Citation
N-acetylglutathione mono(glycyl)ethyl ester

Material Information

Title:
N-acetylglutathione mono(glycyl)ethyl ester synthesis, characterization and transport into a human cancer cell line
Creator:
Xu, Mei
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
62 leaves : ; 28 cm

Subjects

Subjects / Keywords:
N-acetylglutathione mono(glycyl)ethyl ester ( lcsh )
NAGME ( lcsh )
Synthesis ( lcsh )
Transportation ( lcsh )
Cisplatin -- Toxicology ( lcsh )
Cancer -- Treatment ( lcsh )
Antineoplastic agents -- Toxicology ( lcsh )
Glutathione -- Physiological transport ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 60-62).
Thesis:
Department of Chemistry
Statement of Responsibility:
by Mei Xu.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
40326534 ( OCLC )
ocm40326534
Classification:
LD1190.L46 1998m .X8 ( lcc )

Downloads

This item has the following downloads:


Full Text
N-Acetylglutathione Mono(glycyl)ethyl Ester:
Synthesis, Characterization and Transport into A Human Cancer Cell Line
by
MeiXu
B. S., Sichuan University, 1986
M.S., Southwest Agriculture University, 1989
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Chemistry
1998


This thesis for the Master of Science
degree by
Mei Xu
has been approved
by
SjLjqt
Date
Teresa Audesirk


Xu, Mei (M.S., Chemistry)
N-Acetylglutathione Mono(glycyl)ethyl Ester: Synthesis, Characterization and
Transport into A Human Cancer Cell Line
Thesis directed by Assistant Professor Ellen J. Levy
ABSTRACT
In this thesis N-acetylglutathione mono(glycyl)ethyl ester (NAGME) was
synthesized and characterized by DTNB, HPLC and NMR. In addition, the uptake of
this compound in J82 (human bladder carcinoma) cells was explored.
Glutathione Mono(glycyl)ethyl Ester Hemihydrosulfate (GEET/2H2S04) was
first synthesized through the acid-catalyzed esterification of GSH. The sulfate in this
product was removed by an anion exchanger DE 53 to produce free base GEE. The
overall percent yield for free base GEE synthesis was 42-50%. NAGME was then
synthesized through the acetylation of free base GEE. The percent yield was 56-57%.
The DTNB assay verified that the free thiol content of NAGME was 84-89.5%. The
HPLC spectrum of NAGME showed that there was no other thiol compounds
present. The H1 NMR spectrum of NAGME indicated that there was little diethyl
ether present.
In NAGME cell uptake experiment, GSH level was not elevated in both
GSH-treated and NAGME-treated cells. Besides GSH no other thiol was detected in
m


NAGME-treated cells. These results indicated that J82 cells did not take up
NAGME. NAGME could not increase the level of GSH in J82 cells under the
experimental condition.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed
IV


ACKNOWLEDGMENTS
I wish to express my sincere thanks to Dr. Levy for her advice, support,
understanding, and patience throughout this project.
I would like to also thank my committee members, Dr. Dyckes and Dr.
Audesirk, for their guidance and support during the completion of my degree.
Finally, I would like to thank all those who contributed in any way by their
moral support, patience, and kind words, especially Dr. Kimbrough, Dr. Zapien, Dr.
Damrauer, Dr. Anderson, Marc-Andrew Donsky and all of my wonderful friends.


DEDICATION
I dedicate this thesis to my husband William Tran, my parents Yuxun Xu,
Hongmei Ou and my parents-in-law Anh V. Tran, Muoi Lam for their continual love,
support and encouragement.


CONTENTS
Figures.....................................................................x
Tables.....................................................................xi
Abbreviations.............................................................xii
Chapter
1. Introduction.............................................................1
1.1 Cisplatin...............................................................1
1.2 Glutathione............................................................3
1.3 Interaction between Glutathione and Cisplatin..........................4
1.4 Glutathione Derivatives................................................6
2. Objective and Hypothesis................................................8
3. Materials..............................................................11
4. Instrumentation.........................................................12
5. Methods.................................................................13
5.1 Synthesis Procedure....................................................13
5.1.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester Hemihydrosulfate.....13
5.1.2 Synthesis of Free Base Glutathione Mono(glycyl)ethyl Ester...........14
5.1.3 Synthesis of N-acetylglutathione Mono(glycyl)ethyl Ester.............15
5.2 Analysis for the Synthesis.............................................16
vii


