Inhibition and Functional Characterization of Asparagine Synthetase

Material Information

Inhibition and Functional Characterization of Asparagine Synthetase
GUTIERREZ, JEMY A. ( Author, Primary )
Copyright Date:


Subjects / Keywords:
Active sites ( jstor )
Ammonia ( jstor )
Diphosphates ( jstor )
Enzyme substrates ( jstor )
Enzymes ( jstor )
Isotope effects ( jstor )
Kinetics ( jstor )
pH ( jstor )
Purification ( jstor )
Signals ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Jemy A. Gutierrez. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
Resource Identifier:
649814499 ( OCLC )


This item is only available as the following downloads:

Full Text




Copyright 2006 by Jemy A. Gutierrez


This work is dedicated to my parents and Fairland.


iv ACKNOWLEDGMENTS This project was funded by the Florida Bi omedical Research Program, NIH Grant DK61666, and NIH Training Grant in Cancer Biology T32 CA-09126. I also acknowledge the financial support from the Rueg amer Fellowship of the Department of Chemistry. I am grateful to my doctoral dissertati on committee, Dr. Ronald K. Castellano, Dr. Nicole A. Horenstein, Dr. Michael S. Kilb erg, Dr. Jon D. Stewart, and especially my research advisor, Dr. Nigel G. J. Richards, fo r allowing me to work on this project, for his guidance and encouragement, and for constan tly pushing the entire research group to do better. Thanks go to my collaborators, Dr. Wall ace W. Cleland and Jeremy Van Vleet of the University of Wisconsin in Madison for the work on heavy atom isotope effects, and Dr. Yuan-Xiang Pan of Dr. Mich ael KilbergÂ’s laboratory at th e UF College of Medicine for the experiments involving MOLT-4 cells. To the former members of the Richar dsÂ’ group, Drazenka Svedruzic, Stefan Jonsson, Christopher Chang, Mihai Ciustea, an d Luke Koroniak, I am grateful for the practical training, advice, and camaraderie. Special thanks go to Susan Abbatiello and Cory Toyota, whose technical skills, scie ntific insights and unconditional friendship made the laboratory such a positive place for work. To my Filipino friends here in Gaines ville, Jen, Joy, Jhoana, Cris, Dodge, the Pabits, the Javelosas, and the rest of th e Pinoy_UF group, I express my heartfelt thanks


v for the friendship and the amazing sense of family that they have created for me. Specials thanks go to my parents, Eve lio and Gloria, whose hard work and countless sacrifices gave us a chance at a better life, and to whom I owe whatever success I may stumble upon. I thank my sister Leny for her strength, my brother Vjay for his thoughtfulness, and my sister Ivy and her family for their support and prayers. Finally, I thank my husband Fairland, fo r being my number one support system, and for bringing unbelievable joy to my life.


vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 Acute Lymphoblastic Leukemia (ALL).......................................................................1 Asparagine Synthetase, AS...........................................................................................2 Research Goals.............................................................................................................7 2 EXPRESSION, PURIFICATION, AND STEADY STATE KINETIC CHARACTERIZATION OF C-TERMINALLY TAGGED RECOMBINANT HUMAN ASPARAGINE SYNTHETASE..................................................................8 Introduction................................................................................................................... 8 Results and Discussion...............................................................................................10 Experimental Section..................................................................................................15 Optimization of Expression Parameters..............................................................15 Expression and Purification.................................................................................15 Glutaminase Activity...........................................................................................17 Synthetase Activity..............................................................................................17 3 INHIBITION OF RECOMBINANT HUM AN ASPARAGINE SYNTHETASE.....19 Introduction.................................................................................................................19 Results and Discussion...............................................................................................23 In vitro Characterization of the N -acylsulfonamide 1 as an Inhibitor of Human AS........................................................................................................23 In vitro Characterization of the Adenylated Sulfoximine 2 as an Inhibitor of Human AS........................................................................................................25 In vivo Characterization of the Effects of the Adenylated Sulfoximine on the Proliferation of an Asparaginase-Resistant MOLT-4 Cell Line......................33


vii Experimental Section..................................................................................................34 Materials..............................................................................................................34 Expression and Purification of Recombinant, C-Terminally Tagged AS...........35 Steady-State Kinetic Assa ys and Data Analysis.................................................35 Enzyme Binding Assay.......................................................................................38 Enzyme Reactivation Assay................................................................................38 Mass Spectrometric Analysis of Tryptic Digests................................................39 Cell-Based Assays...............................................................................................40 4 PROBING THE CATALYTIC RESIDUE S IN THE SYNTHETASE ACTIVE SITE OF ASPARAGINE SYNTHETA SE B IN ESCHERICHIA COLI..................42 Introduction.................................................................................................................42 Results and Discussion...............................................................................................45 Residues and Their Predicted Roles....................................................................45 Asp384 and Arg387.....................................................................................45 Glu352..........................................................................................................46 Glu348 and Ser 346......................................................................................46 Site-Directed Mutagenesis...................................................................................47 Expression and Purification of Wild-type and Mutant AS-B..............................49 Glutaminase Activity...........................................................................................51 Synthetase Activity..............................................................................................53 Chemical Rescue.................................................................................................58 13C Kinetic Isotope Effects..................................................................................59 Experimental Section..................................................................................................64 Site-Directed Mutagenesis...................................................................................64 Expression and Purification of Wild-type and Mutant AS-B..............................65 Glutaminase Activity...........................................................................................66 Dinitrophenyl-Asparagin e (DNP-Asn) Assay.....................................................66 Chemical Rescue.................................................................................................67 13C Kinetic Isotope Effects..................................................................................67 5 ROLE OF GLU348 IN CATALYSIS........................................................................69 Introduction.................................................................................................................69 Results and Discussion...............................................................................................70 18O Labeling of L-Aspartate................................................................................70 Control Experiments for 31P NMR......................................................................71 Some Unexpected Results...................................................................................72 18O Transfer Studies............................................................................................74 Michaelis-Menten Kinetics for ATP and Asp.....................................................80 The Proposed Role of Glu348.............................................................................83 Experimental Section..................................................................................................85 18O labeling of Aspartic acid...............................................................................85 31P NMR Assay...................................................................................................86 ATP and Asp Km................................................................................................86


viii 6 CONCLUSIONS........................................................................................................88 Inhibition of Recombinant Human Asparagine Synthetase........................................88 Functional Characterization of E. coli Asparagine Synthetase B...............................89 APPENDIX A PRIMERS USED IN AS-B MUTAGENESIS EXPERIMENTS...............................91 B MASS SPECTROMETRIC ANALYSIS OF 18O ASPARTIC ACID.......................93 LIST OF REFERENCES...................................................................................................99 BIOGRAPHICAL SKETCH...........................................................................................106


ix LIST OF TABLES Table page 2-1. Steady-state kinetic parameters for ammonia-dependent and glutaminedependent synthetase activity of the C-terminally tagged, recombinant human AS............................................................................................................................. 13 3-1. Results of the enzyme-inhi bitor binding experiment................................................28 3-2. Steady-state parameters for inhibi tion of human AS by the adenylated sulfoximine...............................................................................................................29 4-1. Steady state parameters for the glutamin ase activity of the AS-B wild type and mutant enzymes........................................................................................................52 4-2. Comparison of pyrophosphate versus asparagine production for ammoniadependent synthetase activ ity of AS-B mutants.......................................................54 4-3. Asparagine versus glut amate production for glutaminedependent synthetase activities of wild type, E348D , and E348A AS-B enzymes....................................56 4-4. 13C kinetic isotope effects on -COOof aspartate...................................................63 4-5. PCR parameters used for site-d irected mutagenesis experiments............................64 5-1. Percentage of 18On-labeled aspartic acid in the incubated sample............................71 5-2. List of 31P NMR signals present in the reaction set-ups...........................................72 5-3. Amount of asparagine (in mM) produced at 37 ºC for 3 hours under various conditions..................................................................................................................78 5-4. Kinetic constants for the substrat es ATP and aspartate under steady state conditions, evaluated for wild-type AS-B, and the two mutants..............................83 A-1. Sequence of primers used in AS-B mutagenesis experiments..................................84


x LIST OF FIGURES Figure page 1-1. Hypothetical mechanism for the ATPdependent conversion of aspartate to asparagine catalyzed by asparagine synthetase......................................................3 1-2. Cartoon representation of the C1A mutant of E. coli AS-B.......................................6 2-1. Dependence of recombinant AS activity on time and MOI.....................................11 2-2. Purification of His-ta gged human AS using IMAC.................................................16 3-1. Reactions catalyzed by asparagine synthetase.........................................................21 3-2. Chemical structures of the adenylated sulfoximine 2 ..............................................21 3-3. Computational evaluation of the ade nylated methylsulfoximine moiety as a stable analog of the transition state for asparagine formation...............................22 3-4. Progress curves showing the effect of N -acylsulfonamide 1 on PPi production in the ammonia-dependent synthetase reaction catalyzed by recombinant, C-terminally tagged human AS................................................................................24 3-5. Kinetic model used in th e quantitative analysis of the steady-state progress curves...............................................................................25 3-6. Steady-state kinetic behavior of hAS in the presence of the adenylated sulfoximine 2 ....................................................................................26 3-7. Comparison of asparagine and PPi production incubated with 2, and 10 µM sulfoximine...............................................................................................................27 3-8. Reactivation of inhibited hAS as a function of time................................................30 3-9. Progress curves showing PPi formati on as a function of incubation time for glutamine-dependent synthetase activ ity at various concentrations of 2 .................31 3-10. Effect of the adenylated sulfoximine 2 on the steady-state glutaminase activity of hAS.....................................................................................33 3-11. Effect of the adenylated sulfoximine 2 on MOLT-4 proliferation in the presence and absence of 1 U L-asparaginase...........................................................34


xi 4-1. Multiple sequence alignment showi ng conserved active site residues.....................43 4-2. Comparison of the chemistry and structure of BLS and AS-B................................44 4-3 Detailed view of the st ructural model for the enzyme-Asp-AMP complex using the CAChe software (Fujitsu).........................................................................45 4-4. Electrophoresis of DNA from AS-B mu tants, in 0.8% agarose gel and stained with ethidium bromide.............................................................................................48 4-5. SDS-PAGE of purification fr actions for wild type AS-B........................................50 4-6. SDS-PAGE of purification fractions for AS-B E348A and E348D mutants...........50 4-7. Ammonia-dependent synthe tase activity for R387A...............................................57 4-8. Model of proposed kinetic mechanism for AS-B.....................................................61 4-9. Strategy for the conversion of aspartate -carboxylate to CO2................................62 5-1. Scheme for 18O aspartate exchange experiment.......................................................69 5-2. 31P NMR spectra for the reactions catalyzed by wild-type AS-B............................74 5-3. 31P NMR spectra for the reactions catalyzed by E348D mutant of AS-B................76 5-4. Comparison of the relative amount of each P nuclei detected in the wild-type and E348D reactions................................................................................................77 5-5. 31P NMR spectra for the reactions catalyzed by E348A mutant of AS-B................79 5-6 31P NMR spectra for the reactions catalyzed by E348Q mutant of AS-B................80 B-1. (+)ESI produced an m/z 134-142 [M+H]+ ion cluster.............................................95 B-2. HPLC/(+)ESI-MS mass chromatograms of each of the 12C-[M+H]+ ions, m/z 134, 136, 138, 140, and 142...............................................................................96 B-3. HPLC/(+)ESI-MS mass chromatograms of the Asp (m/z 142) and an m/z 186 ion-peak which eluted at RT 2.41 min.....................................................................97 B-4. The m/z 164, 186 ions of the earlier elu ting peak are likely the sodiated forms of Asp as indicated...................................................................................................98


xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy INHIBITION AND FUNC TIONAL CHARACTERIZATION OF ASPARAGINE SYNTHETASE By Jemy A. Gutierrez August 2006 Chair: Nigel G. J. Richards Major Department: Chemistry There is substantial evidence to support an inverse correlation between circulating concentrations of asparagine and the suscep tibility of T-cell leukemia to treatment with anti-cancer drugs. L-asparaginase is an e ffective part of chemotherapeutic protocols because it works by breaking down asparagine to aspartate in the blood. The reverse reaction is catalyzed by asparagi ne synthetase (AS), an enzy me implicated in the emergence of L-asparaginase resistance in acute lymphoblastic leukemia (ALL) cell line and patients. The underlying molecular mechanis ms for this drug resistance are poorly understood. Inhibitors specific to AS presen t new tools in the i nvestigation of this phenomenon. They could also represent potential drugs in the treatment of Lasparaginase resistant ALL. This work describes the efficient expre ssion and purification of recombinant human asparagine synthetase, based on baculovi rus expression vector technology. The recombinant human enzyme is correctly processed and exhibits catalytic activity.


xiii The development of this efficient expressi on system greatly faci litated the kinetic characterization of two potent inhibitors of human AS. N -acylsulfonamide analog of Asp-AMP intermediate, and N -adenylated S -methyl-L-cysteine sulfoximine are slow binding inhibitors with overall dissoc iation constants of 728 nM and 2.46 nM, respectively. More importantl y, sulfoximine was shown to have a cytostatic effect in MOLT-4 human leukemia cells in the presence of asparaginase. The work on Escherichia coli asparagine synthetase B (AS-B) focused in the characterization of catalytic residues in th e synthetase active site of the enzyme. A computational model with the -aspartyl-AMP intermediate placed in the active site revealed several conserved residues that c ould have catalytic func tions. These residues were tested by site-directed mutagenesis a nd steady state kinetics. Of these, Glu348 showed the most interesting results. Substitu tions to this residue perturbed the glutamate to asparagine product ratio in glutaminedependent synthetase activity, as well as abolishing the stimulating effect of ATP on glutaminase activity. In addition, Glu348 plays an important part in the formation of -aspartyl-AMP intermediate.


1 CHAPTER 1 INTRODUCTION Acute Lymphoblastic Leukemia Acute lymphoblastic leukemia (ALL) is a cancer of white blood cells that originates from the bone marrow. It is the most common cancer in children, and is predicted* to be diagnosed in some 3970 cases in the United States during 2006, 38% of which will result in death. Chemotherapy is the primary treatment for ALL, which involves multiple drugs given in precise schedul es and is not the same for all patients. Treatment of childhood ALL over the past 35 years has been a major success story, from less than 5% survival rate to 85% of children with ALL who live five years or longer. Despite improvements in treatment, ALL rema ins a leading cause of cancer death in children and young adults. L-asparaginase is an effec tive tool in therapeutic prot ocols for ALL in children, as it breaks down asparagine to aspartate in the blood. Ther e is substantial evidence to support an inverse correlation between circula ting concentrations of asparagine and the susceptibility of T-cell leukemia to treatme nt with anti-cancer drugs, hence, the therapeutic efficacy of L-asparaginase. Unfo rtunately, its use is limited by significant side effects that are fatal in some patients. One of these is the emer gence of tumors that are resistant to further L-asparaginase trea tment in patients who suffer a relapse after initial remission of the disease (Barr et al. 1992, Kiriyama et al. 1989). * from the American Cancer Society


2 Asparagine synthetase (AS) is an enzy me implicated in the emergence of L-asparaginase resistance in ALL. In 1969, Haskell and Canellos showed higher AS enzyme activity in five Lasparaginase resistant leukem ia patients compared to four patients who were L-asparagi nase sensitive. Work on MOLT-4† cell line has demonstrated that L-asparaginase resistance can be induced by short-term treatment of the drug, and this is marked by elevat ed concentrations of AS (Aslanian et al. 2001). Even more compelling were the results from an experiment where drug-sensitive MOLT4 cells were transformed using a retrovirus that contained the human AS gene under a constitutively active promoter. In a sim ilar manner, another batch of drug-sensitive MOLT-4 cells was transformed with the same vector lacking the human AS gene. Drug resistance was observed only in the MOLT-4 cells in which irreversible AS expression was taking place. These results show that ove rexpression of the AS alone is sufficient to induce L-asparaginase resistance in MOLT-4 human leukemia cells. Asparagine Synthetase Asparagine synthetase (AS) is an AT P-dependent glutamine-hydrolyzing amidotransferase that catalyzes the conversion of Laspartate to L-asparagine (Zalkin and Smith 1998). It utilizes L-glutamine or ammonia as source of nitrogen. Overall, asparagine synthetase catalyzes three di stinct reactions (Boehlein et al. 1994a): † cell line derived from the blood of a 19 year old male ALL patient Gln + H2O Glu + NH3 (1-1) ATP + Asp + NH3 Asn + AMP + PPi (1-2) ATP + Asp + Gln Asn + AMP + PPi + Glu (1-3)


3 Equation 1-1 represents glut aminase activity, 1-2 is amm onia-dependent synthetase activity, and 1-3 is glutaminedependent synthetase activity. Glutaminase and synthetase activities occur at structurally distinct site s, namely, the glutamine amide transfer (GAT) domain in the N-terminus where glutamine is bound and hydrolyzed to produce free ammonia and glutamate, and the synthetase domain in the C-terminus where ATP and aspartate are bound and transformed into aspa ragine, AMP, and pyrophosphate. The two active sites are linked together by an intram olecular tunnel where ammonia liberated in the GAT domain travels to the synthetase act ive site where ATP-activated aspartate is poised for nucleophilic attack. The current chemical mechanism is shown in Figure 1-1. Figure 1-1. Hypothetical mechanism for the ATP-dependent conversi on of aspartate to asparagine catalyzed by asparagine synthetase. Many different kinetic mechanisms have been proposed for the glutaminedependent synthetase active of AS fr om a variety of sources (Fresquet et al. 2004, Hongo and Sato 1985, Markin et al. 1981, Milman et al. 1980, Rognes 1975). However diverse, N O N N N NH2 OH O O P O O -O -O P O P O -O -O O O NH3 -O2C H O N O N N N NH2 OH O O P O O O -O2C O NH3 H N O N N N NH2 OH O O P O O O -O2C O NH3 H NH2 -O2C NH3 H CONH2 -O2C NH3 H CO2H + H + H + + + ATP -PPi NH3 -AMP + Glu Gln + H2O L-Asp L-Asn -Asp-AMP


4 the mechanisms all specify that aspartate attacks the -phosphate of ATP to form a aspartyl-AMP intermediate and pyrophosphate. Evidence for this adenylated intermediate was obtained in 1985 for bovine pancreatic asparagine synthetase. 18Olabeled aspartate was incubated with ATP and enzyme, and incorporation of 18O into AMP was detected by 31P NMR (Luehr and Schuster 1985). In 1998 the same 18O transfer to AMP was observed for Escherichia coli asparagine syntheta se B (AS-B) in an attempt to prove the kinetic competence of the -Asp-AMP intermediate (Boehlein et al. 1998). The goal of the isotope exchange experime nt was to show that as the intermediate is formed and pyrophosphate is added ATP will ev entually be regenerated faster than the intermediate can be hydrolyzed by water, all in the absence of the amide source. The work demonstrated that ATP a nd aspartate bind sequentially and -Asp-AMP hydrolysis occurs 1500-fold slower in th e absence of nitrogen source. AS belongs to the family of Cla ss II or Ntn (N-terminal nucleophilic) amidotransferases that also include gl utamine 5Â’-phosphoribosyl-1-pyrophosphate amidotransferase (GPATase) (Tso et al. 1982a, Tso et al. 1982b), glutamine fructose-6phosphate amidotransferase (GFAT) (Badet et al. 1987), and glutamate synthase (Vanoni and Curti 1999). These enzymes all possess a catalytic cysteine residue in the Nterminus that attacks the amide group of gl utamine to form a tetrahedral adduct that eventually breaks down to form products (Zalkin and Smith 1998). The C-terminal synthetase domain belongs to the family of ATP pyrophosphatases including guanosine5Â’-monophosphate synthetase (GMPS). The sequence motif SGGXDS is thought to be present in this family of enzymes th at convert ATP to AMP and pyrophosphate (Denessiouk and Johnson 2000). The intram olecular tunnel linking the active sites has


5 been investigated in some systems such as carbamoyl phosphate synthetase II (CPS), imidazole glycerol phosphate (ImGP) synthase , as well as others, leading to characterization of structural features, functional role in coordinating activities in the two active sites, and evolution (Huang et al. 2001, Raushel et al. 2003). Asparagine synthetase has been isolated from bacteria (Deng et al. 2002, McClelland et al. 2001, Ravel et al. 1962, Reitzer and Magasanik 1982, Scofield et al. 1990), archaebacteria (Bhatia et al. 1997, Bult et al. 1996), yeast (Dang et al. 1996, Ramos and Wiame 1979), plants (Chevalier et al. 1996, Lam et al. 1994, Waterhouse et al. 1996, Yamagata et al. 1998), and mammals (Andrulis et al. 1987, Arfin 1967, Hongo 1983, Luehr and Schuster 1985). Low abundance and instability of AS from mammalian sources however (Huang and Knox 1975, Mehlhaff et al. 1985), hindered the detailed kinetic and structural characterization of this enzyme. Human asparagine synthetase is made up of 560 amino acids. The recombinant enzyme has been expressed in Saccharomyces cerevisiae (Sheng et al. 1992, Vanheeke and Schuster 1990) and in Escherichia coli (Vanheeke and Schuster 1989b). E. coli asparagine synthetase B (AS-B) bel ongs to one out of two structurally unrelated families of asparagine synthetases, which are encoded by the genes asnA and asnB . asnA encodes for strictly ammonia-dependen t asparagine synthetase and is found only in prokaryotes (Cedar and Schwartz 1969a, Cedar and Schwartz 1969b) whereas asnB is found in both prokaryotes and euka ryotes and shows both glutamineand ammonia-dependent activities (Scofield et al. 1990). AS-B is made up of 553 amino acids with a molecular weight of 62.5kDa a nd is believed to function as homodimers (Richards and Schuster 1998).


