure the concentration of glutamate formed (Bernt &
Bergmeyer, 1974). Reactions (100 pi) containing lOOmM Bis-Tris and
Tris-HCl (pH 7.5) each, 8mM MgCl2 and various concentrations of
glutamine, LGH, or glutamic acid-y-hydrazide were initiated by the
addition of 1.5 pg wild-type AS-B and incubated at 37C. After 10
minutes, the reactions were terminated by boiling for 2 minutes.
Reactions in which the enzyme was added to the mix during the
termination period of boiling were used to monitor any glutamate
formed from spontaneous breakdown. A coupling reagent (300 mM
glycine, 250 mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2
units of glutamate dehydrogenase) was added to the reaction and the
resulting mixture was incubated 10 minutes. The absorbance at 340
nm was measured and the concentration of glutamate was
determined by comparison to a standard curve. Initial velocities
were measured at 9 different substrate concentrations and each
velocity was an average of 2-4 measurements.
The values of Km and kcat were determined from least-squares
fit to v0 vs [S] plots using the Prism software package supplied by
Graphpad, Inc., (San Diego, CA).
Determination of the K. for glutamine and LGH
The rate of inactivation by the glutamine analog, 6-diazo-5-
oxonorleucine (DON), was determined by incubating 30 pg wt AS-B in
a 300 pi solution containing 100 mM Bis-Tris and Tris-HCl, pH 8 and

72
varying concentrations of DON at 5C. At various time points 20 pi
aliquots were removed from the incubation mixture and the residual
activity was measured in 200 pi buffer containing 12 mM glutamine.
Control experiments measuring the activity in the presence of diluted
inhibitor concentrations verified that DON did not interfere in the
assay of residual activity.
Glutamine protection of DON inhibition was examined as
described above except that wt AS-B was incubated with varying
concentrations of DON in the presence of either 20 pM or 50 pM
glutamine. Alternatively, LGH protection of DON inactivation of the
wild type enzyme and glutamine protection of R30A inactivation
were examined by incubating wild-type or R30A AS-B with a
constant concentration of DON (6 pM for wild type and 650 pM for
R30A) and varying concentrations of LGH or glutamine, respectively.
An analysis of glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time.
Glutamine hydrolysis catalyzed by wt AS-B was examined at
5C, a temperature at which the reactions were slow enough to mix
by hand. The reactions were initiated by adding 0.37 or 0.74 nmol
wt AS-B to a 50 pi solution containing 100 mM Bis-Tris and Tris-HCl
(pH 8) and [14C]-glutamine (2mM, SA = 9 nCi/pl). After the
appropriate amount of time, the reactions were terminated with 20
pi of 2N acetic acid. Glutamate was separated from glutamine on
Whatman DE-81 ion-exchange chromatography paper using
deionized distilled water as the mobile phase. After 6 hours, elution
was stopped, and the paper was dried and cut into two strips to
separate glutamine from glutamate. The radioactivity associated

73
with each strip was counted in ScintiVersII* scintillation fluid on a
Beckman LS 61C scintillation counter. Control experiments were
carried out in an identical manner except that the enzyme was
excluded from the reaction mixture. The separation of the amino
acids was visualized in a non-radioactive control by spraying the
paper with 0.2% ninhydrin in 95% ethanol. Glutamine was found to
migrate with the solvent front while glutamate remained at the
origin. The number of nanomoles of 14C-labeled glutamine in each
sample was determined by subtracting the percentage of total
radioactive counts located at the origin in the control reactions from
the percentage in each sample. All values are derived from the
average of four separate reactions.
To examine the effect of glutamate on the burst, wild-type AS-B
was incubated on ice either alone or with 100 mM L-glutamate, or D-
Glutamate for 3 hours prior to the reaction. The enzyme/glutamate
mixture then was added to the reaction and product formation was
determined at various time points as described above.
Hvdroxvlamine Partitioning of the Thioester
A reaction mixture containing either wt AS-B (9.3 pg), or R30A
(18.5 pg), and 50 mM glutamine, 100 mM Bis-Tris and Tris-HCl (pH
8) and a variable concentration of hydroxylamine in a total volume
of 300 pi was incubated at 37C for 45 minutes. The reaction was
terminated by the addition of 100 pi of 16% TCA and a 50 pi aliquot
was removed to determine the amount of glutamate formed using
the previously mentioned GDH assay. LGH formation was determined
in a final volume of 500 pi by adding a solution containing 80% TCA,
6N HC1, and 10% FeCl? in 0.02 N HC1 to the remaining reaction,

74
and measuring the absorbance of the hydroxamate-FeCl3 complex at
540 nm. A standard curve using a stock solution of authentic LGH
was used to quantitate the product formed in these reactions.
Results
Partitioning of a thioester intermediate to v-glutamyl hvdroxamate
The minimal mechanism, given in equation 1, requires the
formation of a covalent intermediate. Due to the highly reactive
nature of hydroxylamine towards acyl groups, this agent has been
used to provide evidence for an acylenzyme intermediate and to
determine whether its formation or breakdown is rate limiting in
hydrolysis reactions catalyzed by serine proteases (Caplow et al.,
1964; Epand et al., 1963) as well as by the amidotransferase CAD
(Chaparan et al., 1991). Since hydroxylamine competes with H20 in
the deacylation step, its presence will cause an increase in the amount
of total product formed (hydroxamate plus glutamate) only if this
same step is rate limiting (reaction 2). In contrast, if the acylation
step is very slow then the presence of hydroxylamine will not have
an effect on the amount of total product produced as the slow rate of
the acylation step will limit the amount of thioester available for
hydrolysis and hydroxaminolysis.
H20 E + Glu
K,
E + Gln
(Reaction 2)
E + LGH

75
In an attempt to show that AS-B catalyzed hydrolysis occurs
via a thioester intermediate and to determine which step in
hydrolysis is rate limiting, the effect of various concentrations of
hydroxylamine on the glutaminase reaction was examined. As
shown in Figure 4.1, even very low concentrations of hydroxylamine
were sufficient to partition the reaction towards the production of
glutamate hydroxamate. Unexpectedly however, an increase in
hydroxylamine concentration in the AS-B catalyzed glutaminase
reaction caused a decrease in total product formed (figure 4.1).
One potential cause for this decrease in total product formed in
the presence of hydroxylamine is a slow release of L-glutamate
hydroxamate due to its tight binding to the enzyme. To test this
idea, the partitioning experiment was repeated using an AS-B mutant
(R30A) which exhibits poor substrate binding. A role in substrate
binding has been implied for Arginine-30 in AS-B (Boehlein et al.,
1994b) and the cognate arginine in GFAT (Isupov et al., 1996) in
studies using site-directed mutagenesis and x-ray crystallography,
respectively. Use of R30A in the partitioning experiments resulted in
an increase in total product with increasing concentrations of
hydroxylamine (figure 4.2).
Comparison of AS-B catalyzed hydrolysis of three substrates
The kinetic constants kcat, Km, and kcat/Km for the AS-B catalyzed
hydrolysis of L-glutamine, L-glutamic acid-y-monohydroxamate, and
L-glutamic acid-y-hydrazide are given in Table 4.1. As shown in the
table, the kcat values for the three substrates were similar while each
displayed different kcat/Km values.

76
[NH2OH] (mM)
Figure 4.1: Partitioning of the thioester intermediate to L-glutamic-
y-monohydroxamate using hydroxylamine in a reaction catalyzed by
wt AS-B. A mixture containing wt AS-B, 50 mM glutamine, 100 mM
Bis-Tris and Tris-HCl (pH 8), and the indicated concentrations of
hydroxylamine was incubated at 37C for 45 min. The reactions
were terminated by the addition of 4% TCA. An aliquot was removed
to determine glutamate formed () by an end-point assay using
glutamate dehydrogenase. LGH formation (A) was measured by
adding a FeCl2 solution and measuring the absorbance at 540 nm as
described in materials and methods. The nmol of LGH and glutamine
formed at each concentration of hydroxylamine were added to give
the total product formed (sa).

77
[NH2OH] mM
Figure 4.2: Partitioning of the thioester to glutamate hydroxamate
using hydroxylamine in the glutaminase reaction catalyzed by AS-B
R30A. A mixture containing wt AS-B, 80 mM glutamine, 100 mM
Bis-Tris and Tris-HCl (pH 8), and the indicated concentrations of
hydroxylamine was incubated at 37C for 45 min. The reactions
were terminated by the addition of 4% TCA. An aliquot was removed
to determine glutamate formed (A) by an end-point assay using
glutamate dehydrogenase. LGH formation () was measured by
adding a FeCl, solution and measuring the absorbance at 540 nm as
described in materials and methods. The nmol of LGH and glutamine
formed at each concentration of hydroxylamine were added to give
the total product formed (+).

78
TABLE 4.1:
Kinetic constants for wt AS-B-catalyzed hydrolysis of L-glutamine, L-
glutamic acid-y-monohydroxamate, and L-glutamic acid-y-hydrazide.
Substrate
k^. (sec'1)
K,an;iM)
KJK (M-1 sec1)
L-glutamine
1.9 0.03
1.0 0.06
1900
L-glutamic-y-monohydroxamate
2.5 0.02
0.65 0.02
3846
L-glutamic-y-hydrazide
2.0 0.05
2.3 0.20
869

79
Determination of the Dissociation constants for glutamine and LGH
using the covalent modifier. 6-diazo-5-oxonorIeucine (DON).
The diazo ketone analogue of glutamine, DON, has been shown
to inhibit the glutaminase activity of bovine AS in such a manner
that the effects could be eliminated only by the addition of glutamine
and not by desalting (Mehlhaff & Schuster, 1991). Furthermore, the
x-ray structure of the glutamine active site labeled by DON in
another class II amidotransferase, E. coli GPA (Kim et al., 1996), has
provided evidence that this analog alkylates the cys-1 residue in a
reaction which is comparable to the formation of the glutamyl-
enzyme intermediate. Using the method of Meloche (1967), the
affinity analogue, DON, has been used in the determination of the
glutamine dissociation constant for the glutaminase reaction
catalyzed by anthranilate synthetase (Goto et al., 1976), CAD
glutaminase (Chaparan & Evans, 1991) and the amidotransferase
PabA/PabB complex (Roux & Walsh, 1992). Likewise, the ability of
DON to inactivate wild-type and the R30A mutant of AS-B was
examined and the effect of glutamine or LGH on the inactivation was
used to determine the dissociation constant of glutamine and LGH.
Incubation of wild type AS-B with DON prior to the measurement of
the glutaminase activity caused a time-dependent loss of activity
over a period of 20 minutes. The semilogrithmic plot of the residual
activity versus time shows that DON inactivates AS-B with pseudo-
first order kinetics (figure 4.3). This plot also shows that inactivation
is irreversible since the residual activity of the enzyme after it had
been incubated with DON was determined using 20 mM glutamine,
providing a 1400-fold dilution. The inactivation half-times, x, were

80
determined from these plots and were used in a replot to determine
the K¡ for DON according to the following equation described by
Meloche (1967):
(1)
T= (TK:) + T
[I]
where x is the time required for the inhibitor to cause a 50% loss of
activity (x= ln2/first order rate constant for inactivation); [I] is the
inhibitor concentration; T is the time required for one-half inhibition
at infinite inhibitor concentration; and K¡ is the inhibition constant for
the inhibitor. When measuring the residual AS-B glutaminase
activity after incubating the enzyme with DON, the replot (x vs 1/[I])
showed a straight line which extrapolated to a minimum inactivation
half-time, T, of 1.22 minutes (figure 4.4). The K¡ for DON determined
from the slope of this replot was 34 pM.
The dissociation constant for glutamine, Ks, was determined by
measuring the competitive binding of glutamine and DON. Therefore,
DON inactivation was examined in the presence of 20 pM and 50 pM
glutamine (Figures 4.5 and 4.6). Using the following equation
(Meloche, 1967):
(2)

81
time (min)
[DON]
a 4 (iM
a 6 (i M
8 uM
o 1 0 (iM
1 2 (iM
1 4 (iM
Figure 4.3: DON inactivation of AS-B in the absence of glutamine.
Wild type AS-B was incubated with the concentrations of DON given
in the figure. At time points ranging from 0-20 minutes, an aliquot
was removed from the mixture and tested for residual activity in a
reaction containing 12 mM glutamine. The glutaminase activities of
the DON modified enzymes were measured as described in materials
and methods.

82
Figure 4.4: Replot of DON inactivation of AS-B in the absence () and
presence of glutamine (A, 20 pM; #, 50 pM).

83
a replot of x vs 1/[I] at each concentration of glutamine gave straight
lines with a different slope for each glutamine concentration (Figure
4.4). The slope of the line resulting from 20 pM glutamine is 50.73
and that from 50 pM glutamine is 65.02 corresponding to Ks values of
140 |iM and 200 pM respectively.
In a similar manner, DON inactivation and LGH protection
against inactivation was examined for wild type AS-B, and DON
inactivation and glutamine protection against inactivation was
examined for the AS-B mutant R30A (Figures 4.7-4.14). Using
equation (1) and the replot of t vs l/[DON] the minimum inactivation
half-time, T, for DON inactivation of wild type catalyzed LGH
hydrolysis (Figure 4.8) and R30A catalyzed glutamine hydrolysis
(Figure 4.12) were 1.4 sec and 3.5 sec, respectively. The K¡ for DON in
these two reactions, determined from the slope of the replot was 28
pM for LGH hydrolysis and 8.7 mM for inactivation of R30A catalyzed
glutamine hydrolysis. The following equation was used to determine
the dissociation constant for LGH in the wild type catalyzed
glutaminase reaction and for glutamine in the R30A catalyzed
reaction:
(3)
The Ks, derived from the slope of the line in a replot of x vs
[substrate] for LGH in the wild type catalyzed hydrolysis reaction
(Figure 4.10) was 70 pM and for glutamine in the R30A catalyzed
reaction (Figure 4.14) was 9.5 mM

84
Time (min)
[DON]
a 4 jllM
a 6 fiM
8 jiM
* 10 fiM
12 [iM
14 jiM
Figure 4.5: DON inactivation of AS-B in the presence of 20 pM
glutamine. Wild type AS-B was incubated with the concentrations of
DON given in the figure. At time points ranging from 0-20 min, an
aliquot was removed from the mixture and tested for residual
activity in a reaction containing 12 mM glutamine. The glutaminase
activities of the DON modified enzymes were measured as described
in materials and methods.

85
[DON]
a 4 (iM
a 6 ¡aM
A 8 llM
10 pM
12 llM
14 (iM
Figure 4.6: DON inactivation of AS-B in the presence of 50 pM
glutamine. Wild type AS-B was incubated with the concentrations of
DON given in the figure. At time points ranging from 0-20 min, an
aliquot was removed from the mixture and tested for residual
activity in a reaction containing 12 mM glutamine. The glutaminase
activities of the DON modified enzymes were measured as described
in materials and methods.

86
[DON]
4 jliM
6 jiM
a 8 jiM
* 10
Figure 4.7: DON inactivation of wild-type catalyzed LGH hydrolysis in
the absence of LGH. Wild type AS-B was incubated with the
concentrations of DON indicated in the figure. At time points ranging
from 0-18 min an aliquot was removed from the mixture and tested
for residual activity in a reaction containing 12 mM glutamine. The
glutaminase activities of the DON modified enzymes were measured
as described in materials and methods.

87
1/[D0N] |iM1
Figure 4.8: Replot of DON inactivation of AS-B catalyzed LGH
hydrolysis. The K¡ for DON, determined from the slope of this plot and
equation (1), was 28 pM.

88
[LGH]
O jiM
a 20 |iM
a 40 jiM
v 60 jiM
80 jiM
100 |XM
Figure 4.9: The effect of various concentrations of LGH on DON
inactivation of AS-B catalyzed LGH hydrolysis. Wild-type AS-B was
incubated with 6 pM DON in the presence of LGH (concentrations
varied as indicated in the figure). At time points ranging from 0-18
min, an aliquot was removed from the mixture and tested for
residual activity in a reaction containing 12 raM LGH. The LGH
hydrolysis activity of the DON modified enzymes were measured as
described in materials and methods.

89
Figure 4.10: LGH protection against DON inactivation of AS-B
catalyzed LGH hydrolysis. The Ks determined from the slope of the
line and equation (3), was 70 pM.

90
[DON] jiM
+ 350
450
a 500
f 600
O 650
. 750
Time (min)
Figure 4.11: DON inactivation of the R30A mutant of AS-B in the
absence of glutamine. R30A was incubated with the concentrations
of DON given in the figure. At time points ranging from 0-18 min, an
aliquot was removed from the mixture and tested for residual
activity in a reaction containing 12 mM glutamine. The glutaminase
activities of the DON modified enzyme were measured as described
in materials and methods.

O d
91
Figure 4.12: Replot of DON inactivation of R30A in the absence of
glutamine. The K¡ for DON, determined from the slope of the line and
equation (1), was 8.7 mM.

92
Time (min)
[Gin] mM
a
ir
o
2
A
5
+-
7
10

15

20
X
25
V
30
Figure 4.13: Glutamine protection from DON inactivation of the R30A
catalyzed glutaminase reaction. R30A was incubated with 0.65 mM
DON and various concentrations of glutamine (given in the figure).
At time points ranging from 0-18 min, an aliquot was removed from
the mixture and tested for residual activity in a reaction containing
12 mM glutamine. The glutaminase activities of the DON modified
enzyme were measured as described in materials and methods.

93
Figure 4.14: Replot of glutamine protection of DON inactivation of
R30A catalyzed glutaminase. The Ks, determined from the slope of
the line and equation (3), was 9.5 mM.

94
A step after the formation of the thioester intermediate is rate
determining in glutamine hydrolysis catalyzed bv AS-B
In order to determine which step is rate limiting in the
minimal mechanism, the production of glutamate was studied as a
function of time. A burst of product formation during the first
turnover of the enzyme is an indication that the slow step follows the
appearance of the thioester intermediate.
The time-course of glutamate production was examined at 5C,
a temperature at which the reaction was slow enough to mix by
hand. A graph of nmol of glutamate formed versus reaction time
(figure 4.15) shows that the rate limiting step occurs after the
formation of the thioester intermediate as extrapolation of the line to
time zero intercepts the y-axis at a point above zero. This point of
intersection should represent the concentration of the dominant
intermediate, possibly the thioester, and therefore a time course
using a higher concentration of enzyme should intercept the y-axis at
a higher value. In fact, a higher y-axis intercept is observed with
double the concentration of enzyme (Figure 4.15). The burst of
product formed in the first turnover of an enzymatic reaction occurs
because the steps leading to product formation are fast relative to
those which follow. In the minimal mechanism proposed for AS-B,
the first appearance of glutamate occurs upon the formation of the
thioester intermediate. Therefore, it should be possible to eliminate
the appearance of the burst by forcing the reaction to begin at an
enzyme form which would require passing through the rate
determining step prior to the first appearance of glutamate. As
shown in Figure 4.16, the burst in glutamate formation was abolished

95
Figure 4.15: Glutamate formation in AS-B catalyzed glutamine
hydrolysis as a function of time. The progress curve shows the nmol
of glutamate formed versus time using 2 mM [14C]-glutamine
(SA=9nCi/pl), and either 0.37 nmol () or 0.74 nmol (A) of wt AS-B
in 100 mM Bis-Tris and Tris-HCl (pH 8). After terminating the
reaction with acetic acid, the glutamine and glutamate were
separated by anion exchange chromatography as described in the
materials and methods and the radioactivity associated with each
amino acid was counted. All values are derived from the average of
four separate determinations.

96
Time (sec)
Figure 4.16: Effect of the presence of L-glutamate on the burst in
product formation in AS-B catalyzed glutamine hydrolysis. The
figure shows the nmol of glutamate formed versus time in reactions
where wild-type AS-B was incubated alone (+) or with 100 mM L-
glutamate () or D-glutamate (A) prior to addition into the reaction.

97
Figure 4.17: Mechanism of glutamine hydrolysis catalyzed by the
class II glutamine amidotransferases as proposed by Mei and Zalkin
(1989).