5.2.1 Derivatization for High-performance Liquid Chromatography (HPLC)
16
5.2.2 HPLC Analysis of the Thiols..........................................17
5.2.3 DTNB Assay for Free -SH (Thiol) in Unknown Compounds.................17
5.2.4 NMR Analysis for N-acetylglutathione Mono(glycyl)ethyl Ester.........19
5.2.5 BaCl2 Test...........................................................19
5.2.6 DTNB Test............................................................19
5.3 Cell Uptake Experiment................................................20
5.4 Analysis for Cell Uptake Experiment...................................22
5.4.1 Derivatization for HPLC..............................................22
5.4.2 Cell Counting........................................................23
5.4.3 Calculations.........................................................23
6. Results and Discussion..................................................25
6.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester......................24
6.1.1 Reaction Mechanism...................................................24
6.1.2 The Percent Yield....................................................29
6.1.3 HPLC Analysis........................................................30
6.1.4 BaCl2 Test...........................................................36
6.1.5 DTNB Assay...........................................................36
6.2 Synthesis of N-acetylglutathione Mono(glycyl)ethyl Ester..............39
Vlll
6.2.1 Reaction Mechanism
39


6.2.2 The Percent Yield
40
6.2.3 HPLC Analysis.........................................................41
6.2.4 DTNB Assay............................................................46
6.2.5 NMR Analysis..........................................................46
6.3 Cell Uptake Experiment..................................................51
6.3.1 HPLC Analysis for the Media...........................................51
6.3.2 HPLC Analysis for Cell Lysates........................................52
7. Conclusion and Further Study.............................................54
8. Failures.................................................................55
8.1 Synthesis ofNAGME ffomNAGDI.............................................55
8.2 Preparation of GEE Free Base with AG1X2 (Bicarbonate Form) Resin......56
8.3 Direct Preparation of GEE Free Base....................................56
8.4 Reduction ofNAGME with DTT.............................................57
References..................................................................60
IX


FIGURES
Figure
1.1 The Structure of Cisplatin [Cis-diamminedichloroplatinum (II)].........1
1.2 The Structure of Glutathione (y-glutamylcysteinylglycine)..............3
6.1 The Mechanism of the Acid-catalyzed Esterification of GSH............28
6.2 The Reaction of A Thiol with mBBr....................................31
6.3 The HPLC Spectrum for Synthesis of GEET/2H2S04.......................33
6.4 The HPLC Spectrum for GEET/2H2S04 Product............................34
6.5 The HPLC Spectrum for Free Base GEE Product..........................35
6.6 The Structure of 5,5-Dithiobis(2-nitro)benzoic Acid (DTNB).........37
6.7 The Mechanism of the Reaction between DTNB and A Free Thiol..........37
6.8 The Mechanism of the Acetylation of GEE..............................39
6.9 The HPLC Spectrum for Synthesis of NAGME.............................43
6.10 The HPLC Spectrum for Mixture of NAGME Reaction Solution and GEE
Standard Solution....................................................44
6.11 The HPLC Spectrum for NAGME Product.................................45
6.12 H1 NMR Spectrum at 200 MHz of NAGME Product in D20.................49
6.13 The Structure of NAGME..............................................50


TABLES
Table
6.1 NMR Data for NAGME Product...........................................47
6.2 The Levels of Thiols in Medium Control Wells..........................51
6.3 The Levels of External Thiols in Experimental Wells...................51
6.4 The Levels of Cellular Thiols in Experimental Wells...................52
XI


ABBREVIATIONS
Abbreviation Name
a-MEM a-Minimal Eagle Medium with GlutaMax
ATP Adenosine Triphosphate
BSO Buthionine Sulfoximine
DIGEE Glutathione Diethyl Ester
DTNB 5, 5-Dithiobis(2-nitro)benzoic Acid
DTPA Diethylenetriamine-pentaacetic Acid
DTT Dithiothreitol
GEE Glutathione Mono(glycyl)ethyl Ester
GSH Glutathione (y-glutamylcysteinylglycine)
GS-X Glutathione S-conjugate Export
HPLC High-performance Liquid Chromatography
rnBBr Monobromobimane
NAG N-acetylglutathione
NAGDI N-acetylglutathione Diethyl Ester
NAGME N-acetylglutathione Mono(glycyl)ethyl Ester
NMR Nuclear Magnetic Resonance
PBS Dulbeccos Phosphate-buffered Saline
Xll