6 In 1999 the crystal structure of AS-B (C1A mutant) was solved to 2.0Å resolution (Larsen et al. 1999) (Figure 1-2). Figure 1-2. Cartoon representa tion of the C1A mutant of E. coli AS-B. Shown are (A) the entire enzyme complexed with glutamine ( blue space-filling model ) and AMP ( green space-filling model ) showing the domain organization of the enzyme, and (B) the putative pathway by which ammonia ( light blue spheres ) travels between the glutaminase ( top ) and the synthetase ( bottom ) active sites in AS-B. The side chai ns of residues defining the ammonia tunnel are colored green and red. Bound glutamine ( top ) and AMP ( bottom ) are rendered as gray-white “ball-andstick” models. Figure generated by Dr. Nigel G. J. Richards. The structure showed the overall shape of the enzyme delineating the two domains and the active sites were identified with glut amine and AMP in place. It was possible to trace the path of the intramol ecular channel which stretched ~19Å in length. Disorder in the C-terminal domain of the C1A/Gl n/AMP complex, unfortunately, did not permit observation of two loop regions (Ala250 to Le u267 and Cys422 to Ala426), and the final forty residues of the enzyme. Since then, there has not been much luck in obtaining structural information for AS-B, or any AS for that matter. Computational modeling has been useful in obtaining a more comple te picture of the enzyme, with the missing (A) (B) (A) (B)


7 elements modeled into the structure. It ma kes it possible to look at the active sites more closely and draw hypotheses a bout residues that could be catalytically important. Research Goals A major long-term goal of the project is to develop clinically us eful inhibitors of human asparagine synthetase. One of the bi ggest hurdles in achieving this goal is the lack of routine access to sufficient quantitie s of reproducibly active human enzyme. The recombinant enzyme expression systems developed from yeast and E. coli have their own limitations. Ultimately, the challenge is the inhibitor it self. In the past, compounds that have been characterized for AS have only shown micr omolar affinities for the enzyme that are nowhere near clinically useful values. The design of next generation inhibitors for AS will be greatly facilitated by a more detailed knowledge of the important struct ure-function relationshi ps in the enzyme, specifically at the synthetase active site. This kind of information will reveal critical interactions between enzyme a nd substrates, intermediates, or transition states that can be exploited in designing better inhibitors. The specific goals of the presented research were therefore (1) development of an efficient expression system for active reco mbinant human AS; (2) characterization of intermediate and transition state analog inhibi tors of the human enzyme; (3) evaluation of the catalytic significance of the synthetase active s ite residues Ser 346, Glu348, Glu352, Asp384, and Arg387 of E. coli AS-B.


8 CHAPTER 2 EXPRESSION, PURIFICATION, AND STEADY STATE KINETIC CHARACTERIZATION OF C-TERMINALLY TAGGED RECOMBINANT HUMAN ASPA RAGINE SYNTHETASE Introduction The need for an efficient expression syst em for human asparagine synthetase has long been recognized because of the demandi ng requirements of kinetic and structural studies. As the native enzyme is clearly not easily accessible, means to obtain large quantities of active recombinant human enzyme has been a major challenge. In 1989b, Van Heeke and Schuster reported the expression of recombinant human AS in Escherichia. coli , which despite showing both amm oniaand glutamine-dependent synthetase activities, exhibited major sol ubility problems. They acknowledged the limitations of expressing a eukaryotic enzyme in a prokaryotic system. Expression of recombinant human AS was also carried out in Saccharomyces cerevisiae (Sheng et al. 1992, Vanheeke and Schuster 1990). The yeast-based expression system produced correctly processed human AS, but the protocol appears limited for large-scale work by (i) variability in the steady-state kinetic parameters reported for different batches of the recombinant enzyme (Sheng et al. 1993, Vanheeke and Schuster 1990), and (ii) the use of a tedious immunoa ffinity-based purification strategy, which constrains the amount of human AS that can be routinely obtained. Much success has been achieved in recent y ears with the expre ssion of recombinant proteins from eukaryotic sources by empl oying baculovirus expression vector system (BEVS) (Farrell 1998, Lenhard 1997, Sanghani 2000). This technology takes advantage


9 of the efficient infection and reproduction of genetic material exhibited by viruses from their host. Baculoviruses specifically target insect cells and are nonpathogenic to mammals. One of the more commonl y used baculovirus isolates is Autographa californica multiple nuclear polyhedrosis virus (AcMNPV), propagated in cell lines such as that derived from fall armyworm Spodoptera frugiperda (e.g., Sf9 , Sf11 ) (Vaughn 1977). The expression system works by replacing the naturally occurring polyhedrin gene ( polh ) in the wild-type baculovirus genome with the recombinant gene or cDNA of interest placed under the transcriptional co ntrol of the strong polyhedrin promoter. At very late stages of infection, instea d of expressing polyhedrin – a protein that assembles into a crystalline matrix that pr otects the virus particle s during cell lysis and death – the cells express the recombinant product, which are processed, modified and targeted to the appropriate cellular locations. The versatility of BEVS is enhanced by th e functionality that can be engineered into the transfer vector, such as fusion tags that can facilitate purification, for instance. The use of poly-histidine tags (a sequence of six or more c onsecutive histidines) has been recognized as an efficient tool in purifica tion protocols (Sulkowski 1985) because of the high affinity of the imidazole moiety to certain transition metal ions (e.g., Zn2+, Cu2+, and Ni2+). Immobilized metal affinity chroma tography (IMAC) was developed in 1975 to exploit this property in the pur ification of recombinant, his tidine-tagged proteins (Porath 1975). A metal chelating ligand is linked with a long alkyl ch ain to a stable resin which may be packed in a column or suspended in solution. The chelating ligand has multiple sites for coordination of the metal ion, which is “charged” onto the resin.


10 Protein that contains the poly histidine tag is then applied to the resin (on a column for column purification, or in solution for ba tch purification). Wash ing the resin with a low concentration imidazole solution elimin ates nonspecific binding, and with a higher concentration imidazole solution elutes the po ly histidine tagged protein, almost exclusively as naturally occu rring poly histidine sequen ces are rare in nature. This work reports a detailed experiment al protocol for obtaining a recombinant form of human AS in which the enzyme c ontains an additional C-terminal segment composed of c myc and poly-histidine tags. This b aculovirus-based expression system reproducibly yields multi-milligram quantitie s of highly active, human AS that (i) exhibits reproducible steadystate kinetic behavior, and (i i) is sufficiently stable for use in identifying and characterizing inhibitors. Results and Discussion‡ A key problem in obtaining fully active , C-terminally tagged human AS was associated with the relatively large number of unoxidized cystei ne residues in the enzyme. In particular, since AS belongs to the Ntn amidohydrolase enzyme superfamily (Brannigan et al. 1995), glutamine-dependent AS activ ity requires that Cys-1 be present as a free thiol (Boehlein et al. 1994a, Boehlein et al. 1994b, Cameron et al. 1989, Vanheeke and Schuster 1989a), and previous studies of mammalian asparagine synthetases were severely hampered by the instability of these enzymes during purification (Hongo 1983, Jayaram et al. 1976, Patterson 1967). For example, native murine pancreatic AS exhibited a near comple te loss in its ability to utilize glutamine ‡ Reprinted from Archives of Biochemistry and Biophysics, Vol 440, Ciustea, M., Gutierrez, J. A., Abbatiello, S. E., Eyler, J. R., and Richards, N. G. J, Efficient expression, purifi cation, and characterization of C-terminally tagged, recombinant human asparagine synthetase, Pages 22-26, Copyright (2005), with permission from Elsevier.


11 as a nitrogen donor when stored at -87°C in the presence of 1 mM DTT (Milman and Cooney 1979). A detailed investigation of the activity of the recomb inant AS present in the lysate as a function of expressi on conditions was undert aken (Figure 2-1). Figure 2-1. Dependence of recombinant AS ac tivity on time and MOI. Specific activity was assayed by monitoring pyrophos phate production. Key: () MOI = 1.0, ( ) MOI = 0.1, ( ) MOI = 0.01 pfu/cell, where pfu stands for plaque forming units. In these experiments, which employed va rious MOI values (multiplicity of infection in units of pfu/cell, the ra tio of virus particles to inse ct cells), the time course of tagged human AS expression was monitored by (i) Western blotting experiments using antic-myc and anti-HisTag antibodies, and (i i) ammonia-dependent pyrophosphate formation in Ni–agarose purified cell lysates on the addition of ammonia, aspartic acid, and ATP. Asparagine and pyrophosphate were reported as being formed with a 1:1 stoichiometry for the recombinant, hu man AS obtained from yeast (Sheng et al. 1993, Sheng et al. 1992). Appropriate control experiment s were performed to eliminate the possibility that the observed activity was due to the presence of any contaminating pyrophosphatases. These results showed that AS expression began 24 h post-infection,


12 with pyrophosphatase activity gr adually increasing to reach a maximum value after an additional 36 h. Thus, the amount of expre ssed hAS did not parall el the profile of synthetase activity. Further anal ysis revealed that the subseq uent decrease in the specific activity of the recombinant enzyme was coincident with virally induced cell lysis. Since expression of the “tagged” human AS was under the control of polh promoter, which results in high levels of enzyme production du ring the very late phase of viral infection (24-72 h), this observation s uggested that exposure of th e recombinant enzyme to the oxygenated medium after cell lysis was probabl y resulting in cystei ne oxidation and/or disulfide formation. A strategy of employing a high MOI was adopted in order to infect the highest number of Sf9 cells and then harvested after a period of 60 h such that only approximately 25% cell lysis had occurred. When this protocol was followed, the recombinant enzyme exhibited high levels of specific activity and increased stability during purification. It is likely that expression strategies simila r to that reported here can be employed to obtain other intracellular prot eins lacking disulfide bonds in baculovirusbased systems (Broschat et al. 2002). Despite the observation that there is consid erable variation in the C-terminal region of the deduced primary structures for a variety of asparagine syntheta ses, the presence of the two “tag” sequences in the recombinant, human AS was a source of concern as it could adversely impact the steady-state kine tic properties of the enzyme. The human enzyme was therefore assayed for synthetase activity us ing a standard pyrophosphate assay. These experiments showed that the recombinant enzyme is able to employ ammonia as a nitrogen source in asparagine formation, with steady-state kinetic parameters similar to those reported previously fo r the wild type human AS obtained from a


13 yeast-based expression system (Sheng et al. 1993, Sheng et al. 1992). For example, the KM app values for ATP and aspartate in Table 2-1 were 0.11 and 1.3 mM, respectively, for ammonia-dependent asparagine synthesis rather than 0.84 and 1.9 mM as reported previously (Sheng et al. 1993). In a study of the ammonia-dependent synt hetase activity of recombinant human AS expressed in yeast, a value of 0.55 s 1 was reported for the turnover number of the enzyme (Sheng et al. 1993), corresponding to a specifi c activity of 0.155 µmol/min/mg. A specific activity of 0.115 for this reaction was obtained for recombinant, human AS expressed in E. coli (Vanheeke and Schuster 1989b). Table 2-1. Steady-state kine tic parameters for ammonia-dependent and glutaminedependent synthetase activity of th e C-terminally tagged, recombinant human AS. a KM app values given in parentheses refer to those obtained from recombinant, human AS expressed in yeast. b The value shown here for ammonia has been corrected from that originally reported, which was based on the amount of NH4OAc present in the assay (pH 8) rather than free ammonia. In addition, the tagged, human AS obtained using the baculovirus-based approach exhibited stable, reproducible glutaminase (KM = 1.7 ± 0.05 mM; kcat = 1.8 ± 0.4 s 1; kcat/KM = 1058 M 1 s 1) and glutamine-dependent aspa ragine synthetase activities (Table 2-1), suggesting that the presence of the additional residues at the C-terminus does not perturb the protein stru cture significantly. In this re gard, it should be noted KM app (mM)a kcat (s 1) kcat/KM (M 1 s 1) Ammonia-dependent synthetase Ammonia 1.7 ± 0.1 (1.49)b 1.8 ± 0.04 1059 Aspartic acid 1.3 ± 0.04 (1.19) 1.7 ± 0.01 1308 ATP 0.11 ± 0.01 (0.84) 1.6 ± 0.03 14545 Glutamine-dependent synthetase Glutamine 1.9 ± 0.1 1.7 ± 0.04 895 Aspartic acid 0.38 ± 0.03 1.3 ± 0.02 3421 ATP 0.08 ± 0.01 1.7 ± 0.03 21250


14 that no detailed steady-state kinetic studies on the glutaminase ac tivity of human AS have been reported, although a low specifi c activity for AS-catalyzed glutamine hydrolysis of 0.152 U was obtained for the r ecombinant enzyme obtained by expression in yeast (Sheng et al. 1993). The use of an HPLC-based assay for m onitoring asparagine production, which has been previously described for char acterizing the kinetic mechanism of E. coli AS-B (Tesson et al. 2003), confirmed that the asparagine and pyrophosphate are formed in a 1:1 ratio for the recombinant, human AS. Gi ven that (i) kinetic parameters were not reported for the glutamine-dependent activit ies of human AS obtained by expression in yeast (Sheng et al. 1993), and (ii) the na tive enzyme has not been purified and characterized from human liver cells, it is difficult to evaluate the activity of the C-terminally tagged, human AS obtained from the baculovi rus-based expression system. In this regard, it should be noted that the current protocol yields recombinant enzyme with a turnover number similar to that observed for the glutamine-dependent synthetase activity of E. coli AS-B (Boehlein et al. 2001). Moreover, the en zyme obtained using the baculovirus-based system described herein e xhibits significantly hi gher specific activity for both glutamine-dependent reactions than those reported for the human AS obtained by expression in yeast (Vanheeke and Schuster 1990) (Table 2-1). In summary, an efficient pr otocol for preparing large quantities of C-terminally tagged, wild type human AS in a baculovirusbased expression system was developed. The recombinant enzyme is correctly proce ssed, exhibits high activ ity and is stable on prolonged storage at 80°C. Not only do these studies offer the possibility for investtigating the kine tic behavior of biochemically in teresting mammalian AS mutants


15 (Gong and Basilico 1990), but such ready acces s to substantial amounts of enzyme also represents a major step in the development a nd characterization of inhibitors that might have clinical utility in treating as paraginase-resistant ALL (Aslanian et al. 2001). Experimental Section Optimization of Expression Parameters Sf9 insect cells were grown to 95% viability in serum-free medium. Once it reached a density of 3 x 106 live cells/mL, human AS viral stock (2.5 x 108 pfu/mL) was added in separate cell cultures at MO I values of 0.01, 0.10, and 1.0. Recombinant, C-terminally tagged human AS was isolat ed at each time point (24-96 hours after infection) using Ni–agarose spin columns, and the eluate added to a solution comprised of 5 mM ATP, 10 mM as partic acid, 100 mM NH4Cl, and 10 mM MgCl2 in 100 mM EPPS, pH 8, containing reconstituted pyr ophosphate reagent (Sigma) that had been preincubated at 37°C for 10 min (total react ion volume 1 mL). The rate of pyrophosphate production (µmol/min/mg) was then determined by monitoring absorbance at 340 nm over a period of 20 min. Expression and Purification A high titer viral stock (2.5x108 pfu/ml) containing the human AS gene was prepared by Mihai Ciustea in the laboratory. This was done by cloning the cDNA for human AS (obtained from Dr. Michael Ki lberg, UF) into a pBAC-1 transfer plasmid which has an additional coding sequence for C-terminal c-myc tag and a polyhistidine tag. Sf9 insect cells were grown in serum-free medium up to a density of 3x106 cells/ml and viability of 95%. A liter of cell culture was tran sfected with the AS viral stock to get an MOI of 1. At 60 hours post infecti on, the cells were harvested by centrifugation and the pellet was resuspende d in lysis buffer (50 mM EPPS, pH 8, 300 mM NaCl,


16 10 mM imidazole, 1% Triton-X, 0.5 mM DTT) . After two rounds of sonication, cell debris was pelleted out using centrifugation and the cell free extract was passed through a 0.45 m membrane filter. The clear, yellowis h extract was then applied on a Ni-NTA column (Qiagen, Valencia, CA) equilibrated in lysis buffer. The column was then washed with wash buffer (50 mM EPPS, pH 8, 300 mM NaCl, 20 mM imidazole, 0.5 mM DTT), followed by the elution buffe r (50 mM EPPS, pH 8, 300 mM NaCl, 250 mM imidazole, 0.5 mM DTT). All purification fractions we re collected and kept in ice. Active fractions were pooled together followed by ammonium sulfate precipitation to 70% saturation. The slurry was centrifuged and the pellet was resuspended in storage buffer (50 mM EPPS, pH 8, 5 mM DTT) and di alyzed against the same buffer overnight at 4 C. The dialysis retentate was assayed for pyrophosphate activity and 20% glycerol was added before flash freezing. Aliquots were stored at -80 C. SDS-PAGE showed a single band with the correct molecular wei ght (Figure 2-2) and concentration was determined to be 4 mg/ml by Coomassie assay (Pierce) using bovine albumin standards. Up to 30 mg of pure recombinant human AS was obtained from one liter of cell culture. Figure 2-2. Purification of His-tagged human AS using IM AC. Lanes 1-2, cell lysate; Lane 3, lysate flowthrough; Lane 4, column wash; Lane 5, pooled active elution fractions; Lane 6, dialysis rete ntate; MW, molecular weight markers.


17 Glutaminase Activity L-glutamate dehydrogenase assay foll ows the production of NADH from NAD+ as glutamate is converted to -ketoglutarate and ammonia (Sig ma Chemical Co., Technical Bulletin No. GLN-1). In this assay, 1 00 mM EPPS buffer, pH 8.0, 0.1 – 100 mM Gln*, 100 mM NaCl, 8 mM MgCl2, 0.5 mM DTT were all premixed. Enzyme was added and the mix was incubated at 37 C for 10 minutes and then quenched with 20% TCA. This was added to a solution of 300 mM glycin e 250 mM hydrazine buffer, pH 9.0, 1 mM ADP, and 1.5 mM NAD+ after which 2.2 U of L-glutamate dehydrogenase (Sigma) was added to initiate the reaction. Abs340 was monitored for 30 minutes and compared to that of a set of glutamate standards. KM and kcat values were derived from the MichaelisMenten plot , using KaleidaGraph (Synergy Software) to fit the data. Synthetase Activity Commercially available pyrophosphate assa y kit was used (Sigma Chemical Co., Technical Bulletin No. BI-100) . This assay takes pyrophosph ate (produced from the AS-catalyzed reaction) through a series of enzymatic reactio ns found in the glycolytic pathway which eventually oxidizes two mol ecules of NADH, monitored at 340 nm. The reaction mix consists of 100 mM EPPS buffer, pH 8, 10 mM MgCl2, and substrates ATP, aspartic acid, and ammonia (or glutamine). For KM determination of each substrate, concentrations of the other two substrates we re held constant (saturating at 5 mM for ATP, 20 mM for aspartic acid, 100 mM for ammonia or 20 mM for glutamine) while varying the concentration for the substrate of interest. For ATP, concentrations range from 0.07 to 5.0 mM, for aspartic acid 0.3 to 20 mM, for ammonia 1.5 to 100 mM and for * Glutamine stock was recrystallized in aqueous ethanol before use.


18 glutamine 0.3 to 20 mM. The reaction was done at 37 C and initiated by adding 4 g of purified human AS and then following th e absorbance at 340 nm for 10 minutes. Absorbance is converted to NADH concentr ation, and pyrophosphate production is obtained directly from the amount of NADH consumed. Apparent KM and kcat values were obtained from Michaelis-Menten plot s, and data fitti ng using KaleidaGraph (Synergy Software).


19 CHAPTER 3 INHIBITION OF RECOMBINANT HUM AN ASPARAGINE SYNTHETASE Introduction L-asparaginase (ASNase), an enzyme that catalyzes the hydrolysis of L-asparagine (Aghaiypour et al. 2001), is a component of most ther apeutic protocols for the treatment of acute lymphoblastic leukemia (ALL) (Barr et al. 1992, Ertel et al. 1979, Sutow et al. 1971). The opposing reaction is catalyzed by as paragine synthetase (AS), which converts L-aspartic acid into L-asparagine in a tran sformation requiring ATP and a nitrogen source that is L-glutamine in eukaryotic cells (R ichards and Schuster 1998). The human enzyme (hAS) is of clinical intere st (Richards and Kilberg 2006) because several lines of evidence suggest that the development of L-as paraginase resistance in ALL is correlated with up-regulation and expression of this enzyme (Aslanian et al. 2001, Chakrabarti and Schuster 1997, Haskell and Canellos 1969, Kiriyama et al. 1989). The clinical significance of this observation has been challenged (Fine et al. 2005, Pieters et al. 1997, Stams et al. 2005), however, and new strategies ar e therefore needed to probe the molecular mechanisms by which cellular hAS contributes to ASNase-resistant ALL. In particular, cell-permeable compounds capable of specifically i nhibiting the enzyme should be valuable tools in evaluating whether increased levels of in tracellular asparagine biosynthesis might be the key change in cellular metabolism that underlies the appearance of drug resistance. Assays employing small molecule libraries to identify potent AS inhibitors have not yet been reported, and early studies of substr ate analogs failed to yield any compounds exhibiting mi cromolar affinity against the form of AS present in


20 either murine lymphoblasts and pancreatic cells (Jayaram and Cooney 1979), or rodent neoplasm leukemia 5178Y/AR cell lines (Cooney et al. 1976). Based on the crystal structure of the glutamine-dependent AS from Escherichia coli (Larsen et al. 1999), the human enzyme is likely bu ilt from two domains, each of which contains a catalytic site. The N-terminal s ite catalyzes the conversion of glutamine into glutamic acid and ammonia, and aspartate is re acted with ATP in the C-terminal site to yield the reactive intermediate -aspartyl-AMP (Figure 3-1), the existence of which has been demonstrated by isotopic labeling experiments (Boehlein et al. 1998, Luehr and Schuster 1985). As is the case in other gl utamine-dependent amidotransferases (Huang et al. 2001), ammonia released in the N-terminal domain of AS travels through an intramolecular tunnel linking the active sites, a nd reacts with the ac tivated acyladenylate moiety to form asparagine (Richards and Kilberg 2006). N -acylsulfonamide 1 was synthesized§ and purified based on literature procedures for the preparation of isoleucyland ty rosyl-tRNA synthetase inhibitors (Koroniak et al. 2003). It was designed to mimic the -Asp-AMP intermediate in the reaction catalyzed by AS (Figure 3-1). The mixed anhydride in th e intermediate is replaced with a sulfonamide linkage, possessing a carbonyl group more stable against nucle ophilic attack and cleavage. The adenylated sulfoximine, as a mixture of diastereoisomers 2a and 2b (Koizumi et al. 1999) (Figures 3-1, 3-2) wa s first reported as an inhi bitor of the ammonia-dependent variant of Escherichia coli asparagine synthetase (AS-A), an enzyme that has been § Reproduced in part with permission from Organic Letters, Vol. 5 (12), Koroniak, L.; Ciustea, M.; Gutierrez, J. A.; Richards, N. G. J., Synth esis and Characterization of an N-Acylsulfonamide Inhibitor of Human Asparagine Synthetase, Pages 2033-2036. Copyright 2003 American Chemical Society.