98
in a time course experiment in which the enzyme was incubated with
L-glutamate prior to its addition to the reaction. The observation
that D-giutamate has no effect on the burst suggests that salt effects
are not a cause of the disappearance of burst in the presence of L-
glutamate.
Discussion
This report describes the characterization of the AS-B catalyzed
glutaminase reaction by a comparison of its kinetic behavior to that
of papain and CAD. The mechanism previously proposed for
glutamine hydrolysis catalyzed by the class II amidotransferases
(Mei & Zalkin, 1989) (figure 4.17), in analogy to that of papain,
involves a nucleophilic attack of the N-terminal cysteine on the side
chain of glutamine to form a tetrahedral intermediate (1).
Subsequent protonation of the leaving group allows the release of
ammonia and the formation of a thioester intermediate (2). An
attack of a water molecule on the thioester then causes the formation
of a second tetrahedral intermediate (3) and finally the production
of free enzyme and glutamate (4). Using the minimal model in
equation 1, it has been found that for papain the formation of the
thioester (k2) is rate limiting when amide substrates are used. This
observation is probably due to the greater leaving group ability of
the thiol group relative to the amine or the hydroxyl group of the
first and second tetrahedral intermediate, respectively (OLeary et
al., 1974; Wells et ak, 1963). This situation causes the reaction to
favor return from the first tetrahedral intermediate to free enzyme
and substrate whereas it favors forward progression from the second
tetrahedral intermediate to free enzyme and product.

99
In contrast to the above observations for papain, evidence is
presented in this study for a rate limiting thioester breakdown for
AS-B catalyzed glutamine hydrolysis. The similarity in the kcat
values for three substrates with different leaving groups suggests
that a step following the release of the leaving group and thus after
the appearance of a common intermediate is rate limiting (Table 4.1).
Moreover, a burst in glutamate production with time indicates that
an intermediate containing a glutamate moiety is formed quickly
followed by a slow breakdown to product and free enzyme (Figure
4.15). The presence of the burst in the production of glutamate was
verified by the ability to eliminate it when the enzyme was forced
into an enzyme-product form (Figure 4.16). It is possible that
glutamate could bind the enzyme and form a thioester intermediate,
therefore the disappearance of the burst in the presence of
glutamate does not reveal which step after thioester formation is
rate limiting. On the other hand, under the acidic conditions of the
reaction termination (see materials and methods), it is likely that the
acylenzyme intermediate remained intact and precipitated in the
acidic solution. Therefore, the radioactive glutamate counted after
paper chromatography represented only the glutamate which was
not covalently attached to the enzyme. If this is true, then the burst
suggests that a step occurring after deacylation is rate limiting.
Further work to address this issue will be presented in subsequent
chapters.
While the experiments described above suggest a rate
determining step following the formation of a thioester intermediate,
the results from hydroxylamine partitioning studies were more

100
complicated. Indeed, additional evidence for the presence of a
thioester intermediate along the reaction pathway of AS-B catalyzed
glutamine hydrolysis is given in the formation of LGH upon addition
of hydroxylamine to the glutaminase reaction. However, an increase
in the hydroxylamine concentration in the reaction was expected to
either increase the total product formed, if thioester breakdown (k3)
were rate limiting, or to have no effect on the amount of product
formed, if thioester formation (k2) were rate limiting. For example,
hydroxylamine partitioning experiments have been used to show
that deacylation is rate limiting in glutamine hydrolysis catalyzed by
a class I amidotransferase, CAD (Chaparian & Evans., 1991). In
contrast, an increase in hydroxylamine concentration in the AS-B
catalyzed glutaminase reaction resulted in an unexpected decrease in
total product formed (Figure 4.1). This observed decrease in total
product may have been caused by hydroxylamine inhibition of the
enzyme. However, the Km for hydroxylamine in an LGH synthesis
reaction (see chapter 3) has been shown to be several times higher
than the concentration used in these experiments. Alternatively,
LGH release from the enzyme may be very slow thus hindering the
regeneration of free enzyme for further hydrolysis. Consistent with
this hypothesis, an R30A mutant of AS-B with impaired substrate
binding (shown in its Ks, Figure 4.2) exhibited an increase in total
product formed with increasing concentrations of hydroxylamine
indicating a rate limiting deacylation step (Figure 4.2).
The findings described in this report represent mechanistic
differences between substrate hydrolysis catalyzed by the class II
amidotransferase, AS-B, and hydrolysis catalyzed by the thiol

101
protease, papain, and the class I amidotransferase, CAD. While both
AS-B and CAD exhibit a rate determining step occurring after the
formation of the acyl enzyme, a feature unlike the slow acylation
step in amide hydrolysis catalyzed by papain, AS-B may have an
additional step which contributes to the rate of the reaction and is
unlike the sole rate determining deacylation which characterizes CAD
catalysis (Chaparan & Evans, 1991). Additional effort will be
required to understand the details of the mechanisms of these
reactions which cause the differences seen in the rate determining
steps. One step towards accomplishing this goal is a knowledge of
the rates of the individual steps in the reactions. In this regard, the
Ks for glutamine determined in this work not only reveals that the
glutaminase reaction occurs under rapid equilibrium conditions (k.
,>>k2) but also contributes a rate constant to the overall model of the
reaction which will be presented in Chapter 7. In addition, Chapters
5 and 6 represent the first attempt at the examination of an
individual step in the reaction.

CHAPTER 5
EVIDENCE OF A THIOESTER INTERMEDIATE FORMED DURING
GLUTAMINE HYDROLYSIS CATALYZED BY ASPARAGINE
SYNTHETASE B
Introduction
Numerous clinical and basic studies have correlated the
importance of asparagine biosynthesis with the ability of some
leukemia cells to escape the toxic effects of treatment with L-
asparaginase (Chakrabarti & Schuster, 1997). As a result of this
relationship, a great deal of effort has been directed towards
understanding the chemical and kinetic mechanism of asparagine
synthetase, the enzyme mediating asparagine formation (Richards &
Schuster, 1997). It has been the goal of several research groups to
obtain specific, and potent, inhibitors of AS that might prevent
asparagine synthesis in tumor cells, and therefore act as either
adjuncts to L-asparaginase therapy or replacements for this drug
(Cooney et al., 1976; Cooney et al., 1980). Such efforts have been
hampered, however, by an incomplete understanding of the chemical
mechanism by which nitrogen is removed from the side chain amide
of glutamine and transferred to B-aspartyl-AMP to form asparagine
(Richards & Schuster, 1997).
Escherichia coli asparagine synthetase B (AS-B) (Boehlein et al.,
1994a; Scofield et ah, 1990) is a member of the class II (formerly
purF) amidotransferase superfamily (Smith. 1995; Zalkin, 1993), that
102

103
includes glutamine 5'-phosphoribosylpyrophosphate amido-
transferase (GPA) (Tso et al., 1982), glutamine fructose-6-phosphate
amidotransferase (GFAT) (Badet-Denisot et al., 1993), and glutamate
synthase (Vanoni et al., 1991). These enzymes are characterized by
a highly conserved N-terminal cysteine residue in their mature,
active form. AS-B catalyzes the ATP dependent synthesis of
asparagine utilizing either ammonia (Reaction 1) or glutamine
(Reaction 2) as a nitrogen source. When aspartate is absent, AS-B
catalyzes the hydrolysis of glutamine to glutamate and ammonia
(Reaction 3), in a reaction that is stimulated by ATP (Boehlein et al.,
1994b; Boehlein et al., 1996).
Reaction 1: L-Asp + NH? + ATP > L-Asn + AMP + PPi
Reaction 2: L-Asp + L-Gln + ATP > L-Asn + L-Glu + AMP + PPi
Reaction 3: L-Gln -> L-Glu + NH?
AS-B has been employed in an extensive series of studies aimed at
delineating (i) the molecular mechanism of asparagine synthesis, and
(ii) the functional roles of conserved residues in glutamine binding
and activation (Boehlein et al., 1994a; Boehlein et al., 1994b; Boehlein
et al., 1996; Richards & Schuster, 1997). The high level of
glutaminase activity exhibited by AS-B has also been exploited
(Boehlein et al., 1997b) to show a mechanistic relationship between
Class II amidotransferases and thiol proteases, such as papain (Storer
& Menard, 1994). On the other hand, and despite substantial effort
(Badet, B., personal communication), the thioester intermediate that
must therefore be formed during glutamine breakdown has not yet

104
been characterized for any member of the Class II amidotransferase
family. We now report experimental conditions for the isolation of a
y-glutamyl AS-B adduct, together with evidence supporting the
hypothesis that this covalent derivative is the putative thioester
intermediate formed during AS-B catalyzed glutamine hydrolysis.
The characterization of this intermediate is likely to allow further
insight into the details of the kinetic and chemical mechanisms of
nitrogen transfer in AS-B, and other Class II amidotransferases.
Materials and Methods
Enzymes and reagents.
Recombinant wild-type AS-B and the CIA AS-B mutant, in
which Cys-1 is substituted by alanine, were constructed, expressed
and purified using published procedures (Boehlein et al., 1994a).
Protein concentration was determined using an assay kit supplied by
BioRad and immunoglobulin G as a standard. Recent examination of
the protein concentration by amino acid analysis of AS-B, performed
by the Protein Core Facility of ICBR at the University of Florida, gave
a value for protein concentration which was 2.7-fold lower than that
obtained with the previous method. Therefore, the enzyme
concentration has been corrected by a factor of 2.7. [U-14C]-L-
Glutamine (277 mCi/mmol) was purchased from Amersham
(Arlington Heights, IL). 6-Diazo-5-oxo-L-norleucine (DON) and
Sephadex G-50 Fine were obtained from Sigma Chemical Company.
Isolation of a Covalent Adduct between AS-B and P4Cl-glutamine
Wild-type AS-B (0.74 nmol) or the CIA AS-B mutant (0.74
nmol) was incubated with 1 mM [U-14C]-F-glutamine (SA 22,000

105
dpm/nmol) in 100 mM Tris-HCl (pH 8.0) at room temperature (total
reaction volume 100 pi). After 30 s, the reaction was quenched in 1
ml of 8% TCA on a 2.5 cm nitrocellulose filter (0.45 mm porosity) and
100 pi BSA (10 mg/ml) was added. After 2 minutes in the quench
solution the samples were filtered under vacuum on a 2.5 cm
nitrocellulose filter (0.45 pm porosity), and the filter was washed
with 50 ml of 1 N HC1. The filter was transferred to 5 ml of
scintillation fluid (ScintiVers IP) and the 14(y activity was measured
on a Beckman model LS6000IC scintillation counter. This procedure
was repeated in the absence of enzyme as a control for non-specific
binding of L-glutamine to the filter.
Gel Filtration of the Intermediate formed between AS-B and
Glutamine
(A) 1 mM [U-14-C]-L-glutamine (SA 22 000 dpm/nmol) was
incubated with either wt AS-B (0.74 nmol) or the CIA AS-B mutant
(0.74 nmol) in 100 mM Tris-HCl (pH 8.0) at room temperature (total
reaction volume 100 pi). After 30 s, reactions were terminated using
22 pi of 0.1 M NaOAC (pH 4.0) containing 5% SDS, and the protein was
separated from free glutamine by gel filtration using Sephadex G-50
Fine spin column equilibrated with 1% SDS in 0.1 M NaOAc (pH 4.0)
(Penefsky, 1979). The amount of radiolabeled enzyme was then
measured by liquid scintillation counting. (B) In a separate gel
filtration experiment, wt AS-B (2 nmol) was incubated with 2 mM
[U-14C]-L-glutamine (SA 11,000 dpm/nmol) in 100 mM Tris-HCl (pH
8.0) for 30 s at room temperature (total volume 100 pi). After
termination by dilution with 100 pi of 8M guanidinium-HCl in 0.1M
NaOAc (pH 4.0), the protein was isolated by gel filtration on a

106
Sephadex G-50 Fine spin column (Penefsky, 1979) equilibrated in the
quench solution. The amount of radiolabeled enzyme was measured
by liquid scintillation counting.
Lability of the intermediate formed between AS-B and 14C1-
Glutamine at high pH
The gel filtration procedure was used to determine the stability
of the intermediate at low and high pH. Wild-type AS-B (4.8 nmol)
was incubated 30 s at room temperature with 1.5 mM [14C] -
Glutamine (25,633 dpm/nmol) in a 100 pi reaction containing 100
mM Bis-Tris and Tris-HCl (pH 8). The reactions were terminated by
the addition of 22 pi 0.5 M sodium acetate (pH 4) in 5% SDS. The
protein was separated from free glutamine by the method of
Penefsky (1979) using a Sephadex G-50 fine spin column
equilibrated in either: 1) 0.1 M sodium phosphate (pH 12) in 1% SDS
or 2) 0.1 sodium phosphate (pH 2) in 1% SDS. The amount of
radiolabeled enzyme in 70 pi of the solution which passed through
the column was measured by liquid scintillation counting, and the
protein concentration was measured using an assay kit provided by
BioRad. Values for the protein concentration and the percent of
enzyme labeled at each pH were derived from seven separate
experiments.
Results
The partitioning studies presented in chapter 4 provided
evidence for a thioester formed on the reaction pathway of
glutamine hydrolysis catalyzed by wt AS-B. If the breakdown of the
thioester intermediate is slower than its formation, as indicated in

107
the previous chapter, then accumulation of this intermediate may
allow an opportunity for its isolation.
Two different methods were used to isolate a covalent adduct
formed between wt AS-B and glutamine. In the initial approach,
labeled protein was obtained by incubating wt AS-B with [44c]-L-
glutamine, denaturing the protein with SDS in acid solution, and
separating the enzyme from free glutamine by gel filtration (Table
1). The radioactivity associated with the enzyme most likely was not
due to non-specific glutamine binding as wt AS-B that had been
covalently modified by 6-diazo-5-oxonorleucine (DON), prior to
incubation with [14c]-L-glutamine under these conditions, did not
give the radiolabeled covalent adduct (Table 5.1). Furthermore,
radiolabeled protein was not formed when [14c]-L-glutamine was
incubated with the CIA AS-B mutant in place of the wild-type
enzyme, despite the observation that this AS-B mutant binds L-
glutamine with high affinity (Boehlein et al., 1994a).
While gel filtration could be used successfully to isolate the
putative y-glutamyl AS-B adduct, quantitative measurements of its
kinetic properties were hampered by a reduction in the amount of
radiolabeled enzyme as a function of the time during which samples
were maintained in the quench solution (Table 5.2). After
approximately fifteen minutes, the observed reduction in
radiolabeled protein stabilized at 59% of the value observed at short
incubation times. Efforts to address this problem by employing 8 M
guanidine-HCl as a denaturant, instead of SDS, did not affect the rate
of loss of radiolabeled enzyme.

108
In contrast to the slow disappearance of the y-glutamyl-
enzyme adduct in the gel filtration approach, the radiolabeled
protein was stable when isolated by filter binding (Lusty, 1992), and
could be used in quantitative measurements. In these experiments,
enzyme was combined with [14c]-L-glutamine, denatured in TCA,
and the precipitate filtered and washed with HC1. The radioactivity
remaining on the filter was measured using liquid scintillation
counting. Approximately 35% of the enzyme was radiolabeled in
reaction mixtures containing wild-type AS-B (Table 5.3).
Examination of the effect of glutamine concentration on the steady-
state level of the glutamyl-enzyme complex revealed a saturation
curve which gave a maximum molar ratio of intermediate to enzyme
of 0.14 and a KM of 0.14 mM for glutamine (Figure 5.1). In control
studies, radioactivity was not associated with the filter when
incubation solutions lacked the enzyme, or contained either the CIA
AS-B mutant or wild-type AS-B that had been covalently modified
with DON (Table 5.3).
The stability of thioesters decreases with increasing pH,
therefore if the isolated intermediate is a thioester the radioactivity
associated with the enzyme should diminish under alkaline
conditions. Initially, an attempt was made to determine the stability
of the intermediate at various conditions of pH using the filter
binding assay. Unfortunately, the enzyme redissolved under the
alkaline conditions of the wash and passed through the filter. Thus,
the observed disappearance of radioactivity associated with the
membrane was due to the decrease in enzyme on the filter rather
than instability of a thioester intermediate. As an alternative

109
approach, the adduct was formed, the reaction was terminated, and
the resulting solution was added to a gel filtration column either
equilibrated at pH 2 or pH 12. As shown in Table 5.4, 35% of the
enzyme was labeled after gel filtration at pH 2 whereas only 10% of
the enzyme was labeled after gel filtration at pH 12. While a
decrease in radiolabeled enzyme isolated by gel filtration was
mentioned above, this problem was alleviated by maintaining a
constant time interval of 20 s between terminating the reaction and
addition of the solution to the column. As shown through the
statistics at each pH. variability in the amount of radiolabeled
enzyme due to slow termination does not seem to be a factor.
Discussion
Glutamine amidotransferases can be divided into two families
based on multiple sequence alignment of the glutamine amide
transfer (GAT) domains (Zalkin. 1993). Class I (formerly TrpG)
enzymes, such as GMP synthetase (Tesmer et al., 1996) and
carbamoyl phosphate synthetase (CPS) (Thoden et al., 1997) are
characterized by conserved Cys, His and Asp active site residues that
are thought to function as a catalytic triad in a manner analogous to
thiol proteases (Brocklehurst et al., 1987). Thus, in analogy to the
papain mechanism (Brocklehurst et al., 1987), glutamine hydrolysis
proceeds by initial attack of the Cys thiolate on the side chain amide
to yield a tetrahedral intermediate (Scheme 1). Breakdown of this
intermediate after N-protonation, by the conserved active site
histidine to transform the nitrogen into an efficient leaving group,
then yields a thioester. Subsequent hydrolysis of this intermediate

110
would then regenerate the enzyme and produce glutamate. Direct
support for this mechanistic hypothesis for Class I amidotransferases
has been provided by isolation of the thioester in studies on CAD
(Chaparian & Evans, 1991), Escherichia coli PabA (Roux & Walsh,
1992) and CPS (Lusty, 1992; Wellner et al., 1973). This
accomplishment has contributed significantly to understanding the
mechanism of glutamine hydrolysis in these enzymes by allowing
use of a minimal model to examine the rate determining step of the
reaction, as well as the effects of the other substrates on this step.
Class II amidotransferases, on the other hand, possess a
conserved N-terminal cysteine that has been demonstrated to be
essential for all glutamine-dependent activities of these enzymes
(Badet et al., 1987; Boehlein et al., 1994a; Mei & Zalkin, 1989; Van
Heeke & Schuster, 1989). Although early experiments suggested the
presence of a catalytically important histidine in GPA (Mei & Zalkin,
1989) site directed mutagenesis of histidines within the GAT domain
of AS-B has almost no effect on the kinetic properties of the resulting
mutant enzymes (Boehlein et al., 1994a) and recent X-ray crystal
structures of Bacillus subtilis GPA (Smith et al., 1994), Escherichia
coli GPA (Kim et al., 1996) and an N-terminal fragment of Escherichia
coli GFAT (Isupov et al., 1996) do not reveal a histidine in the
appropriate position to participate in acid-base catalysis. Class II
amidotransferases have therefore been proposed to be members of
the N-terminal nucleophile (Ntn) hydrolase superfamily (Brannigan
et al., 1995), that includes the 20S proteasome (Seemuller et al.,
1996) and penicillin acylase (Duggleby et al., 1995). Ntn hydrolases
are characterized by an N-terminal catalytic nucleophile (hydroxyl or