SSA
TMSPA
TNB
Tris
5-Sulfosalicylic Acid
3-(Trimethylsilyl)propionic-2,2,3,3-d4 Acid, Sodium Salt
Thionitrobenzoate
Tris(hydroxymethyl) Aminomethane
Xlll


1. Introduction
1.1 Cisplatin
Cis-diamminedichloroplatinum (cisplatin), an inorganic compound (Figure
1.1), was first synthesized in 1845. Its biological activity was discovered
serendipitously in 1965 when Barnett Rosenberg investigated the effect of an electric
current on Escherichia coli (1). The cisplatin, which was synthesized in the electric
field, was found to induce filamentous growth in E. coli and also to be active against
implanted tumors in mice. Now cisplatin is one of the largest selling cancer
chemotherapeutic agents and has been used successfully in treating many
malignancies, including testicular, ovarian, head and neck, bladder, esophageal and
small cell lung cancers (2).
Figure 1.1 The Structure of Cisplatin [Cis-diamminedichloroplatinum (II)]
The cytotoxicity of cisplatin is believed to be due to the formation of DNA
adducts, which include DNA-protein cross-links, DNA monoadducts, and interstrand
and intrastrand DNA cross-links. The intrastrand cisplatin cross-link produces a
severe local distortion in the DNA double helix, leading to unwinding and kinking
which eventually causes cancer cell death (3). Besides DNA, cisplatin also has
Pt
1


affinity for sulfur in methionine and cysteine because of its platinum center. Thus all
peptides and proteins containing methionine or cysteine are potential ligands of
cisplatin in biological systems. The ability of cisplatin to bind to some these species,
thus potentially affecting enzyme activity or protein conformation, may also play a
role in systemic toxicity (4).
Although cisplatin plays an important role in cancer therapy, unfortunately, it
leads to severe multiorgan toxicities, such as nephrotoxicity (5). Additionally, after
an initially favorable response, there can be rapid growth of a cisplatin-resistant cell
population (6). The acquired resistance is a serious barrier to chemotherapeutic
success. Numerous potential mechanisms of resistance to cisplatin have been
elucidated. They include decreased drug accumulation, increased levels of
intracellular thiols, increased DNA repair, and overexpression of metallothioneins
(7). Actually there has been a continuous search for biological and pharmacological
strategies to prevent cisplatin-induced nephrotoxicity and drug-resistance. These
strategies include modification of administration modes to dilute cisplatin in the
tubule and decrease systemic peak concentrations, the use of antidotes to antagonize
the activity of cisplatin, and development of new galenic forms (a pharmaceutical
preparation of a drug of animal or plant origin) (8).
2


1.2 Glutathione
R1NH-CH-CH2-CH2-CO-NH-CH-CO-NH-CH2-COOR3
coor2 ch2sh
Figure 1.2 The Structure of Glutathione (y-glutamylcysteinylglycine) Ri=R2=R3=H
Glutathione (GSH, Figure 1.2) is a tripeptide thiol, y-
glutamylcysteinylglycine. GSH is not required in the diets of animals. It is
synthesized in many types of cells from glutamate (Glu), cysteine (Cys), and glycine
(Gly), which are formed as products of cell metabolism and are also obtained from
the diet. GSH is synthesized in a two-step pathway [(1) Glu + Cys - y-Glu-Cys; (2)
y-Glu-Cys + Gly -> y-Glu-Cys-Gly] involving the ATP-dependent enzymes y-
glutamylcysteine synthetase and glutathione synthetase, respectively. The first step is
rate-limiting and inhibited by glutathione itself and by buthionine sulfoximine
(BSO), a transition-state analog (9).
Glutathione has a variety of physiologically important functions, for
example: (a) maintenance of the thiols of proteins (and other compounds) and of
antioxidants, (b) reduction of ribonucleotides to form the deoxyribonucleotide
precursors of DNA, (c) protection against oxidative damage, free radical damage,
and other types of toxicity, and (d) synthesis and transport of biologically active,
endogenous substances, such as cysteine moieties (9).
3