21 identified only in prokaryotes (Cedar a nd Schwartz 1969a) and does not share a common ancestor with glutamine-dependent AS (Nakatsu et al. 1998). Figure 3-1. Reactions catalyzed by aspara gine synthetase. Ammonia may replace glutamine as a nitrogen source for the AS-catalyzed synthetase reaction. N-acylsulfonamide 1 , and adenylated sulfoximine 2 , are analogs of -AspAMP intermediate, and the transiti on state after ammonia attack, respectively. Figure 3-2. Chemical structures of the adenylated sulfoximine 2 . Based on the crystal structure of the comp lex between AS-A and the inhibitor, the adenylated sulfoximine moiety was proposed to be a stable analog of the transition state for the attack of ammonia on -Asp-AMP (Koizumi et al. 1999). In order to verify this N O N N N NH2 OH O O S O O N -O2C O NH3 H H N O N N N NH2 OH O O P O O N S -O2C NH3 H CH3 O H + -O2C NH3 H CO2N O N N N NH2 OH O O P O O O -O2C O NH3 H H + -O2C NH3 H CONH2 N O N N N NH2 OH O O P O O O -O2C NH3 H NH3 -O2C NH3 H CONH2 -O2C NH3 H CO2H + + Mg2+-ATP PPi + + AMP H + O + + NH3 -Asp-AMP N -acylsulfonamide1Adenylatedsulfoximine2Transition State N O N N N NH2 OH O O S O O N -O2C O NH3 H H N O N N N NH2 OH O O P O O N S -O2C NH3 H CH3 O H + -O2C NH3 H CO2N O N N N NH2 OH O O P O O O -O2C O NH3 H H + -O2C NH3 H CONH2 N O N N N NH2 OH O O P O O O -O2C NH3 H NH3 -O2C NH3 H CONH2 -O2C NH3 H CO2H + + Mg2+-ATP PPi + + AMP H + O + + NH3 -Asp-AMP N -acylsulfonamide1Adenylatedsulfoximine2Transition State O N HOOH N N N NH2 O P N O O S O OH3N+ -O H H3C 2a O N HOOH N N N NH2 O P N O O S O OH3N+ -O H H3C 2b


22 hypothesis, the electrostatic a nd steric properties of a m odel phosphorylated sulfoximine 3 (Figure 3-3A) and the transition state for the addition of ammoni a to the acylphosphate 4 (Figure 3-3B) were compared using semi empirical calculations (Cramer and Truhlar 1999) at the PM3 level of theory (Stewart 1989). These calculations also employed a continuum solvation potential (Cramer and Truhlar 1999, Klamt 1995) so as to model the effects of a polarizable medium on the ch arge distributions in the transition state and sulfoximine. Figure 3-3. Computational eval uation of the adenylated methylsulfoximine moiety as a stable analog of the transition state fo r asparagine formation. (A) Structure of the model sulfoximine 3 (left), and graphical re presentations of the PM3optimized molecular structure (middle) a nd electrostatic properties (right) of this compound. (B) Structure of the transition state for the attack of ammonia on the model acylphosphate 4 (left), and graphical representations of the PM3-optimized molecular stru cture (middle) and electrostatic properties (right) of this transition state. Dott ed lines represent weak interactions, and the electrostatic potential is mapped on the isodensity surface in both figures. Model and fi gure generated by Dr. Nigel Richards. These studies confirm that (i) the me thyl substituent of a phosphorylated sulfoximine is an excellent steric mimic of ammonia, and (ii) the tetrahedral sulfur OCH3P N O O S H3C OH3C 3 OCH3P O O O H3C 4 ON HH H (A) (B)OCH3P N O O S H3C OH3C 3 OCH3P O O O H3C 4 ON HH H (A) (B)


23 atom is a good model for the rehybridization of the carbonyl carbon in th e transition state. A more interesting finding, however, is that polarization of the C-H bonds in the methyl group, which presumably reflects the inductive effect of the phosphorylated sulfoximine, causes the methyl substituent to exhibit electro static properties that are strikingly similar to those of the ammonia molecule in the transition state for nuc leophilic attack on the carbonyl group of an acylphosphate. This work describes in vitro experiments showing that the N -acylsulfonamide 1 and adenylated sulfoximine 2 are slow-onset, tight-binding inhi bitors of hAS. Perhaps more importantly, treatment of a drug resistan t MOLT-4 cell line (Srivasta and Minowada 1973) with adenylated sulfoximine 2 has a cytostatic effect. This observation is the first direct evidence that AS inhibitors repres ent interesting compounds for the clinical treatment of ASNase-resistant ALL, as initially proposed almost forty years ago (Cooney and Handschumacher 1970). Results and Discussion In vitro Characterization of the N -acylsulfonamide 1 as an Inhibitor of Human AS The ability of N -acylsulfonamide to inhibit the ammonia-dependent activity of the recombinant, C-terminally tagged human AS was examined. Hence, progress curves were generated by incubating the purified enzy me (4 µg) in reaction mixtures containing 100 mM NH4Cl, 0.5 mM ATP, 10 mM aspartate, and 10 mM MgCl2 in 100 mM EPPS, pH 8, together with the inhib itor (0-50 µM) (1 mL total volu me). Since asparagine and PPi are formed in a 1:1 stoichiometric ra tio in the enzyme-catalyzed reaction, the synthetase activity of human AS under th ese conditions was determined by spectrophotometric monitoring of PPi production. The results showed that compound 1 did inhibit the recombinant human enzyme (Figur e 3-4), presumably by binding within the


24 C-terminal, synthetase site of human AS in a fashion similar to the -Asp-AMP intermediate that is formed in the catalytic mechanism of asparagine production. Importantly, control experiments establishe d that the pyrophosphate reagent is not affected by the presence of the N -acylsulfonamide. 0 2 4 6 8 10 12 14 20040060080010001200[PPi] ( M)Time (s) Figure 3-4. Progress curves showing the effect of N -acylsulfonamide 1 on PPi production in the ammonia-dependent synthetase reaction catalyzed by recombinant, C-terminally tagged human AS. Key: [ 1 ] = 0 µM, filled circles; [ 1 ] = 5 µM, open circles; [ 1 ] = 10 µM, filled squares; [ 1 ] = 25 µM, open squares; [ 1 ] = 50 µM, filled triangles. Solid lines represent the theoretical curve computed from eq 3-1 that best fit the experimental data, and error bars correspond to the standard deviation of the [PPi] concentration measured at a given time point. Analysis of the steady-state kinetics of AS inhibition by N -acylsulfonamide 1 was undertaken to determine the mechanism by whic h the inhibitor exerted its effects. The data were best fit (Figure 3-4) using the following equation, corres ponding to slow-onset inhibition (Morrison and Walsh 1988), ) 1 ( ) (0 kt ss sse k v v t v P (3-1) where vo is the initial velocity of the reaction, vss is the velocity at large t , and k is a


25 parameter that depends on the inhibitor concentration. Figure 3-5. Kinetic model used in the quant itative analysis of the steady-state progress curves. E and I are enzyme and inhibitor, respectively. Further experiments demonstrated that elev ated levels of ATP decreased the ability of 1 to inhibit recombinant human AS. This implies that ATP and 1 compete for the free enzyme, an observation consistent with the assumption that 1 binds within the Cterminal, synthetase site of AS. The i nhibition mechanism is consistent with the following model for which the variation of k with inhibitor concen tration is given by: ) / / 1 ( /5 6 i a iK I K A K I k k k (3-2) Analysis of the progress curves usin g equation 3-1 then gave values of k as a function of the concentration of 1, and a replot of k versus [1] then yielded values for Ki, k5, and k6 of 21 µM, 3.6 × 10-2 s-1, and 1.3 × 10-3 s-1, respectively. The overall dissociation constant (K* i) for the slow-onset inhibitor 1 is therefore 728 nM. While still not sub-nanomolar in potency, 1 inhibits human AS at a level that is 1000-fold greater than for any compound yet reported, and hence, this N -acylsulfonamide represents a useful lead compound for futu re studies in this area. In vitro Characterization of the Adenylated Sulfoximine 2 as an Inhibitor of Human AS The ability of the adenylated sulfoximine (as a mixture of diastereoisomers 2a and 2b) to inhibit the ammonia-dependent activity of C-terminally, tagged recombinant hAS E E.I k1[ATP] k3[ 1 ] k2k4k5k6 E.ATP EI* k7E


26 was initially assayed. Using a coupled a ssay to detect inorganic pyrophosphate (PPi) (O'Brien 1976), which is formed in a 1:1 st oichiometry with aspa ragine by the human enzyme (Horowitz and Meister 1972), timedependent inhibition was observed when reactions were initiated by the addition of hAS to a mixture of substrates containing the transition state analog (Figure 3-6). 0 2 4 6 8 10 20040060080010001200[PPi] ( M)Time (s) Figure 3-6. Steady-st ate kinetic behavior of hAS in the presence of the adenylated sulfoximine 2. (A) Progress curves showing PPi formation as a function of incubation time for ammonia-depende nt synthetase activity at various concentrations of 2: (o) 0 µM; ( ) 1 µM; ( ) 2 µM; ( ) 4 µM; ( ) 6 µM; ( ) 10 µM. Lines represent the fit used to obtain the kinetic parameters for inhibition. Control experiments established that the sulfoximine did not affect the PPi detection assay, but there was always the possibility that the presence of 2 had decoupled asparagine and PPi formation. The Asn:PPi ratio was therefore checked using an HPLCbased end-point assay in which recombinant AS was incubated with substrates at various


27 sulfoximine concentrations before reaction was terminated by the addition of trichloroacetic acid. The resulting so lution was then assayed for PPi, and the asparagine formed was quantitated by HPLC after conversion in to its 2,4-dinitrophe nol adduct (Tesson et al. 2003). The results showed that the product st oichiometry was not significantly affected by the inhibitor, at least within experimental error (Figure 3-7). 0 5 10 15 20 25 30 0210 Asn PPi[Product], M[Sulfoximine], M Figure 3-7. Comparison of asparagine and PPi production incubated with 2, and 10 µM sulfoximine, using DNP-Asparagine a ssay and Sigma pyrophosphate assay, respectively. In previous studies of the ability of th e adenylated sulfoximine to inhibit the glutamine-dependent AS from Escherichia coli (AS-B), it was reported that the adenylated sulfoximine did not bi nd to the free enzyme (Boehlein et al. 2001). It was therefore examined whether this was the cas e for its interaction with hAS by measuring the residual activity of the human enzyme after incubation with 2 for 10 minutes and subsequent filtration through a Sephadex G-50 spin column. These experiments showed that AS synthetase activity was reduced by in cubation with the adenylated sulfoximine in


28 a concentration-dependent manner (Table 3-1) , suggesting that the inhibitor was capable of binding to the free enzyme. Computer simulations of the kinetic mode l (Figure 3-5) using GEPASI software ver.3.30 and constants derived from the progre ss curves predicted much lower values for activities compared to experimental results (T able 3-1, in parentheses). The discrepancy may be due to several reasons, one of whic h is the assumption made in the simulation that the “on” rate for enzyme and inhi bitor binding is diffusion rate limited (108 M-1 s-1), which in this case could be an over approxima tion. Nevertheless, they follow the same trend for enzyme deactivation in the presence of the adenylated sulfoximine that supports inhibitor binding to the free enzyme. Table 3-1. Results of the enzy me-inhibitor binding experiment. Values are expressed as % activity rela tive to the uninhibited reactions. In parentheses are values predicted from Gepasi simulations using kinetic constants obtained from the progress curves. This contrasts with its observed kinetic behavior with AS-B under similar reaction conditions, and in the absence of a crystal stru cture for hAS, it is difficult to assess the structural basis for this apparent difference in inhibition kinetics. Given the multidomain structure of the two glutamine-depende nt enzymes, it is possible that differences in the conformational states of the two form s of AS may result in modified accessibility of the adenylated sulfoximine to its binding site . The fact that it could bind to free hAS, however, permitted quantitative analysis of the progress curve data for the ammoniadependent reaction using a standard kinetic model for slow-onset, tight-binding inhibition % Activity, (relativ e to uninhibited) without G-50 with G-50 hAS 100 100 hAS + 5 M I 83 (52) 82 (59) hAS + 10 M I 64 (17) 56 (25)


29 (Figure 3-5) (Morri son and Walsh 1988). Curve fitting gave values of 280 ± 43 nM and 2.5 ± 0.3 nM for Ki and Ki*, respectively (Table 3-2). This value of Ki* is 100-fold smaller than that observed for the inhibition of hAS by the N -acylsulfonamide 1 (Figure 3-4), which is an analog of the -Asp-AMP intermediate (Koroniak et al. 2003), and compares very favorably to the Ki* value of 67 nM reported for the interaction of the dias tereoisomeric mixture 2a and 2b with the bacterial, ammonia-dependent AS-A (Koizumi et al. 1999). Table 3-2. Steady-state pa rameters for inhibition of human AS by the adenylated sulfoximine. Parameters are those obta ined by fitting to a kinetic model in which the inhibitor binds to free enzyme, and assume a Ka value for ATP of 0.2 mM. Reversibility of in vitro hAS inhibition using standard protocols (Cha 1975) was investigated. Thus, after complete inactiva tion of the enzyme by incubating hAS with ATP, aspartate and ammonium chloride in th e presence of the adenylated sulfoximine (as judged by the cessation of PPi production), the reaction mixt ure was subjected to gel filtration on a Sephadex G-50 column. Fracti ons containing hAS were diluted into a solution containing substrates at saturating concentration and PPi formation was monitored spectroscopically over a peri od of several hours (Figure 3-8). Over the time course of the experiment , the inactivated sample of human AS regained 56% of its activity relative to a c ontrol sample of the enzyme that had been subjected to identical treatment in the abse nce of sulfoximine, and quantitative analysis Nitrogen source Ki (nM) k5 (s-1) k6 (s-1) kobs/[I]tot (M-1s-1) Ki* (nM) NH4Cl 285 2.98 x 10-3 2.6 x 10-5 436 2.46 L-Glutamine 985 2.81 x 10-3 7.1 x 10-5 219 24.3


30 gave a value of 7.4 h for the ha lf-life of reactivation, which is consistent with the estimate of k6 obtained from analysis of progress curv e data (Table 3-2). The demonstrated reversibility of hAS inhibition again contrasts with reported observations on the bacterial enzymes given that both AS-A and AS-B do not regain activity after being inactivated by the adenylated sulfoximine 2 (Boehlein et al. 2001, Koizumi et al. 1999). 20 25 30 35 40 45 50 55 60 12345678% ActivityTime (h) Figure 3-8. Reactivation of inhibited hAS as a function of time. The line shows the exponential of best fit to the data. One possibility is that 2 was exerting its effects by covalently modifying the enzyme. A series of experiments was carried out to evaluate whether protein adenylation was indeed occurring when hAS was incubated with 2 using conditions that had been developed for employing electrospray TOF mass spectrometry to observe tryptic peptides from over 90% of the sequence of recombinant hAS. The enzyme was incubated with the adenylated sulfoximine 2 (10 µM) until the rate of pyrophosphate production became zero, and the protein was then subjected to in-gel digestion with trypsin. Mass spectrometric analysis of the peptide fragments from the sample of fully inhibited hAS showed


31 no fragments with an increased mass correspondi ng to that expected if adenylation of any protein side chains had taken pl ace. In this regard, all pep tides containing residues that are located in the synthetase active site of the enzyme can be observed using electrospray TOF mass spectrometry. Given that L-glutamine is the likely physiologi cal nitrogen source for asparagine biosynthesis in eukaryotic cells (Ric hards and Kilberg 2006), the ability of 2 to inhibit the glutamine-dependent synthetase activity of hAS at a physiologically relevant ATP concentration (5 mM) and pH 8 (Figure 3-9) was also investigated. 0 2 4 6 8 10 12 14 02004006008001000[PPi] ( M)Time (s) Figure 3-9. Progress curves showing PPi formation as a function of incubation time for glutamine-dependent synthetase ac tivity at various concentrations of 2: (o) 0 µM; ( ) 1 µM; ( ) 2 µM; ( ) 4 µM; ( ) 6 µM; ( ) 10 µM. Lines represent the fit used to obtain the kinetic parameters for inhibition. Once again, slow-onset tight -binding inhibition was obser ved, and the quantitative analysis of the progress curves for PPi production yielded values of 1000 ± 176 nM and 24 ± 2.8 nM for Ki and Ki*, respectively, under these conditions (Table 3-2). The 10-fold


32 decrease in the ability of 2 to inhibit the glutamine-dependent synthetase reaction may arise from conformational differences in th e C-terminal domain of the human enzyme that are associated with occupancy of the N-terminal glutaminase site by substrate, L-glutamate or the thioester intermediate (Schnizer et al. 1999, Schnizer et al. 2002). The effects of the adenylated sulfoximine 2 on the glutaminase activity of the enzyme hAS was therefore evaluated usi ng a continuous assay in which glutamate formed by AS-catalyzed glutamine hydrolysis is coupled to th e production of NADH (Bernt 1974). The rationale for these experiments came from previous studies showing that the presence of ATP stimulat es the glutaminase activity of Escherichia coli AS-B (Boehlein et al. 1994a), an effect that was used to demonstrate the functional roles of conserved residues in the N-termin al, glutaminase domain (Boehlein et al. 1994). In the presence of 10 µM adenylated sulfoximine 2, kcat/KM for the glutaminase activity of hAS was 982 M-1s-1, which was relatively unchanged from the value of 795 M-1s-1 determined for the reaction in the absence of the inhibitor (Figure 3-10). This finding was unexpected, however, on th e basis of ATP-dependent stimulation of the glutaminase activity of AS-B, and so the glutaminase activity of hAS was assayed to evaluate if it is affected by ATP (Liaw and Eisenberg 1994). In another interesting observation, the presence of 3 mM ATP had almost no impact on kcat/KM for the glutaminase activity of hAS, which was determined to be 674 M-1s-1. The lack of effect on the steady-state rate of AS-catalyzed glutamine hydrolysis for either ATP or the sulfoximine derivative 2 contrasts with the stimulation of glutaminase activity observed for other members of the Class II glutamine-dependent amidotransferase family (Zalkin 1993, Zalk in and Smith 1998), including glutamine 5’-


33 phosphoribosyl-pyrophosphate amidotransferase (Bera et al. 2000) and glutamine fructose-6-phosphate amidotransferase (Bad et-Denisot 1993), when nitrogen-accepting substrates are bound in the C-terminal synthetase domains of these enzymes. 0 20 40 60 80 100 0102030405060[Glu] ( M)[Gln] (mM) Figure 3-10. Effect of th e adenylated sulfoximine 2 on the steady-state glutaminase activity of hAS. Glutamine-dependence of the glutaminase activity in the presence ( ) and absence (o) of 10 µM 2. Lines represent the fit to the Michaelis-Menten equations used to obtain the steady state kinetic parameters. In vivo Characterization of the Effects of the Adenylated Sulfoximine on the Proliferation of an Asparagina se-Resistant MOLT-4 Cell Line After demonstrating that th e adenylated sulfoximine 2 was the first AS inhibitor with low nanomolar affinity for the enzyme, the effects of incubating this compound with a MOLT-4 leukemia cell line (Srivasta and Mino wada 1973) were investigated. This cell line has been used extensively in previous studies of the molecular mechanisms that mediate ASNase resistance (Aslanian et al. 2001, Aslanian and Kilberg 2001, Hutson et al. 1997, Richards and Kilberg 2006). He nce, MOLT-4(R) cells, which had been selected for their ability to grow in the presence of ASNase (Aslanian et al. 2001, Hutson


34 et al. 1997), were grown on media to which the hAS inhibitor 2 had been added at concentrations of 0.1-1 mM (Figure 3-11). These high concentrations were chosen because it can be anticipated that the charged functional groups on the molecule, which are known to be important in mediating recognition, would negatively impact cell permeability. Incubations were performed for a total of 48 hours, and the effects of the AS inhibitor on cell proliferation were a ssayed using a dye-based (WST-1) method for estimating the number of live MOLT-4 cells (Ishiyama et al. 1996). In a concentrationdependent manner, the adenylated sulfoximine 2 inhibited the ability of the MOLT-4(R) cells to proliferate. Interestingly, this inhibitory effect required the simultaneous presence of L-asparaginase in the growth medium (Figure 3-11). Figure 3-11. Effect of th e adenylated sulfoximine 2 on MOLT-4 proliferation in the presence and absence of 1 U L-asparaginase. Cell pro liferation is defined as the number of viable cells after 48 h expressed as the ratio to the initial number of cells (t = 0). Error bars represent the standard deviation of triplicate experiments. Figure generated by Dr. Yuan-Xiang Pan. Experimental Section Materials N -acylsulfonamide was synthesized by Luke Koroniak following literature procedure (Koroniak et al. 2003). All samples of the adenylated sulfoximine 2 (as a 1:1