Ill
thiol) that is thought to be activated by transfer of its proton to the
free N-terminal amino group (Brannigan et al., 1995). Thus, in the
case of AS-B and other Class II amidotransferases, the a-amino group
of Cys-1 would serve as a general acid-base catalyst in place of
histidine. No unambiguous kinetic evidence, however, has yet been
reported that supports such an assumption. The observation of a
significantly smaller amide 15N (Vmax/Km) kinetic isotope effect for
AS-B catalyzed glutamine hydrolysis compared to that observed for
papain catalyzed amide hydrolysis, however, may reflect this
difference in the nature of protonation of the leaving group in the
tetrahedral intermediate (Stoker et al., 1996). Whereas in papain,
His-159 functions as a general acid catalyst (Brocklehurst, et al.,
1987) it has been suggested that, at least for AS-B, N-protonation
occurs by specific acid catalysis (Stoker et al., 1996). On the other
hand, Class II amidotransferases do exhibit many characteristics of
thiol proteases, including an oxyanion hole (Boehlein et al., 1994a;
Boehlein et al., 1997; Isupov et al., 1996; Kim et al., 1996).
In light of these structural and kinetic differences, it is
important to validate the assumption of a thioester intermediate in
AS-B catalyzed glutamine hydrolysis. Toward this goal, the
demonstration of the formation of LGH upon addition of
hydroxylamine to the glutaminase reaction presented in this report
supports the notion that a thioester is formed on the main pathway
of the hydrolysis reaction catalyzed by AS-B. Since the amide of
glutamine is resonance stabilized, hydroxylamine attack requires its
conversion to a more activated carbonyl derivative. Several
intermediates other than a thioester could serve as this activated

112
intermediate. For example, a carboxyl group on the enzyme could
attack the side chain of glutamine to form an anhydride
intermediate. In this case, hydroxylamine could attack on either the
glutamyl or enzyme side of the central oxygen of this intermediate.
Therefore, each attack of the hydroxylamine on the enzyme side
would result in a covalently labeled enzyme which would be
unavailable for further turnover thus resulting in inhibition of the
reaction with time. Since a time dependent inhibition of the reaction
in the presence of hydroxylamine was not observed and since an
essential glutamate or aspartate residue which could participate in
forming such an intermediate has never been found in AS-B, a
mechanism involving an anhydride intermediate is unlikely.
Alternatively, LGH could be formed via a pyroglutamate although
this would not be efficient in ammonia transfer. The possibility of an
oxygen ester intermediate rather than a thioester is also unlikely
since hydroxylamine causes partitioning even at pH 6 which is
consistent with a thioester and not the less reactive oxygen ester
which only reacts with hydroxylamine at alkaline pH. The
demonstration of a radiolabeled intermediate upon incubation of wt
AS-B with [l4C]-glutamine provides further evidence for an
acylenzyme intermediate. Several observations support that this
isolated adduct is in fact the thioester. First, a radiolabeled enzyme
was not formed when the wild-type enzyme was labeled with DON
prior to the reaction (Tables 5.1 & 5.3). DON has been shown to react
at the active site of mammalian asparagine synthetases (Mehlhaff &
Schuster, 1991; Larsen & Schuster, 1992) and causes irreversible
enzyme inactivation by modification of the thiol of Cys-1, as

113
observed in the crystal structure of DON-inactivated Escherichia coli
GPA (Kim et al., 1996). Second, radiolabelled enzyme was not found
when the CIA AS-B was incubated with [14C]-L-glutamine in place of
wild-type enzyme (Tables 5.1 & 5.3). This AS-B mutant lacks both
glutamine-dependent synthetase and glutaminase activity but
retains ammonia dependent activity and the ability to bind
glutamine tightly (Boehlein et al., 1994a). Since previous studies
employing asparagine synthetases, as well as other Class II
amidotransferases, have demonstrated the involvement of the Cys-1
side chain in formation of the thioester linkage, the lack of a y-
glutamyl CIA complex strongly suggests that covalent modification
occurs at the GAT-domain active site requiring the Cys-1 thiolate.
The failure of the CIA AS-B mutant to form a radiolabeled complex is
further evidence that non-specific glutamine binding cannot account
for the radioactivity associated with the wild-type enzyme. Third,
the amount of intermediate saturates with increasing glutamine
concentration, the half saturation point being consistent with steady
state parameters observed for the AS-B glutaminase reaction (Figure
5.1). And finally, the intermediate was isolated only under acidic
conditions and was not stable at higher pH which is consistent with
the alkaline lability of thioesters.
Although the covalent y-glutamyl AS-B adduct appeared
unstable under the denaturation conditions used in the gel filtration
assay, possibly reflecting residual activity of slowly denaturing
enzyme in the solutions containing SDS or guanidinium HC1,
approximately 34% of the enzyme was radiolabeled based on data
from filter-binding. The percentage of enzyme in the thioester form,

114
which was similar to that reported for PabA (Roux & Walsh, 1992)
indicates that this intermediate is kinetically important.
The isolation of the intermediate described here is an essential
step in obtaining an accurate model of glutamine hydrolysis catalyzed
by Class II amidotransferases. Furthermore, the ability to detect and
quantitate this intermediate has important implications in
determination of the mechanism of glutamine-dependent nitrogen
transfer for AS-B and other enzymes in this family. For asparagine
synthetases, current experimental evidence suggests that nitrogen
transfer can proceed via direct attack of the tetrahedral intermediate
(Scheme 1) on 8-aspartyl-AMP, or by release of enzyme-bound,
unprotonated ammonia from glutamine (Richards & Schuster, 1997).
While these mechanisms of glutamine amide transfer are
substantially different, the formation of a thioacylenzyme
intermediate during asparagine synthesis is a common requirement.
Methods for isolation of the thioester therefore have potential
application in experiments aimed at resolving which of these two
transfer mechanisms is operative in AS-B. For example, if the
thioester intermediate is formed prior to the breakdown of
8-aspartyl -AMP, then nitrogen transfer mediated by enzyme-bound
ammonia is mechanistically most likely. On the other hand, if no
thioester is formed until after B-aspartyl-AMP breakdown, then it is
most probable that direct amide transfer takes place via covalent
intermediates.

115
TABLE 5.1
Isolation of a covalent glutamyl-enzyme intermediate by gel
filtration.
Enzyme
d p in
wild-type (5 nmol)
5016 100
wild-type (4 nmol) + 1 mM DON
152 52
CIA (4 nmol)
65 2
The intermediate was formed by incubating the indicated
concentration of enzyme with 1 mM [14C] glutamine (S.A. 22,000
dpm/nmol) as described in the materials and methods. The reactions
were terminated with 22 |_il of 5% SDS in 0.1 M sodium acetate (pH
4.0) and the enzyme was separated from free glutamine using a
Sephadex G-50-80 spin column equilibrated in 1% SDS in 0.1 M
sodium acetate (pH 4.0) according to the method of Penefsky (1979).
In the second entry, wt AS-B (1.5nmol) was incubated with ImM
DON for 30 min prior to dilution into the reaction mixture containing
[14C] glutamine. The amount of radiolabeled enzyme in the column
effluent was then measured by liquid scintillation.

116
TABLE 5.2
Time-dependence of y-glutamyl AS-B adduct degradation in SDS
solution.
Time
(min )
d p
m
< 1
4802
77
1
4165
3
5
3593
460
1 5
2828
220
29
3036
72
The intermediate was formed using 4 nmol of wild-type AS-B and 1
mM [14C] glutamine (S.A. 22,000 dpm/nmol). The reactions were
terminated as described in Table I and then incubated for the
amount of time indicated before they were added to the sephadex
column.

117
TABLE 5.3
Trapping the thioester by filter binding
Sample
d p m
- enzyme
384
187
CIA
290
115
wt + DON
185
55
w t
6504
183
The intermediate was formed as described in materials and methods
using 2 nmol of enzyme and 1 mM glutamine (S.A. 22,000
dpm/nmol). To examine the effect of DON on intermediate formation,
ImM DON was incubated with wild-type AS-B for 30 minutes. The
enzyme covalently labeled with DON was then diluted into the
reaction mixture containing [14C] glutamine.

118
[Gin] mM
Figure 5.1 The effect of glutamine on the steady-state concentration
of the thioester. Various concentrations of [14C] glutamine were
combined with wild-type AS-B (0.74 nmol) and the covalent adduct
was isolated using a filter binding method as described in Materials
and Methods.

119
Table 5.4
Base lability of the covalent adduct isolated by gel filtration.
Conditions Protein Concentration Enzyme Labeled
(mg/ml) (%)
pH 2
1.18 0.06
37 2
pH 12
1.6 0.2
10 2
In all reactions, wt AS-B (4.8 nmol) was incubated 30 s at room
temperature with 1.5 inM [I4C]-glutamine (25,633 dpm/nmol) in a
reaction containing 100 inM Bis-Tris and Tris-HCl (pH 8). The
reactions were terminated by the addition of 22 pi 0.5 M sodium
acetate (pH 4) in 5% SDS. To determine the stability of the
intermediate at pH 2 and 12, the enzyme was separated from free
glutamine by the method of Penefsky (1979) using a Sephadex G-50
fine spin column equilibrated in 0.1 M sodium phosphate (either pH
2 or pH 12) in 1% SDS. The amount of radiolabeled enzyme in 70 pi
of the solution which passed through the column was then measured
by scintillation counting, and the protein concentration was
measured using an assay kit supplied by BioRad.

CHAPTER 6
CHARACTERIZATION OF A THIOESTER INTERMEDIATE
I ntroduction
The ability to isolate an intermediate on the reaction pathway
of glutamine hydrolysis catalyzed by AS-B (Chapter 5) will provide
an opportunity to gain a more detailed description of the reaction by
examining the rate of the individual step of deacylation. Previous
chapters have suggested that the rate determining step of the
reaction occurs after the formation of the covalent intermediate, but
that the deacylation step may not be the slow step. Therefore,
knowledge of the rate of deacylation will allow a comparison with
the overall rate of the reaction. If indeed, deacylation is not rate
limiting then the rate of this step should be significantly faster than
that of the overall reaction.
The ability to measure the rate of deacylation also provides a
tool with which to determine the significance of conserved residues
in an individual step of the reaction. The importance of Cys-1 in
making the thioacyl bond has already been implied in chapter 5 by
the inability to capture a glutamyl-enzyme intermediate when this
cysteine residue is replaced with an alanine (Chapter 5). Two other
residues, conserved among the class II amidotransferases, which
may participate in either the acylation or deacylation process are
Asn-74 and Arg-30. Evidence is mounting that Asn-74 plays a role,
120

121
very similar to Gin-19 of papain, as part of an oxyanion hole which
stabilizes the tetrahedral intermediates thought to be formed during
the acylation and deacylation steps in the hydrolysis reaction
(Boehlein et al., 1994b; 1996; 1997). Therefore, such an involvement
should be reflected in a difference in either the steady-state
concentration of the thioester or the rate of deacylation catalyzed by
the AS-B mutant, N74A, relative to those values measured for the wt
enzyme.
Several roles have been suggested for Arginine-30 such as an
involvement in the stabilization of the glutamine binding site, the
communication between the GAT and synthetase domains, and the
release of nitrogen from glutamine (Boehlein et al., 1994b). These
ideas were supported by structural information obtained for GPA
(Kim et al., 1996) and GFAT (Isupov et al., 1996) which revealed a
network of hydrogen bonding between the cognate arginine residue
in these enzymes and other residues involved in catalysis in the
glutamine site including Cys-1 and Asn-101(Asn-74 in AS-B). While
chapter 4 presented evidence which supports the active participation
of Arg-30 in glutamine binding, additional exploration into its
interactions and functions will be necessary to obtain a more precise
and comprehensive picture of its purpose. For example, partitioning
studies (Chapter 4) suggested that the deacylation step is rate
limiting in glutamine hydrolysis catalyzed by the R30A mutant of
AS-B. This conclusion can now be tested by comparison of the rate of
deacylation (determined by direct measurement) with that of the
overall reaction.

122
In the previous chapter, the steady-state concentration of the
putative thioester intermediate, formed during glutamine hydrolysis
catalyzed by wt AS-B, was determined to reach 35% of the total
enzyme concentration. These studies are continued in this chapter
which presents the measurement of the deacylation step for the
wild-type enzyme, and the effect of mutations on the steady-state
concentration of the thioester and its deacylation rate.
Materials and Methods
Enzymes and reagents
The construction of mutants in which asparagine-74 is replaced
by alanine (N74A) or arginine-30 is replaced with an alanine (R30A)
has been described previously (Boehlein et al., 1994b) Recombinant
AS-B, N74A, and R30A were expressed and purified using standard
procedures (Boehlein et al., 1994a). [U-14C]-L-Glutamine (277
mCi/mmol) was purchased from Amersham. The nitrocellulose
filters were purchased from Bio-Rad. Other reagents including L-
glutamine were purchase from SIGMA Chemical Company (St. Louis,
MO) and were of the highest commercial purity.
Effect of Glutamine on the Steady-State Concentration of the
Thioester Intermediate
The radiolabeled enzyme complex was isolated by filter
binding as described in chapter 5. A glutamyl-enzyme complex was
formed by incubating 0.74 nmol of either wt, N74A, or R30A with
various concentrations of L-glutamine (SA 22,000 dpm/nmol) in a
100 pi reaction containing 100 mM Tris-HCl (pH 8.0) for 30 sec at 5C.
The reactions were quenched in 1 ml of 8% TCA and 100 pi BSA (10

123
mg/ml) was added. After 2 minutes in the quench solution the
samples were filtered by vacuum through a 2.5 cm nitrocellulose
filter (0.45 mm porosity) and the filter was washed with 50 ml of IN
HC1. The filter was transferred to 5 ml of scintillation fluid
(ScintiVers II*) and the 14C activity was measured.
Rate of deacvlation of the thioester intermediate
A glutamyl-enzyme adduct was formed at 5C by incubating
0.74 nmol of either wt, N74A, or R30A with 1.5 mM L-glutamine (SA
20,000 dpm/nmol) in a 100 pi reaction containing 100 mM Bis-Tris
and Tris (pH 8.0). After incubating the reaction for 1 min, 100 pi of
200 mM non-radioactive glutamine was added to each sample to
dilute the unreacted radioactive glutamine. The samples were
quenched at various time intervals after the dilution as described
above and the amount of radioactivity associated with the protein
was determined by scintillation counting.
Results
Characterization of the deacvlation of wild-tvpe AS-B
Isolation of a thioester intermediate formed between wild-type
AS-B and glutamine and the steady-state concentration of this
intermediate have been described in chapter 5.
A plot of thioester concentration versus time of incubation with
unlabeled glutamine followed a mono-exponential decay and the rate
constant for breakdown of the thioester determined from this plot
was 0.16 sec1 (Figure 6.1). This value is four times higher than the
kcat (Table 6.1).

124
Previous studies of the glutaminase reaction have indicated
that the presence of ATP stimulates the hydrolysis of glutamine
(Boehlein et al., 1994), therefore it was of interest to determine
whether or not ATP affected the rate of the individual deacylation
step. An examination of the effect of glutamine on the steady-state
concentration of the thioester in the presence of 1 mM ATP suggests
that ATP causes a slight decrease in the maximum concentration of
thioester formed at saturating glutamine concentrations (Figure 6.2,
Table 6.1). However, the addition of 1 mM ATP has no effect on the
rate of deacylation (Figure 6.3, Table 6.1).
The effect of NH4C1 on the deacylation rate was examined. If
the step involving C-N bond cleavage to release ammonia is readily
reversible it is expected that addition of ammonia to the reaction
would shift the equilibrium away from the thioester and thus result
in a decrease in the concentration of thioester formed and the rate of
its deacylation. However, measurement of the deacylation of the
thioester intermediate showed that the presence of 10 mM NH4C1 had
little effect on the rate of deacylation (Figure 6.4, Table 6.1).
The effect of glutamine on the steady state concentration of the
glutamvl-enzvme adduct formed with the AS-B mutant. R30A
Studies in chapter 4 demonstrated that the rate limiting step of
glutamine hydrolysis catalyzed by an AS-B mutant impaired in
glutamine binding, R30A, was deacylation. Therefore, it would be
expected that the deacylation rate would be the same as the kcat
value determined in glutamine hydrolysis. This comparison could be
used in contrast to the slightly varying rates obtained for the wild-
type enzyme. Unfortunately, the measurement of the deacylation

125
constant was impractical due to the high concentration and therefore
the high specific activity of glutamine necessary to saturate this
mutant enzyme. However, it was possible to examine the effect of
glutamine on the steady-state concentration of the thioester formed
with R30A. As shown in Figure 6.5 and Table 6.1, the maximum
percentage of R30A which was labeled was higher than that obtained
with the wild-type enzyme.
Characterization of a glutamvl-enzyme adduct formed with the AS-B
mutant N74A
Examination of the effect of glutamine concentration on the
steady-state level of the glutamyl-enzyme complex formed with
N74A revealed a saturation curve which gave a maximum
percentage of enzyme labeled with radioactivity of 57% and a Km of
0.06 mM for glutamine (Figure 6.6 and Table 6.1). The thioester
concentration in reactions using N74A followed a mono-exponential
decay when plotted against time of incubation with unlabeled
glutamine. The rate constant for deacylation determined from this
plot was 0.062 sec'1, significantly slower than that of wild-type
(Figure 6.7 and Table 6.1).

TE (nmol)
126
Figure 6.1: Rate of deacylation of the thioester intermediate formed
during wild-type AS-B catalyzed glutamine hydrolysis.

127
[Gin] mM
Figure 6.2: The steady-state concentration of the thioester formed in
the presence of ATP.

128
Time (sec)
Figure 6.3: Effect of ATP on the rate of deacylation. A glutamyl-
enzyme adduct was formed at 5C by incubating 0.74 nmol wild-type
AS-B with 1.5 mM glutamine (SA 20,000 dpm/nmol) and 1 mM ATP,
O in 100 mM Bis-Tris and Tris (pH 8.0). After incubating the
reaction for 1 min, the radioactive glutamine was diluted by the
addition of 200 mM non-radioactive glutamine. The samples were
quenched at various time intervals after the dilution as described in
materials and methods and the amount of radioactivity associated
with the protein was determined by scintillation counting. The
deacylation in the presence of ATP, O, was compared to that in the
absence of ATP, .

129
Table 6.1: Characterization of the glutamyl-enzyme intermediate.
Enzyme
Max % of
Enzyme labeled
(%)
Deacylation
rate
(sec'1)
lc
iVcat
(sec-1)
w t
35.1 0.9
0.16 0.02
0.043 0.0007
wt + 1 mM ATP
26 1.1
0.18 0.01
0.076 0.002
wt + 10 mM NH4C1
n d
0.11 0.01
nd2
R30A
50 3
nd1
0.042 0.001
N74A
57 1
0.062 0.005
0.024 0.0005
All values in the table were determined at 5C. The deacylation rate
for the reaction catalyzed by R30A could not be obtained due to the
high concentrations of glutamine necessary to saturate the enzyme.
2Ammonia has been shown to have little effect on the kcat for the
glutaminase reaction catalyzed by wt AS-B at 37C.

130
Time (sec)
Figure 6.4: Effect of ammonia on the rate of deacylation. A
glutamyl-enzyme adduct was formed at 5C by incubating 0.74 nmol
wild-type AS-B with 1.5 mM glutamine (SA 20,000 dpm/nmol) in
100 mM Bis-Tris and Tris (pH 8.0) in the presence, B, and absence,
O, of 10 mM NH4C1. After incubating the reaction for 1 min, the
radioactive glutamine was diluted by the addition of 200 mM non
radioactive glutamine. The samples were quenched at various time
intervals after the dilution as described in materials and methods
and the amounts of radioactivity associated with the protein was
determined by scintillation counting.

131
Figure 6.5: Effect of glutamine on the steady state concentration of
thioester formed in the glutaminase reaction catalyzed by the AS-B
mutant R30A. A glutamyl-enzyme complex was formed by
incubating 1.8 nmol of R30A with various concentrations of L-
glutamine (SA 22,000 dpm/nmol) in a reaction containing 100 mM
Tris-HCl (pH 8.0) for 30 sec at 5C. A radiolabeled enzyme complex
was captured by quenching the reaction in 8% TCA over a
nitrocellulose filter as described in materials and methods and the
radioactivity associated with the filter was measured by scintillation
counting.