Under normal steady-state conditions, the majority of GSH exists in the
reduced form (0.5 to 10 mM) in mammalian cells. NADPH-dependent reduction of
glutathione disulfide (GSSH) to GSH produces a GSH:GSSH ratio of over 100:1,
which permits GSH to function as an intracellular reducing agent (10). In situations
in which GSH is consumed by intracellular reactions, for example in detoxification
reactions, therapeutic delivery of GSH to the cell might be sufficiently rapid to make
up for the constant utilization of GSH and enhance the function of GSH. It was
reported that elevation of GSH levels by addition of GSH monoethyl ester
effectively protected human lymphoid cells against the lethal effects of irradiation,
while depletion of GSH increased cell sensitivity to radiation under hypoxic
conditions (9). This suggests that manipulation of GSH metabolism may be a
rewarding therapeutic strategy. Since definite conclusions about normal,
maximal, and minimal cellular GSH levels cannot yet be drawn, there must also
be concern about the possibility that very high levels would lead to toxic effects. For
example, high thiol levels in cell culture media may lead to toxicity due to peroxide
formation (9).
1.3 Interaction between Glutathione and Cisplatin
Glutathione interacts with a wide variety of drugs that include cisplatin. It is
reported (11) that in human leukemia HL-60 cells, each GSH molecule acts as a
bidentate chelating ligand, coordinating to platinum via cysteinyl sulfur and nitrogen
atoms. After a GSH- platinum complex is formed in a 2:1 molar ratio, the complex is
4


eliminated from the cell by an ATP-dependent glutathione S-conjugate export pump
(GS-X pump). Although the structure of GSH-cisplatin complexes needs to be
further explored, it has been reported by many sources that GSH plays a role in the
resistance of normal and tumor cells to cisplatin toxicity. Elevated levels of cellular
GSH were observed in human ovarian cancer cell lines that have high resistance to
cisplatin (12). Similarly, depleting GSH levels can enhance the toxicity of cisplatin
in a lung cancer cell line (13).
The exact mechanism by which GSH influences cisplatin cytotoxicity is not
yet completely clear. There are several ways in which glutathione could affect
cisplatin toxicity, for example, by preventing it from reacting with DNA, by
quenching monofunctional adducts with DNA before they rearrange to the more
toxic bifunctional lesions, or by increasing the rate of cellular elimination of cisplatin
by direct binding and efflux through a GS-X pump. The possible roles and means of
glutathione in modifying the biological activity of cisplatin, and thus its roles in
cisplatin resistance and toxicity, are clearly varied and complex.
Elevated intracellular GSH is a component of the cisplatin-resistant
phenotypes, but the significance of elevated glutathione in cisplatin resistance is
unclear. If elevated glutathione is responsible for cisplatin resistance, then
experimentally lowering the glutathione concentration should reverse resistance.
Buthionine sufoximine (BSO) is the agent of choice for reducing glutathione
concentration, because it is a potent and specific inhibitor of glutathione synthesis,
5


specifically at the y-glutamylcysteine synthetase step. It has been shown that GSH
depletion with BSO can sensitize some cisplatin-resistant cell lines to the drug by
increasing both the amount of Pt bound to DNA and the Pt-GpG adduct (12, 14). In
other studies, little or no sensitization was observed in other cisplatin-resistant cell
lines after BSO treatment (15, 16). This suggests that there are other cellular
mechanisms of resistance. Andrews et al. demonstrated that only after a prolonged
decrease in the glutathione levels did human ovarian carcinoma cells become
sensitized to cisplatin or more sensitized to cisplatin, and this was true for either
cisplatin-resistant or cisplatin-sensitive cells (17).
The intracellular concentration of glutathione is generally in the range of 0.5-
10 mM. At 1 mM, only a 20% reduction in the platination of DNA was observed in
an in vitro incubation (18). At 10 mM, a 70% reduction in platination of DNA was
observed. Hence, if the major role of glutathione is to reduce platination of DNA, it
might be possible to modulate toxicity at the higher ranges of glutathione
concentrations, but not at intracellular concentrations of less than 1 mM. This may
explain the ambiguity in the results obtained after experimentally reducing
glutathione concentration, and this suggests that clinical alteration of cellular
glutathione levels would be one of means of influencing cisplatin activity.
1.4 Glutathione Derivatives
Because elevated glutathione has been shown to be a component of drug-
resistant phenotypes in a number of cisplatin-resistant cells, manipulation of cellular
6