35 mixture of diastereoisomers 2a and 2b) we re obtained from Dr. Jun Hiratake (Kyoto University) by chemical synthesis follo wing literature procedures (Koizumi et al. 1999). Unless otherwise stated, all other chemicals and reagents, including authentic samples of dinitrobenzene derivatives of L-asparagine, Laspartate and L-glutamate, were purchased from Sigma-Aldrich, and were of the highest available purity. Pr otein concentrations were determined by a modified Bradford assay (Pierce) based on standard curves constructed using bovine serum albumin (B radford 1976). L-glutamine was purified by recrystallization prior to us e in all assays containing this reagent. Extreme care was taken when handling solutions of 2,4-dinitr ofluorobenzene in organic solvents because this reagent is a potent allergen and w ill penetrate many types of laboratory gloves (Thompson and Edmonds 1980). Expression and Purification of Re combinant, C-Terminally Tagged AS Multi-milligram amounts of wild type, C-terminally tagged human AS were expressed in Sf9 cells and purified using procedures described in the previous chapter. Steady-State Kinetic Assays and Data Analysis Progress curves were obtained under stea dy-state conditions using a continuous assay in which the formation of inorga nic pyrophosphate (PPi) is measured by monitoring the consumption of NADH (340 nm) (Sigma Technical Bulletin BI-100). It has been shown that the asparagine a nd PPi are produced in a 1:1 ratio by the recombinant, C-terminally tagged form of hA S. In these experiments, purified hAS (2 µg) was incubated at 37°C with substrates in 100 mM EPPS buffer, pH 8, containing 5 mM ATP, 10 mM MgCl2, 100 mM NH4Cl, 10 mM L-aspartic acid, and varying concentrations of N -acylsulfonamide 1 (0 50 µM) or adenylated sulfoximine 2 (0-10 µM) over a period of 20 min (1 mL final volume). Reactions were initiated by addition of the


36 enzyme. The resulting progress curves were analyzed by fitting the data to equation 3-1 (Morrison and Walsh 1988) using the Kaleidagraph v3.5 (Synergy software). The values of k at different inhibitor concentrat ions were therefore determined by fitting, and used to find k6 from equation 3-3. 0 6v kv kss (3-3) The average value of k6 could then be computed, and this was subsequently used to obtain estimates of Ki and k5 by fitting to equation 3-2. For experiments in which L-glutamine re placed ammonia as the nitrogen source, purified hAS (2 µg) was incubated at 37°C in EPPS buffer, pH 8, (1 mL final volume) containing 5 mM ATP, 10 mM MgCl2, 25 mM L-glutamine, 10 mM L-aspartic acid, and varying concentrations of 2 (0-10 µM) over a period of 20 min. Control experiments using known amounts of PPi demonstrated that the assay reagent is not affected by the presence of the adenylated sulfoximine 2. In order to check that the inhibitor 2 did not affect the 1:1 stoichiometry of Asn:PPi, an alternate HPLC-based assay (Tesson et al. 2003) was employed in control experiments to measure the amount of L-aspa ragine directly. Thus, recombinant hAS (4 µg) was incubated at 37C with 5 mM ATP, 10 mM L-aspartic acid, 10 mM MgCl2 and 100 mM NH4Cl in 100 mM EPPS buffer, pH 8, and varying concentrations of the adenylated sulfoximine 2 (0, 2 or 10 µM) over a period of 20 min (1 mL final volume). Reactions were initiated by a ddition of the enzyme. After quenching with glacial AcOH, and neutralization with aq. NaOH, an aliquot of each mixture (40 µL) was diluted (200 µL final volume) with 400 mM aq. Na2CO3, pH 9, containing 10% DMSO and 30% dinitrofluorobenzene (DNFB) (as a saturated solution in EtOH). The resulting solutions


37 were heated at 50C for 45 min to permit reaction of DN FB with the amino acids to yield their dinitrophenyl (DNP) deri vatives (Morton and Gerber 1 988, Orth 2001). Aliquots of each assay mixture (40 µL) were analyzed by reverse-phase HPLC (RP-HPLC) using a C18 column (Varian Inc.). The DNP-derivati zed amino acids were eluted with a step gradient of 40 mM formic acid buffer, pH 3.6, and CH3CN. In this procedure, the initial concentration of th e organic phase (CH3CN) was 13.5%, which was increased to 14.5% over a period of 21 min. Af ter this time, amount of CH3CN was increased to 80% over a period of 30 s, and elution continued fo r a further 20 min. Eluted amino acid DNP derivatives were monitored at 365 nm a nd identified by comparison to authentic standards. Under these conditions, DNP-a sparagine exhibited a retention time of approximately 21 min, and could be quantified on the basis of its peak area. Calibration curves were constructed usi ng solutions of pure L-asparagine derivatized in the same manner as the samples. The amount of asparagi ne detected in this assay was compared to PPi as determined using the coupled enzyme assay (Sigma). These experiments were also repeated for mixtures containing 0.5 mM ATP. The glutaminase activity of hAS was assayed by determining L-glutamate formation with glutamate dehydroge nase in the presence of NAD+ (Bernt 1974). Assay mixtures (200 µL total volume) contained varying concentrations of L-glutamine (150 mM) and 8 mM MgCl2 in 100 mM EPPS buffer, pH 8, containing 0.5 mM DTT and 100 mM NaCl, and reaction was initiated by the addition of recombinant hAS (2.6 µg). The resulting solution was then incubated at 37C for 20 min before being terminated by addition of 20% trichloroacetic acid (30 µL). The mixture was added to 770 µL of the coupling reagent (300 mM glycine, 250 mM hydrazine, pH 9, c ontaining 1 mM ADP,


38 1.6 mM NAD+ and 2.2 units of glutamate dehydrogena se) and incubated for 30 min (final volume 1 mL). The absorbance at 340 nm of the resulting solution was measured, and the amount of L-glutamate determined from a standard curve. Steady-state parameters were obtained by fitting the data to the Michaelis-Menten equation using Kaleidagraph v3.5 (Synergy software). These experiments were repeated in the presence of ATP (5 mM) or the adenylated sulfoximine 2 (10 µM). Enzyme Binding Assay Purified hAS (4 µg) was incubated in the presence (5 M or 10 M) or absence of the adenylated sulfoximine 2 for 10 minutes in 100 mM EPPS, pH 8 (20 L total volume). The resulting reaction mixtures we re then loaded onto a Sephadex G-50 spin column and washed with 100 L 100 mM EPPS buffer, pH 8. The activity of the enzyme and its concentration in each of the three solutions were determined using the coupled assay for detection of PPi and the modified Br adford assay, respectively. Experiments to monitor the synthetase activity employed satura ting levels of ATP and L-aspartic acid, 10 mM MgCl2 and 100 mM NH4Cl, and PPi production was measured over a time of twenty minutes. Enzyme Reactivation Assay Recombinant, wild type hAS (at a final concentration of 0.25 M) was added to a solution of 5 M sulfoximine 2, 0.5 mM ATP, 10 mM MgCl2, 10 mM Asp, and 100 mM NH4Cl in 100 mM EPPS, pH 8 (1 mL total vo lume). The solution was incubated until synthetase activity ceased, as measured by PPi production (O'Brien 1976). The reaction mixture was then filtered through a Sephade x G-25 column using elution with 100 mM EPPS, pH 8, so as to remove substr ates, products and the unbound sulfoximine 2. The


39 reactivation of AS activity was then m onitored by measuring ammonia-dependent asparagine formation in aliquots (500 µ L) of the fractions co ntaining human AS, which were diluted 20-fold into assay mixt ures containing 5 mM ATP, 10 mM MgCl2, 10 mM aspartate and 100 mM NH4Cl in 100 mM EPPS buffer, pH 8 at 37°C. Mass Spectrometric Analysis of Tryptic Digests Recombinant, wild type hAS (at a final concentration of 0.25 M) was added to a solution of 5 M sulfoximine 2, 5 mM ATP, 10 mM MgCl2, 10 mM Asp, and 100 mM NH4Cl in 100 mM EPPS, pH 8 (1 mL total vo lume). The solution was incubated until synthetase activity ceased as measured by PPi production. The resulting protein was isolated and a sample purified by SDS-PAGE on a 9% resolving gel, at 110 V, including a lane for molecular weight markers. Staini ng with Coomassie brilliant blue dye revealed a protein band of 66 kDa molecular weight, which was excised and destained by soaking the gel pieces in 50% methanol (1 mL) overnight, followed by several additional washings. Cysteine residues were reduced with dithiothreitol and alkylated with iodoacetamide to give the carbamidomethylated-modi fied protein, which was then digested in-gel with trypsin (625 ng trypsin per gel ba nd) on ice for 45 minutes and overnight at 37°C. The trypsin solution was replaced with 20 mM ammonium bicarbonate, and the reaction quenched with 5 µL glacial AcOH. After centrifugation, the sample was desalted prior to mass spectrometr ic analysis by elution through C18 ZipTips (Millipore) using a solution of 50% aq. CH3CN containing 1% HCO2H. The separation of tryptic peptides produced in the protein digest was performed by capillary RP-HPLC on a 15 cm x 75 µm i.d. PepMap C18 column (LC Packings) in combination with an Ultimate Capillary HPLC System (LC Packings) operating at a flow


40 rate of 200 nL/min. On-line mass spectrometric analysis of the column eluate was accomplished using a hybrid quadrupole ti me-of-flight (TOF) instrument (QSTAR, Applied Biosystems) equipped with a nano-el ectrospray source. Fragment ion data obtained on the TOF instrument were search ed against the NCBI non-redundant sequence database using the Mascot database sear ch engine (Matrix Science). Peptide masses were calculated based on the mono-isotopic peak a nd charge state of each ion cluster, and compared to those expected for tryptic peptides, and their corresponding adenylated derivatives, from recombinant hAS. Probabi lity-based MOWSE scores above the default significant value were considered for peptid e identification in addition to validation by manual interpretation of tandem MS/MS da ta. Control experiments employed hAS treated in an identical manner except that the adenylated sulfoximine 2 was absent from the initial incubation mixture. Cell-Based Assays The effect of the adenylated sulfoximine 2 was tested on cell proliferation by Dr. Yuan-Xiang Pan. MOLT-4/R resistant cells were seeded into 96-well plates at a density of 4000 cells per well using RPMI 1640 medium with 10% FBS and incubated in 95% air with 5% CO2 at 37°C. MOLT-4/R is a deriva tive of a human acute lymphoblastic leukemia cell line (MOLT4) (Srivasta and Minowada 1973) that is resistant to the presence of ASNase in the medium and s hows up-regulated levels of AS expression (Aslanian et al. 2001, Hutson et al. 1997). The sulfoximine 2 was diluted to give 0.1, 0.5 and 1 mM concentrations in culture medi um, and cells were incubated with the compound in the presence or absence of 1 U of ASNase. Cell viability was determined 48 h after treatment using the WST-1 Cell Prol iferation Assay (Roche Diagnostics). The optical density was read at 450 and 690 nm by an ELX800 Universal Microplate Reader


41 (Bio-Tek Instruments, Inc.). The mean cell titer of treated samples relative to control cells prior to treatment (time = 0) was calcula ted, and the data expressed as the mean ± SD of triplicate experiments.


42 CHAPTER 4 PROBING THE CATALYTIC RESIDUES IN THE SYNTHETASE ACTIVE SITE OF ASPARAGINE SYNTHETASE B IN ESCHERICHIA COLI Introduction An essential step in the rational design of inhibitors for human asparagine synthetase is probing the enzyme active site, iden tifying residues that have catalytic roles, and looking at how these residues make cont acts with the bound substrates, and/or products. In order to study the relevant structure-functi on relationships though, one has to work with a system that allows mutagenesis studies to be easily carried out, and for the efficient baculovirus expre ssion system developed for recombinant human AS, this involves going through the tedious task of pr eparing a unique viral stock for each desired mutation. This difficulty is overcome by employing Escherichia coli asparagine synthetase B (AS-B) in these mutagenesis studies. The only crystal structure solved for aspara gine synthetase was from a C1A mutant of AS-B complexed with L-glutamine and AMP (Larsen et al. 1999) and attempts to solve the structure of the human enzyme have been futile. The availability of structural information and the straightforward manner in which mutagenesis experiments can be carried out are advantages to working with the E. coli system. In addition, the active site identified from the crystal structure contai ns residues that are completely conserved within known AS sequences. A multiple sequence alignment [DARWIN software, (Gonnet 1991)] highlighting some of the resi dues that are conserved across the AS sources is shown in Figure 4-1.


43 Figure 4-1. Multiple sequence alignment s howing conserved active site residues. Highlighted are residues chosen for mutagenesis studies. E. coli numbering was used. These residues present good targets for mu tagenesis studies. Predictions can be made about their potential roles in catalysis, which in turn w ill be useful in the design of the assays to be used to test these predic tions. This process is greatly facilitated by having a clear picture of the active site, show ing where and how the residues interact with bound substrates, intermediate, and transition state. In 2002, Ding generated a computational m odel of asparagine synthetase with -Asp-AMP intermediate bound in the synthetase active site, that addressed problematic issues with the original solved AS-B structur e, such as the missing last 40 residues in the crystal structure , as well as two other albe it short regions of the sequence. This was accomplished by using homology modeling and molecular dynamics simulation methods, as well as structural information from the AS-B crystal structure and from the evolutionarily related ß-lactam synthase (BLS) (Bachmann et al. 1998, McNaughton et al. 1998). BLS and AS-B share remarkable simila rities not only with th eir active site and overall structure, but also the type of ch emistry they carry out (Figure 4-2) (Miller et al. 2001, Miller et al. 2002). 346 ..359 -SGECADEIFGGYPW -SGEGSDELTQGYIY -SGEGSDELTQGYIY -SGEGSDELTQGYIY -SGEGADELFGGYLY -SGEGSDEVFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEILGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY 382 ..387 LLD__R LFDVLR LFDVLR LFDVLR YADCLR MYDCAR QYDCLR RYDCLR QFDCLR LYDCLR LYDCLR YYDVLR LSDCLR LADCLR bacillus mus rattus homo aedes ecoli helianthus pisum arabidopsis zea oryza plasmodium schizosaccharomyces saccharomyces 346 ..359 -SGECADEIFGGYPW -SGEGSDELTQGYIY -SGEGSDELTQGYIY -SGEGSDELTQGYIY -SGEGADELFGGYLY -SGEGSDEVFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEILGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY -SGEGSDEIFGGYLY 382 ..387 LLD__R LFDVLR LFDVLR LFDVLR YADCLR MYDCAR QYDCLR RYDCLR QFDCLR LYDCLR LYDCLR YYDVLR LSDCLR LADCLR bacillus mus rattus homo aedes ecoli helianthus pisum arabidopsis zea oryza plasmodium schizosaccharomyces saccharomyces


44 Figure 4-2. Comparison of the chemistry a nd structure of BLS and AS-B. (A) Reaction catalyzed by AS-B involves substrate (a spartate) activation by adenylation to form -Asp-AMP intermediate, followed by intermolecular attack by ammonia to form asparagine. (B) BLS activates substrate carboxyethylarginine by adenylation to form a similar acyl-AMP intermediate, followed by intramolecular attack by nitrogen to form the -lactam product. (C) BLS shown with bound substrate and ATP an alog (green space-filling model). (D) AS-B with bound AMP (green sp ace-filling model). The striking similarity in secondary structure elem ents and overall fold suggest a common ancestry for the two enzymes. Structural figures generated by Dr. Nigel G. J. Richards. The computational model generated by Di ng revealed the following conserved residues within interacting distance from the -Asp-AMP intermediate: Ser346; Glu348; Glu352; Asp384; and R387 among others (Figur e 4-3) (Ding 2002). Mutagenesis and steady state kinetic characterization will not only reveal the catalytic roles of the chosen residues but also test the correctne ss of the computational model. (C) (D) (C) (D)


45 Figure 4-3 Detailed view of the structural model for the enzyme-Asp-AMP complex using the CAChe software (Fujitsu). Shown here are the interactions of -Asp-AMP with conserved residu es Arg387, Asp384, Glu352, Glu348, and Ser346 in the synthetase active site . [Color coding: C-grey; H-white; O-oxygen; N-blue; P-purple]. Inset pi cture shows the entire enzyme with bound substrates in blue and green spacefilling models, for the glutaminase and synthetase active sites, respectively. Results and Discussion Residues and Their Predicted Roles Asp384 and Arg387 In the model (Figure 4-3), Asp384 and Arg387 are found within H-bonding distance from the aminoand -carboxylgroups of ß-Asp-AMP, respectively. These residues could potentially play a role in binding aspa rtic acid and holdi ng it in the correct position to react with ATP in forming the in termediate. The interesting Asp384 mutants are the ones incorporatin g Asn and Ala, and for Arg387, Lys and Ala. If the residues are involved in aspartate binding, it will be reflected in the KM app for the substrate. Asp384 Glu352 Glu348 Ser346 ß-Asp-AMP Arg387 Asp384 Glu352 Glu348 Ser346 ß-Asp-AMP Arg387


46 Glu352 In addition to Asp384 and Arg387, a number of other conserved residues are found very close to the aspartate moiety of ß-Asp-AMP. Nota ble is the network of hydrogen bonded residues including Lys376 (also hydrogen-bonded to Asp384), Glu352, and Tyr357. The working hypothesis is that this group of residues is important in stabilizing the protonated, unreactive form of the aspartate -amino group. In the evolutionarily related enzyme ß-lactam syntha se (BLS), this group is actu ally unprotonated, capable of intramolecular nucleophilic attack on the adenylated carbonyl. The fact that AS-B does not go by this intramolecular route is perhaps dictated by the subtleties in the active site that impart specificity to the chemistry it ca rries out. Substitution of Glu352 to Asp, Gln and Ala should reveal important features of this residue. Glu348 and Ser 346 The intramolecular tunnel from the glutamin ase active site to the synthetase active site is lined with mainly hydrophobic residue s with the startling exception of Glu348, which happens to be conserved within the know n sequences. Located very close to it is Ser346. Both residues lie at the bottom of th e tunnel, at the C-terminal end and within close proximity to the ß-Asp-AMP/PPi in the m odel. One critical issue with the current chemical mechanism for AS-B is that after ammonia is channeled through and attacks the carbonyl of the ß-Asp-AMP intermediate, the amide has to be deprotonated in the tetrahedral transition state which then colla pses to form the products asparagine and AMP. Glu348 is strategically close to this s ite and may act as the critical base that deprotonates the amide. Ser346 on the other ha nd, could play a role in coordinating the two active sites during glutamine-depende nt synthetase activity. Glu348 will be substituted with Asp, Gln, and Ala, wher eas Ser346 will be changed to Thr and Ala.


47 Site-Directed Mutagenesis In the design of primers for site-directed mutagenesis, several factors had to be considered. Both sense and anti-sense pr imers should contain the desired mutation and anneal to the same sequence on opposite strands of the plasmid template. They should be between 25 to 45 bases in length, with a melting temperature (Tm) not lower than 78°C. The desired mutation must be in the middle of the primers with about 10 to 15 bases of correct sequence on both sides. In addition, the GC content ha s to be greater than 40% of the total primer sequence, which should terminate in one or more C or G bases to facilitate better annealing. The organism’s (E. coli) codon usage should also be carefully considered. Primers designed in the manner describe d above were analyzed on Gene Runner ver3.05 (Hastings Software), using the Oligo Analysis tool. They were analyzed for possible secondary structures (e.g., hairpin loops, dimers, bul ge loops, internal loops) that could pose potential problems. Prim er sequences are listed in Appendix A. The DNA template used in mutagenesis st udies is a pET-21c(+) plasmid (Novagen) containing the asnB gene from E. coli inserted into the plasmid’s Nde I restriction site. Its length is 5441 bps which gets 1662 bps longer with asnB, and it carries a marker for ampicillin resistance. Site-directed mutagenesis is carried out by Polymerase Chain Reaction (PCR), using the plasmid DNA containing asnB gene as template, and a pair of mutagenic primers targeting the complementary sequences at the same site. This is done in the presence of dNTPs, which are the build ing blocks for the amplified DNA, and PfuTurbo DNA polymerase (Stratagene) which re plicates both strands with high


48 fidelity. The DNA product is then treated with Dpn I, an endonucl ease that targets methylated and hemimethylated DNA such as the parental (template) DNA, as well as DNA derived from most E. coli strains. Mutagenic DNA product is hence selected for, and is transformed into XL1-Blue superc ompetent cells (Stratagene). Successful transformants were grown in cell culture and the DNA was isol ated by a method of alkaline lysis and polyethelene glycol (PEG) precipitation. Figure 4-4. Electrophoresis of DNA from AS-B mutants, in 0.8% agarose gel and stained with ethidium bromide. Lanes: 2 E348A; 3 E348D; 4 R387K; 5 R387A; 1, 6 DNA marker with MW values on the left. Figure 4-4 shows results of a typical DNA el ectrophoresis gel for AS-B mutants, in this case for E348A, E348D, R387A, and R 387K. The gel displays multiple bands on a single lane, one corresponding to the expected molecular weight of about 7100 bps for the plasmid containing the asnB insert. The other bands may represent the same DNA exhibiting a different tertiary structure, such as the kinds caused by supercoiling. Nevertheless, the mutant DNA is sent fo r sequencing (DNA Seque ncing Laboratory, ICBR, UF), which confirms the mutation at the correct location. The DNA is stored in -20°C where it is stable until use. 7.1kbp 1 2 3 4 5 6 21.2 kbp5.1 kbp 4.9 kbp 4.2 kbp 7.1kbp 1 2 3 4 5 6 7.1kbp 1 2 3 4 5 6 21.2 kbp5.1 kbp 4.9 kbp 5.1 kbp 4.9 kbp 4.2 kbp


49 Expression and Purification of Wild-type and Mutant AS-B The pET-21c(+) plasmid places the asnB gene under the control of T7 promoter. BL21(DE-3) (Novagen) is the expression strain from E. coli that has T7 DNA dependent RNA polymerase under the control of lacP. In the expression cells transformed with the asnB vector, transcription by both E. coli RNA polymerase and T7 RNA polymerase is inducible by addition of isopropylthiogalact opyranoside (IPTG), which allows expression of T7 RNA polymerase and AS-B, respectively. Cells were lysed by sonication, and AS-B in the soluble fraction was purified. Dithiothreitol (DTT) was included in all the buffers used, so as to protect the cr itical Cys1 residue from oxidation. Some mutants proved toxic to the expression cells and did not allow for colonies to grow under any condition, even with the use of a pLysS host strain. This strain is used to suppress basal expression of T7 RNA polym erase prior to IPTG induction and thus stabilize pET recombinants encoding target pr oteins that could be toxic to the cells. These constructs include D384N, E352D, and E352Q. Purification was carried out primarily with anion exchange chromatography using a MonoQ column (Amersham Biosciences), f unctionalized with quaternary polyamine (ammonia) species. AS-B has a theoretical pI of about 5.5 and would be negatively charged at much higher pH conditions. Eluti on was done with a gradient of 1 M sodium chloride, with AS-B emerging between 0.2 a nd 0.5 M. This was followed by ammonium sulfate precipitation, which serv es as an extra purification st ep as well as a concentration step. Dialysis was carried out to remove all the salts from the elution step to ammonium sulfate precipitation. The purified enzyme was quickly frozen in liquid nitrogen and stored in 10% glycerol at -80° C until use. Bradford assay ga ve a concentration of 8 to 10 mg/mL for wild-type and mutant enzymes, using bovine serum albumin as standard.