132
Figure 6.6: Effect of glutamine on the steady-state concentration of
the thioester formed during the glutaminase reaction catalyzed by
the AS-B mutant N74A. A glutamyl-enzyme complex was formed by
incubating 0.74 nmol of N74A with various concentrations of L-
glutamine (SA 22,000 dpm/nmol) in a reaction containing 100 mM
Tris-HCl (pH 8.0) for 30 sec at 5C. A radiolabeled enzyme complex
was captured by quenching the reaction in 8% TCA over a
nitrocellulose filter as described in materials and methods and the
radioactivity associated with the filter was measured by scintillation
counting.

133
Figure 6.7: The rate of deacylation of the AS-B mutant N74A during
glutamine hydrolysis. A glutamyl-enzyme adduct was formed at 5C
by incubating 0.74 nmol N74A with 1 mM glutamine (SA 20,000
dpm/nmol) in 100 mM Bis-Tris and Tris (pH 8.0). After incubating
the reaction for 30 sec, the radioactive glutamine was diluted by the
addition of 200 mM non-radioactive glutamine. The samples were
quenched at various time intervals after the dilution as described in
materials and methods and the amounts of radioactivity associated
with the protein was determined by scintillation counting.

134
Discussion
Isolation of the thioester has allowed the examination of an
individual step in glutamine hydrolysis, deacylation. The results
presented here describe the use of this tool to show that the
deacylation step is not solely rate determining in glutamine
hydrolysis catalyzed by AS-B. The possibility of multiple steps
making significant contributions to the overall rate rather than a
single rate determining step was implied in chapter 4. In this
chapter, not only was the deacylation rate 4-fold faster than kcat for
glutamine hydrolysis catalyzed by wild type AS-B but it was not
affected by the presence of ATP (Table 6.1) which has been shown to
stimulate the glutaminase reaction (Boehlein et al., 1994b). An
increase in the rate of a reaction would require a change in rate in
the slowest step or a step with a comparable rate to the slowest step
of the reaction. Therefore, the above results suggest that a step in
addition to deacylation is rate limiting and contributes to the overall
rate constant for the reaction catalyzed by wild type AS-B. Further
evidence for the lack of ATP involvement in the deacylation step
comes from the absence of ATP stimulation of glutamine hydrolysis
when deacylation is rate limiting such as the case which has been
demonstrated for the glutaminase reaction catalyzed by the AS-B
mutant, R30A (see chapter 4).
The characterization of the deacylation rate occurring in the
AS-B catalyzed glutaminase reaction reveals some interesting
contrasts in relation to amide hydrolysis catalyzed by the thiol
protease papain and the amidotransferase. CAD (Chaparan & Evans,

135
1991). In support of the contrasts mentioned in chapter 4, CAD
(Chaparan & Evans, 1991) and now AS-B have been shown to posses
a different rate limiting step than found in amide hydrolysis
catalyzed by papain. However, the present study suggests that while
the deacylation step is the slowest step in glutamine hydrolysis
catalyzed by CAD, AS-B catalyzed hydrolysis seems to have an
additional step which has a rate comparable to deacylation.
Secondly, high concentrations of ammonia essentially abolished the
appearance of the thioester intermediate formed during CAD-
catalyzed glutamine hydrolysis demonstrating that the acylation step
is a reversible process (Chaparan & Evans, 1991). In contrast,
ammonia seems to have a minimal effect on the AS-B-catalyzed
thioester deacylation rate (Figure 6.4) which suggests a greater
difficulty in reversing the formation of the thioester intermediate in
the AS-B catalyzed reaction. Consistent with this observation, studies
in chapter 3 show that ammonia does not inhibit in the range of
concentrations used and that asparagine synthesis does not occur to a
significant degree upon mixing the wild type enzyme with glutamate
and ammonia. Hydroxylamine can be used in the synthesis of LGH
therefore the lack of glutamine synthesis using ammonia is probably
not due to the inability of glutamate to bind to the enzyme and form
a thioester intermediate.
With the knowledge of the steady-state concentration of the
thioester formed and the rate of deacylation in the glutaminase
reaction catalyzed by wild type AS-B it is now possible to determine
the involvement of individual residues in this step. This report
describes an initial characterization of the deacylation step during

136
the glutaminase reactions catalyzed by R30A and N74A. While the
deacylation rate was not obtainable for R30A, the steady state
concentration of the thioester formed with this mutant was found to
be slightly higher than that with the wild type enzyme which may be
diagnostic of a slower deacylation step. Indeed, earlier studies
provided evidence that the deacylation step was rate limiting for this
mutant (partitioning studies in chapter 4). Furthermore, previous
structural studies have indicated that the cognate arginine residue in
GPA (Kim et al., 1996) and GFAT (Isupov et al., 1996), R26, is
involved in positioning the conserved asparagine residue (N74 in AS-
B). Since it is thought that this asparagine residue is important in
forming an oxyanion hole in the stabilization of the tetrahedral
intermediates, a decrease in deacylation step in the R30A catalyzed
glutamine hydrolysis could reflect an indirect effect on the
interactions made by this asparagine residue resulting in a decrease
in stability of the second tetrahedral intermediate which occurs
during deacylation. Further examination is necessary to confirm this
idea. However, replacement of asparagine-74 with alanine (N74A)
causes similar changes the amount of thioester formed and the
deacylation rate (Table 6.1). Glutamine hydrolysis catalyzed by the
N74A mutant, like R30A, exhibited a greater concentration of
thioester than the wild type enzyme (Figure 6.6 & Table 6.1).
Moreover, N74A exhibited a slightly slower rate of deacylation than
the wild-type (Figure 6.7 & Table 6.1). Once again, this slower rate
of deacylation may indicate the inability of an enzyme lacking this
important asparagine to stabilize the second tetrahedral
intermediate. Thus the deacylation step is slowed and with a lesser

137
change in the acylation step the amount of thioester that is formed
increases.
The ability to measure the rate constants of the individual
steps in glutamine hydrolysis will allow a complete characterization
of the wild type enzyme and thus will provide a model of comparison
for the determination of mutations on catalysis. This report
represents the first step toward achieving this goal. The deacylation
rate for the wild type AS-B now can be used as part of a model in the
simulation of the glutaminase reaction presented in the following
chapter. Furthermore, an extension of these studies using rapid
quench analysis will provide rates of both acylation and deacylation
thus furthering the development of the model and understanding the
mechanism of glutamine hydrolysis.

CHAPTER 7
DISCUSSION AND FUTURE DIRECTIONS
Understanding the mechanism of nitrogen transfer catalyzed
by asparagine synthetase will require the knowledge of the integral
parts of the process such as glutamine utilization. The past few years
of research on AS-B have seen substantial progress towards this goal
especially with respect to the catalytic roles played by the conserved
residues in the glutamine active site (See Boehlein et al., 1994a,b;
Boehlein et al., 1996; Boehlein et al., 1997a,b; Stoker et al., 1996) .
From the knowledge gained in the investigations of AS-B and other
class II amidotransferases, two schemes have developed which are
used to describe glutamine utilization and nitrogen transfer. First, as
described by Mei and Zalkin (1989) (see Figure 1.1), nitrogen
transfer occurs via a free ammonia released from glutamine by a
mechanism similar to thiol protease catalyzed amide hydrolysis.
Alternatively, the nucleophilic attack of the N-terminal cysteine
residue on the amide carbon of glutamine activates the side chain
nitrogen for a direct attack on the acceptor molecule (Stoker et al.,
1996) (see Figure 1.2). One key difference between these two
schemes may lie in the requirement of a common mechanism for
glutamine dependent asparagine synthesis and glutaminase reactions
in the scheme involving free ammonia and the lack of a common
mechanism for the two reactions in the scheme involving the direct
138

139
transfer of nitrogen. Therefore, one way to discriminate between
which mechanism describes asparagine synthesis catalyzed by AS-B
is to obtain a detailed model, including the rates of the individual
steps, of both glutamine dependent reactions. Such a model also
would expedite the course of understanding how individual amino
acid residues participate in catalysis as it would allow us to visualize
which individual step is affected by a mutation. Finally, a detailed
model of the glutaminase reaction is necessary in order to complete
the description of the overall kinetic mechanism of AS-B.
The goal of this dissertation was to define both chemical and
kinetic characteristics of the glutaminase reaction which could aid in
building this much needed model. Certainly as shown here, the
initial stages are complete. Due to some similarities between the
class II amidotransferases and the thiol proteases, such as the
presence of a catalytically important cysteine, the original model for
the glutaminase reaction was based on the mechanism of amide
hydrolysis catalyzed by the thiol protease, papain. Therefore, as a
first step, it was important to ascertain that the major features of
this mechanism, such as the cys-his dyad and the formation of a
thioester intermediate, could be found in AS-B catalyzed glutamine
hydrolysis.
For papain, the importance of Cys-25 and His-159 is first
apparent through their complete conservation among all members of
the family of thiol proteases (Brocklehurst et al., 1987). The
conserved histidine plays a number of essential roles in the
hydrolysis reaction including activation of the Cys-25, protonation of
the leaving group, and activation of a water molecule for nucleophilic

140
attack on the thioester (Storer & Menard, 1994). If there is a
histidine playing an equivalent role in AS-B catalysis, then it would
be expected that its removal would not only dramatically affect the
rate of the glutaminase reaction but would also change the pH
profile. However, as shown in chapter 2, replacing histidine-47 with
an asparagine only causes a four-fold decrease in the kcat for
glutamine hydrolysis. Moreover, the pH profile of the glutaminase
reaction catalyzed by this mutant is similar to that of the wild-type.
These results suggest that a role for His-47 in acid-base catalysis is
unlikely. On the other hand, the 10-fold drop in activity should not
be ignored, especially in light of the hypothesis presented here that a
conformation change is rate limiting. One further test for the
participation of His-47 in acid-base capacity would be to examine the
individual steps catalyzed by H47N. The work presented in this
dissertation has provided a method to examine the deacylation step
in the glutaminase reaction. Therefore, a comparison of the pH-
dependence of the deacylation rate catalyzed by wild-type AS-B and
its H47N mutant might provide the information necessary to reveal
whether His-47 plays a role in activating a water molecule for a
nucleophilic attack on the thioester.
While there may be differences in acid-base catalysis between
AS-B and papain, the overall pathway of hydrolysis may be similar.
This means that there must be an alternative for the deprotonation
of Cys-1 and the protonation of the leaving group during hydrolysis
catalyzed by AS-B. In fact, such an alternative has been offered in
regards to the class II amidotransferases (Kim et al., 1996).
Structural studies have placed the class II amidotransferases in a

141
family of hydrolytic enzymes, the Ntn-hydrolases, which use the N-
terminal residue (Ser, Thr, or Cys) as the nucleophilic catalyst.
Members of this family, including the 20S proteasome (Brannigan et
al., 1995; Seemuller et al., 1996), aspartylglucosaminidase (Oinonen
et al., 1995), and penicillin acylase (Duggleby et al., 1995), are
thought to catalyze general acid-base reactions through their N-
terminal amino group. Whether or not the class II amidotransferases
use such a mechanism remains to be established since previous
studies, employing site-directed mutants of GPA and GFAT which
contain an additional residue N-terminal to the cysteine, have given
ambiguous results (Isupov et al., 1996; Kim et al, 1996). Another
alternative for a histidine could be specific acid-base catalysis.
Indeed, this mechanism would explain the unexpected differences
between the V/K 15N KIEs for the glutaminase reaction catalyzed by
AS-B (Stoker et al., 1996) and that for amide hydrolysis catalyzed by
papain (OLeary et al., 1974). For papain, theoretical studies have
suggested that the transfer of the proton from His-159 to the leaving
group amine is concerted with the nucleophilic attack of Cys-25
(Arad et al., 1990). While this conclusion remains to be confirmed,
the large V/K 15N KIEs implied that C-N bond cleavage was the slow
step in the acylation process. On the other hand, the absence of a
V/K 15N KIE in the glutaminase reaction catalyzed by AS-B suggested
that a step during acylation other than C-N bond cleavage was rate
limiting. This observation may reflect a slow protonation of the
leaving group regulated by a specific acid catalyst. This issue could
be addressed by a study of the pH dependence of the 15N V/K KIE. If
the rate of acylation is dependent on the protonation of the leaving

142
group via specific acid catalysis, then the 15N V/K KIE might increase
at low pHs where it is easier to protonate the leaving group. The
limitation to such a study is the narrow pH range in which AS-B is
stable.
The finding that there are differences in acid-base catalysis
between papain and AS-B raises the question of whether the AS-B
catalyzed glutaminase reaction follows a pathway which has any
similarities to that of papain. Therefore, the existence of an acyl-
enzyme intermediate in the pathway of glutamine hydrolysis
catalyzed by AS-B was investigated. Chapter 5 presents evidence for
such an intermediate as the combination of [14C] glutamine with the
wild-type AS-B resulted in a radiolabeled enzyme which could be
trapped by precipitation on a nitrocellulose filter. Evidence that the
glutamyl-enzyme adduct is a thioester includes: 1) the inability to
form a glutamyl-enzyme adduct with the AS-B mutant, CIA, which
can bind glutamine but lacks the cysteine thought to make the thiol
linkage (Chapter 5) 2) the inability to form an adduct with wild-
type enzyme that had been incubated with a covalent modifier (DON)
which was shown to be competitive with glutamine and to cause
irreversible inhibition of the glutaminase activity (Chapter 4) 3) the
observation that the amount of intermediate saturates with
increasing glutamine concentration and the half saturation point is
consistent with steady state parameters observed for the AS-B
glutaminase reaction (Chapter 5), and 4) the instability of the
intermediate in a basic environment as would be expected of a thiol
ester intermediate (Chapter 5). Future studies to support the
conclusion that the covalent intermediate is a thioester include the

143
examination of the enzyme structure in the presence of 15N and 14C
double labeled glutamine using solid state nmr.
Computer Simulations of glutamate formation as a function of time:
The evidence for the existence of a thioester intermediate
given in chapters 5 and 6 provides a starting point in building a
model for the AS-B glutaminase reaction. In this dissertation the
model has been built from the simplest case which involves the
formation of a covalent intermediate (shown in scheme (1)), and
additions have been made according to the minimal requirements
necessary to make the best fit to the actual data from the
examination of glutamate formation as a function of time.
(1)
k-2 + p k-3
In this model, k, and k, represent the steps leading to and from the
formation of the Michaelis complex; k2 and k 2 the steps involved in
the acylation process; and k3 and k_3 the steps involved in
deacylation. The experiments in chapters 4 and 6 yielded values for
several of the rate constants in this reaction which now can be used
in a simulation of the glutaminase reaction using the computer
program KINSIM (Barshop et al. 1983). The values for the rate
constants to be used in the simulation of scheme (1) are given in
Table 7.1. The dissociation constant (Ks) for glutamine can be used to
obtain the value of k., using the relationship, Ks= k ,/k, and assuming
the value of k, approaches diffusion. The direct measurement of

144
deacylation demonstrated in chapter 6 provides a rate constant for
k3. The observation of a burst in glutamate formation with time
presented in chapter 4 suggests that k3 is less than k2 and therefore
k2 will be arbitrarily set at 1 sec'1. Preliminary studies using rapid
quench techniques have confirmed that this value of k2 is reasonable.
The acylation step will be made irreversible since the backwards
reaction seems unlikely. Recall that ammonia has no effect on the
rate of deacylation (chapter 6, Figure 6.4), it does not seem to act as a
substrate in the synthesis of glutamine (chapter 3), and it does not
inhibit the glutaminase reaction (chapter 3). The value for k3 is
estimated from the observation that the rate of LGH synthesis is 12%
of the rate of its hydrolysis. Partitioning studies suggest that
hydroxylamine readily attacks the thioester, therefore it is probably
not the rate determining step for the reverse reaction. Rather, the
step most likely to govern the rate of the reverse reaction is
formation of the thioester from glutamate (k_3). Thus, an upper limit
can be set to 12% of the forward rate for k3. However, the value of
k_3 was varied in order to attempt to fit the simulation to the actual
data. These values were applied to a simulation of the formation of
glutamate with time. It is apparent from the fit shown in Figure 7.1
that scheme (1) is not sufficient to describe the experimental data.
First, such a scheme results in the lack of a burst in glutamate
formation. Second, the slope of glutamate formation as a function of
time and thus kca[ are too high in the simulated data when compared
to the actual values (see Table 7.4).

145
Time (sec)
Figure 7.1: Comparison of the simulated to the actual data
representing glutamate formation as a function of time. The dashed
line () represents the simulation of equations (1) and (2). The
solid line () that parallels the actual data points (E) represents the
simulation from equation (3).

146
Table 7.1: Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by wild type AS-B according to schemes (1) and (2).
step
forward (kj
reverse He n)
1
1 x 10s NTsec'1
1.75xlO4 sec'1
2
1 sec1
1 x 10"6M" sec'1
3
1.6x10 sec'1
1,6xl0'2 sec"1
4
1.82xl07 sec'1
lxlO8 M"1 sec'1
Steps 1 through 3 were used in the simulation of scheme (1) whereas
all 4 steps were used in scheme (2).

147
The examination of the deacylation of the thioester, presented
in chapter 6, provided evidence that the deacylation was not the sole
rate determining step. Therefore, the additional step which may
influence the overall rate of the reaction should be added to the
model above. The knowledge of both the deacylation rate and the
for glutamate obtained in this work allow this expansion of the
minimal model to include both deacylation and product release as
separate steps as shown in scheme (2):
(2)
ES
where EP2 represents the enzyme-glutamate complex and step 4
represents product release. For the simulation of this model, steps 1-
3 were set at the same values as in scheme (1) (Table 7.1) and k.3was
varied. The value for k4 was derived (Table 7.1) from the K¡ for
glutamate assuming that k.4 approaches diffusion just as the case for
ki-
The high value for K¡ leads to the calculation of a very fast
product release, therefore the extra step added to the model had
little effect on the fit of the simulated values to the actual data
(Figure 7.1). A much more precise fit of the data was obtained by
adding an additional step to account for a slow conformational
change prior to the fast product release as shown in scheme (3)
below:

148
kl k2 k3 kd k5
S + E =^= ES E-TE EP2 FP2 E + P2
k-l k-2 + P¡ k-3 k-4 k-5
(3)
The rate constants of the first three steps of this scheme are the
same as in scheme (2) (Table 7.2). The inhibition constant for
glutamine will now be used in step 5 instead of step 4. The value for
k_4 was arbitrarily set to 1 sec1. Results in chapter 4 suggest that
another step occurring after the formation of the thioester
intermediate may be rate limiting in glutamine hydrolysis. In
scheme (3) the slow step could either be k4 or k5. Since inhibition
studies and the simulation of scheme (2) rule out k5 as being slow, k4
was given the slowest rate of scheme (3) (Table 7.2) with kcal setting
the lower limit to 0.04 sec'1. As shown in figure 7.1 and Table 7.3,
these estimates suggest that scheme (3) is sufficient to describe
glutamine hydrolysis catalyzed by wild-type AS-B.
The work in this dissertation has been combined to form a
minimal model of the glutamine hydrolysis reaction in which the
formation of the thioester is fast relative to its breakdown, and in
which two steps (deacylation and rate limiting conformational
change) regulate the rate of the reaction. This model will provide a
foundation on which to begin forming a complete description of the
reaction mechanism. In fact, efforts are now underway to obtain
experimental values using rapid quench techniques for some of the
estimated rate constants (such as k2) in this model. Rapid quench
combined with the thioester trapping techniques described in
chapter 5 could be used to compare the rates of thioester formation

149
Table 7.2: Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by AS-B according to equation (3).
step
forward (k^)
reverse (kn)
1
lxlO8 IVT'sec"1
1.75xl04 sec"
2
1 sec1
lxlO"6M' sec'
3
l.xlO'1 sec'1
l.xlO'2 sec'1
4
7xl0'2sec'1
1 sec'1
5
1.82xl07 sec1
lxlO8 M'1 sec'

150
Table 7.3: Comparison of the actual and simulated values obtained
for the kcat, slope, intercept, and [TE] from Figure 7.1.
Experimental
Simulated Values
values
(1)
(2)
(3)
\r
^ cat
0.040
0.135
0.135
0.043
slope
0.015
0.050
0.050
0.016
Y-intercept
0.34
-0.04
-0.04
0.15
[TE]
2.51xl0'6
6.29x10
6 6.29x10
2.46x 1 O'6

151
between the glutaminase and glutamine dependent asparagine
synthesis reactions. If the mechanism of nitrogen transfer involves a
direct attack of the amine of the tetrahedral intermediate on the
acceptor substrate then, at least at high pH, the glutamine dependent
asparagine synthesis reaction might have a faster acylation rate than
the glutaminase reaction since this reaction would not necessarily
require a protonation step.
Another feature of the reaction that will require experimental
testing is the existence of a rate determining conformational change
which has been predicted by the model. Studies are already
underway to measure the rate of fluorescence change during AS-B
catalyzed conversion of glutamine to glutamate using stop-flow
techniques. If the rate is found to be consistent with a rate
determining conformational change then it would be interesting to
compare this rate with that for the AS-B mutant R30A since the
reaction catalyzed by this mutant seems to have a different rate
limiting step (see partitioning experiments in chapter 4).
The studies summarized here represent the first step towards
completing a detailed chemical and kinetic description of glutamine
hydrolysis catalyzed by AS-B. With further experimental support,
the model developed in this work can be used as a basis for
comparison in examining the effects of mutations on glutamine
hydrolysis in efforts to determine the roles of the individual amino
acid residues in the utilization of glutamine. Ultimately, these results
may be significant in forming a bridge toward understanding the
overall mechanism of nitrogen transfer.