glutathione levels is probably one of means of influencing cisplatin activity. But
glutathione is not effectively transported into most types of cells under a variety of
experimental conditions (9). So administration of glutathione does not result in
substantial cellular elevation. Fortunately, it has been shown that simple organic
derivatives of GSH can be transported into cells more effectively than GSH itself and
can be converted to GSH after transport into the cells and increase GSH levels. Two
of these derivatives are glutathione mono(glycyl)ethyl ester (GEE, R3=CH2CH3 in
figure 1.2) and glutathione diethyl ester (DIGEE, CH2CH3 in figure 1.2).
They can be transported into cells and intracellularly hydrolyzed to yield GSH and
also result in large increase in total cellular thiols (including cysteine and the esters
themselves) (19, 20). On the other hand, whether N-acetylglutathione derivatives are
effectively transported into cells and yield similar increases in GSH has been only
minimally explored. Studies on suspensions of human erythrocytes have shown
evidence for uptake of N-acetyl-GSH diethyl ester (NAGDI, Ri=CH3CO,
R2=R3=CH2CH3 in figure 1.2) and for partial deesterification of this compound
intracellularly; no cleavage of the N-acetyl moiety was found. Initial studies on
hamsters indicate that administration of N-acetyl-GSH diethyl ester leads to some
increase in the level of GSH in the liver and kidney of buthionine sulfoximine-
treated animals (21). There is no report about the influence of the transport of N-
acetyl-GSH monoester (NAGME, Ri=CH3CO, Rr=H, R3=CH2CH3 in figure 1.2) into
cells on GSH levels in cells.
7


2. Objective and Hypothesis
The mono (glycyl) ethyl ester of glutathione has been shown to offer
protection against cisplatin toxicity to both cells and mice (22, 23). The effect of the
diethyl ester on cisplatin activity is not known. However, the diester has been shown
to protect K-562 cells (a kind of cancer cell line) against the cytotoxicity of the
potential cancer drug, 4-hydroperoxycyclophosphamide (4-HC) (24). The ability of
glutathione esters to elevate cellular glutathione levels is well-established (25). The
ability of N-acetyl-GSH monoester to influence cellular GSH levels is unexplored.
The objectives of this study are: (1) to synthesize N-acetylglutathione
mono(glycyl)ethyl ester (NAGME, Ri=CH3CO and R.3=CH2CH3 in figure 1.2) using
a modification of the literature method and then characterize the product by NMR,
HPLC, and the reaction with DTNB; (2) to investigate uptake of NAGME into a
human cancer cell line to determine the influence of NAGME on the level of GSH
and other thiols in cells. If NAGME could be transported into cells, the effect of
NAGME on the cytotoxicity of cisplatin would be explored in further study. This
study will therefore ultimately contribute to safer and more effective cancer
chemotherapy.
Although NAGME like GSH has a net charge of -1 at physiological pH,
acetyl and ethyl groups in NAGME may make it different from GSH in
hydrophobicity. The hypothesis for NAGME cell uptake was that NAGME could be
8


transported into cells because it was more hydrophobic than GSH. After transported
into the cells NAGME might be hydrolized to yield NAG or deacetylized to yield
GEE or GSH and then increase the levels of intracellular thiols.
Cislpatin is toxic to bladder cancer cells (2). In this study J82 cells (human
bladder carcinoma) have therefore been chosen for cell uptake experiment. If
NAGME could be taken up by J82 cells, it might then be like GSH, in which it could
protect cancer cells against cisplatin toxicity. Although protection of cancer cells
against cisplatin toxicity by elevation of thiol level is obviously not a goal, the
selection of fast-growing cancer cells over normal cells is helpful for the initial study
on NAGME uptake.
In addition to the elevation of cellular glutathione, NAGME as a thiol may
offer unique mechanisms for influencing the biochemistry of cisplatin. Four of these
mechanisms, specific for cisplatin, might be: (a) direct binding of NAGME to
cisplatin extracellularly, preventing transport of cisplatin across cell membranes.
This mechanism may decrease the toxicity of cisplatin; (b) direct binding of
NAGME to cisplatin intracellularly, yielding complexes which may interact with
DNA or proteins in a manner different from that of glutathione-cisplatin complexes;
(c) competition with GSH for intracellular cisplatin complexes, potentially
decreasing the amount of GSH-platinum and then decreasing cisplatin efflux through
GSH-specific pumps; (d) competition of NAGME-platinum complexes with GSH-
platinum complexes for the GSH-specific pumps, potentially weakening the ability
9


of GSH-specific pumps and then decreasing the amount of cisplatin efflux. The latter
three mechanisms could potentially enhance toxicity. The exploration of cisplatin
interactions with glutathione derivatives and the effects of such interactions on
cellular cisplatin activity will yield information critical to designing adjunct therapies
for cisplatin cancer treatment. The ultimate purpose is to make chemotherapy safer
and more effective, by lessening toxicity of cisplatin toward healthy cells and by
overcoming the resistance of cancer cells to cisplatin.
10