50 Purified wild-type AS-B was resolved using SDS-PAGE (Figure 4-5), with the correct molecular weight of around 62 kDa. AS-B mutants showed SDS-PAGE profiles similar to that of the wild-type enzyme, with a single band in the final purification step fraction corresponding to the expected molecular weight (Figure 4-6). Figure 4-5. SDS-PAGE of pur ification fractions for wild type AS-B. Lanes: 1, 9 molecular weight marker, 2 lysate, 3 column flowthrough, 4 column wash, 5 elution fraction, 6 pooled fractions, 7 supernatant after (NH4)2SO4 precipitation, 8 in storage buffer. On 12% polyacrylamide gel. Figure 4-6. SDS-PAGE of pur ification fractions for AS-B E348A and E348D mutants. Lanes (2-5 for E348A, 7-10 for E348D): 1, 6 molecular weight marker, 2 E348A crude extract, 3 column flowthrough, 4 column wash, 5 pooled fractions, 7 E348D crude extract, 8 column flowthrough, 9 column wash, 10 pooled fractions. On 12% polyacryl amide gel, stained with Coomassie blue dye. 123456789 AS-B AS-B 1 2 3 4 5 6 7 8 9 10


51 Glutaminase Activity One important issue that arises during mu tagenesis experiments is the possibility that changes occur in the gross structure of the enzyme and not just at the site of mutation. AS-B has an advantage in that it possesses two distinct active sites that could catalyze independent reactions. Looking at the glutaminase activity is therefore a good strategy for evaluating domain misfolding and gross structural disruptions brought about by mutations in the synthetase active site. L-glutamate dehydrogenase assay foll ows the production of NADH from NAD+ as glutamate is converted to -ketoglutarate and ammonia (Sigma Chemical Co., Technical Bulletin No. GLN-1). Hence, the AS-B en zyme (wild-type or mu tant) was incubated with different concentrations of glutamine in the presence of DTT, and glutamate produced after some time was assayed with L-glutamate dehydrogenase in the presence of ADP and NAD+. The change in absorbance at 340 nm reflects the amount of glutamate produced by the AS-B reaction. Glutamine can cyclize into pyroglutamate with ease, and presents a serious problem in assays that involve the amino acid. For this reason, glutamine used in these studies with AS-B was recrystallized in a queous ethanol and prepared fresh before each assay. Table 4-1 shows the glutaminase kine tic constants obtained for the AS-B wild type and mutant enzymes. In the absence of ATP, all enzymes behave almost similarly, in giving comparable KM and kcat values, also consistent with valu es previously reported (Boehlein et al. 1994b, Tesson et al. 2003). Even though kcat values do not reveal the effect of mutagenesis on non-rate limiting steps in the entire glutaminase reaction, similar


52 turnover numbers for WT AS-B and mutants sugg est that the substitu tions did not destroy the structural integrity of the enzymes. Table 4-1. Steady state parameters for the glutaminase activity of the AS-B wild type and mutant enzymes. Assays were conducted in the presence a nd absence of 5 mM ATP. The assays were conducted in the pres ence and absence of ATP, which although not “necessary” was previously observed to e nhance the glutaminase activity of wild-type enzyme (Boehlein et al. 1994b). This effect was suspect ed to result from the coordination between the two active sites and/or the need for bound ATP at the synthetase active site for structural integrity of the enzyme. In the presen ce of ATP, wild-type enzyme was still consistent with literature values for ATP-enhanced glutaminase activity, KM, mM kcat, s-1 kcat/Km, M-1s-1 Without ATP WT 1.5 ± 0.1 3.03 ± 0.05 2020 S346T 1.3 0.2 2.6 0.1 2000 S346A 1.6 0.6 3.1 0.3 1938 E348D 1.6 0.1 1.79 0.03 1119 E348Q 1.6 0.5 1.7 0.1 1063 E348A 1.6 ± 0.3 2.0 ± 0.1 1250 E352A 1.1 0.5 1.7 0.2 1545 D384A 1.9 0.4 3.1 0.1 1632 R387K 1.7 ± 0.1 2.3 ± 0.1 1359 R387A 1.1 ± 0.1 2.3 ± 0.1 2091 With ATP WT 1.0 ± 0.1 5.7 ± 0.1 5700 S346T 1.1 0.3 3.7 0.2 3364 S346A 0.9 0.1 4.7 0.2 5054 E348D 0.7 0.1 4.1 0.1 5857 E348Q 1.2 0.4 2.0 0.1 1667 E348A 1.3 ± 0.2 2.3 ± 0.1 1770 E352A 1.2 0.3 3.0 0.1 2500 D384A 1.2 0.1 4.8 0.1 4000 R387K 1.1 ± 0.1 2.6 ± 0.1 2364 R387A 0.7 ± 0.1 2.9 ± 0.1 4143


53 where evidently the presence of ATP increased kcat to 188%, and the kcat/KM to 282%. All the mutants showed enhanced glutam inase activities in the presence of ATP, although the ones exhibited by E348Q a nd E348A were marginal, especially in comparison with E348D. Stimulation on kcat dropped from 229% to 118% to 115% in going from E348D, to E348Q, to E348A, respect ively. Whatever the mechanism of ATP stimulation on the glutaminase activity is, ther e appears an apparent correlation between the functionality present in the 348 pos ition and the level of observed activity enhancement. Synthetase Activity Under conditions of saturati ng substrates (ATP, Asp, NH4 +), and within initial velocity reaction times, synthetase activity was measured for wild-type AS-B and mutants. The pyrophosphate product of a 10 minute reaction was measured continuously using a commercially available kit containing a number of reagents and enzymes (Sigma Chemical Co., Technical Bulletin No. BI-100) , in the presence of all substrates and initiated by addition of AS-B. This assa y takes pyrophosphate (produced from the AS-B catalyzed reaction) through a series of enzymatic reactions found in the glycolytic pathway which eventually oxidizes two mol ecules of NADH per mole of pyrophosphate, monitored at 340 nm. The asparagine product, on the other hand, wa s measured with an end-point assay. The enzyme is incubated with all the substrates for 10 minutes, and then quenched. An aliquot of the asparagine produ ced in the reaction (as well as other amino acids present in solution) is converted to a dinitrophenyl deriva tive that can be sepa rated by reverse-phase HPLC (RP-HPLC) and detected spectr ophotometrically at 365 nm (Tesson et al. 2003).


54 Results for all mutants are listed in Table 4-2. Table 4-2. Comparison of pyrophosphate versus asparagine production for ammonia-dependent synthetase activity of AS-B mutants. In parentheses are values obtained when corrected for unexp ected pyrophosphate activity in the absence of aspartat e, discussed further in the text. n/a represents value lower than detectable limit. Immediately striking was the disparity in pyrophosphate and asparagine production for the mutants, with the exception of S346A. In the case of wild type AS-B, there was always a 1:1 ratio for these two products. For the pyrophosphate ac tivity, a control set-up containing no enzyme was always included in the assay to identify and correct for any anomalous outcomes. To investigate further, another control was set-up, this time excluding aspartate from the reaction mixture, and similarly initia ted with addition of en zyme. Surprisingly for most of the mutants, there was apparent activity based on the decrease in absorbance of NADH during the reaction. A few interpretati ons can be drawn from this result, one is that for these enzymes, ATP hydrolysis occu rs in the absence of aspartate, hence, pyrophosphate is formed without producing aspara gine. Another interpretation is that the decrease in absorbance has nothing to do wi th pyrophosphate at all, and represents a contamination in the enzyme stock aff ecting some other reaction by the coupling enzymes. This is not impossible, esp ecially since the commercially available Mutant PPi (as % WT) Asn (as % WT) S346T 65.9 (52.4) 53.5 S346A 95.0 95.0 E348D 66.7 (38.0) 34.5 E348Q 35.6 (1.9) n/a E348A 76.7 (n/a) 1.4 E352A 63.0 (13.6) 11.6 D384A 48.7 (4.9) 3.8 R387K 31.1 (4.3) 4.3 R387A 32.8 (3.2) 2.6


55 pyrophosphate reagent used in this assay is essentially a black box, made up of four enzymes and all the reagents necessary for these reactions to make pyrophosphate production (by AS-B) rate limiting. When corrections were made to incorporat e the results for the absence of aspartate (Table 4-2, in parentheses), the new numb ers are in good agreement with Asn production. S346T gave promising results at about 52 53% wild type activity. The change from Ser to Thr was a subtle one, with an additional methyl group and the hydroxyl functionality retained, which can explain why it was partially active. Mutants that retain some activity become more useful than ones th at lose it completely because the former can be assayed further, albeit at a much slow er rate. This potential for the Thr mutant however, was contradicted by the results obtai ned from the Ala mutant. S346A was 95% as active as wild type AS-B, indicating that the hydroxyl gr oup may not be important for function after all. The enzy me could better tolerate the loss of an –OH group at that position than an additional methyl group. S ubstitution with the bulkier Thr side chain may have pushed critical inter actions apart which may be the cau se of the loss of activity. The importance of S346 theref ore appears to be one of structure, rather than functionality. E348 mutants provided a different set of resu lts, with loss of ac tivity that could be correlated with the degree of substitution in that position. Perhaps more importantly, E348D was not completely dead in its ab ility to produce asparagine, and it allows measurement of extent of reaction rather than just a positive or a negative. E348A retained very little activity, as well as E348Q, supporting the importa nce of not just the correct size but the corre ct (carboxylate) function ality at this position.


56 With glutamine as nitrogen source, wild type AS-B, E348D, and E348A were again assayed for asparagine production, in the pres ence of saturating ATP and aspartate. The asparagine and glutamate products were DNP -derivatized and compared to a set of standards for quantitation. Table 4-3 lists the results. Table 4-3. Asparagine versus glutamate production for glutaminedependent synthetase activities of wild type, E348D, and E348A AS-B enzymes. Values represent the amount of asparagine produced during the 10 minute incubation period with units of micromolar (µM). Last column shows the ratio of Glu/Asn production. For the glutamine dependent activity, the wild type enzyme shows close to a 1:1 Asn:Glu ratio. This was expected, for a system that has two active si tes to be coordinated in their activities. Every a mmonia molecule produced by glutamine hydrolysis in one active site is used in the formation of one mo lecule of asparagine in the other active site. This was not the case for the E348D muta nt, which only produced a third of the asparagine as wild type AS-B, but about twi ce as much glutamate. The Glu:Asn ratio is altered, and active site coordination is perturbed. The result is a more “loose” enzyme that wastes 6 molecules of amm onia for every asparagine formed. Product ratio is even more disturbed in the case of E348A, which did not produce any detectable asparagine at all. Further characterization of E348 for its potential role in catalysis is described in the next chapter. uM Asn uM Glu Glu : Asn WT AS-B 92.3 112.6 1.2 : 1 E348D 31.3 229.8 7.3 : 1 E348A no Asn 105.6 -


57 D384A, R387K, and R387A gave greatly re duced activities, between 2.6 4.9% that of wild type enzyme, which were en couraging if the residues were critical for function. In order to explor e their involvement in binding of aspartate, the enzymes were each assayed for ammonia-dependent synthe tase activity at saturating ATP and NH4 +, and various concentrations of aspartate, m onitoring pyrophosphate production. Wild type AS-B has a KM app value of around 1 mM for aspartate, and initially the ra nge of aspartate concentrations used for the mutant assay cen tered around this value, with 10 mM as the highest concentration. The Michaelis-Menten plot for all three enzymes failed to show saturation, nor dependence on aspartate conc entration. Even when the aspartate concentration range was extended to as high as 100 mM, the results still failed to show any trend. Figure 4-7 shows the plots for R387A. Figure 4-7. Ammonia-depe ndent synthetase activity for R387A at aspartate concentrations of (A) 0.15 -10 mM, and (B) 1.6 100 mM. One interpretation to these re sults is that the range of concentrations used above was much higher than what was necessary to observe normal saturation kinetics. This would suggest however, that the KM app is significantly lower than 1 mM (that of wild type enzyme) which would approximate a much tighter binding for the substrate. 0.002 0.004 0.006 0.008 0.01 0.012 0.014 024681012 M PPi / s[Asp], mM 0.002 0.004 0.006 0.008 0.01 0.012 0.014 020406080100120 [Asp], mM M PPi / s(A) (B) 0.002 0.004 0.006 0.008 0.01 0.012 0.014 024681012 M PPi / s[Asp], mM 0.002 0.004 0.006 0.008 0.01 0.012 0.014 020406080100120 [Asp], mM M PPi / s(A) (B)


58 Since it is highly unlikely that the cha nges in the mutants would cause a higher affinity for aspartate, it appears that a s econd possible explanation may be true. The mutants could be dead and the measured activit y represents the error of the method used. This is highly reasonable especially in th e case of the Lys and Ala mutants of R387, which exhibited the same low level of activit y despite major difference in size and charge of the side chain. D384 and R387 may be critical in catalysi s (i.e., any changes in these positions will result in a dead enzyme), but with no means to assay the mutants further, this conclusion is difficult to make. Chemical Rescue The rescue of catalytic activity of inac tive or impaired mutant enzyme by the addition of exogenous compounds has already been described in the past (Carlow et al. 1995, Frillingos and Kaback 1996, Ha rpel and Hartman 1994, Newmyer and deMontellano 1996, Toney and Kirsch 1989). Th e recovery is reasonably attributed to the incorporation of a critical functionality in itially absent in the system, which validates the importance of the said functionality. In AS-B, a similar experiment had been carried out with R325A, which regained 15% of wild type activity after incubation with exogenous guanidinium hydrochloride (GdmHCl) (Boehlein et al. 1997). Using the same rationale, a chemical rescue experiment was attempted for the R387A mutant. Pyrophosphate assay was used to monitor bo th glutamineand ammoniadependent synthetase activ ities. In the presence of saturating ATP and NH4 + (or Gln), and up to 200 mM Asp, the mutant enzyme did not show a change in activity upon addition of GdmHCl. Even at a concentra tion of 25 mM, GdmHCl failed to rescue the


59 activity of the mutant. One possible reason is accessibility, an important prerequisite for this kind of chemical rescue. R387 may be bur ied much deeper in the enzyme structure, where exogenous compounds could not easily access. In order to make sure that GdmHCl did not affect the assay other than intended, a control experiment was conducted in whic h pyrophosphate solution of known concentration was assayed with th e pyrophosphate reagent in th e presence and absence of 25 mM GdmHCl. There was no difference in th e results, indicating that this concentration of GdmHCl does not alte r the pyrophosphate assay by itself. 13C Kinetic Isotope Effects Isotope effects are powerful tools that reve al important mechanistic information in enzymatic reactions that steady state kineti cs alone can not provide. Kinetic isotope effects reflect the changes in the reaction rate when a normal isotope is substituted with a heavier one. One method to determine isotope effects on V/K for a given substrate is called internal competition (Cleland 1982, Cleland 1999, O'Leary 1980), where both heavy and light isotopes of a particular subs trate are present in th e reaction mixture and the isotope effect is calculated from the mass ratio (or specific activ ity) of the product or the residual substrate, and is the method of choice when using natural abundance of 13C, 15N, and 18O. An isotope effect on V/K reflects changes coveri ng substrate binding up to the first irreversible step of the reaction and is described by the following equations. ) / 1 log( ) 1 log( ) / (0R fR f K Vp x (4-1) )] / )( 1 log[( ) 1 log( ) / (0R R f f K Vs x (4-2)


60 R0 is the specific activity or isotope ratio of the initial substrate or the product at 100% reaction, Rp is the isotope ratio of the product at a fraction f of the reaction, and Rs is the isotope ratio of the residual substrate at a fraction f of the reaction. In experiments using natural abundance 13C, 15N, or 18O in the “labeled” substrate, the carbon and oxygen are converted to CO2, and the nitrogen to N2 in order to determine the mass ratio using isotope ratio mass sp ectrometry (IRMS) (Klinman 1978, O'Leary 1989). Important factors that have to be c onsidered include possible contaminants from similar atoms in the substr ate during conversion to CO2 or N2 (i.e., it is imperative to single out the atom at a specific location), as well as solvent exchange of 18O in CO2 (i.e., it is necessary to use high vacuum conditions, or derive CO2 from I2 oxidation of formate or oxalate, or employ the remote la bel method) (O'Leary and Marlier 1979). In any case, if the isotope-sensitive step is the rate-limiting step, isotope effects can be measured. However, since the chemistry step is rarely fully rate-limiting, the isotope effect on V/K may be evaluated by the following e quation, which relates the measured isotope effect, x(V/K) , to the intrinsic (true) isotope effect on the chemical step, xk , and to the isotope effect on the equilibrium constant, xKeq. ) 1 ( ) ( ) / (r f eq x r f x xc c K c c k K V (4-3) The expressions cf and cr are commitment factors in the forward and reverse directions, respectively, and reflect the pa rtitioning of a reactan t between the isotope sensitive step versus its release from the enzyme. High commitments mask the intrinsic isotope effect and present a major obstacle in its measurement. There are several ways to determine values for commitments and to estimate intrinsic isotope effects but are however, beyond the scope of this work.