REFERENCES
Amuro, N, Paluh, JL, & Zalkin, H (1985) Replacement by site-directed
mutagenesis indicates a role for histidine 170 in the glutamine amide
transfer function of anthranilate synthase. J. Biol. Chem. 260:
14844-14849
Arad, D, Langridge, R & Kollman, PA (1990) A simulation of the
sulfur attack in the catalytic pathway of papain using molecular
mechanics and semiempirical quantum mechanics. J. Am. Chem. Soc.
112: 491-502
Badet, B, Vermoote, P, Haumont, P-Y, Lederer, F, Le Goffic, F (1987)
Glucosamine synthetase from Escherichia coli: purification,
properties, and glutamine-utilizing site location. Biochemistry 26:
1940-1948
Badet-Denisot, MA, Badet, B (1992) Chemical modification of
glucosamine-6-phosphate synthase by diethyl pyrocarbonate.
Evidence of a histidine requirement for enzymatic activity. Arch.
Biochem. Biophys. 292: 475-478
Badet-Denisot, MA, Badet, B (1993) Mechanistic investigations on
glucosamine-6-phosphate synthase, Bull. Soc. Chim. Fr. 130: 249-
255
Barshop, BA, Wrenn, RF, Frieden, C (1983) Analysis of numerical
methods for computer simulation of kinetic processes; development
of KINSIM, a flexible portable system. Anal. Biochem. 130: 134-145
Bernt, E & Bergmeyer, HU (1974) in Methods of Enzymatic Analysis
(Bergmeyer, H.U., ed) pp. 1704-1708, Academic Press, New York
152

Boehlein, SK, Richards, NGI, and Schuster, SM (1994a) Glutamine-
dependent Nitrogen Transfer in Escherichia coli Asparagine
Synthetase B. J. Biol. Chem. 269: 7450-7457
153
Boehlein, SK, Richards, NGJ, Walworth, ES, Schuster, SM (1994b)
Arginine 30 and Asparagine 74 Have Functional Roles in the
Glutamine Dependent Activities of Escherichia coli Asparagine
Synthetase B. J. Biol. Chem. 269: 26789-26795
Boehlein, SK, Schuster, SM, Richards, NGJ (1996) Glutamic acid-y-
monohydroxamate and hydroxylamine are alternate substrates for
Escherichia coli asparagine synthetase B. Biochemistry 35: 3031-
3037
Boehlein, SK, Walworth, ES, Richards, NGJ, Schuster, SM (1997a)
Mutagenesis and chemical rescue indicate residues involved in (}-
aspartyl-AMP formation by Escherichia coli asparagine synthetase B.
J. Biol. Chem. 272: 12384-12392
Boehlein, SK, Rosa-Rodriquez, JG, Schuster, SM, Richards, NGJ (1997b)
Catalytic activity of the N-terminal domain of Escherichia coli
asparagine synthetase B can be reengineered by a single mutation. J.
Am. Chem. Soc. 119: In press
Bork, P & Koonin, EV (1994) A P-loop-like motif in a widespread ATP
pyrophosphatase domain: implications for the evolution of sequence
motifs and enzyme activity. Prot. Struct. Funct. Genet. 20: 347-355
Bradford, MM (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
Brannigan, JA, Dodson, G, Duggleby, HJ, Moody, PCE, Smith, JL,
Tomchick, DR, Murzin, AG (1995) A protein framework with an N-
terminal nucleophile is capable of self-activation. Nature 378: 416-
419
Brocklehurst, K (1994) A sound basis for pH-dependent kinetic
studies on enzymes. Prot. Eng. 7: 291-299

154
Brocklehurst, K, Kowlessur, D, Patel, G, Templeton, W, Quigley, K,
Thomas, EW, Wharton, CW, Willenbrock, F, Szawelski, RJ (1988a)
Consequences of molecular recognition in the S,-S2 intersubsite
region of papain for catalytic-site chemistry. Biochem J. 250: 761-
772
Brocklehurst, K, Brocklehurst, SM, Kowlessur, D, ODriscoll, D, Patel, G,
Salih, E, Templeton, W, Thomas, E, Topham, CM, Willenbrock, F
(1988b) Supracrystallographic resolution of interactions contributing
to enzyme catalysis by use of natural structural variants and
reactivity-probe kinetics. Biochem. J. 256: 543-558
Brocklehurst, K, Willenbrock, F, & Salih, E (1987) in Hydrolytic
Enzymes (Neuberger, A, Brocklehurst, K, eds) Elsevier, Amsterdam
pp 39-158
Broome, J.D. (1961) Evidence that the L-asparaginase activity of
guinea pig serum is responsible for its antilymphoma effects. Nature
(London) 191: 1114-1115
Broome, J.D. (1963a) Evidence that the L-asparaginase activity of
guinea pig serum is responsible for its antilymphoma effects. I.
Properties of the L-asparaginase of guinea pig serum in relation to
those of the antilymphoma substance. J. Exp. Med. 118: 99-120
Broome, J.D. (1963b) Evidence that the L-asparaginase activity of
guinea pig serum is responsible for its antilymphoma effects. II.
Lymphoma 6C3HED cells cultured in a medium devoid of L-
asparaginase lose their susceptibility to the effects of guinea pig
serum in vivo. J. Exp. Med. 118: 121-147
Burchall, JJ, Reichelt, EC, Wolin, MJ (1964) Purification and properties
of the asparagine synthetase of Streptococcus bovis. J. Biol. Chem.
239: 1794-1798
Capizzi, RL, Bertino, JR, Skeel, RT, Creasey, WA, Zanes, R, Olayon, C,
Petterson, RG, Handschumacher, RE (1971) L-asparaginase: clinical,
biochemical, pharmacological, and immunological studies. Ann.
Intern. Med. 74: 893-900

Caplow,M & Jencks, WP (1964) The chymotrypsin-catalyzed
hydrolysis and synthesis of N-acetyl-L-tyrosine hydroxamic acid.
J. Biol. Chem. 239: 1640-1652
155
Carlomagno, MS, Chiariotti, L, Alifano, P, Nappo. AG, Bruni, CB (1988)
Structure and function of the Salmonella typhimurium and
Escherichia coli K-12 histidine operons. J. Mol. Biol. 203: 585-606
Cedar, H. & Schwarz, JH (1969a) The asparagine synthetase of
Escherichia coli. I. Biosynthetic role of the enzyme, purification, and
characterization of the reaction products. J. Biol. Chem. 244: 4122-
4127
Cedar, H. & Schwarz, JH (1969b) The asparagine synthetase of
Escherichia coli. II. Studies on mechanism. J. Biol. Chem. 244: 4122-
4127
Chakrabarti, R, & Schuster, SM (1997) L-Asparaginase: Perspectives
on the mechanism of action and resistance. Internat. J. Ped.
Hem./One. In Press
Chaparian, MG & Evans, DR (1991) The catalytic mechanism of the
amidotransferase domain of the Syrian hamster multifunctional
protein CAD. J. Biol. Chem. 266: 3387-3395
Cleland, WW (1963) The kinetics of enzyme catalyzed reactions with
two or more substrates or products. Biochim. Biophys. Acta 67: 104-
137
Cleland, WW (1982) An analysis of Haldane relationships. Methods
Enzymol. 87: 366-369
Cooney, DA & Handschumacher, RE (1970) L-asparaginase and L-
asparagine metabolism. Ann. Rev. Pharmacol. 10: 421-440
Cooney, DA, Driscoll, JS, Milman, HA, Jayaram, HN, Davis, RD (1976)
Inhibitors of L-asparagine synthetase in vitro. Cane. Treat. Rep. 60:
1493-1557

156
Cooney, DA, Jones, MT, Milman, HA, Young, DM, & Jayaram, HN (1980)
Regulators of the metabolism of L-asparagine. A search for
endogenous inhibitors. Int. J. Biochem. 11: 519-539
Crowther, D (1971) L-asparaginase and human malignant disease.
Nature 229: 168-171
Drenth, J, Kalk, KH, & Swen, HM (1976) Binding of chloromethyl
ketone substrate analogues to crystalline papain. Biochemistry 15:
3731-3738
Duggleby, HI, Tolley, SP, Hill, CP, Dodson, EJ, Dodson, G, Moody, PCE
(1995) Penicillin acylase has a single-amino-acid catalytic centre.
Nature 373: 264-268
Epand, RM & Wilson, IB (1963) Evidence for the formation of
hippuryl chymotrypsin during hydrolysis of hippuric acid esters. J.
Biol. Chem. 238: 1718-1723
Fisher, KJ (1993) Post-Translational Processing and Thr-206 are
required for glucosylasparaginase activity. FEES Lett. 323: 271-275
Gantt, IS, & Arfin, SM (1981) Elevated level of asparagine synthetse
activity in physiologically and genetically derepressed Chinese
hamster ovary cells due to increase rates of enzyme synthesis. J.
Biol. Chem. 256: 7311-7315
Goto, Y, Zalkin, H, Keim, PS, Heinrikson, RL (1976) Properties of
anthranilate synthetase component II from Pseudomonas putida.
J. Biol. Chem. 251: 941-949
Hanahan, D (1983) Studies on transformation of Escherichia coli with
plasmids. J. Mol. Biol. 166: 557-580
Hinchman, SK, Henikoff, S, Schuster, SM (1992) A relationship
between Asparagine Synthetase A and Aspartyl tRNA synthetase. J.
Biol. Chem. 267: 144-149

157
Hongo, S, Matsumoto, T, & Sato, T (1978) Purification and properties
of asparagine synthetase from rat liver. Biochim Biophys Acta 552:
258-266
Humbert, R & Simoni, RD (1980) Genetic and biochemical studies
demonstrating a second gene coding for asparagine synthetase in
Escherichia coli. J. Bacteriol. 142: 212-220
Hutson, RG, Kitoh, T, Moraga-Amador, DA, Cosic, S, Schuster, SM,
Kilberg, MS (1997) Amino acid control of asparagine synthetase:
relation to asparaginase resistence in human leukemia cells. Am. J.
Physiol. 272: C1691-C1699
Isupov, MN, Obmolova, G, Butterworth, S, Badet-Denisot, MA, Badet,
B, Polikarpov, I, Littlechild, JA, Teplyakov, A (1996) Substrate
binding is required of the assembly of the active conformation of the
catalytic site in Ntn amidotransferases: Evidence from the 1.8 A
crystal structure of the glutaminase domain of glucosamine 6-
phosphate synthase. Structure 4: 801-810
Kidd, S.T. (1953a) Regression of transplanted lymphomas induced in
vivo by means of normal guinea pig serum. I. course of
transplanted cancers of various kinds in mice and rats given guinea
pig serum, horse serum, or rabbit serum. J. Exp. Med. 98: 565-581
Kidd, S.T. (1953b) Regression of transplanted lymphomas induced in
vivo by means of normal guinea pig serum. II. Studies on the
nature of the active serum constituent. J. Exp. Med. 98: 583-606
Kim, JH, Krahn, JM, Tomchick, DR, Smith, JL, Zalkin, H (1996)
Structure and function of the glutamine
phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 271:
15549-15557
Kiriyama, Y, Kubota, M, Takimoto, T, Kitoh, T, Tanizawa, A, Akiyama,
Y, Mikawa, H (1989) Biochemical characterization of U937 cells
resistant to L-asparaginase: the role of asparagine synthetase.
Leukemia 3: 294-297

158
Knowles, JR (1976) The intrinsic pKa-values of functional groups in
enzymes: improper deductions from the pH-dependence of steady-
state parameters. CRC Crit. Rev. Biochem. 4: 165-173
Larsen, MC & Schuster (1992) The topology of the glutamine and ATP
binding sites of human asparagine synthetase. Arch. Biochem.
Biophys. 299: 15-22
Lobel, L, OBrian, RT, McIntosh, S, Aspnes, GT, Capizzi, RL (1979)
Methotrexate and asparaginase combination chemotherapy in
refractory ALL of childhood. Cancer 43: 1089-1094
Lowe, J, Stock, D, Jap, B, Zwickl, P, Baumeister, W, Huber, R (1995)
Crystal structure of the 20S proteasome from the Archaeon T.
acidophilum at 3.4 resolution. Science 268: 533-539
Luehr, CA & Schuster, SM (1985) Purification and properties of beef
pancreatic asparagine synthetase. Arch. Biochem. Biophys. 237:
335-346
Lusty, CJ (1992) Detection of an enzyme bound g-glutamyl acyl ester
of carbamyl phosphate synthetase of Escherichia coli. FEBS LETT
314: 135-138
MacPhee, K, Nelson, R, & Schuster, SM (1983) Neurospora crassa
mutants deficient in asparagine synthetase. J. Bacteriol. 156: 475-
478
Mantsala, P & Zalkin, H (1984) Glutamine Amidotransferase Function.
J. Biol. Chem. 259: 14230-14236
Markin, RS and Schuster, SM (1979) Multiple forms of rat liver L-
asparagine synthetase. Biochem. Biophys. Res. Commun. 88: 583-
588
Markin, RS, Luehr, CA, & Schuster, SM (1981) The kinetic mechanism
of beef pancreatic L-asparagine synthetase. Biochemistry 20: 7226-
7232

159
Mehlhaff, PM & Schuster, SM (1991) Bovine pancreatic asparagine
synthetase explored with substrate analogs and specific monoclonal
antibodies. Arch. Biochem. Biophys. 284: 143-150
Mei, B, and Zalkin, H (1989) Cysteine-Histidine-Aspartate catalytic
triad is involved in glutamine amide transfer function in purF-type
glutamine amidotransferases. J. Biol. Chem. 264: 16613-16619
Meister, A, Soben, H, Tice, SV, Frasen, PE (1952) Transamination and
associated deamidation of asparagine and glutamine. J. Biol. Chem.
197: 319-330
Meloche, HP (1967) Bromopyruvate inactivation of 2-Keto-3-deoxy-
6-phosphogluconic aldolase. I. Kinetic evidence for active site
specificity. Biochemistry 6: 2273-2280
Menard, R, Khouri, HE, Plouffe, C, Dupras, R, Ripoll, D, Vernet, T,
Tessier, DC, Laliberte, F, Thomas, DY, Storer, AC (1990) A protein
engineering study of the role of aspartate 158 in the catalytic
mechanism of papain. Biochemistry 29: 6706-6713
Menard, R, Khouri, HE, Plouffe, C, Laflamme, P, Dupras, R, Vernet, T,
Tessier, DC, Thomas, DY, Storer, AC (1991) Importance of hydrogen
bonding interactions involving the side chain of Aspl58 in the
catalytic mechanism of papain. Biochemistry 30: 5531-5538
Menard, R., Plouffe, C., Laflamme, P., Vernet, T., Tessier, D.C., Thomas,
D.Y., Storer, A.C. (1995) Modification of the Electorstatic Environment
is tolerated in the oxyanion hole of the cysteine protease papain.
Biochemistry 34: 464-471
Milman, HA & Cooney, DA (1979) Partial purification and properties
of L-asparagine synthetase from mouse pancreas. Biochem. J. 181:
51-59
Miran, SG, Chang, SH, & Raushel, FM (1991) Role of four histidine
residues in the amidotransferase domain of carbamoyl phosphate
synthetase. Biochemistry 30: 7901-7907

160
Nichols, BP, Miozzari, GF, van Cleemput, M, Bennett, GN, Yanofsky, C
(1980) Nucleotide sequences of the trpG regions of Escherichia coli,
Shigella dysenteriae, Salmonella typhimurium, and Serratia
marcescens. J. Mol. Biol. 142: 503-517
Oinonen, C, Tikkanen, R, Rouvinen, J,Peltonen, L (1995) Three-
dimensional structure of human lysosomal aspartylglucosamidase.
Nature Struct. Biol. 2: 1105-1108
OLeary, MH, Urberg, M, Young, AP (1974) Nitrogen isotope effects on
the papain-catalyzed hydrolysis of N-benzoyl-L argininamide.
Biochemistry 13: 2077-2081
Patterson, MK, & Orr, G (1969) Regeneration, tumor, dietary, and L-
asparaginase effects on asparagine biosynthesis in rat liver. Cancer
Res. 29: 1179-1183
Peller, L, and Alberty, RA (1959) Multiple Intermediates in steady
state enzyme kinetics. I. The mechanism involving a single substrate
and product. J. Am. Chem. Soc. 81: 5907-5914
Penefsky, H.S. (1979) A centrifuged-column procedure for the
mechanism of ligand binding by beef heart F,. Methods Enzymol.
56: 527-530
Pfeiffer, NE, Mehlhaff, PM, Wylie, DE, Schuster, SM (1986) Monoclonal
antibodies specific for bovine pancreatic asparagine synthetase. J.
Biol. Chem. 261: 1914-1919
Pfeiffer, NE, Mehlhaff, PM, Wylie, DW, & Schuster, SM (1987)
Topographical separation of the catalytic sites of asparagine
synthetase explored with monoclonal antibodies. J. Biol. Chem. 262:
11565-11570
Piette, J, Nyunoya, H, Lusty, CJ, Cunin, R, Weyens, G, Crabeel, M,
Charlier, D, Glansdorff, N, Pierard (1984) DNA sequence of the carA
gene and the control region of carB: Tandem promoters, respectively
controlled by arginine and the pyrimidines, regulate the synthesis of
carbamoyl-phosphate synthetase in Escherichia coli K-12. Proc. Natl.
Acad. Sci. U.S.A. 81:4134-4138

161
Pike, DC & Beevers, L (1982) Kinetic studies on asparagine synthetase
from mung beans (vigna radiata (L) wilczek) with a phosphonate
analogue of a proposed reaction intermediate. Biochim. Biophys.
Acta. 708: 203-209
Ramos, F & Wiame, JM (1979) The genetics of asparagine synthesis
activity in yeast. Eur. J. Biochem. 94: 409-417
Ramos, F & Wiame, JM (1980) Two asparagine synthetases in
Saccharomyces cerevisiae. Eur. J. Biochem. 108: 373-377
Richards, NGJ, & Schuster, SM (1992) An alternative mechanism for
the nitrogen transfer reaction in asparagine synthetase. FEBS Lett.
313: 98-102
Richards, NGJ, & Schuster, SM (1997) Mechanistic issues in
Asparagine Synthetase catalysis. Adv. Enymol. Relat. Areas Mol. Biol.
In Press
Rognes, SE (1970) A glutamine-dependent asparagine synthetase
from yellow lupine seedlings. FEBS lett. 10: 62-66
Roux, B & Walsh, CT (1992) Para-aminobenzoate synthesis in
Escherichia coli: Kinetic and mechanistic characterization of the
amidotransferase PabA. Biochemistry 31: 6904-6910
Sanz, GF, Sanz, MA, Rafecas, FJ, Martinez, JA, Martin-Aragon, G,
Marty, ML (1986) Teniposide and cytarabine combination
chemotherapy in the treatment of relapsed adolescent and adult
acute lymphoblastic leukemia. Cancer Treat. Rep. 70: 1321-1324
Schendel, FJ, Mueller, E, Stubbe, J, Shiau, A, Smith, JM (1989)
Formylglucinamide ribonucleotide synthetase from Escherichia coli:
cloning, sequencing, overproduction, isolation, and characterization.
Biochemistry 28: 2459-2471
Scofield, MA, Lewis, WS, Schuster, SM (1990) Nucleotide Sequence of
Escherichia coli asnB and deduced amino acid sequence of asparagine
synthetase B. J. Biol. Chem. 265: 12895-12902