3. Materials
Glutathione (98-100% reduced form) was from Sigma. Pre-swollen
microgranular anion exchanger DE 53 [Diethylaminoethyl cellulose, cellulose-O-
(CH2)2N(C2H5)2] was from Whatman. AG1X2 resin (chloride form, 200-400 mesh)
was from Bio-Rad Laboratories. 5, 5-Dithiobis(2-nitro)benzoic acid (DTNB), 3-
(Trimethylsilyl)propionic2,2,3,3-d4 acid sodium salt (TMSPA) and D2O (99.9 atom
% D) were from Aldrich Chem. CO. Monobromobimane (mBBR) was from
Molecular Probes. Absolute ethanol was from AAPER Alcohol and Chemical CO.
Methanol (HPLC grade), Acetic anhydride and Sulfuric acid (trace metal grade, min
95.0%-max 98.0%) were from FisherChemical. Dulbeccos phosphate-buffered
saline (PBS), a-Minimal Eagle Medium with GlutaMax (a-MEM), Trypsin (0.25%),
Trypan Blue Stain (0.4%), antibiotic/antimycotic mixture and Fetal Bovine Serum
were from Gibco/BRL. All other reagents were the highest grade available.
11


4. Instrumentation
The UV-Visible spectrophotometer (cary IE) was from Varian Australia Pty.
Ltd. The HPLC system consisted of a Beckman Ultrasphere 0.46x25 cm Cl8
reverse-phase column, Waters 501 HPLC pump (Millipore), and a Waters U6K
injector (Millipore). The detector was a model 121 fluorometer (Gilson). The
Nuclear Magnetic Resonance spectrometer (200 MHz) was from Varian.
12


5. Methods
5.1 Synthesis Procedure
5.1.1 Synthesis of Glutathione Mono(glycyI)ethyl Ester Hemihydrosulfate
Esterification conditions were essentially based on those described by Levy
et al. (26). Reduced GSH (lOg, 32.5 mmol) was added to an ice-cold solution of 100
mL absolute ethanol containing 2.74 mL concentrated H2SO4 in a 1-liter flask The
reaction was started in an ice bath and stirred for 18 to 22 hr while warming to room
temperature. Progress of the esterification was monitored by HPLC (see below).
After 90% of GSH was gone, and over 80% of the glutathione monoester and 10% of
diester appeared, diethyl ether was slowly added to the flask with swirling. When a
white product formed, diethyl ether was poured up to the top of the flask. The flask
was placed in an ice bath overnight. The supernatant was then decanted, and the
syrup on the bottom of flask was dried under a flow of N2 to remove the residual
ether. The syrup was dried overnight under vacuum over KOH. The fluffy dry white
product (about 13 g) was dissolved in the minimal amount of warm Milli Q water
(35-40 C, about 3 mL), and absolute ethanol (75 mL) was slowly added with
swirling. This solution deposited crystalline product after standing in the -20 C
freezer overnight. The white crystals were filtered on a sintered glass funnel and
washed with ice-cold absolute ethanol (3 x 80 mL) and diethyl ether (5 x 80 mL).
The product was transferred to a large watchglass and air-dried in the hood to
13


remove the last traces of ether. The product was weighed, analyzed by HPLC and
DTNB, and then stored in a desiccator. The synthesis yielded 7.2-7.9 g product. The
percent yield was 58%-63%.
5.1.2 Synthesis of Free Base Glutathione Mono(glycyI)ethyl Ester
The process of removing acid from GEET/2H2SO4 was carried out in a cold
box. 60 g of DE 53 was washed with 5 x 100 mL Milli Q water and divided into
portions of about 5 g. 4 g of GEET/2H2SO4 was dissolved in 60 mL cold water. This
solution was tested by 2% BaCl2 and 6 mM DTNB (in 100 mM phosphate buffer at
pH 7.4) to check for the presence of SO42' and -SH (see 5.2.5 BaCl2 test and 5.2.6
DNTB test). The pH of this solution was checked by spotting 5 p,L of the solution on
a pH stick (pH range: 4.5-10). The BaCl2 test and the DTNB test for this initial
solution were strong positive and weak positive respectively. The pH of this solution
was <4.5. Two portions of clean DE53 were then added to the solution, and the
mixture was stirred in a cold box to keep the resin suspended. After 5 min, stirring
was stopped, and 0.5 mL of the supernatant was removed and centrifuged for 3 min
at 15,000 rpm. The resulting clear supernatant was then tested by 2% BaCl2, 6mM
DTNB and a pH stick. If BaCl2 test was positive (white precipitate appeared),
another portion of DE53 was added to the solution to remove sulfate. This procedure
was repeated until two consecutive negative BaCL tests were obtained (usually 6-7
portions were needed). During the DE 53 treatment all DTNB tests were positive and
the pH of the solution was changed from <4.5 to 6.0. After DE 53 treatment was
14