61 Figure 4-8 shows a proposed kinetic m echanism for AS-B, modeling in the observed glutamate/asparagine stoichiometry for glutamine-dependent synthetase activity (Tesson et al. 2003). From this model, equations for kinetic isotope effects can be derived for each substrate and used to calculate both forward (cf) and reverse (cr) commitment factors for the reaction of either Asp or ammonia. Figure 4-8. Model of proposed kinetic m echanism for AS-B. The mechanism assumes active site coor dination after -Asp-AMP formation, and no ammonia leakage from the intramolecular tunnel (Tesson et al. 2003). Preliminary work on heavy atom isotope effects for the ammonia-dependent synthetase activity of AS-B involved investigation of the -carboxylate of aspartic acid. This functional group is involved in many step s in the synthetase mechanism, including the attack on ATP to form the -Asp-AMP intermediate (i.e., steps described by k3 and k12), the nucleophilic attack of ammonia to form the tetrahed ral inter-mediate (i.e., step described by k7), and lastly, the departure of AMP to fo rm asparagine (i.e., step described by k7). 18O primary kinetic isotope effects (PKIE) will be sensitive to the first and third steps mentioned above, whereas 13C PKIE will be sensitive to the second and third steps. To determine the isotope effect on V/K for aspartic acid, wild type AS-B was incubated with aspartic acid containing natural abundance 13C (1.1%) and 18O (0.2%), in EE.ATPE.ATP.AspE. -Asp-AMP.PPi E.GlnE.ATP.Gln E.ATP.Asp.Gln E. -Asp-AMP.PPi.Gln E+Glu+NH3E.ATP+Glu+NH3E.ATP.Asp+Glu+NH3E+Asn+Glu+AMP+PPi k4[Gln] k5[Gln]k11[Gln] k6[Gln] k1[ATP]k2[Asp] k3k12k-4k-5k-11k-6k8k9k10k7k-2k-1


62 the presence of ATP and ammoni a carried out to various degrees of conversion. Residual aspartic acid was converted to oxaloacetat e using the enzyme glutamate oxaloacetate transaminase (GOT), which was then decar boxylated using oxaloacetate decarboxylase (OAD) (Figure 4-9). The CO2 product was then isolated via a high vacuum gas distillation line and analyzed by IRMS to dete rmine the isotope ratio in aspartic acid as a function of the extent of reaction. This method has been used to determine 13C and 18O isotope effects in reactions catalyzed by oxaoacetate decarboxylase and pyruvate kinase (Waldrop et al. 1994). Figure 4-9. Strategy for the conversion of aspartate -carboxylate to CO2. Wild type AS-B gave a very small average 13C isotope effect of 1.0014 0.004 (0.14%), a very small value considering 13( V/K ) can reach up to 4%. It appears that considerable commitments prevent the measurem ent of isotope effects at this position. The use of mutant enzymes with slower tu rnover rates can facilitate in making the isotope-sensitive step in this reaction at leas t partially rate limiting. At 34% activity of the wild type enzyme, E348D exhibited much slower turnover rate, yet allowed significant product conversion ne cessary for isotope effect measurements. However, the 13C isotope effect turned out even sma ller than that of wild-type, at 1.0006 0.0003. Attempts at another mutant, E352A, gave a sli ght inverse isotope eff ect but was still too small for interpretation. -O2C CO2NH3+ -O2C CO2O CO2 + -ketoglutarate + glutamate + pyruvate glutamate oxaloacetate transaminase oxaloacetate decarboxylase


63 The 13C isotope effects for wild-type, E348D , E352A AS-B, and even human AS are listed in Table 4-4. Table 4-4. 13C kinetic isotope effects on -COOof aspartate, n represents number of measurements. It therefore appears that the reaction of ATP and aspa rtate may commit the latter substrate to undergo reaction, and the use of slower turnove r mutants could not make the ammonia attack on the -carboxylate of the aspartyl moiety in -Asp-AMP the rate limiting step in catalysis. Other technique s are now being explored in the Cleland laboratory to address this pr oblem, including the potential us e of alternate substrates that are less “sticky”, and the use of the “d ribble-drip” method as in studies on aspartate transcarbamoylase (Parmentier et al. 1992a, Parmentier et al. 1992b), which introduces the substrate(s) at a controlled rate that maintains the concentration at low levels. Initial experiments with 18O isotope effects were problematic mainly because of the high vacuum conditions necessary for the GOT-OAD reactions. After driving the ammonia-dependent synthetase reaction to va rious degrees of completion, the residual aspartate had to be converted to CO2 for IRMS. This required the reactions with GOT and OAD to be carried out under high vacuum to prevent oxygen exchange. Under these conditions, it became physically difficult to introduce the GOT-OAD enzyme mixture into the flask containing the residual aspartate. In addition, the c onstant stirring (with a spin bar) necessary for this technique (Waldrop et al. 1994) also ended up denaturing the enzymes before the reactions were complete . Steps are currently being taken in the pH 13(V/K) SE n WT AS-B 8.0 1.0014 0.0004 5 E348D 8.0 1.0006 0.0003 3 E352A 8.0 0.9979 0.0011 3 Human AS 8.0 1.0006 0.0005 2


64 Cleland laboratory to optimize this techni que for AS, such as changing the order of addition of components in the reaction flas k. The changes have shown encouraging outcomes. Experimental Section Site-Directed Mutagenesis The same pET-21c(+) plasmid containing asnB gene from E. coli was used. A pair of primers was designed for each mutant, see Appendix A for more details. Polymerase Chain Reaction (PCR) was carried out following the QuikChange Mutagenesis Kit protocol (Stratagene), in which the reacti on mixture consisted of 10x reaction buffer, double stranded DNA template (plasmid containing E. coli asnB gene), primers, dNTPs (Stratagene), and dH2O to a total volume of 50 µL in a PCR tube. The reaction tubes were placed in a thermal cycler. Once the temperature reached 95°C, 2.5 U of Pfu Turbo® DNA polymerase (Stratagene) was added onto the reaction mixture to initiate the reaction. Table 4-5 lists the cycling parameters. Table 4-5. PCR parameters used for site-directed mutagenesis experiments. The PCR product was treated with Dpn I (Stratagene), which digests the parental DNA template, and then transformed into XL1Blue supercompetent cells (Stratagene) following manufacturer instructions. These were inoculated on Luria-Bertani agar plates containing 100 g/ml of ampicillin, and grown overnight at 37 C. Segment Cycles Temperature, °C Time 1 (hold) 1 95 30 s 2 16 95 30 s 55 1 min 68 10 min 3 (hold) 1 4 30 min


65 DNA from successful transformants was prepared from 10 mL overnight cell culture (in terrific broth + ampicillin) using alkaline lysis/polyethylene glycol (PEG) precipitation. Approximate DNA concentra tion was measured using UV spectroscopy at 260 nm and the following relationship: factor dilution ml g Abs DNA . / 50 ] [260 assuming that 1 absorbance unit at 260 nm is equivalent to 50 µg/mL of double stranded DNA. Expression and Purification of Wild-type and Mutant AS-B The pET-21c(+) plasmid containing the asnB gene from E. coli was available in the laboratory. Competent BL21(DE -3) cells were transformed with the pET-21c(+) vector and successful transformants were grown in a liter of M9 minimal salts medium containing tryptone (10 g/L) , glucose (0.75%) and 100 g/ml ampicillin, shaken at 250 rpm and at 37 C until OD600 was between 0.6 – 1.0 -D-isopropyl thiogalactopyranoside (IPTG) was added and shaking wa s continued for another 2.5 hours. Cells were harvested in a refrigerated centrifuge at 9k rpms and the pellet was resuspended in 50 mL lysis buffer (50 mM EPPS, pH 7.4, 1.0 mM DTT, 0.5 mM EDTA, 10% glycerol) and sonicated for 15 seconds th ree times with 1 minute cooli ng between bursts. Lysate was centrifuged at maximum speed to remove cell debris and the resulting supernatant was applied to a MonoQ column (Amersham Biosciences) equilibrated in 50 mM BisTris, 0.5 mM DTT, pH 6.5. Elution was carried out with runn ing buffer and 0-100% 1 M NaCl gradient at a flow rate of 2 mL/min , and 1 ml fractions were collected and kept in ice. Abs280 was obtained for each tube, and frac tions containing the active enzyme were pooled together. To concentrate the enzyme, solid (NH4)2SO4 was added to 70%


66 saturation, the solution was centrifuged, and the pellet was resuspended in storage buffer (50 mM EPPS pH 7.5, 1 mM DTT, 0.5 mM EDTA , 20% glycerol). Aliquots were taken, snap-frozen, and stored at -80°C. Glutaminase Activity In this assay, 100 mM EPPS buffer, pH 8.0, 0.1-100 mM Gln, 100 mM NaCl, 8 mM MgCl2, 0.5 mM DTT were all premixed with and without 5 mM ATP. Enzyme was added and the mix was incubated at 37 C for 10 minutes and then quenched with 20% trichloroacetic acid. This was adde d to a solution of 300 mM glycine-250 mM hydrazine buffer, pH 9.0, 1 mM ADP, and 1.5 mM NAD+, after which 2.2 U of L-glutamate dehydrogenase (Sigma) was a dded to initiate the reaction. Abs340 was monitored for 30 minutes and the difference be tween the final and initial measurements was compared to that of a set of glutamate sta ndards. Initial velocity values were plotted against glutamine concentration from which KM values were derived. Vmax and the known protein concentration determine kcat. Data fitting on Michaelis-Menten equation was done on Kaleidagraph (Synergy Software). Dinitrophenyl-Asparagine (DNP-Asn) Assay Wild type and mutant enzymes were inc ubated with saturating substrates (ammonia or L-gln as nitrogen source) at pH 8 for 10 minutes at 37 C and then quenched with glacial acetic acid. After neutralizing with base, an aliquot was incubated with DNFB for the derivatization reaction. A fraction of the reaction mix was injected into a C18 column (Varian) and separation was achieved with a gradient of acetonitr ile and 40 mM formic acid pH 3.6. Signal detection was at 365 nm. Standard calibration curves were


67 constructed from solutions of pure asparagine and glutamate derivatized in the same manner as the samples. Chemical Rescue Reaction mix included 100 mM EP PS at pH 8.0, 15 mM MgCl2, 100 mM NH4Cl or 12.5 mM Gln, 10 200 mM Asp, 10 mM AT P, 2-25 mM guanidinium hydrochloride, 350 L reconstituted pyrophosphate reagent (Sigma) and dH2O to a total volume of 1 mL. Reaction was initiate d by addition of enzyme. Pyrophosphate production was monitored spectrophotometrically ( 340 nm) at 37°C for up to 1 hour. 13C Kinetic Isotope Effects R0 values of the -carboxylate of aspartate was determined by setting up 3 mL reactions containing 100 mM HEPES pH 7.5, 10 mM MgCl2, 20 mM -ketoglutarate and 10 mM aspartate. The reaction mixture was sparged with N2 gas passed through an ascarite column and 2 N H2SO4 for 3 hours. Conversion to CO2 was carried out with the addition of 120 units of GOT and 150 units of OAD via needle and gas-tight syringe to the reaction solution, and allowed to incubate overnight. The reaction was quenched with 300 µ L of conc. H2SO4 the next day. The liberated CO2 was isolated and collected on a high vacuum gas distillation line using two dr y ice/isopropyl traps a nd a liquid nitrogen trap, and analyzed on an IRMS system (Finni gan-Mat Delta E) to determine the isotopic content. Partial isotope effect reacti ons were carried out in 3 mL reactions composed of 100 mM EPPS pH 8, 100 mM NH4 +, 10 mM ATP, 10 mM aspa rtate, and 20 mM MgCl2. The reactions were initiated by addition of the enzyme (wild-type AS-B, E348D AS-B, E352A AS-B and human AS), and allowed to proceed to 20-50% completion at room


68 temperature. Boiling at 100oC for 10 min quenched the reactions, and the dead enzyme was pelleted out by spinning th e reaction tube at maximum speed for 5 minutes. The sample was then dried by rotary evaporation and stored at -20oC until further use. Fraction of reaction was determined by meas uring the amount of asparagine and/or residual aspartate in an aliquot of the que nched reaction using the DNP-derivative HPLC assay. The dried reaction containi ng residual aspartate was resuspended in 4 mL dH2O containing 8 mM MgCl2 and 20 mM -ketoglutarate, and then sparged with N2 for four hours. Conversion to CO2 using GOT and OAD, and isotopic analysis were performed as described above. Control experiments containing all react ion components except for aspartate, carried out in exactly the same wa y produced no measurable amount of CO2. Even when asparagine was purposely added to the reaction mixture, no CO2 was detected by IRMS, indicating that asparagine doe s not decompose to aspartate during the lifetime of the reaction.


69 CHAPTER 5 ROLE OF GLU348 IN CATALYSIS Introduction The previous chapter reveals that substitutions to some of the synthetase active site residues results in dramatic loss of activity, which makes it difficult to assay the enzymes even further. An exception is Glu348 whose Asp mutant retained 34% of the wild-type synthetase activity. In addition, substitution with Asp resulted in nonstoichiometric glutamate to asparagine produc tion during glutamine-dependent synthetase turnover. It apparently perturbed the mechanism that al lows for the two active sites to coordinate or couple their activities more efficiently. It was proposed that Glu348 acts as a general base that abstracts a proton from ammonia after its nucleophili c attack on the carbonyl carbon. One important question then is the involvement of Glu348 in the formation and/or breakdown of the -Asp-AMP intermediate. In 1998, Boehlein et al. employed 31P NMR to follow the formation of -Asp-AMP in wild type AS-B in the absen ce and presence of ammonia. Following the scheme in Figure 5-1, they used 18O aspartate labeled at all fo ur positions, and monitored the AMP product at different time points. Figure 5-1. Scheme for 18O aspartate exchange experiment. Filled O represents 18O atom, whereas empty O represents 16O.


70 31P NMR is a valuable tool in analyzi ng bonding changes to phosphorus nuclei, and was used to investigate the AMP product. The 31P resonance shifts 0.02 ppm upfield when bonded to 18O rather than 16O and can therefore indicate changes in phosphorus bonding and location of 18O labels (Cohn and Hu 1980). The presence of the 0.02 ppm upfield signal distinguis hed AMP generated by the -Asp-AMP intermediate from the product of straightforward ATP hydrolysis. Using the same experimental scheme on Glu348 mutants and comparing with wild type AS-B will reveal if the residue plays a role in the formation of the -Asp-AMP intermediate. In addition, steady state kine tics will again be em ployed to explore the effect of Glu348 in substrate binding. Results and Discussion 18O Labeling of L-Aspartate 18O-labeled aspartic acid was synthesized by incubating with 18O water (95%, Cambridge Isotopes) at 80°C under acidic conditions for two weeks. Aspartic acid was only partially soluble during the entire incuba tion period. The solid was recovered by microdistillation on a Kugelrohr apparatus (Büchi), and wa s brought into solution at pH 8. To determine the extent of 18O incorporation, the sample, as well as unlabeled aspartic acid standard were analyzed via C18 HPLC and electrospray ionization-mass spectrometry (ESI-MSn) in (+) and (-) mode (Mass Spectrometry Facility, Chemistry Department, UF). Results show that the predom inant form of aspartic acid in the sample was 18O4, yielding a molecular weight of 141 g/mo l. Positive ESI-MS (zoom-scan mode) was used to estimate the percentages of each of the isotopes in the Table 5-1. The data suggests that the labeled aspartic acid was ~ 80% enriched in 18O.


71 Table 5-1. Percentage of 18On-labeled aspartic acid in the incubated sample, where n goes from 0 to 4. Control Experiments for 31P NMR In order to identify the 31P NMR signals for ATP and AMP, control experiments were carried out including 50 mM ATP, 50 mM AMP, and a 50/50 solution of the two prepared in 10% D2O and analyzed on a 31P NMR spectrometer. The instrument was VXR 300 (Varian) with a 7 T (300 MHz) magn et, which is routinely used for direct observation of H1, F19, C13 and P31 wit hout the need for tu ning (NMR Facility, Chemistry Department, UF). The spectra we re obtained using 128 scans with acquisition time of 1.6 s, and with broadband proton dec oupling. The ATP standard produced three peaks: –5.656 (d, -P), -10.037 (d, -P), -20.772 (t, -P). AMP gave one singlet at 4.785 ppm. Due to the ease of identification, the -phosphorus was used as reference for the ATP/AMP solution and the frequencies are: 25.199 (s, AMP), 15.688 (d, ATP -P), 10.512 (d, ATP -P), 0.000 (t, ATP -P). This suggests that at comparable concentrations, phosphorus species in ATP a nd AMP can be easily detected using 31P NMR. A sample AS-B reaction was set-up to i nvestigate the conditions that will produce enough AMP for the NMR experiment. The r eaction mixture consists of 100 mM EPPS, pH 8, 10 mM ATP, 20 mM MgCl2, 10 mM NH4Cl, 10 mM Asp, and 100 µg of WT AS-B. Control set-ups where no enzyme and no aspartate were also prepared, for blank and for straightforward ATP hydrolysis, respect ively. The tubes were incubated in a Ion # 18O Area % Total m/z 134 0 2,049,072 6.49 m/z 136 1 1,292,907 4.09 m/z 138 2 2,436,826 7.72 m/z 140 3 8,691,437 27.52 m/z 142 4 17,106,937 54.18 Total Area =31,577,179 100.00


72 37ºC water bath for 3 hours and the reaction wa s quenched with trichl oroacetic acid to a final concentration of 4%. The tubes were centrifuged to isolate th e precipitated protein and the supernatant was transferred to a se parate tube containi ng EDTA, glycine, and 20% D2O, and the pH was adjusted to 9.5 with sodium hydroxide [based on (Boehlein et al. 1998)]. Finally, the soluti on was transferred to an NMR tube for analysis. The spectra were obtained using 128 scans with acq uisition time of 1.6 s, and with broadband proton decoupling, and using the triplet signal for -phosphorus of ATP as reference. The results are summarized in Table 5-2. Table 5-2. List of 31P NMR signals present in the reaction set-ups [(A) no AS-B, (B) ASB reaction, and (C) no Asp] afte r the 3-hour incubation period. Based on the chemical shift, AMP was id entified and as expected, was detected only in (B), which contains all the substrates and AS-B. This result suggests that the assay conditions used in (B) are suitable to produce adequate AMP for detection. The peak at 17.9, present in both (A) and (C) was identified as a spurious signal produced by the spectrometer and does not correspond to any real phosphorus spec ies present in the reaction mixture. Some Unexpected Results What was immediately noticeable apart from the AMP and ATP signals was the one observed at 23.7, present in the AS-B reac tion (B) and where no Asp (C) was added. This was assigned as a Pi signal based on previous literature (Vanwazer et al. 1956). A related observation is the absence of a signal for pyrophosphate, which would have appeared at around 15 ppm. This wa s unexpected as one would anticipate finding ATP ( -, -, -P) AMP Pi other signals A 15.486(d), 10.298(d), 0(t) ----17.891 B 15.493(d), 10.298(d), 0(t) 24.858( s) 23.717(s) 15.147(d), 10.755(d) C 15.483(d), 10.303(d), 0(t) --23.722(s) 17.916, 15.031(d), 10.728(d)


73 the pyrophosphate signal if AMP was being produced, whether from the AS-B reaction or from straightforward ATP hydrolysis. Despite the fact this NMR data was not fit for integration, a direct comparison of peak heig hts reveals that there is at least twice as much Pi as AMP. This suggests that the Pi signal possibly arose from pyrophosphate via a pyrophosphatase contaminant in the enzyme stock. Another unexpected result observed in (B ) and (C) was the emergence of two doublets ( 15.147 and 10.755 in B) close to the and -phosphorus of ATP. These are potentially signals for ADP (Lerman and Cohn 1980), which can either come from: (i) direct hydrolysis of ATP to ADP and Pi; (ii) contaminating adenylate kinase activity in the enzyme stock, or (iii ) both. Adenylate kinase ta kes AMP and ATP, and converts them into two molecules of ADP, requiring only Mg2+ as cofactor. If adenylate kinase was a contaminant in the AS-B stock solution, it could take the AMP produced by the AS-B reaction and the unreacted ATP in the mi xture to produce ADP, in the presence of Mg2+. The small signals for Pi and ADP in (C) where no aspartate was included in the reaction mix, supports the fact that ATP hydrolys is to ADP and Pi occur to a small extent (i.e., ADP is produced even in the absence of AMP), and perhaps represents the basal level of this activity when enzyme is presen t. However, the much bigger signals for ADP detected in (B) containing enzyme and a ll substrates, suggests that ADP production was also dependent on AS-B turning over, and s upports adenylate kinase contamination. From these two pieces of data, it appears that both mechanisms for ADP production are at play, although adenylate kinase contamina tion presents a much bigger complication in future quantitation efforts. Despite the complications presented by cont aminants in the AS-B enzyme stock,


74 the conditions used for the enzymatic reaction were suitable for the detection of AMP, and should prove useful in the experiments employing 18O aspartate as substrate. 18O Transfer Studies Following the procedure outlined in the E xperimental Section, and using 10 mM of 18O-labeled aspartate in place of the regular solution, the spectra in Figure 5-2 were results obtained for wild-type AS-B. Figure 5-2. 31P NMR spectra for the reactions cata lyzed by wild-type AS-B in the (A) presence, and (B) absence of 18O labeled aspartate. Inset picture shows a blown up version of the AMP signal, marked by the red asterisk in the complete spectrum. Arrows in (B) indicate Pi and ADP. Signals obtained from the AS-B reaction w ith labeled aspartate (Figure 5-2A) were: 0(t), 10.718(d), and 15.113(d), for the -, -, and -P of ATP, respectively; 10.237 (d), and 15.463 (d), for the -, and -P of ADP, respectively; 23.650 for Pi; and 24.758, 24.783 for AMP. Similar to what was observed in the control reactions, (A)* (B) (A)* (B) 24.783 — —24.758 (A)* (B) (A)* (B) (A)* (B) (A)* (B) 24.783 — —24.758 24.783 — —24.758


75 no PPi signal was observed, and there was s ubstantial ADP production, both of which were taken as a result of contaminating enzy me activities in the AS-B stock solution. More importantly, AMP gave two distinct peaks, 0.025 ppm from each other. This suggests the formation of -Asp-AMP intermediate, as expected, from which the 18O label was transferred to AMP. In the absence of aspartate (Figure 5-2B), ATP hydrolyzed to ADP and Pi to a very small extent. The set-up and results are exac tly the same as the control reaction where aspartate was omitted. The definitive experiment for the invol vement of E348 in the formation of -AspAMP intermediate was the 18O transfer studies performed on th e mutants of this residue. E348D would have introduced a smaller side chain in the acti ve site, while still retaining the negative charge with the carboxylate group. Following the procedure in the Experimental Section and with 10 mM 18O-labeled aspartate as one of the substrates, the 31P NMR experiment was done using 100 µg of E348D mutant, with and without 100 mM NH4Cl. Figure 5-3 shows the results. Figure 5-3A shows the spectrum for the complete reaction of E348D (i.e., in the presence of all substrates). All the peaks a ppear similar to that of the wild-type enzyme, except for their heights. There was much more ATP left unreacted (i.e., peaks were taller), and much less AMP, Pi (hence, PPi), and ADP produced. This agrees well with the slower turnover observed for this mutant for asparagine formation under steady state conditions, reported in the previous chapter. The AMP product also exhibited two peaks separated by 0.020 ppm, which indicates the involvement of -Asp-AMP intermediate in the breakdown of products.