162
Seemuller, E, Lupas, A, Baumeister, W (1996) Autocatalytic
processing of the 20S proteasome. Nature 382: 468-470
Sheng, S, Moraga-Amador, DA, van Heeke, G, Allison, RD, Richards,
NGJ, Schuster, SM (1993) Glutamine inhibits the ammonia-dependent
activities of two cys-1 mutants of human asparagine synthetase
through the formation of an abortive complex. J. Biol. Chem. 268:
16771-16780
Smith, IL, Zaluzec, EJ, Wery, J-P, Niu, L, Switzer, RL, Zalkin, H, Satow,
Y(1994) Structure of the allosteric regulatory enzyme of purine
biosynthesis. Science 264: 1427-1433
Smith, JL (1995) Structures of glutamine amidotransferases from the
puring biosynthetic pathway. Biochem. Soc. Trans. 23: 894-898
Spiro, RG (1969) Glycoproteins: their biochemistry, biology, and role
in human disease. New Eng. J. Med. 281: 991-1000
Stoker, PW, OLeary, MH, Boehlein, SK, Schuster, SM, Richards, NGJ
(1996) Probing the mechanism of nitrogen transfer in Escherichia coli
asparagine synthetase by using heavy atom isotope effects.
Biochemistry 35: 3024-3030
Storer, AC and Menard, R (1994) Catalytic mechanism in papain
family of cysteine peptidases. Methods Enzymol. 244: 486-500
Tallal, L, Tan, C, Oettgen, H, Wollner, N, McCarthy, M, Helson, L,
Burchenal, J, Karnofsky, D, Murphy, ML (1970) E.coli L-asparaginase
in the treatment of leukemia and solid tumor in 131 children. Cancer
25: 306-320
Terebello, HR, Anderson, K, Wiernik, PH, Cuttner, J, Cooper, RM, Faso,
L, Beerenberg, JL (1986) Therapy of refractory adult lymphoblastic
leukemia with vincristine and prednisone plus tandem methotrexate
and L-asparaginase. Results of a cancer and leukemia group B study.
Amer. J. Clin. Oncol. 9: 411-415

163
Tesmer, JJG, Klem, TJ, Deras, ML, Davisson, VJ, Smith, JL (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
Thoden, JB, Holden, HM, Wesenberg, G, Raushel, FM, Rayment, I
(1997) Structure of carbamoyl phosphate synthetase: A journey of
96 from substrate to product. Biochemistry 36: 6305-6316
Tiedeman, AA, Smith, JM, Zalkin, H (1985) Nucleotide sequence of the
guaA gene encoding GMP synthetase of Escherichia coli K12. J. Biol.
Chem. 260: 8676-8679
Tso, J.Y., Zalkin, H., van Cleemput, M, Yanofsky, C, Smith, JM (1982)
Nucleotide sequence of Escherichia coli purF and deduced amino acid
sequence of glutamine phosphoribosylpyrophosphate
amidotransferase. J. Biol. Chem. 257: 3525-3531
Uren, JR, & Handschumacher, RE (1977) Enzyme therapy. Cancer 5:
457-487
Van Heeke, G, & Schuster, SM (1989a) Expression of human
asparagine synthetase in Escherichia coli. J. Biol. Chem. 264: 5503-
5509
Van Heeke, G, & Schuster, SM (1989b) The N-terminal cysteine of
human asparagine synthetase is essential for glutamine dependent
activity. J. Biol. Chem. 264: 19475-19477
Van Heeke, G & Schuster, SM (1990) Expression of human asparagine
synthetase in Saccharomyces cerevisiae. Protein Eng. 3: 739-744
Vanoni, MA, Edmondson, DE, Rescigno, M, Zanetti, G, Curti, B (1991)
Mechanistic studies on Azospirillum hrasil.ense glutamate synthetase.
Biochemistry 30: 11478-11484
Vauquelin, LN, Robiquet, PJ (1806) Ann. Chim. (Paris) 57: 88

164
Vollmer, SJ, Switzer, RL, Hermodson, MA, Bower, SG, Zalkin, H (1983)
The glutamine-utilizing site of Bacillus subtilis glutamine
phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 258:
10582-10585
Wellner, VP, Anderson, PM, & Meister, A (1973) Interaction of
Escherichia coli carbamoyl phosphate synthetase with glutamine.
Biochemistry 12: 2061-2066
Wells, PR (1963) Linear free energy relationships. Chem. Rev. 63:
171
Weng, M, Makaroff, CA, Zalkin, H (1986) Nucleotide sequence of
Escherichia coli pyrG encoding CTP synthetase. J. Biol. Chem. 261:
5568-5574
Zalkin, H (1993) The Amidotransferases. Adv. Enzymol. Relat. Areas
Mol. Biol. 66: 203-309

BIOGRAPHICAL SKETCH
Holly Schnizer was born on April 11, 1969 to Dr. Flint and Dottye
Gray of Kingsport, Tennesse. In 1991, Holly received her Bachelor of
Science degree in chemistry from Centre College in Danville,
Kentucky. The following fall, Holly began her graduate studies at the
University of Florida in the Department of Biochemistry and
Molecular Biology under the supervision of Dr. Sheldon Schuster. In
the spring of 1996, Holly married Dr. Richard Schnizer whom she met
while working in the laboratory. She expects to receive her Ph.D. in
Biochemistry and Molecular Biology in the fall of 1997.
165

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
Sheldon M. Schuster, Chair
Professor of Biochemistry and
Molecular Biology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
//
/
-1/.
///S'/t
Harry S. Nffk
Professor or Neuroscience
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
Thomas W. OBrien
Professor of Biochemistry and
Molecular Biology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
Thomas P. Yang j
Associate Professor of Biochemistry
and Molecular Biology ¡

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
R. Donald Allison
Associate Scientist of Biochemistry
and Molecular Biology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a thesis for the degree of
Doctor of Philosophy.
\ N,
[ //-> J -V <
7 |
Richard W. Moyer ( j
Professor of Moleculai^Genetics
and Microbiology
This Dissertation was submitted to the Graduate Faculty of the
College of Medicine and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
1 f
a
-i-L*
December, 1997
Dean, College of Medicine
Dean, Graduate School



26
mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2 units of
glutamate dehydrogenase) was added and incubated 10 minutes at
room temperature. The absorbance at 340 nm was measured and
the concentration of glutamate was determined by comparison to a
standard curve. Initial velocities were measured at 9 different
substrate concentrations and each velocity was an average of 2-4
measurements. Stability of the wild-type and mutant AS-B enzymes
at each pH was confirmed by incubation in buffer of the appropriate
pH for 20 minutes followed by an assay of activity at pH 8. Initial
velocity conditions were confirmed at each pH by linear plots of v
versus time and v versus enzyme concentration.
Data Analysis
The values of Km and kcat were determined from least-squares
fit to v0 vs [S] plots using the Prism software package supplied by
Graphpad, Inc., (San Diego, CA) according to equation 1:
Vmax [S]
v = (1)
Km + [S]
Where v is the maximal velocity, [S] is the concentration of st
Vmax is the maximal velocity, and Km is the Michaelis constant. The
pH dependence of kcat/Km obtained from equation 1 were fit to the
bell-shaped curve described by equation 2:
Yiim
Y= (2)
l + !0(Pkl-pH) + 10(pH-pK2)


148
kl k2 k3 kd k5
S + E =^= ES E-TE EP2 FP2 E + P2
k-l k-2 + P¡ k-3 k-4 k-5
(3)
The rate constants of the first three steps of this scheme are the
same as in scheme (2) (Table 7.2). The inhibition constant for
glutamine will now be used in step 5 instead of step 4. The value for
k_4 was arbitrarily set to 1 sec1. Results in chapter 4 suggest that
another step occurring after the formation of the thioester
intermediate may be rate limiting in glutamine hydrolysis. In
scheme (3) the slow step could either be k4 or k5. Since inhibition
studies and the simulation of scheme (2) rule out k5 as being slow, k4
was given the slowest rate of scheme (3) (Table 7.2) with kcal setting
the lower limit to 0.04 sec'1. As shown in figure 7.1 and Table 7.3,
these estimates suggest that scheme (3) is sufficient to describe
glutamine hydrolysis catalyzed by wild-type AS-B.
The work in this dissertation has been combined to form a
minimal model of the glutamine hydrolysis reaction in which the
formation of the thioester is fast relative to its breakdown, and in
which two steps (deacylation and rate limiting conformational
change) regulate the rate of the reaction. This model will provide a
foundation on which to begin forming a complete description of the
reaction mechanism. In fact, efforts are now underway to obtain
experimental values using rapid quench techniques for some of the
estimated rate constants (such as k2) in this model. Rapid quench
combined with the thioester trapping techniques described in
chapter 5 could be used to compare the rates of thioester formation


135
1991). In support of the contrasts mentioned in chapter 4, CAD
(Chaparan & Evans, 1991) and now AS-B have been shown to posses
a different rate limiting step than found in amide hydrolysis
catalyzed by papain. However, the present study suggests that while
the deacylation step is the slowest step in glutamine hydrolysis
catalyzed by CAD, AS-B catalyzed hydrolysis seems to have an
additional step which has a rate comparable to deacylation.
Secondly, high concentrations of ammonia essentially abolished the
appearance of the thioester intermediate formed during CAD-
catalyzed glutamine hydrolysis demonstrating that the acylation step
is a reversible process (Chaparan & Evans, 1991). In contrast,
ammonia seems to have a minimal effect on the AS-B-catalyzed
thioester deacylation rate (Figure 6.4) which suggests a greater
difficulty in reversing the formation of the thioester intermediate in
the AS-B catalyzed reaction. Consistent with this observation, studies
in chapter 3 show that ammonia does not inhibit in the range of
concentrations used and that asparagine synthesis does not occur to a
significant degree upon mixing the wild type enzyme with glutamate
and ammonia. Hydroxylamine can be used in the synthesis of LGH
therefore the lack of glutamine synthesis using ammonia is probably
not due to the inability of glutamate to bind to the enzyme and form
a thioester intermediate.
With the knowledge of the steady-state concentration of the
thioester formed and the rate of deacylation in the glutaminase
reaction catalyzed by wild type AS-B it is now possible to determine
the involvement of individual residues in this step. This report
describes an initial characterization of the deacylation step during


concentration and therefore all kcat and kca[/Km values given in this
chapter have been corrected by dividing the protein concentration
by the factor 2.7.
25
TABLE 2.1
Oligonucleotides used in construction of site-directed mutants of
AS-B
Oligo
Number
Oligonucleotide Sequence
ss65
5 A GCT TCC CAT ATG TGT TCA ATT TTT GGC GTA
TTC GAT 3
ss350R
5 CGC CCC CGC GTT AAC GTC GAC AAT TGA CAG ACG
TTC GGC GGC GAG A 3
ss349R
5 CGC CCC CGC GTT AAC GTC GAC AAT TGA CAG ACG
TTC GGC GGC GAG A 3
Glutaminase Assays
Initial rates of L-glutamine and L-glutamic-y-monohydroxamate
(LGH) hydrolysis were determined by a modified procedure which
uses glutamate dehydrogenase in the presence of NAD+ to measure
glutamate concentration (Bernt and Bergmeyer, 1974). Reactions
(100 pi) containing lOOmM Bis-Tris and Tris-HCl (pH 6.0-9.0) each
(the kinetic constants appear to be independent of the nature and
concentrations of these buffers), 8mM MgCl2 and various
concentrations of glutamine or LGH were initiated by the addition of
wild-type (0.7-2 pg) or mutant (4 pg) AS-B and incubated at 37C. A
control containing denatured enzyme was used to monitor any
glutamate formed from spontaneous breakdown. The reactions were
terminated by boiling and a coupling reagent (300 mM glycine, 250


100
complicated. Indeed, additional evidence for the presence of a
thioester intermediate along the reaction pathway of AS-B catalyzed
glutamine hydrolysis is given in the formation of LGH upon addition
of hydroxylamine to the glutaminase reaction. However, an increase
in the hydroxylamine concentration in the reaction was expected to
either increase the total product formed, if thioester breakdown (k3)
were rate limiting, or to have no effect on the amount of product
formed, if thioester formation (k2) were rate limiting. For example,
hydroxylamine partitioning experiments have been used to show
that deacylation is rate limiting in glutamine hydrolysis catalyzed by
a class I amidotransferase, CAD (Chaparian & Evans., 1991). In
contrast, an increase in hydroxylamine concentration in the AS-B
catalyzed glutaminase reaction resulted in an unexpected decrease in
total product formed (Figure 4.1). This observed decrease in total
product may have been caused by hydroxylamine inhibition of the
enzyme. However, the Km for hydroxylamine in an LGH synthesis
reaction (see chapter 3) has been shown to be several times higher
than the concentration used in these experiments. Alternatively,
LGH release from the enzyme may be very slow thus hindering the
regeneration of free enzyme for further hydrolysis. Consistent with
this hypothesis, an R30A mutant of AS-B with impaired substrate
binding (shown in its Ks, Figure 4.2) exhibited an increase in total
product formed with increasing concentrations of hydroxylamine
indicating a rate limiting deacylation step (Figure 4.2).
The findings described in this report represent mechanistic
differences between substrate hydrolysis catalyzed by the class II
amidotransferase, AS-B, and hydrolysis catalyzed by the thiol


128
Time (sec)
Figure 6.3: Effect of ATP on the rate of deacylation. A glutamyl-
enzyme adduct was formed at 5C by incubating 0.74 nmol wild-type
AS-B with 1.5 mM glutamine (SA 20,000 dpm/nmol) and 1 mM ATP,
O in 100 mM Bis-Tris and Tris (pH 8.0). After incubating the
reaction for 1 min, the radioactive glutamine was diluted by the
addition of 200 mM non-radioactive glutamine. The samples were
quenched at various time intervals after the dilution as described in
materials and methods and the amount of radioactivity associated
with the protein was determined by scintillation counting. The
deacylation in the presence of ATP, O, was compared to that in the
absence of ATP, .


11
thioester. Initially, it was thought, from results of site-directed
mutagenesis (Mei & Zalkin, 1989), that His-102 was involved in such
acid-base catalysis for GPA and covalent modification studies using
diethyl pyrocarbonate suggested that GFAT (Badet-Denisot & Badet,
1992) also possessed this important histidine. However, site-
directed mutagenesis of each of the conserved histidines in the GAT
domain of AS-B revealed little or no change in the kinetic parameters
(Boehlein et al., 1994a). Subsequent structural studies of GPA (Kim
et ah, 1996; Smith et ah, 1994) and the N-terminal domain of GFAT
(Isupov et ah, 1996) did not reveal a histidine in the appropriate
position to participate in acid-base catalysis. Alternatively, these
structural studies led to the proposal that the a-amino group of cys-1
serves as the general acid-base catalyst (Smith, 1995). Subsequent
structural comparisons of GPA with other hydrolytic enzymes placed
the class II enzymes into a larger superfamily called the N-terminal
nucleophile (Ntn) hydrolases which includes the 20S proteasome and
penicillin acylase (Brannigan et ah, 1995; Duggleby et ah, 1995;
Seemuller et ah, 1996). This family is characterized by an N-terminal
catalytic nucleophile which is thought to be activated by proton
transfer from its side chain to the free N-terminus (Seemuller et ah,
1996). However, experiments aimed at establishing the functional
role of the N-terminal amine using mutants of GPA and GFAT with an
additional N-terminal residue have yielded contradictory results
(Isupov et ah, 1996; Kim et ah, 1996).
Further evidence against an ammonia mediated nitrogen
transfer in AS-B catalyzed asparagine synthesis was revealed in an
investigation of the kinetic isotope effects (KIE) (Stoker et ah 1996)


139
transfer of nitrogen. Therefore, one way to discriminate between
which mechanism describes asparagine synthesis catalyzed by AS-B
is to obtain a detailed model, including the rates of the individual
steps, of both glutamine dependent reactions. Such a model also
would expedite the course of understanding how individual amino
acid residues participate in catalysis as it would allow us to visualize
which individual step is affected by a mutation. Finally, a detailed
model of the glutaminase reaction is necessary in order to complete
the description of the overall kinetic mechanism of AS-B.
The goal of this dissertation was to define both chemical and
kinetic characteristics of the glutaminase reaction which could aid in
building this much needed model. Certainly as shown here, the
initial stages are complete. Due to some similarities between the
class II amidotransferases and the thiol proteases, such as the
presence of a catalytically important cysteine, the original model for
the glutaminase reaction was based on the mechanism of amide
hydrolysis catalyzed by the thiol protease, papain. Therefore, as a
first step, it was important to ascertain that the major features of
this mechanism, such as the cys-his dyad and the formation of a
thioester intermediate, could be found in AS-B catalyzed glutamine
hydrolysis.
For papain, the importance of Cys-25 and His-159 is first
apparent through their complete conservation among all members of
the family of thiol proteases (Brocklehurst et al., 1987). The
conserved histidine plays a number of essential roles in the
hydrolysis reaction including activation of the Cys-25, protonation of
the leaving group, and activation of a water molecule for nucleophilic


113
observed in the crystal structure of DON-inactivated Escherichia coli
GPA (Kim et al., 1996). Second, radiolabelled enzyme was not found
when the CIA AS-B was incubated with [14C]-L-glutamine in place of
wild-type enzyme (Tables 5.1 & 5.3). This AS-B mutant lacks both
glutamine-dependent synthetase and glutaminase activity but
retains ammonia dependent activity and the ability to bind
glutamine tightly (Boehlein et al., 1994a). Since previous studies
employing asparagine synthetases, as well as other Class II
amidotransferases, have demonstrated the involvement of the Cys-1
side chain in formation of the thioester linkage, the lack of a y-
glutamyl CIA complex strongly suggests that covalent modification
occurs at the GAT-domain active site requiring the Cys-1 thiolate.
The failure of the CIA AS-B mutant to form a radiolabeled complex is
further evidence that non-specific glutamine binding cannot account
for the radioactivity associated with the wild-type enzyme. Third,
the amount of intermediate saturates with increasing glutamine
concentration, the half saturation point being consistent with steady
state parameters observed for the AS-B glutaminase reaction (Figure
5.1). And finally, the intermediate was isolated only under acidic
conditions and was not stable at higher pH which is consistent with
the alkaline lability of thioesters.
Although the covalent y-glutamyl AS-B adduct appeared
unstable under the denaturation conditions used in the gel filtration
assay, possibly reflecting residual activity of slowly denaturing
enzyme in the solutions containing SDS or guanidinium HC1,
approximately 34% of the enzyme was radiolabeled based on data
from filter-binding. The percentage of enzyme in the thioester form,


137
change in the acylation step the amount of thioester that is formed
increases.
The ability to measure the rate constants of the individual
steps in glutamine hydrolysis will allow a complete characterization
of the wild type enzyme and thus will provide a model of comparison
for the determination of mutations on catalysis. This report
represents the first step toward achieving this goal. The deacylation
rate for the wild type AS-B now can be used as part of a model in the
simulation of the glutaminase reaction presented in the following
chapter. Furthermore, an extension of these studies using rapid
quench analysis will provide rates of both acylation and deacylation
thus furthering the development of the model and understanding the
mechanism of glutamine hydrolysis.