done, the DE 53 was immediately removed from the sulfate-free solution by
filtration. The filtrate was tested with 2% BaCk, 6 mM DTNB and a pH stick. The
resin was washed with 25 mL cold water, and then filtered. The filtrate was also
tested with 2% BaCl2, 6 mM DTNB and a pH stick. If DTNB test was positive (the
solution was brown-yellow), another 25 mL of cold water was used to wash DE 53.
This procedure was repeated until most of the -SH was washed out of the resin
(DTNB test for the filtrate was pale-yellow). All filtrates with positive DTNB results
were combined in a one-liter flask (if the filtrates were not clear, they were filtered
again), shell frozen in a dry ice-ethanol bath and lyophilized overnight. The dried
GEE free base was weighed and analyzed by HPLC, DTNB and BaCl2. The
purification yielded 2.5-2.8 g free base GEE. The percent yield was 73%-80%. The
overall percent yield for GEE synthesis was 42%-50%.
5.1.3 Synthesis of N-acetylglutathione Mono(glycyI)ethyl Ester
NAGME was prepared from the free base form of GSH monoethyl ester
using a method based on the method of Levy et al. (21) with modification. GEE free
base (2g, 6 mmol) was dissolved in the minimal amount of warm water (35-40 C,
about 6 mL). While water was added dropwise to GEE, the solution was warmed in a
water bath (35-40 C) to make GEE dissolve in water quickly. This solution was
added over a period of at least 40 min to a 500 mL flask containing 20 mL of glacial
acetic acid and 20 mL of acetic anhydride in an ice bath. The reaction solution was
stirred during addition. After the addition was completed an additional 10 mL of
15


acetic anhydride was added. The covered reaction flask was placed in an ice bath and
allowed to come to room temperature overnight. The volume of the reaction solution
was then reduced to about 15 mL under vacuum. Diethyl ether was slowly added to
the mixture with swirling. When the white precipitate formed, diethyl ether was
poured to the top of the flask. This solution, on standing overnight in an ice bath,
gradually deposits the crystalline product. The product was filtered, washed with
diethyl ether and air-dried (if needed, the wet product was kept drying in a vacuum
desiccator over KOH). The dry product was weighed and characterized by HPLC,
DTNB and NMR. The synthesis yielded 1.26-1.29 g NAGME. The percent yield was
56%-57%.
5.2 Analysis for the Synthesis
5.2.1 Derivatization for High-performance Liquid Chromatography (HPLC)
HPLC was used for monitoring the progress of the esterification reaction,
acetylation reaction and product purity (verifying if there was any other thiol
compound). For monitoring esterification reaction progress, 5 p,L of the reaction
solution was diluted with 95 pL of 3.33% 5-sulfosalicylic acid (SSA). To 5 pL of the
dilution were added 3.33% SSA (65 pL), 4 mM diethylenetriamine-pentaacetic acid
(DTPA, 192.5 pL), Tris-HCl (2 M, pH 9.0, 20 pL), and monobromobimane (mBBr,
0.1 M in acetonitrile, 10 pL). This mixture was placed in the dark at room
temperature. After 20 min, the reaction was quenched by addition of glacial acetic
acid (7.5 pL). To monitor acetylation reaction progress, 5 pL of the reaction solution
16