76 Figure 5-3. 31P NMR spectra for the reactions catalyzed by E348D mutant of AS-B in the (A) presence, and (B) absence of a mmonia. Inset picture shows a blown up version of the AMP signal, marked by the red asterisk in the complete spectrum. Arrows in (B) indicate Pi and ADP. Shown in Figure 5-3B is the reaction wit hout ammonia, which displays the same features as that of wild-type AS-B in the ab sence of aspartate (Figure 5-2B). The arrows point at ADP and Pi produced at basal leve ls of ATP hydrolysis. AMP was not detected under these conditions, which suggests that in the absence of a nitrogen source, either: (i) the formation of the -Asp-AMP intermediate becomes more difficult; or (ii) the intermediate is formed just as easily but the subsequent hydrolysis and release of AMP to bulk solution becomes more difficult. Whatever the underlying reason is, it can be generalized that for the E348D mutant, ammonia is required for AMP to form in detectable levels, unlike the wild-type enzy me which forms AMP even in the absence of a nitrogen source (Boehlein et al. 1998). A direct comparison of the peak heights (representing the amount of each P nuclei) for the wild-type and E348D mutant is outline d in Figure 5-4. On the x-axis are the (A) (B)* (A) (B)* 24.773 — —24.753 (A) (B)* (A) (B)* 24.773 — —24.753


77 phosphorus signals for each 31P detected in the NMR experi ment, and on the y-axis are the relative amounts. The smallest p eak height detected (g-ATP for the -P of ATP) was set a value of 1. Figure 5-4. Comparison of the relative amount of each P nuclei detected in the wild-type and E348D reactions, colored blue and red, respectively. g-, a-, and bstand for -, -, and P, respectively. The figure demonstrates that despite the fact that AMP is produced in both the wild-type and the E348D mutant, the latter only made about a third as the wild-type enzyme. This is also reflected in the relativ e amount of PPi (from Pi signal) that the two enzymes produced. The same pattern can be s een in the levels of ATP consumed, which suggests that the mutation indeed had an e ffect on the manner in which ATP was being used up. In this case, ATP is consumed towards the formation of the -Asp-AMP intermediate, in both enzymes, and the re sults suggest that changing Glu348 to Asp reduces the enzymeÂ’s ability to form the -Asp-AMP intermediate, necessary for the entire AS-B reaction to proceed. WT vs E348D0 1 2 3 4 5 6 7 8 9 10 g-ATPa-ATPb-ATPb-ADPa-ADPAMPPi31PRelative amount


78 The experiments for wild-type enzyme and E 348D were repeated in the presence of sodium hypophosphite as internal standard ( 28.470, -P of ATP as reference). Results were consistent with the co mparison already presented. Asparagine production was determined unde r NMR assay conditions for the wildtype and E348D enzymes to validate the reduced activity observed for AMP and PPi production. A large quantity of enzyme (100 µg for each) was therefore incubated with substrates (see 31P NMR assay, Experimental Section) in the absence and presence of aspartate, at 37°C for three hours and then quenched with acid. An aliquot of the neutralized reaction was deri vatized using dinitrofluor obenzene (see DNP-Asparagine assay, Experimental Section, Chapter 4) , and analyzed using RP-HPLC along with standards. Results are summarized in Table 5-3. Table 5-3. Amount of asparagine (in mM) produced at 37 ºC for 3 hours under various cond itions. Asparagine was separated and identified as the DNP derivative. E348D gave 37% activity relative to the w ild-type reaction. Th e values obtained for the set-ups where aspartate was omitted ar e within the error of the method and should not be taken as real. Results for asparagine production were consis tent with those from the 31P NMR experiments. E348D shows about a third of the wild-type activity, in the production of AMP, PPi, and asparagine. The Ala mutant of E348 provides further suppo rt for the role of this residue in the formation of the -Asp-AMP intermediate. When assa yed under similar conditions as the mM Asn WT 3.69 WT Asp 0.0019 E348D 1.36 E348D Asp 0.0028


79 two previous enzymes for the 31P NMR experiment, E348A gave no evidence for AMP formation at all (see Figure 5-5), or perhaps at levels much lower than the detection limit of the NMR method employed. The spectra ge nerated in the absence and presence of ammonia appeared almost identical, which bot h featured basal leve ls of ATP hydrolysis to ADP and Pi. Figure 5-5. 31P NMR spectra for the reactions catalyzed by E348A mutant of AS-B in the (A) presence, and (B) absence of ammonia. An identical set of results was obtained for E348Q (Figure 5-6). The series of 31P NMR experiments demonstrated the significance of E348 in the formation of -Asp-AMP, contrary to what was initia lly predicted for this residue. The 18O transfer studies revealed not only the extent of AMP pro duction but also the fate of ATP as it is broken down. (B) (A) (B) (A)


80 Figure 5-6. 31P NMR spectra for the reactions catalyzed by E348Q mutant of AS-B in the (A) presence, and (B) absence of ammonia. Michaelis-Menten Kinetics for ATP and Asp One avenue that had to be explored was the potential role of Glu348 in binding the substrates. The task quickly became imperative in light of the results obtained from the 18O transfer studies demonstrating how the residue is critical in the formation of an important intermediate. It was easy to im agine how a weakening in substrate binding can slow down everything el se that follows, including -Asp-AMP formation. Steady state kinetics provides a means to physically evalua te the effect of mutations on the enzymeÂ’s kinetic properties. In an enzymatic reaction that involves one or several intermediates, there are conditions th at allow for the rate of formation of these species to equal the rate at which they are consumed, and the reaction is said to be at steady state. One of these conditions is to ha ve the substrates in far greater excess than the enzyme, such that changes in substrat e concentration as th e reaction progresses (A) (B) (A) (B)


81 become negligible. Another requirement is to measure “initial rate s” of the reaction, where substrate conversion (or product formati on) is not considerable . This allows for the initial rate equation to be simplified (i n equations 5-1 and 5-2) as follows, known as the Michaelis-Menten eq uation (Michaelis 1913): S K k S E vM cat 0 (5-1) max 0V E kcat (5-2) where v is initial rate, [E]0 is total enzyme concentration, [S] is substrate concentration, kcat is the turnover number, KM is the Michaelis constant, and Vmax is the maximum limiting value for v . At low [S] ([S] <

82 variation in concentratio n of that one substrate, as all the others were held constant. This gives a value for an ‘apparent’ KM, KM app, instead of a true constant. The experiment was carried out by having al l the substrates present at saturating levels (i.e., at least 10-fold greater than their reported KM values), except for the substrate whose KM was to be determined, in which case, a range of concentrations is tested centering on the predicted or literature value. Hence, the reaction mixture consisted of EPPS buffer at pH 8, ATP, aspartate, MgCl2, and NH4Cl as nitrogen source, as well as the pyrophosphate reagent (Sigma) that will allow for monitoring of the said product through the course of the r eaction (see Experimental Se ction). The mixture was incubated at 37°C prior to addition of enzyme , which initiated the reaction. Absorbance at 340 nm was then monitored for the oxidation of NADH, and the absorbance/ time was taken and converted to mM PPi/s with the following equation derived from Beer’s Law: 2 22 . 6 ] [ absorbance PPi (5-3) where 6.22 is the millimolar absorptivity of NADH at 340 nm, and 2 is the number of moles of NADH oxidized per mole of pyrophos phate consumed by the coupled enzyme reactions. Plotting the values for the reaction rate, v , versus the substrate concentration, [S], would produce a hyperbolic curve, mark ed by a linear portion at low [S] and an asymptotic curve at high [S] values. Data fitting into equation 5-1 was done and the values for KM app and Vmax were obtained, which when applied to equation 5-2 with the known [E]0 gave the value for kcat. Kinetic constants for ATP and aspartate with wildtype and the two mutants ar e listed in Table 5-4.


83 Table 5-4. Kinetic constant s for the substrates ATP a nd aspartate under steady state conditions, evaluated for wild-typ e AS-B, and the two mutants. It was striking how for the three enzymes the KM app values for both substrates were remarkably close to each other. If KM were to be taken as an approximation of the dissociation constant for the enzyme-substrat e complex, the results suggest that ATP and aspartate were able to bind invariably to the three enzymes, despite differences in the active site where Glu348 is located. Th e dramatic difference was seen in the turnover number, which was not surprising from the NMR results showing the mutantsÂ’ diminished capacity to make the products (A MP, PPi, and asparagine), although, in no means were under steady state conditions. The decreasing trend observed for the specificity constant therefore, was a result of a reduced kcat, with no considerable contribution from the KM values. From these findings, it is apparent that Glu348 does not exert its effect in binding of ATP and as partate, and suggests a role somewhere else. Unfortunately, kcat information will not provide any means to further scrutinize the mechanism. It is important to highlight at this po int however, that the kinetic parameters obtained for the wild-type en zyme was in good agreement with published results for AS-B, and reflects consistency in enzy me preparation and characterization. The Proposed Role of Glu348 In summary, Glu348 is an intriguing residue that appears to be important in the coordination of the glutaminase and synthetase activities in AS-B. It does not influence KM app, mM kcat, s-1 kcat/KM app, M-1s-1 ATP Asp ATP Asp WT 0.19 0.03 0.95 0.23 3.0 15789 3158 E348D 0.17 0.01 0.92 0.24 1.0 5882 1087 E348A 0.13 0.01 0.94 0.34 <0.3 <2308 <319


84 binding of ATP and aspartate, yet is involve d in the mechanism of ATP stimulation of glutaminase activity. In addition, it play s a significant role in the formation of -AspAMP intermediate. In an attempt to put all the above piece s together, the following arguments were generated: If Glu348 is involved in -Asp-AMP formation, it has to be in the events that follow ATP binding and before aspa rtate binds or reacts with AT P, to be able to explain the results for ATP stimulation of glutam inase activity. ATP binding perhaps causes a conformational change as it gets ready for aspartate attack that signals the glutaminase site to go much faster, hence, the observed ATP stimulation. If E348 really is involved in giving the enzyme the “correct” structural c onformation, this woul d explain why the Ala mutant’s glutaminase activity was not affect ed by ATP. The E348A enzyme may have been able to take ATP in ju st the same (i.e., invariant KM app for ATP for all enzymes) but the “necessary” conformational change could not take place without the correct functionality at the 348 position, and gl utaminase stimulation is not observed. For the E348D mutant, ATP binds, conformational change follows, and glutaminase activity is “switched on” or activated fo r the synthetase reaction. However, because of the slower rate of -Asp-AMP formation and the overall reaction, the glutaminase site stays “on” and doubles glutamat e production without equivalent asparagine formation. Hence, the observed decoupling of the two activities. For the wild-type enzyme, the same thing happens initially with ATP binding, followed by conformational change, and then gl utaminase activity being “switched on”. Because the enzyme is working efficiently, glutamine hydrolysis is better coordinated


85 with the faster turnover in the synthetase site and will only make enough ammonia for the -Asp-AMP intermediate being formed. T hus, glutamine hydrolysis is limited by the synthetase reaction (E + Gln + H2O + ATP + Asp E + Glu + Asn + AMP + PPi), resulting in better coordi nation of the two sites. For the E348A, the conformational change necessary to switch on the glutaminase site may not occur. Glutamine hydrolysis th en will only arise from the glutaminase activity (E + Gln + H2O E + Glu + NH3) and nothing else. The role of Glu348 may be to give the enzyme the “correct” structural conformation after ATP binding necessary for: (A) the formation of the -Asp-AMP intermediate; and (B) the coordination of the two active sites. Experimental Section 18O labeling of Aspartic acid 18O-labeled Asp was prepared by ta king 0.0597 g L-Asp, adding 700 µL 18O-water, adjusting the pH to ~ 1 by adding a drop of 6 M HCl , transferring to a clear 2 mL glass ampule and sealing the end with heat. Lasp was only partially soluble under these conditions. The sealed ampul e was then placed in an 80 C heating block for 14 days. At no point during the incubati on period did the solid completely go into solution. The ampule was broken at the neck and inserted into the inner glass bulb of a Kugelrohr microdistillation apparatus (B chi). Temperature was set to 130 C with spinning, at atmospheric conditions (i.e., non-vacuum). Drying took 1 hour 10 minutes, at which point 18O-water collected in the distillation bulb wa s transferred to a screw cap glass vial and stored in –20C. Solid L-asp was resu spended in 70 µL 10 M NaOH and deionized water to a total volume of 177.3 µL. This makes a stock of 2.5 M and pH 8. The


86 solution was filtered through a Microcon spin fi lter to remove impurities and stored in a screw cap vial at -20 C. The sample and unlabeled aspartic acid st andard were analyzed via isocratic C18 HPLC/ESI-MSn with both (+) and (-) modes (Mass Spectrometry Facility, Chemistry Department, UF). Positive ESI-MS (zoom scan mode) was used to estimate the percentages of each of the isotopes. Mass spectra are shown in Appendix B. 31P NMR Assay The reaction mixture consists of 100 mM EPPS, pH 8, 10 mM ATP, 20 mM MgCl2, 10 mM NH4Cl, 10 mM Asp, and 100 µg of enzy me in 1 mL total volume. The tubes were incubated in a 37 ºC water ba th for 3 hours and the reaction was quenched with 250 µL of 20% TCA (4% final concentr ation). After spi nning the tubes in a centrifuge, 700 µL was transferred to a sepa rate tube containing 0.0287 g EDTA, 0.045 g glycine, 200 µL D2O, and 12 drops of 1 M NaOH to adjust the pH to 9.5. Finally, 750 µL of the solution was transferred to an NMR t ube for analysis. The instrument used was VXR 300 (Varian) with a 7 T (300 MHz) magn et (NMR Facility, Chemistry Department, UF). The spectra were obtained using 128 scan s with acquisition time of 1.6 s, and with broadband proton decoupling, and using the triplet signal for -phosphorus of ATP as reference. ATP and Asp KM Pyrophosphate assay was used to check th e activity of the mutant enzyme. All other substrates were saturati ng except for the one whose KM is to be determined at 100 mM EPPS pH 8, 10 mM Mg2+, 100 mM NH4 +, (5 mM ATP, 100 mM Asp). Buffer, substrates, and pyrophosphate reagent were mixed, incubated at 37 C, and the reaction

PAGE 100

87 initiated by the addition of enzyme. Abso rbance at 340 nm was monitored for 10 mins and the slope was taken and converted to units of mM PPi/s. This was plotted and fitted to obtain Vmax and KMa pp. Data fitting was done using KaleidaGraph ver. 3.5 (Synergy Software).

PAGE 101

88 CHAPTER 6 CONCLUSIONS Inhibition of Recombinant Hu man Asparagine Synthetase Efforts at finding effective inhibitors fo r human AS in the past were greatly hindered by limitations in obtaining milligram qu antities of active, correctly processed enzyme. The development of an efficient expression system for C-terminally tagged, recombinant human AS, that reproducibly yiel ds 25-30 mg of active enzyme per liter of cell culture, represents a milestone in AS res earch. It has allowed a detailed characterization of the two most potent inhibitors ever reported for human AS, and will facilitate the discovery of more drug-like compounds in the future. N -acylsulfonamide analog of -Asp-AMP intermediate, and N -adenylated S -methyl-L-cysteine sulfoximine are slow bi nding inhibitors of human asparagine synthetase. With overall dissociation cons tants of 728 nM and 2.46 nM, the sulfonamide and sulfoximine, respectively, represent th e most tightly bound inhibitors for any asparagine synthetase ever reported. Sulfoximine binds to the free form of the enzyme and is shown to inhibit asparagine and pyrophosphate production to the same extent. Inhibition is shown to be reversible, with enzyme reac tivation having a half-life of 7.4 hours, and is found to enhance glutaminase activity by 138%. More importantly, sulfoximine has a cytostatic effect in MOLT-4 human leukemia cells in the presence of asparaginase. Although sulfoximine may not be a suitable drug candi date, it provides a good starting point for the next generation of inhibitors that will be designed and synthesized for human AS.

PAGE 102

89 The implications of this work are far more significant than just the discovery of a nanomolar inhibitor though. It describes the development of a system – of protein expression and purification, of activity assa ys, and cell-based work – that allows the characterization of human AS inhibitors to a certain level of scrutiny. More importantly, it provides direct evidence that inhibitors of human AS re present interesting compounds for the treatment of asparaginase-resistant ALL, as initially proposed some forty years ago. Functional Characterization of E. coli Asparagine Synthetase B Considerable work was done in the 1990’s on E. coli AS-B to find out important structural and mechanistic features of the enzyme. This was marked by efforts to map out the two active sites using only sequence information and mutagenesis studies, with much of the work focused on the glutaminase active site. The first and only crystal structure for AS-B (C1A mutant) was solved much later on, when st ructural studies had ceased. To date, the fundamental question of how the synthetase active site works remains an important objective, especially si nce: (1) it imparts the functional specificity to this glutamine amidotransferase; and (2) some residues in this site are conserved throughout various species (inclu ding humans), and may prov ide clues to making better inhibitors. The computational model generated for this enzyme highlights several residues as possibly catalytically important. Of these, Gl u348 showed the most in teresting results. It was observed that substitution of this re sidue with Ala abolishes ATP stimulation on glutaminase activity. It decouples gl utamate and asparagine production during glutamine-dependent synthetase turnover. 18O transfer studies from aspartate to AMP

PAGE 103

90 employing 31P NMR demonstrate that Glu348 a ffects the formation of the -Asp-AMP intermediate without any evidence th at it alters ATP and Asp binding. From these findings, it is clear that Glu348 is catalytically important, and the proposed role for this residue is to give th e enzyme the “correct” structural conformation after ATP binding necessary for the formation of the -Asp-AMP intermediate, and the coordination of the glutaminase and the synthetase active sites. There are more residues to be invest igated, and work on Glu348 provided a good place to start. The results we re different from the initial predictions, but proved to be more interesting. This emphasizes that characterization of the functional elements in the synthetase active site of AS-B remains an e ssential objective of research in this field.

PAGE 104


PAGE 105


PAGE 106

93 APPENDIX B MASS SPECTROMETRIC ANALYSIS OF 18O ASPARTIC ACID Mass Spectrometry: ThermoFinnigan (San Jo se, CA) LCQ in elec trospray ionization (ESI) mode ESI-Parameters: sheath gas (N2) = 65; aux ga s (N2) = 2; spray voltage = 3.0 kV; cap temp = 250C; cap volt = 40; tube lens offset = 3V HPLC: Agilent (Palo Alto, CA) 1100 series binary pump Column: Phenomenex (Torrace, CA) Synergi 4u Hydro-RP 80A (2 x 150 mm; 4 µm; S/N=106273-5) plus C18 guard column (2mm x 4 mm) Mobile Phases: A = H2O (Burdick & Jackson, B&J), B = MeOH (B&J) Flow rate: 0.15 mL/min Injector: manual Rheodyne 7125 injector, 25 µL injection loop All analyses were obtained with following conditions: HPLC isocratic at 0.15 mL/min A:B = 100:0 PCM = 20 µL/min MeOH (B & J) 18O-labeled aspartate sample, 1 mg/50 µL concentration a. 2.5 µL injected b. (+)ESI-MS (m/z 50-300); dependent MS /MS of base peak; Zoom-MS(m/z 129-149)

PAGE 107

94 The predominant form of Asp in the sample was 18O4 yielding a molecular weight of 141 g/mol. Positive (+)ESIMS (zoom scan mode) was us ed estimate the percentages of each of the isotopes in the following table. N H2O18O H18O18O H18 Molecular formula = C4 H7 N 18O4Molecular weight = 141 u

PAGE 108

95E:\0-Spec\0-Data\Seq-5364-0310/04/05 07:12:59 AMJAG-ASP, 1 mg/50 uL; 2.5 uL injected Hydro-RP;0.15;isocratic A:B=100:0/PCM=20ul/m MeOH/ESI SEQ-5364-03 # 134-172 RT: 3.40-4.24 AV: 13 SB: 22.84-2.99 NL: 1.60E5 F: + c ESI Full ms [ 50.00-300.00] 60 80 100 120 140 160 180 200 220 240 260 280 300 m/z 0 20 40 60 80 100Relative Abundance 141.9 140.0 137.9 91.9 180.8 93.9 164.0 241.0283.0 90.0 262.8 77.9 142.9 102.1 204.7287.9 209.3 222.5255.1 237.1 194.7266.8 72.9 122.0 64.9 58.3 SEQ-5364-03 # 135-175 RT: 3.38-4.29 AV: 14 NL: 5.07E3 F: + ESI Z ms [ 131.00-151.00] 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 m/z 0 20 40 60 80 100Relative Abundance 142.0 140.0 139.9 138.0 134.0 136.0 141.8 141.1 149.1 139.8 143.0 137.9 139.1 150.2 136.5 135.1 133.1 146.0 145.2148.7 144.7 146.9 132.9 147.9 144.1 131.9 SEQ-5364-03 # 135-175 RT: 3.43-4.27 AV: 13 NL: 3.94E4 T: + c d Full ms2 141.89@37.50 [ 25.00-295.00] 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 m/z 0 20 40 60 80 100Relative Abundance 121.8 91.9 119.7 89.8 77.9 47.8 139.5 153.7 118.9 123.5 77.1 151.7 110.0 164.4 132.3 84.4101.8 55.8 161.6 95.9 Figure B-1. (+)ESI produced an m/z 134-142 [M+H]+ ion cluster (top and middle). The m/z 140/142 ions underwent loss of 20 u (H2 18O) to form m/z 120, 122 and loss of 50u (HC18O2H) to form the m/z 90, 92 ions.