56
Table 3.1: Thermodynamic properties of glutamine and LGH
hydrolysis catalyzed by wild-type AS-B and the R30A mutant.
Substrate/
AG*
AH*
AS*
TAS*
enzyme
kJ mol1
kJ mol 1
J mol 'K1
kJ mol"1
(37C)
(37C)
(5-40C)
(37C)
gln/wt
1 67
89
-252
-78
LGH/wt
152
82
-226
-70
gln/R30A
147
77
-226
-70
The initial rates were determined at pH 8.0 in 100 mM Bis-Tris and
Tris-HCl at temperatures ranging from 5-40C, under conditions
described in the materials and methods.


122
In the previous chapter, the steady-state concentration of the
putative thioester intermediate, formed during glutamine hydrolysis
catalyzed by wt AS-B, was determined to reach 35% of the total
enzyme concentration. These studies are continued in this chapter
which presents the measurement of the deacylation step for the
wild-type enzyme, and the effect of mutations on the steady-state
concentration of the thioester and its deacylation rate.
Materials and Methods
Enzymes and reagents
The construction of mutants in which asparagine-74 is replaced
by alanine (N74A) or arginine-30 is replaced with an alanine (R30A)
has been described previously (Boehlein et al., 1994b) Recombinant
AS-B, N74A, and R30A were expressed and purified using standard
procedures (Boehlein et al., 1994a). [U-14C]-L-Glutamine (277
mCi/mmol) was purchased from Amersham. The nitrocellulose
filters were purchased from Bio-Rad. Other reagents including L-
glutamine were purchase from SIGMA Chemical Company (St. Louis,
MO) and were of the highest commercial purity.
Effect of Glutamine on the Steady-State Concentration of the
Thioester Intermediate
The radiolabeled enzyme complex was isolated by filter
binding as described in chapter 5. A glutamyl-enzyme complex was
formed by incubating 0.74 nmol of either wt, N74A, or R30A with
various concentrations of L-glutamine (SA 22,000 dpm/nmol) in a
100 pi reaction containing 100 mM Tris-HCl (pH 8.0) for 30 sec at 5C.
The reactions were quenched in 1 ml of 8% TCA and 100 pi BSA (10


112
intermediate. For example, a carboxyl group on the enzyme could
attack the side chain of glutamine to form an anhydride
intermediate. In this case, hydroxylamine could attack on either the
glutamyl or enzyme side of the central oxygen of this intermediate.
Therefore, each attack of the hydroxylamine on the enzyme side
would result in a covalently labeled enzyme which would be
unavailable for further turnover thus resulting in inhibition of the
reaction with time. Since a time dependent inhibition of the reaction
in the presence of hydroxylamine was not observed and since an
essential glutamate or aspartate residue which could participate in
forming such an intermediate has never been found in AS-B, a
mechanism involving an anhydride intermediate is unlikely.
Alternatively, LGH could be formed via a pyroglutamate although
this would not be efficient in ammonia transfer. The possibility of an
oxygen ester intermediate rather than a thioester is also unlikely
since hydroxylamine causes partitioning even at pH 6 which is
consistent with a thioester and not the less reactive oxygen ester
which only reacts with hydroxylamine at alkaline pH. The
demonstration of a radiolabeled intermediate upon incubation of wt
AS-B with [l4C]-glutamine provides further evidence for an
acylenzyme intermediate. Several observations support that this
isolated adduct is in fact the thioester. First, a radiolabeled enzyme
was not formed when the wild-type enzyme was labeled with DON
prior to the reaction (Tables 5.1 & 5.3). DON has been shown to react
at the active site of mammalian asparagine synthetases (Mehlhaff &
Schuster, 1991; Larsen & Schuster, 1992) and causes irreversible
enzyme inactivation by modification of the thiol of Cys-1, as


ABBREVIATIONS
ALL: Acute Lymphoblastic Leukemia
AS: asparagine synthetase
AS-A: ammonia dependent asparagine synthetase A from E.coli
AS-B: asparagine synthetase B from E. coli
AMP: adenosine monophosphate
ATP: adenosine 5'-triphosphate
CIA: mutant of AS-B with cysteine-1 replaced by alanine
DON: 6-diazo-5-oxonorleucine
E. coli: Escherichia coli
GAT: glutamine amide transfer
GFAT: glutamine fructose-6-phosphate amidotransferase
GPA: glutamine 5'-phosphoribosyl-l-pyrophosphate
amidotransferase
H47N: mutant of AS-B with histidine-47 replaced by asparagine
KIE: kinetic isotope effect
LGH: L-glutamic-y-monohydroxamate
N74A: AS-B mutant with asparagine-74 replaced by alanine
PP¡: pyrophosphate
R30A: AS-B mutant with arginine-30 replaced by alanine
wt: wild-type
xii


52
be used to calculate the equilibrium constant for the overall reaction
using the Haldane relationship for a one-substrate/two-product
reaction given in equation (6) (Cleland, 1963):
(6)
where Keq is the equilibrium constant for the conversion of substrate
to product, Vf and Vr represent the forward and reverse reaction
rates, respectively, and the indices p, q, and a refer to hydroxylamine,
glutamate and LGH, respectively. Using equation (6) and the
following parameters, obtained at 37C: krcat = 2.4 sec'1, Km(NH20H)= 95
mM, Ki(glu)=182 mM, krcat= 0.29 sec'1, K,
0.83 mM; a value of
m(LGH)-
Keq=172 was calculated. Using this value for Kcq and the equation, AG
= -RTlnKeq, the difference in free energy between substrates and
products, AG, was determined to be -13.2 kJ/mol K.
Product inhibition glutamine hydrolysis catalyzed bv wild-tvpe AS-B
Figures 3.6 and 3.7 show the plots for the inhibition of the
glutaminase reaction by glutamate at 5C and 37C, respectively. The
K¡ values for glutamate determined from the replot of the slope
versus glutamate concentration (Figure 3.8 & 3.9) at 5C and 37C
were 260 mM and 182 mM, respectively. An attempt to obtain the K¡
for ammonia was made. However, NH4C1 concentrations up to 200
mM seemed to have little effect on the rate of the glutaminase
reaction.


114
which was similar to that reported for PabA (Roux & Walsh, 1992)
indicates that this intermediate is kinetically important.
The isolation of the intermediate described here is an essential
step in obtaining an accurate model of glutamine hydrolysis catalyzed
by Class II amidotransferases. Furthermore, the ability to detect and
quantitate this intermediate has important implications in
determination of the mechanism of glutamine-dependent nitrogen
transfer for AS-B and other enzymes in this family. For asparagine
synthetases, current experimental evidence suggests that nitrogen
transfer can proceed via direct attack of the tetrahedral intermediate
(Scheme 1) on 8-aspartyl-AMP, or by release of enzyme-bound,
unprotonated ammonia from glutamine (Richards & Schuster, 1997).
While these mechanisms of glutamine amide transfer are
substantially different, the formation of a thioacylenzyme
intermediate during asparagine synthesis is a common requirement.
Methods for isolation of the thioester therefore have potential
application in experiments aimed at resolving which of these two
transfer mechanisms is operative in AS-B. For example, if the
thioester intermediate is formed prior to the breakdown of
8-aspartyl -AMP, then nitrogen transfer mediated by enzyme-bound
ammonia is mechanistically most likely. On the other hand, if no
thioester is formed until after B-aspartyl-AMP breakdown, then it is
most probable that direct amide transfer takes place via covalent
intermediates.


88
[LGH]
O jiM
a 20 |iM
a 40 jiM
v 60 jiM
80 jiM
100 |XM
Figure 4.9: The effect of various concentrations of LGH on DON
inactivation of AS-B catalyzed LGH hydrolysis. Wild-type AS-B was
incubated with 6 pM DON in the presence of LGH (concentrations
varied as indicated in the figure). At time points ranging from 0-18
min, an aliquot was removed from the mixture and tested for
residual activity in a reaction containing 12 raM LGH. The LGH
hydrolysis activity of the DON modified enzymes were measured as
described in materials and methods.


19
The mechanism of nitrogen transfer remains controversial and poorly
understood. Results from studies using site-directed mutagenesis
suggested the catalytic importance of a conserved N-terminal cysteine,
a histidine, and an aspartic acid in glutamine utilization in GPA (Mei &
Zalkin, 1989). This observation led to the hypothesis that glutamine
amidotransferases catalyze glutamine hydrolysis by a mechanism
similar to that of amide hydrolysis catalyzed by the thiol proteases,
typified by papain (for review see Brocklehurst et al., 1987), as shown
below (1).
Binding Acylation Deacylation
TH I TH II
In analogy to the papain mechanism, a nucleophilic attack of the N-
terminal cysteine on the side-chain amide of glutamine forms a
tetrahedral intermediate (TH I). Protonation of the amide allows C-N
bond cleavage to produce a thioester intermediate and free ammonia.
The ammonia is then transferred to the acceptor molecule and the
thioester intermediate is broken down, via another tetrahedral
intermediate (TH II), to produce glutamate and free enzyme. As
illustrated above, an active site cysteine and an acid-base catalyst (a
role played by his-159 in papain) are vital components to such a
reaction scheme. The cysteine is involved in the activation of
glutamine as described above. In papain, the histidine is thought to


37
Table 2.4:
Kinetic constants for glutamine hydrolysis catalyzed
by the AS-B mutant H47N at 37C
PH
kcat
K.
kcat/Km
(sec'1)
(mM)
(M^sec-1)
6.00
0.19
0.01
18.6 2.8
1 0
6.25
0.22
0.008
13.1 2.0
1 7
6.50
0.30
0.005
8.3 0.9
36
6.75
0.30
0.008
5.4 0.4
55
7.00
0.30
0.005
3.2 0.2
9 4
7.25
0.30
0.003
2.4 0.1
125
7.50
0.38
0.005
2.5 0.1
152
7.75
0.32
0.003
2.5 0.1
128
8.00
0.32
0.003
3.0 0.1
107
8.25
0.40
0.005
4.1 0.2
98
8.50
0.46
0.008
6.1 0.3
75
8.75
0.38
0.01
7.8 0.7
49
9.00
0.43
0.01
10.5 1.1
4 1


glutamine utilizing domain of asparagine synthetase B, the similarity
of pH profiles of wild-type and an asparagine synthetase B mutant in
which a conserved histidine is replaced with an asparagine, suggests
that this histidine is not likely to be involved as part of the cys-his
dyad which is essential to papain-catalyzed hydrolysis. Second, this
study describes the development of a model for glutamine hydrolysis
catalyzed by wild-type asparagine synthetase B from a minimal
mechanism. The use of such a model was confirmed by the
demonstration that one main feature of papain catalysis, the
formation of a thioester, was conserved in asparagine synthetase B
catalyzed glutamine hydrolysis.
The values for the dissociation constants for glutamine and the
inhibition constants for glutamate were determined. Furthermore,
the isolation of a putative thioester intermediate provided the
opportunity to measure the rate of a single step in the reaction,
deacylation. These constants along with the information gained from
the kinetic constants of L-glutamic-y-monohydroxamate synthesis
catalyzed by asparagine synthetase B were used in a computer
simulation of the burst of glutamate production with time which
showed that an additional rate limiting conformational change must
be included to describe the observed kinetic behavior of wild-type
asparagine synthetase B.
xiv


124
Previous studies of the glutaminase reaction have indicated
that the presence of ATP stimulates the hydrolysis of glutamine
(Boehlein et al., 1994), therefore it was of interest to determine
whether or not ATP affected the rate of the individual deacylation
step. An examination of the effect of glutamine on the steady-state
concentration of the thioester in the presence of 1 mM ATP suggests
that ATP causes a slight decrease in the maximum concentration of
thioester formed at saturating glutamine concentrations (Figure 6.2,
Table 6.1). However, the addition of 1 mM ATP has no effect on the
rate of deacylation (Figure 6.3, Table 6.1).
The effect of NH4C1 on the deacylation rate was examined. If
the step involving C-N bond cleavage to release ammonia is readily
reversible it is expected that addition of ammonia to the reaction
would shift the equilibrium away from the thioester and thus result
in a decrease in the concentration of thioester formed and the rate of
its deacylation. However, measurement of the deacylation of the
thioester intermediate showed that the presence of 10 mM NH4C1 had
little effect on the rate of deacylation (Figure 6.4, Table 6.1).
The effect of glutamine on the steady state concentration of the
glutamvl-enzvme adduct formed with the AS-B mutant. R30A
Studies in chapter 4 demonstrated that the rate limiting step of
glutamine hydrolysis catalyzed by an AS-B mutant impaired in
glutamine binding, R30A, was deacylation. Therefore, it would be
expected that the deacylation rate would be the same as the kcat
value determined in glutamine hydrolysis. This comparison could be
used in contrast to the slightly varying rates obtained for the wild-
type enzyme. Unfortunately, the measurement of the deacylation


76
[NH2OH] (mM)
Figure 4.1: Partitioning of the thioester intermediate to L-glutamic-
y-monohydroxamate using hydroxylamine in a reaction catalyzed by
wt AS-B. A mixture containing wt AS-B, 50 mM glutamine, 100 mM
Bis-Tris and Tris-HCl (pH 8), and the indicated concentrations of
hydroxylamine was incubated at 37C for 45 min. The reactions
were terminated by the addition of 4% TCA. An aliquot was removed
to determine glutamate formed () by an end-point assay using
glutamate dehydrogenase. LGH formation (A) was measured by
adding a FeCl2 solution and measuring the absorbance at 540 nm as
described in materials and methods. The nmol of LGH and glutamine
formed at each concentration of hydroxylamine were added to give
the total product formed (sa).


162
Seemuller, E, Lupas, A, Baumeister, W (1996) Autocatalytic
processing of the 20S proteasome. Nature 382: 468-470
Sheng, S, Moraga-Amador, DA, van Heeke, G, Allison, RD, Richards,
NGJ, Schuster, SM (1993) Glutamine inhibits the ammonia-dependent
activities of two cys-1 mutants of human asparagine synthetase
through the formation of an abortive complex. J. Biol. Chem. 268:
16771-16780
Smith, IL, Zaluzec, EJ, Wery, J-P, Niu, L, Switzer, RL, Zalkin, H, Satow,
Y(1994) Structure of the allosteric regulatory enzyme of purine
biosynthesis. Science 264: 1427-1433
Smith, JL (1995) Structures of glutamine amidotransferases from the
puring biosynthetic pathway. Biochem. Soc. Trans. 23: 894-898
Spiro, RG (1969) Glycoproteins: their biochemistry, biology, and role
in human disease. New Eng. J. Med. 281: 991-1000
Stoker, PW, OLeary, MH, Boehlein, SK, Schuster, SM, Richards, NGJ
(1996) Probing the mechanism of nitrogen transfer in Escherichia coli
asparagine synthetase by using heavy atom isotope effects.
Biochemistry 35: 3024-3030
Storer, AC and Menard, R (1994) Catalytic mechanism in papain
family of cysteine peptidases. Methods Enzymol. 244: 486-500
Tallal, L, Tan, C, Oettgen, H, Wollner, N, McCarthy, M, Helson, L,
Burchenal, J, Karnofsky, D, Murphy, ML (1970) E.coli L-asparaginase
in the treatment of leukemia and solid tumor in 131 children. Cancer
25: 306-320
Terebello, HR, Anderson, K, Wiernik, PH, Cuttner, J, Cooper, RM, Faso,
L, Beerenberg, JL (1986) Therapy of refractory adult lymphoblastic
leukemia with vincristine and prednisone plus tandem methotrexate
and L-asparaginase. Results of a cancer and leukemia group B study.
Amer. J. Clin. Oncol. 9: 411-415


CHAPTER 2
pH DEPENDENCE OF THE GLUT AMIN ASE REACTION CATALYZED BY
ASPARAGINE SYNTHETASE B
Introduction
Asparagine Synthetase B (AS-B) catalyzes the ATP dependent
synthesis of asparagine using aspartic acid and either glutamine or
ammonia as a nitrogen source (reactions 1 and 2). In the absence of
aspartate and ATP, this enzyme also is capable of catalyzing
glutamine hydrolysis (reaction 3).
L-Gln + ATP + L-Asp L-Asn + AMP + PPj + L-Glu (reaction 1)
NH3 + ATP + L-Asp -> L-Asn + AMP + PPj (reaction 2)
L-Gln + H20 -4 Glu + NH? (reaction 3)
Amino acid sequence alignment of the N-terminal glutamine utilizing
domain of AS-B indicates that this enzyme is a member of the class II
glutamine amidotransferase family (Zalkin, 1993) which includes
glutamine 5-phosphoribosylpyrophosphate amidotransferase (GPA)
(Tso et al., 1982), glutamine fructose-6-phosphate amidotransferase
(GFAT) (Badet-Denisot & Badet, 1993), and glutamate synthase (Vanoni
et al., 1991). Like AS-B, members of this family of enzymes catalyze
the transfer of the amide nitrogen of glutamine to various acceptor
molecules including amino acids, nucleotides, and carbohydrates.


108
In contrast to the slow disappearance of the y-glutamyl-
enzyme adduct in the gel filtration approach, the radiolabeled
protein was stable when isolated by filter binding (Lusty, 1992), and
could be used in quantitative measurements. In these experiments,
enzyme was combined with [14c]-L-glutamine, denatured in TCA,
and the precipitate filtered and washed with HC1. The radioactivity
remaining on the filter was measured using liquid scintillation
counting. Approximately 35% of the enzyme was radiolabeled in
reaction mixtures containing wild-type AS-B (Table 5.3).
Examination of the effect of glutamine concentration on the steady-
state level of the glutamyl-enzyme complex revealed a saturation
curve which gave a maximum molar ratio of intermediate to enzyme
of 0.14 and a KM of 0.14 mM for glutamine (Figure 5.1). In control
studies, radioactivity was not associated with the filter when
incubation solutions lacked the enzyme, or contained either the CIA
AS-B mutant or wild-type AS-B that had been covalently modified
with DON (Table 5.3).
The stability of thioesters decreases with increasing pH,
therefore if the isolated intermediate is a thioester the radioactivity
associated with the enzyme should diminish under alkaline
conditions. Initially, an attempt was made to determine the stability
of the intermediate at various conditions of pH using the filter
binding assay. Unfortunately, the enzyme redissolved under the
alkaline conditions of the wash and passed through the filter. Thus,
the observed disappearance of radioactivity associated with the
membrane was due to the decrease in enzyme on the filter rather
than instability of a thioester intermediate. As an alternative


129
Table 6.1: Characterization of the glutamyl-enzyme intermediate.
Enzyme
Max % of
Enzyme labeled
(%)
Deacylation
rate
(sec'1)
lc
iVcat
(sec-1)
w t
35.1 0.9
0.16 0.02
0.043 0.0007
wt + 1 mM ATP
26 1.1
0.18 0.01
0.076 0.002
wt + 10 mM NH4C1
n d
0.11 0.01
nd2
R30A
50 3
nd1
0.042 0.001
N74A
57 1
0.062 0.005
0.024 0.0005
All values in the table were determined at 5C. The deacylation rate
for the reaction catalyzed by R30A could not be obtained due to the
high concentrations of glutamine necessary to saturate the enzyme.
2Ammonia has been shown to have little effect on the kcat for the
glutaminase reaction catalyzed by wt AS-B at 37C.


4 DETERMINATION OF THE RATE DETERMINING STEP
IN ASPARAGINE SYNTHETASE B-CATALYZED
GLUTAMINE HYDROLYSIS 67
Introduction 67
Materials and Methods 70
Results 74
Discussion 98
5 EVIDENCE OF A THIOESTER INTERMEDIATE FORMED
DURING GLUTAMINE HYDROLYSIS CATALYZED BY
ASPARAGINE SYNTHETASE B 102
Introduction 102
Materials and Methods 104
Results 106
Discussion 109
6 CHARACTERIZATION OF A THIOESTER INTERMEDIATE I20
Introduction 120
Materials and Methods i22
Results 123
Discussion I34
7 DISCUSSION AND FUTURE DIRECTIONS I38
REFERENCES I52
BIOGRAPHICAL SKETCH
165


118
[Gin] mM
Figure 5.1 The effect of glutamine on the steady-state concentration
of the thioester. Various concentrations of [14C] glutamine were
combined with wild-type AS-B (0.74 nmol) and the covalent adduct
was isolated using a filter binding method as described in Materials
and Methods.


146
Table 7.1: Kinetic parameters at 5C for glutamine hydrolysis
catalyzed by wild type AS-B according to schemes (1) and (2).
step
forward (kj
reverse He n)
1
1 x 10s NTsec'1
1.75xlO4 sec'1
2
1 sec1
1 x 10"6M" sec'1
3
1.6x10 sec'1
1,6xl0'2 sec"1
4
1.82xl07 sec'1
lxlO8 M"1 sec'1
Steps 1 through 3 were used in the simulation of scheme (1) whereas
all 4 steps were used in scheme (2).


109
approach, the adduct was formed, the reaction was terminated, and
the resulting solution was added to a gel filtration column either
equilibrated at pH 2 or pH 12. As shown in Table 5.4, 35% of the
enzyme was labeled after gel filtration at pH 2 whereas only 10% of
the enzyme was labeled after gel filtration at pH 12. While a
decrease in radiolabeled enzyme isolated by gel filtration was
mentioned above, this problem was alleviated by maintaining a
constant time interval of 20 s between terminating the reaction and
addition of the solution to the column. As shown through the
statistics at each pH. variability in the amount of radiolabeled
enzyme due to slow termination does not seem to be a factor.
Discussion
Glutamine amidotransferases can be divided into two families
based on multiple sequence alignment of the glutamine amide
transfer (GAT) domains (Zalkin. 1993). Class I (formerly TrpG)
enzymes, such as GMP synthetase (Tesmer et al., 1996) and
carbamoyl phosphate synthetase (CPS) (Thoden et al., 1997) are
characterized by conserved Cys, His and Asp active site residues that
are thought to function as a catalytic triad in a manner analogous to
thiol proteases (Brocklehurst et al., 1987). Thus, in analogy to the
papain mechanism (Brocklehurst et al., 1987), glutamine hydrolysis
proceeds by initial attack of the Cys thiolate on the side chain amide
to yield a tetrahedral intermediate (Scheme 1). Breakdown of this
intermediate after N-protonation, by the conserved active site
histidine to transform the nitrogen into an efficient leaving group,
then yields a thioester. Subsequent hydrolysis of this intermediate


147
The examination of the deacylation of the thioester, presented
in chapter 6, provided evidence that the deacylation was not the sole
rate determining step. Therefore, the additional step which may
influence the overall rate of the reaction should be added to the
model above. The knowledge of both the deacylation rate and the
for glutamate obtained in this work allow this expansion of the
minimal model to include both deacylation and product release as
separate steps as shown in scheme (2):
(2)
ES
where EP2 represents the enzyme-glutamate complex and step 4
represents product release. For the simulation of this model, steps 1-
3 were set at the same values as in scheme (1) (Table 7.1) and k.3was
varied. The value for k4 was derived (Table 7.1) from the K¡ for
glutamate assuming that k.4 approaches diffusion just as the case for
ki-
The high value for K¡ leads to the calculation of a very fast
product release, therefore the extra step added to the model had
little effect on the fit of the simulated values to the actual data
(Figure 7.1). A much more precise fit of the data was obtained by
adding an additional step to account for a slow conformational
change prior to the fast product release as shown in scheme (3)
below:


61
Figure 3.7: Glutamate inhibition of the glutaminase reaction
catalyzed by wild-type AS-B at 37C. The rate of glutamine
hydrolysis was determined as described in the materials and
methods in the presence of 0 (), 40 (), 80 (A), and 100 () mM
glutamate.


14
ASPm
Figure 1.2 Mechanism of nitrogen transfer via an imide intermediate
3 as proposed by Richards and Schuster, 1992. Attack of the
primary amide occurs directly on the activated L-aspartate 2. Cys-1
and Asp-33 are then involved in the hydrolysis of the imide to yield
L-glutamate and L-asparagine in subsequent steps.


71
Steady State Kinetics
The kinetic parameters of AS-B-catalyzed glutamine, LGH, and
L-glutamine-hydrazide hydrolysis were obtained in an end-point
assay which uses glutamate dehydrogenase in the presence of NAD+
to measure the concentration of glutamate formed (Bernt &
Bergmeyer, 1974). Reactions (100 pi) containing lOOmM Bis-Tris and
Tris-HCl (pH 7.5) each, 8mM MgCl2 and various concentrations of
glutamine, LGH, or glutamic acid-y-hydrazide were initiated by the
addition of 1.5 pg wild-type AS-B and incubated at 37C. After 10
minutes, the reactions were terminated by boiling for 2 minutes.
Reactions in which the enzyme was added to the mix during the
termination period of boiling were used to monitor any glutamate
formed from spontaneous breakdown. A coupling reagent (300 mM
glycine, 250 mM hydrazine, pH 9, 1 mM ADP, 1.6 mM NAD+ 2.2
units of glutamate dehydrogenase) was added to the reaction and the
resulting mixture was incubated 10 minutes. The absorbance at 340
nm was measured and the concentration of glutamate was
determined by comparison to a standard curve. Initial velocities
were measured at 9 different substrate concentrations and each
velocity was an average of 2-4 measurements.
The values of Km and kcat were determined from least-squares
fit to v0 vs [S] plots using the Prism software package supplied by
Graphpad, Inc., (San Diego, CA).
Determination of the K. for glutamine and LGH
The rate of inactivation by the glutamine analog, 6-diazo-5-
oxonorleucine (DON), was determined by incubating 30 pg wt AS-B in
a 300 pi solution containing 100 mM Bis-Tris and Tris-HCl, pH 8 and


57
Figure 3.4: Effect of hydroxylamine concentration on the synthesis of
LGH catalyzed by wild-type AS-B. Wild-type AS-B (18.5 pg) was
incubated with various concentrations of hydroxylamine and
saturating concentrations of glutamate (150 mM) in Tris-HCl, pH 8.0
at 37C for 20 min. The reaction was terminated by the addition of
16% TCA and LGH formation was measured as described in materials
and methods.


13
Schuster, 1992). The first proposal for an alternate mechanism of
nitrogen transfer involved the formation of an imide intermediate
(Richards & Schuster, 1992). According to this hypothesis, a direct
nucleophilic attack of the primary amide of glutamine on the p-
aspartyl-AMP intermediate 2 to form an imide intermediate 3.
Reaction of this intermediate with cys-1 causes the release of
asparagine and the formation of a thioacylenzyme intermediate 1
which forms free enzyme and glutamate in a mechanism similar to
the ammonia mediated mechanism. An attempt to isolate such an
intermediate was made using the AS-B mutant (CIA) which lacks the
cysteine required to catalyze C-N bond cleavage of the imide
intermediate. However, an imide intermediate was not detected by
HPLC using chemically synthesized imide as a standard, when CIA
AS-B was incubated with glutamine, ATP, and aspartic acid. It was
concluded that participation of an imide intermediate in nitrogen
transfer catalyzed by AS-B was unlikely (Boehlein et al., unpublished
observations).
In a second alternative hypothesis for the mechanism of
nitrogen transfer (Stoker et al., 1996), Cys-1 reacts with glutamine to
form the first tetrahedral intermediate 5 (figure 1.3). In contrast to
the ammonia mediated mechanism, however, the tetrahedral
intermediate makes a subsequent direct attack on the p-aspartyl-
AMP intermediate to form an intermediate in which glutamine is
covalently linked to p-aspartyl-AMP 6. Cleavage of the C-N bond in
this intermediate releases the acylenzyme and forms a second
tetrahedral intermediate 8. Asparagine is then formed from a loss of


150
Table 7.3: Comparison of the actual and simulated values obtained
for the kcat, slope, intercept, and [TE] from Figure 7.1.
Experimental
Simulated Values
values
(1)
(2)
(3)
\r
^ cat
0.040
0.135
0.135
0.043
slope
0.015
0.050
0.050
0.016
Y-intercept
0.34
-0.04
-0.04
0.15
[TE]
2.51xl0'6
6.29x10
6 6.29x10
2.46x 1 O'6


92
Time (min)
[Gin] mM
a
ir
o
2
A
5
+-
7
10

15

20
X
25
V
30
Figure 4.13: Glutamine protection from DON inactivation of the R30A
catalyzed glutaminase reaction. R30A was incubated with 0.65 mM
DON and various concentrations of glutamine (given in the figure).
At time points ranging from 0-18 min, an aliquot was removed from
the mixture and tested for residual activity in a reaction containing
12 mM glutamine. The glutaminase activities of the DON modified
enzyme were measured as described in materials and methods.


86
[DON]
4 jliM
6 jiM
a 8 jiM
* 10
Figure 4.7: DON inactivation of wild-type catalyzed LGH hydrolysis in
the absence of LGH. Wild type AS-B was incubated with the
concentrations of DON indicated in the figure. At time points ranging
from 0-18 min an aliquot was removed from the mixture and tested
for residual activity in a reaction containing 12 mM glutamine. The
glutaminase activities of the DON modified enzymes were measured
as described in materials and methods.


TABLE 2.6:
Kinetic parameters for wild-type AS-B and the H29A, H80A, H47N, and R30A mutants at three
values of pH
kcat]
(sec )
pH 6.5
(mM)
VK,
(M'rsec )
kcat
(sec )
pH 7.5
K,,,
(mM)
k,/K
(M sec )
kca.
(sec )
pH 9.0
K,
(mM)
kCj,t/Km
(M sec )
wt
1.89
0.03
2.40
0.10
787
1.92
0.03
1.01
0.06
1901
2.56
0.05
2.33
0.12
1099
H29A
0.94
0.10
34.47
5.23
27
1.08
0.13
10.36
1.84
104
2.29
0.57
115 28
20
H80A
1.57
0.11
30.92
2.59
51
4.54
0.65
5.02
0.78
904
5.89
0.54
58 6
176
H47N
0.30
0.005
8.3 0.9
36
0.38
0.005
2.46
0.11
154
0.43
0.01
10.47
1.15
41


48
and incubated 10 minutes at room temperature. The absorbance at
340 nm was measured and the concentration of glutamate was
determined by comparison to a standard curve, initial velocities were
measured at 9 different substrate concentrations and each velocity
was an average of 3 measurements. Stability of the wild-type and
mutant AS-B enzymes at each temperature was confirmed by
incubation in buffer of the appropriate temperature for various
amounts of time followed by an assay of activity at pH 8. Initial
velocity conditions were confirmed at each pH by linear plots of v
versus time and v versus enzyme concentration. The pH of the buffer
was adjusted to pH 8 at each temperature studied to account for the
pKa change with temperature.
Determination of for glutamate and hvdroxylamine in the AS-B
catalyzed synthesis of LGH
The Michaelis constant, Km, of glutamate in the synthesis of LGH
was determined by incubating wild-type AS-B with various
concentrations of glutamate and a saturating concentration of
hydroxylamine (150 mM) in Tris-HCl, pH 8.0, at 37C for 15 min. In
a similar manner the Km for hydroxylamine was determined by
incubating the wild-type enzyme with various concentrations of
hydroxylamine and a saturating concentrations of glutamate (150
mM) in Tris-HCl, pH 8.0, at 37C for 20 min. In both cases, the 300 pi
reaction was terminated by the addition of 100 pi of 16 % TCA. LGH
formation was determined in a final volume of 500 pi by adding a
solution containing 80% TCA, 6N HC1. and 10% FeCl3 in 0.02 N HC1 to
the remaining reaction, centrifuging the samples in a microcentrifuge
to remove particulates, and measuring the absorbance of the


12
for AS-B catalyzed glutamine dependent reactions. The KIEs
associated with placing 15N in the primary amide of glutamine were
examined for both glutamine dependent reactions and the resulting
values were interpreted using 15N KIE determinations for papain
catalyzed peptide hydrolysis (OLeary et al, 1974). For papain, the
KIE value was shown to be consistent with a mechanism where the
first irreversible step is C-N bond cleavage to form the thioester
intermediate (OLeary et al., 1974). It was expected that C-N bond
cleavage would be rate limiting in the AS-B catalyzed glutaminase
reaction as well and thus give the same 15N (V/K) isotope effect as
papain. Furthermore, an ammonia mediated nitrogen transfer would
require a common pathway for glutamine hydrolysis in the two
glutamine dependent reactions. Therefore, if AS-B followed an
ammonia mediated mechanism, it was expected that the 15N KIEs
determined for both glutamine dependent reactions would be
identical. Surprisingly, the glutaminase 15N (V/K) was significantly
smaller than that obtained for papain catalyzed peptide hydrolysis,
indicating that a step other than C-N bond cleavage may be rate
limiting. Moreover, the KIE for glutaminase was smaller than that
for the glutamine dependent synthesis reaction. These results were
regarded as the first direct evidence against the mediation of
nitrogen transfer by enzyme-bound ammonia.
In addition to the above mentioned inconsistencies between
AS-B catalytic behavior and an ammonia mediated nitrogen transfer,
questions addressing the difficulties of maintaining the free ammonia
in the active site in its unprotonated form led to the proposal of
alternative mechanisms for nitrogen transfer (Figure 1.2) (Richards &


134
Discussion
Isolation of the thioester has allowed the examination of an
individual step in glutamine hydrolysis, deacylation. The results
presented here describe the use of this tool to show that the
deacylation step is not solely rate determining in glutamine
hydrolysis catalyzed by AS-B. The possibility of multiple steps
making significant contributions to the overall rate rather than a
single rate determining step was implied in chapter 4. In this
chapter, not only was the deacylation rate 4-fold faster than kcat for
glutamine hydrolysis catalyzed by wild type AS-B but it was not
affected by the presence of ATP (Table 6.1) which has been shown to
stimulate the glutaminase reaction (Boehlein et al., 1994b). An
increase in the rate of a reaction would require a change in rate in
the slowest step or a step with a comparable rate to the slowest step
of the reaction. Therefore, the above results suggest that a step in
addition to deacylation is rate limiting and contributes to the overall
rate constant for the reaction catalyzed by wild type AS-B. Further
evidence for the lack of ATP involvement in the deacylation step
comes from the absence of ATP stimulation of glutamine hydrolysis
when deacylation is rate limiting such as the case which has been
demonstrated for the glutaminase reaction catalyzed by the AS-B
mutant, R30A (see chapter 4).
The characterization of the deacylation rate occurring in the
AS-B catalyzed glutaminase reaction reveals some interesting
contrasts in relation to amide hydrolysis catalyzed by the thiol
protease papain and the amidotransferase. CAD (Chaparan & Evans,


47
procedures (Boehlein et al., 1994a). Protein concentrations were
determined by the method of Bradford (1976) using an assay kit
supplied by Bio-Rad and mouse immunoglobulin G to construct a
standard curve. During the course of these investigations an
alternate method of determining the protein concentration revealed
that the current use of mouse immunoglobulin G as a standard gives
erroneously high protein concentrations for AS-B. Amino acid
composition analysis of AS-B, carried out by the Protein Chemistry
Core Facility of ICBR at the University of Florida, showed that the
actual value for the concentration of this enzyme is 2.7 times smaller
than that obtained using the Bradford assay. Therefore, all of the
values of enzyme concentration and therefore all kcat and kcat/Km
values given in this chapter have been corrected by dividing the
protein concentration by the factor 2.7.
Glutaminase Assays
Initial rates of L-glutamine and L-glutamic-y-monohydroxamate
(LGH) hydrolysis were determined by a modified procedure which
uses glutamate dehydrogenase in the presence of NAD+ to measure
glutamate concentration (Bernt & Bergmeyer, 1974). Reactions (100
pi) containing lOOmM Bis-Tris and Tris-HCl (pH 8.0) each, 8mM MgCl2
and various concentrations of glutamine or LGH were initiated by the
addition of wild-type or mutant R30A AS-B and incubated at various
temperatures (4C-40C). A control reaction lacking enzyme was used
to monitor any glutamate formed from spontaneous breakdown. The
reactions were terminated by the addition of 20 pi of IN acetic acid. A
coupling reagent (300 mM glycine, 250 mM hydrazine, pH 9, 1 mM
ADP, 1.6 mM NAD+ 2.2 units of glutamate dehydrogenase) was added


58
Figure 3.5: Effect of glutamate concentration on the synthesis of LGH
catalyzed by wild-type AS-B. Wild-type AS-B (18.5 pg) was
incubated with various concentrations of glutamate and a saturating
concentration of hydroxylamine (150 mM) in Tris-HCl, pH 8.0 at 37C
for 15 min. The reactions were terminated by the addition of 16%
TCA and the formation of LGH was measured as described in
materials and methods.


125
constant was impractical due to the high concentration and therefore
the high specific activity of glutamine necessary to saturate this
mutant enzyme. However, it was possible to examine the effect of
glutamine on the steady-state concentration of the thioester formed
with R30A. As shown in Figure 6.5 and Table 6.1, the maximum
percentage of R30A which was labeled was higher than that obtained
with the wild-type enzyme.
Characterization of a glutamvl-enzyme adduct formed with the AS-B
mutant N74A
Examination of the effect of glutamine concentration on the
steady-state level of the glutamyl-enzyme complex formed with
N74A revealed a saturation curve which gave a maximum
percentage of enzyme labeled with radioactivity of 57% and a Km of
0.06 mM for glutamine (Figure 6.6 and Table 6.1). The thioester
concentration in reactions using N74A followed a mono-exponential
decay when plotted against time of incubation with unlabeled
glutamine. The rate constant for deacylation determined from this
plot was 0.062 sec'1, significantly slower than that of wild-type
(Figure 6.7 and Table 6.1).


TE (nmol)
126
Figure 6.1: Rate of deacylation of the thioester intermediate formed
during wild-type AS-B catalyzed glutamine hydrolysis.


46
through similarities in the activation parameters which would
suggest that the same step is rate limiting in both reactions. In
contrast, work in the following chapters suggests that glutamine
hydrolysis catalyzed by the AS-B mutant, R30A, has a different rate
limiting step than that catalyzed by wt AS-B and these differences
might be revealed in different values for the activation parameters.
A complete description of the glutaminase reaction requires 1)
the order of product release, 2) the inhibition constants of the
products, and 3) knowledge of whether or not the reaction is
reversible. Therefore, in order to address these issues, product
inhibition studies were employed and the ability of the enzyme to
catalyze the synthesis of glutamine and LGH was examined.
Materials and Methods
Chemicals and Reagents
[14C]-glutamine was purchased from Amersham. Scintillation
fluid, ScintiVers II* was obtained from Fischer (Orlando, FL). DE-
81 anion exchange chromatography paper was purchased from
Whatman (Hillsboro, OR). Other chemicals, including L-glutamine and
L-glutamic-y-monohydroxamate, and hydroxylamine were obtained
from SIGMA Chemical Company (St. Louis, MO) and were of the
highest commercial purity.
Enzyme Preparation
Construction of recombinant wild-type AS-B and the mutant
AS-B in which R30 has been replaced by alanine (R30A) have been
described elsewhere (Boehlein et al., 1994a,b). Both wild-type and
mutant proteins were expressed, and purified using standard


CHAPTER 6
CHARACTERIZATION OF A THIOESTER INTERMEDIATE
I ntroduction
The ability to isolate an intermediate on the reaction pathway
of glutamine hydrolysis catalyzed by AS-B (Chapter 5) will provide
an opportunity to gain a more detailed description of the reaction by
examining the rate of the individual step of deacylation. Previous
chapters have suggested that the rate determining step of the
reaction occurs after the formation of the covalent intermediate, but
that the deacylation step may not be the slow step. Therefore,
knowledge of the rate of deacylation will allow a comparison with
the overall rate of the reaction. If indeed, deacylation is not rate
limiting then the rate of this step should be significantly faster than
that of the overall reaction.
The ability to measure the rate of deacylation also provides a
tool with which to determine the significance of conserved residues
in an individual step of the reaction. The importance of Cys-1 in
making the thioacyl bond has already been implied in chapter 5 by
the inability to capture a glutamyl-enzyme intermediate when this
cysteine residue is replaced with an alanine (Chapter 5). Two other
residues, conserved among the class II amidotransferases, which
may participate in either the acylation or deacylation process are
Asn-74 and Arg-30. Evidence is mounting that Asn-74 plays a role,
120