PAGE 109

96 Figure B-2. HPLC/(+)ESI-MS mass chromatograms of each of the 12C-[M+H]+ ions, m/z 134, 136, 138, 140, and 142. Each ion-peak was integrated (indicated by the shading) and the areas are given by the MA:# above each peak. These areas were used to estimate the percentage of each (inserted table).E:\0-Spec\0-Data\Seq-5364-0310/04/05 07:12:59 AMJAG-ASP, 1 mg/50 uL; 2.5 uL injected Hydro-RP;0.15;isocratic A:B=100:0/PCM=20ul/m MeOH/ESI RT: 0.00 8.80 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 Time (min) 0 50 100Relative Abundance 50 100Relative Abundance 0 50 100Relative Abundance 0 50 100Relative Abundance 0 50 100Relative Abundance RT: 3.66 MA: 2049072 BP: 134.1 RT: 4.15 MA: 1292907 BP: 136.0 RT: 3.66 MA: 2436826 BP: 138.1 RT: 3.59 MA: 8691437 BP: 140.0 RT: 3.80 MA: 17106937 BP: 142.1NL: 3.58E4 m/z= 133.5-134.5 F: + ESI Z ms [ 131.00-151.00] MS SEQ-5364-03 NL: 2.28E4 m/z= 135.5-136.5 F: + ESI Z ms [ 131.00-151.00] MS SEQ-5364-03 NL: 5.12E4 m/z= 137.5-138.5 F: + ESI Z ms [ 131.00-151.00] MS SEQ-5364-03 NL: 1.22E5 m/z= 139.5-140.5 F: + ESI Z ms [ 131.00-151.00] MS SEQ-5364-03 NL: 2.64E5 m/z= 141.5-142.5 F: + ESI Z ms [ 131.00-151.00] MS SEQ-5364-03 Ion#18-O A rea%Total m/z 13402,049,0726.49 m/z 13611,292,9074.09 m/z 13822,436,8267.72 m/z 14038,691,43727.52 m/z 142417,106,93754.18 Total Area = 31,577,179100.00

PAGE 110

97E:\0-Spec\0-Data\Seq-5364-0310/04/05 07:12:59 AMJAG-ASP, 1 mg/50 uL; 2.5 uL injected Hydro-RP;0.15;isocratic A:B=100:0/PCM=20ul/m MeOH/ESI RT: 0.00 8.80 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 Time (min) 0 20 40 60 80 100Relative Abundance 0 20 40 60 80 100Relative Abundance 10 20 30 40 50 60 70 80 90 100Relative Abundance 3.61 142.0 3.89 142.0 2.41 186.0 2.71 186.0 0.22 180.7 0.83 180.8 1.89 180.7 1.28 180.6 5.13 141.8 5.29 180.9 6.36 180.6 6.98 245.0 5.74 244.9 8.67 180.5 7.28 241.0 7.52 240.8 8.44 244.9 7.98 244.7 MW 163? RT: 2.41 BP: 186.0 2.71 186.0 2.94 185.9 3.46 186.0 3.74 186.0 4.02 185.9 4.39 185.9 0.07 186.1 0.37 185.9 5.21 186.1 5.43 186.1 6.82 186.0 1.58 185.9 7.05 185.9 5.90 186.0 8.29 186.0 7.44 186.0 0.75 185.9 1.05 186.2 6.13 185.9 2.03 186.0 8.59 185.9 8.13 186.2 Aspartic Acid, MW 141 RT: 3.61 BP: 142.0 3.89 142.0 3.02 141.9 5.13 141.8 2.19 142.0 0.83 141.9 2.79 141.8 0.22 141.7 5.36 141.9 5.74 141.9 1.35 142.0 1.51 142.1 6.28 141.6 7.36 142.0 7.05 141.6 6.59 141.9 7.90 141.6 8.44 141.9 NL: 2.23E5 Base Peak F: + c ESI Full ms [ 50.00-300.00] MS SEQ-5364-03 NL: 1.64E5 m/z= 185.5-186.5 F: + c ESI Full ms [ 50.00-300.00] MS SEQ-5364-03 NL: 2.23E5 m/z= 141.5-142.5 F: + c ESI Full ms [ 50.00-300.00] MS SEQ-5364-03 Figure B-3. HPLC/(+)ESI-MS mass chromatograms of the Asp (m/z 142) and an m/z 186 ion-pe ak which eluted at RT 2.41 min. The spectra of this m/z 186 ion-peak suggests it is du e to the sodiated forms of the Asp (next Figure).

PAGE 111

98 Figure B-4. The m/z 164, 186 ions of the earlier eluting peak are likely the sodiated forms of Asp as indicated. E:\0-Spec\0-Data\Seq-5364-0310/04/05 07:12:59 AMJAG-ASP, 1 mg/50 uL; 2.5 uL injected Hydro-RP;0.15;isocratic A:B=100:0/PCM=20ul/m MeOH/ESI SEQ-5364-03 # 90-103 RT: 2.26-2.56 AV: 5 SB: 71.51-1.86, 2.99-3.12 NL: 1.14E5 F: + c ESI Full ms [ 50.00-300.00] 60 80 100 120 140 160 180 200 220 240 260 280 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 186.0 184.0 243.8 164.0 161.9 265.8 241.9 177.9 208.0 253.8 267.9 160.0 114.2 279.0 241.1 225.7 298.0 201.7 82.9 219.4 SEQ-5364-03 # 89-105 RT: 2.21-2.58 AV: 6 NL: 8.15E3 T: + c d Full ms2 185.99@37.50 [ 40.00-385.00] 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 62.8 168.8 166.8 184.4 186.4 165.8 183.7 139.0 169.8 121.8 175.0 199.4 119.8 104.5NH3 +O18O18O18O18Na H NH3 +O18O18O18O18Na Na m/z 164 m/z 186

PAGE 112

99 LIST OF REFERENCES Aghaiypour, K., Wlodawer, A., Lubkowski, J. (2001) Biochemistry 40 , 5655. Andrulis, I. L., Chen, J., Ray, P. N. (1987) Mol. Cell. Biol. 7 , 2435. Arfin, S. M. (1967) Biochim. Biophys. Acta 136 , 233. Aslanian, A. M., Fletcher, B. S., Kilberg, M. S. (2001) Biochem. J. 357 , 321. Aslanian, A. M., Kilberg, M. S. (2001) Biochem. J. 358 , 59. Bachmann, B. O., Li, R. F., Townsend, C. A. (1998) Proc. Natl. Acad. Sci. USA 95 , 9082. Badet, B., Vermoote, P., Haumont, P. Y., Lederer, F., Legoffic, F. (1987) Biochemistry 26 , 1940. Badet-Denisot, M. A., René, L., Badet, B. (1993) Bull. Soc. Chim. Fr. 130 , 249. Barr, R. D., DeVeber, L. L., Pai, K. M., Andrew, M., Halton, J. (1992) Am. J. Pediatr. Hematol. Oncol. 14 , 136. Bera, A. K., Smith, J. L., Zalkin, H. (2000) J. Biol. Chem. 275 , 7975. Bernt, E., Bergmeyer, H. U. (1974) in Methods in enzymatic analysis (Bergmeyer, H. U., Ed.) pp 1704, Academic Press, New York. Bhatia, U., Robison, K., Gilbert, W. (1997) Science 276 , 1724. Boehlein, S. K., Nakatsu, T., Hiratake, J ., Thirumoorthy, R., Stewart, J. D. (2001) Biochemistry 40 , 11168. Boehlein, S. K., Richards, N. G. J., Schuster, S. M. (1994a) J. Biol. Chem. 269 , 7450. Boehlein, S. K., Richards, N. G. J., Schuster, S. M. (1994b) J. Biol. Chem. 269 , 26789. Boehlein, S. K., Stewart, J. D., Walworth, E. S., Thirumoorthy, R., Richards, N. G. J., Schuster, S. M. (1998) Biochemistry 37 , 13230.

PAGE 113

100 Boehlein, S. K., Walworth, E. S., Richards, N. G. J., Schuster, S. M. (1997) J. Biol. Chem. 272 , 12384. Bradford, M. M. (1976) Anal. Biochem. 72 , 248. Brannigan, J. A., Dodson, G., Duggleby, H. J., Moody, P. C. E., Smith, J. L., Tomchick, D. R., Murzin, A. G. (1995) Nature 378 , 416. Broschat, K. O., Gorka, C., Page, J. D., Martin-Berger, C. L., Davies, M. S., Huang, H. C., Gulve, E. A., Salsgiver, W. J., Kasten, T. P. (2002) J. Biol. Chem. 277 , 14764. Bult, C. J., White, O., Olsen, G. J., Zhou, L. X., Fleischmann, R. D., Sutton, G. G., Blake, J. A., FitzGerald, L. M., Clayton, R. A., Gocayne, J. D., Kerlavage, A. R., Dougherty, B. A., Tomb, J. F., Adams, M. D., Reich, C. I., Overbeek, R., Kirkness, E. F., Weinstock, K. G., Merrick, J. M., Glodek, A., Scott, J. L., Geoghagen, N. S. M., Weidman, J. F., Fuhrmann, J. L., Nguyen, D., Utterback, T. R., Kelley, J. M., Peterson, J. D., Sadow , P. W., Hanna, M. C., Cotton, M. D., Roberts, K. M., Hurst, M. A., Kaine, B. P., Borodovsky, M., Klenk, H. P., Fraser, C. M., Smith, H. O., Woese, C. R., Venter, J. C. (1996) Science 273 , 1058. Cameron, I. R., Possee, R. D., Bishop, D. H. L. (1989) Trends Biotechnol. 7 , 66. Carlow, D. C., Smith, A. A., Yang, C. C ., Short, S. A., Wolfenden, R. (1995) Biochemistry 34 , 4220. Cedar, H., Schwartz, J. H. (1969a) J. Biol. Chem. 244 , 4112. Cedar, H., Schwartz, J. H. (1969b) J. Biol. Chem. 244 , 4122. Cha, S. (1975) Biochem. Pharmacol. 24 , 2177. Chakrabarti, R., Schuster, S. M. (1997) Int. J. Pediatr. Hemotol./Oncol. 4 , 597. Chevalier, C., Bourgeois, E ., Just, D., Raymond, P. (1996) Plant J. 9 , 1. Cleland, W. W. (1982) in CRC Crit. Rev. Biochem. pp 385. Cleland, W. W., Northrop, D. B. (1999) in Method Enzymol. pp 3. Cohn, M., Hu, A. (1980) J. Am. Chem. Soc. 102 , 913. Cooney, D. A., Driscoll, J. S., Milman, H. A., Jayaram, H. N., Davis, R. D. (1976) Cancer Treat. Rep. 60 , 1493. Cooney, D. A., Handschumacher, R. E. (1970) Annu. Rev. Pharmacol. 10 , 421.

PAGE 114

101 Cramer, C. J., Truhlar, D. G. (1999) Chem. Rev. 99 , 2161. Dang, V. D., Valens, M., BolotinFuk uhara, M., DaignanFornier, B. (1996) Mol. Microbiol. 22 , 681. Denessiouk, K. A., Johnson, M. S. (2000) Proteins 38 , 310. Deng, W., Burland, V., Plunkett, G., Boutin, A ., Mayhew, G. F., Liss, P., Perna, N. T., Rose, D. J., Mau, B., Zhou, S. G., Schwar tz, D. C., Fetherston, J. D., Lindler, L. E., Brubaker, R. R., Plano, G. V., Straley, S. C., McDonough, K. A., Nilles, M. L., Matson, J. S., Blattner, F. R., Perry, R. D. (2002) J. Bacteriol. 184 , 4601. Ding, Y. (2002) Experimental and theoretical characte rization of asparagine synthetase inhibitors , Ph.D. Thesis, University of Florida. Ertel, I. J., Nesbit, M. E., Hammond , D., Weiner, J., Sather, H. (1979) Cancer Res. 39 , 3893. Farrell, P. J., Lu, M.L., Prevost, J., Br own, C., Behie, L., Iatrou, K. (1998) Biotechnol. Bioeng. 60 , 656. Fersht, A. (1999) Structure and Mechanism in Protein Science , W.H. Freeman and Company, New York. Fine, B. M., Kaspers, G. J., Ho, M., Loonen, A. H., Boxer, L. M. (2005) Cancer Res. 65 , 291. Fresquet, V., Thoden, J. B., Holden, H. M., Raushel, F. M. (2004) Bioorg. Chem. 32 , 63. Frillingos, S., Kaback, H. R. (1996) Biochemistry 35 , 13363. Gong, S. S., Basilico, C. (1990) Nucleic Acids Res. 18 , 3509. Gonnet, G. H., Benner, S. A. (1991) Computational Biochemistry Research at ETH pp 118, Eidgenoessische Technische Hochschule, Zurich. Harpel, M. R., Hartman, F. C. (1994) Biochemistry 33 , 5553. Haskell, C. M., Canellos, G. P. (1969) Biochem. Pharmacol. 18 , 2578. Hongo, S., Sato, T. (1985) Arch. Biochem. Biophys. 238 , 410. Hongo, S., Sato, T. (1983) Biochem. Biophys. Acta 742 , 484. Horowitz, B., Meister, A. (1972) J. Biol. Chem. 247, 6708.

PAGE 115

102 Huang, X. Y., Holden, H. M., Raushel, F. M. (2001) Annu. Rev. Biochem. 70 , 149. Huang, Y.-Z., Knox, E. W. (1975) Enzyme 19 , 314. Hutson, R. G., Kitoh, T., Amador, D. A. M., Cosic, S., Schuster, S. M., Kilberg, M. S. (1997) Am. J. Physiol. Cell Physiol. 272 , C1691. Ishiyama, M., Tominaga, H., Shiga, M., Sasamoto, K., Ohkura, Y., Ueno, K. (1996) Biol. Pharm. Bull. 19 , 1518. Jayaram, H. N., Cooney, D. A. (1979) Cancer Treat. Rep. 63 , 1095. Jayaram, H. N., Cooney, D. A., Milman, H. A., Homan, E. R., Rosenbluth, R. J. (1976) Biochem. Pharmacol. 25 , 1571. Kiriyama, Y., Kubota, M., Takimoto, T., Kitoh, J., Tanizawa, A. (1989) Leukemia 3 , 294. Klamt, A. (1995) J. Phys. Chem. 99 , 2224. Klinman, J. P. (1978) in Adv. Enzymol. Relat. Areas Mol. Biol. pp 415. Koizumi, M., Hiratake, J., Nakatsu, T., Kato, H., Oda, J. (1999) J. Am. Chem. Soc. 121 , 5799. Koroniak, L., Ciustea, M., Gutierrez, J. A., Richards, N. G. J. (2003) Org. Lett. 5 , 2033. Lam, H. M., Peng, S. S. Y., Coruzzi, G. M. (1994) Plant Physiol. 106 , 1347. Larsen, T. M., Boehlein, S. K., Schuster, S. M., Richards, N. G. J., Thoden, J. B. (1999) Biochemistry 38 , 16146. Lenhard, T., Reilander, H. (1997) Biochem. Biophys. Res. Commun. 238 , 823. Lerman, C. L., Cohn, M. (1980) J. Biol. Chem. 255 , 8756. Liaw, S. H., Eisenberg, D. (1994) Biochemistry 33 , 675. Luehr, C. A., Schuster, S. M. (1985) Arch. Biochem. Biophys. 237 , 335. Markin, R. S., Luehr, C. A., Schuster, S. M. (1981) Biochemistry 20 , 7226. McClelland, M., Sanderson, K. E., Spieth, J., C lifton, S. W., Latreill e, P., Courtney, L., Porwollik, S., Ali, J., Dante, M., Du, F. Y., Hou, S. F., Layman, D., Leonard, S., Nguyen, C., Scott, K., Holmes, A., Grewal , N., Mulvaney, E., Ryan, E., Sun, H., Florea, L., Miller, W., Stoneking, T., Nh an, M., Waterston, R., Wilson, R. K. (2001) Nature 413 , 852.

PAGE 116

103 McNaughton, H. J., Thirkettle, J. E., Zhang, Z. H., Schofield, C. J., Jensen, S. E., Barton, B., Greaves, P. (1998) Chem. Commun. , 2325. Mehlhaff, P., Luehr, C. A., Schuster, S. M. (1985) Biochemistry 24 , 1104. Michaelis, L., Menten, M.L. (1913) Biochem. Z. 49 , 333. Miller, M. T., Bachmann, B. O., Townsend, C. A., Rosenzweig, A. C. (2001) Nat. Struct. Biol. 8 , 684. Miller, M. T., Bachmann, B. O., Townsend, C. A., Rosenzweig, A. C. (2002) Proc. Natl. Acad. Sci. USA 99 , 14752. Milman, H. A., Cooney, D. A. (1979) Biochem. J. 181 , 51. Milman, H. A., Cooney, D. A., Huang, C. Y. (1980) J. Biol. Chem. 255 , 1862. Morrison, J. F., Walsh, C. T. (1988) Adv. Enzymol. Relat. Areas Mol. Biol. 61 , 201. Morton, R. C., Gerber, G. E. (1988) Anal. Biochem. 170 , 220. Nakatsu, T., Kato, H., Oda, J. (1998) Nat. Struct. Biol. 5 , 15. Newmyer, S. L., deMontellano, P. R. O. (1996) J. Biol. Chem. 271 , 14891. O'Brien, W. E. (1976) Anal. Biochem. 76 , 423. O'Leary, M. H. (1980) in Method Enzymol. pp 83. O'Leary, M. H. (1989) in Annu. Rev. Biochem. pp 377. O'Leary, M. H., Marlier, J. F. (1979) J. Am. Chem. Soc. 101 , 3300. Orth, D. L. (2001) J. Chem. Educ. 78 , 791. Parmentier, L. E., Oleary, M. H., Sch achman, H. K., Cleland, W. W. (1992a) Biochemistry 31 , 6570. Parmentier, L. E., Weiss, P. M., Oleary, M. H., Schachman, H. K., Cleland, W. W. (1992b) Biochemistry 31 , 6577. Patterson, M. K., Orr, G. (1967) Biochem. Biophys. Res. Commun. 26 , 228. Pieters, R., Klumper, E., Kaspers, G. J. L., Veerman, A. J. P. (1997) Crit. Rev. Oncol. Hematol. 25 , 11.

PAGE 117

104 Porath, J., Carlsson, J., Olss on, I., Belfrage, G. (1975) Nature 258 , 598. Ramos, F., Wiame, J. M. (1979) Eur. J. Biochem. 94 , 409. Raushel, F., Thoden, J. B., Holden, H. M. (2003) Acc. Chem. Res. 36 , 539. Ravel, J. M., Shive, W., Norton, S. J., Humphreys, J. S. (1962) J. Biol. Chem. 237 , 2845. Reitzer, L. J., Magasanik, B. (1982) J. Bacteriol. 151 , 1299. Richards, N. G. J., Kilberg, M. S. (2006) Annu. Rev. Biochem. 75 , 629. Richards, N. G. J., Schuster, S. M. (1998) Adv. Enzymol. Relat. Areas Mol. Biol. 72 , 145. Rognes, S. E. (1975) Phytochemistry 14 , 1975. Sanghani, P. C., Moran, R.G. (2000) Prot. Exp. Purif. 18 , 36. Schnizer, H. G., Boehlein, S. K., Stewart, J. D ., Richards, N. G. J., Schuster, S. M. (1999) Biochemistry 38 , 3677. Schnizer, H. G., Boehlein, S. K., Stewart, J. D ., Richards, N. G. J., Schuster, S. M. (2002) in Method Enzymol. pp 260. Scofield, M. A., Lewis, W. S., Schuster, S. M. (1990) J. Biol. Chem. 265 , 12895. Sheng, S., Moraga-Amador, D. A., Van Heeke, G., Allison, R. D., Richards, N. G. J. (1993) J. Biol. Chem. 268 , 16771. Sheng, S. J., Moraga, D. A., Vanheeke, G., Schuster, S. M. (1992) Prot. Exp. Purif. 3 , 337. Srivasta, B. I., Minowada, J. (1973) Biochem. Biophys. Res. Commun. 51 , 529. Stams, W. A., den Boer, M. L., Holleman, A., Appel, I. M., Beverloo, H. B. (2005) Blood 11 , 4223. Stewart, J. J. P. (1989) J. Comput. Chem. 10 , 209. Sulkowski, E. (1985) Trends Biotechnol. 3 , 1. Sutow, W. W., Garcia, F., Starling, K. A., Williams, T. E., Lane, D. M. (1971) Cancer 28 , 819.

PAGE 118

105 Tesson, A. R., Soper, T. S., Ciust ea, M., Richards, N. G. J. (2003) Arch. Biochem. Biophys. 413 , 23. Thompson, J. S., Edmonds, O. P. (1980) Ann. Occup. Hyg. 23 , 27. Toney, M. D., Kirsch, J. F. (1989) Science 243 , 1485. Tso, J. Y., Hermodson, M. A., Zalkin, H. (1982a) J. Biol. Chem. 257 , 3532. Tso, J. Y., Zalkin, H., Vancleemput, M., Yanofsky, C., Smith, J. M. (1982b) J. Biol. Chem. 257 , 3525. Vanheeke, G., Schuster, S. M. (1989a) J. Biol. Chem. 264 , 19475. Vanheeke, G., Schuster, S. M. (1989b) J. Biol. Chem. 264 , 5503. Vanheeke, G., Schuster, S. M. (1990) Protein Eng. 3 , 739. Vanoni, M. A., Curti, B. (1999) Cell. Mol. Life Sci. 55 , 617. Vanwazer, J. R., Callis, C. F., Shoolery, J. N., Jones, R. C. (1956) J. Am. Chem. Soc. 78 , 5715. Vaughn, J. L., Goodwin, R.H., Tom pkins, G.J., McCawley, P. (1977) In Vitro 13 , 213. Waldrop, G. L., Braxton, B. F., Urbauer, J. L., Cleland, W. W., Kiick, D. M. (1994) Biochemistry 33 , 5262. Waterhouse, R. N., Smyth, A. J., Massonneau, A., Prosser, I. M., Cl arkson, D. T. (1996) Plant Mol. Biol. 30 , 883. Yamagata, H., Nakajima, A., Bowler, C., Iwasaki, T. (1998) Biosci., Biotechnol., Biochem. 62 , 148. Zalkin, H. (1993) Adv. Enzymol. Relat. Areas Mol. Biol. 66 , 203. Zalkin, H., Smith, J. L. (1998) Adv. Enzymol. Relat. Areas Mol. Biol. 72 , 87.

PAGE 119

106 BIOGRAPHICAL SKETCH Jemy A. Gutierrez was born in Cavite, Philippines, on January 15, 1977. She holds a Bachelor of Science degree in chemistry fr om the University of the Philippines in Diliman, Quezon City, where she studied amplification and sequencing of a mitochondrial DNA segment from golden snail under the direction of Dr. Virginia Monje. From 1998 to 2001, Jemy was a research associ ate at the Marine Sc ience Institute of the University of the Philippines, worki ng on the physico-chemical characterization of carrageenan from certain species of seaw eed. In August 2001, she started graduate studies at the Chemistry Depart ment of the University of Florida, where she joined the research group of Dr. Nigel G. J. Rich ards to work on enzy mology of asparagine synthetase. She will continue to do resear ch on transition state analogs as a postdoc under Dr. Vern Schramm, at Albert Einste in College of Medicine in New York.