was directly derivatized as described above. For analysis of the final product, ImM
of the product solution (in 3.33% SSA) was made. To 40 pL of this solution were
added 3.33% SSA (100 pL), 4 mM DTP A (385 pL), Tris-HCl (40 pL), and mBBr
(20 pL) and then placed in the dark at room temperature. After 20 min, the reaction
was quenched by addition of glacial acetic acid (15 pL). All derivatives were stored
in a freezer (-20 C) until HPLC was available.
5.2.2 HPLC Analysis of the Thiols
50 pL of a mBBr-derivatized sample was chromatographed by using
gradients of buffer A (12.8% methanol, 0.25% acetic acid, 86.95% water, pH 3.9
with 10 M NaOH) and buffer B (90% methanol, 0.25% acetic acid, 9.75% water, not
pH adjusted) with flow rate of lmL/min. A 0.46x25 cm fully capped Cl8 reverse-
phase column (Beckman) was used. After sample injection the solvent was changed
from 100% buffer A to 65% buffer A / 35% buffer B by use of a linear gradient over
25 min. Then, over the next 10 min a linear gradient was run to give 100% buffer B.
After 9 min in buffer B, the column was regenerated for another injection by running
buffer A for 14 min. The retention times of the thiol compounds are: GSH 14 min,
GEE 29 min, DIGEE 34 min, NAG 20 min, NAGME 30 min, NAGDI 38min.
5.2.3 DTNB Assay for Free -SH (Thiol) in Unknown Compounds
The percentage of free thiols in the sample was determined using 5, 5-
dithiobis(2-nitro)benzoic acid (DTNB) essentially as described by Ellman (27).
17


(a) Preparation of GSH (reduced form, assuming 100% thiol) standard curve:
36.7 mM GSH solution was made in 10 mL of Dulbecco's Phosphate Buffered Saline
(PBS) and then diluted 10 times with PBS. The dilution (3.67 mM) was diluted to a
series of new dilutions [3.67mM GSH (pL) / PBS (pL): 0/200, 40/160, 80/120,
120/80, 160/40, and 200/0]. 30 pL of each new dilution was transferred to 1470 pL
6mM DTNB (in 0.1 M phosphate buffer, pH 7.4) in a 1 cm-path polystyrene semi-
micro cuvette. The cuvette was covered with parafilm and turned upside down gently
twice to mix the reaction solution. The absorbances of these solutions at 412 nm
were measured by UV-Visible spectrophotometer immediately. Then an absorbance
vs concentration standard curve was generated using Excel 5.0.
(b) Preparation of unknown sample: 36.7 mM GSH derivative solution was
made in 10 mL of PBS and then diluted to 3.67 mM with PBS. 120 pL of 3.67 mM
GSH derivative solution was added to 80 pL of PBS. 30 pL of this dilution was
mixed with 1470 pL of 6 mM DTNB in a cuvette and assayed by the UV-Visible
spectrophotometer as described for the standard curve measurements. Using the
standard curve generated with reduced glutathione, the actual concentration of thiol
in the unknown sample was calculated. The percentage of free thiols in the sample
was obtained by dividing this actual concentration of thiol by the theoretical value.
Each unknown sample was assayed by DTNB in triplicate. The variation among
results was not greater than 5%.
18


5.2.4 NMR Analysis for N-acetylglutathione Mono(glycyl)ethyI Ester
1 mL of 150 mM NAGME in D2O was made. A trace amount of 3-
(trimethylsilyl)propionic-2,2,3,3-d4 acid, sodium salt [TMSPA,
(CH3)SiCD2CD2C02Na] as a standard was dissolved in NAGME solution. 700 pL of
this mixture was transferred to a clean dry NMR tube. An H1 spectrum for the
product was obtained.
5.2.5 BaC^Test
BaCb was used to test the presence of SO4' by forming white precipitate
(BaS04). For the removal of SO42" from GEET/2H2SO4, 75 pL of the solution was
added in a small clean glass tube containing 300 pL of 2% BaCl2 and mixed. For the
free base GEE product, 5 mg of product was dissolved in 5 drops of water in a clean
glass tube and then 5 drops of 2% BaCl2 was added in this solution and mixed.
Appearance of white precipitate indicates BaCl2 test positive. Absence of white
precipitate indicates BaCl2 test negative.
5.2.6 DTNB Test
DTNB was used to test the presence of any free thiol by the change of
solution color from colorless to yellow. For the reaction solution or filtrate, 75 pL of
the sample was added in a test tube containing 75 pL of 6 mM DTNB (in 100 mM
phosphate buffer at pH 7.4) and mixed. If the color of the mixture was changed to
brown-yellow, it indicated DTNB test was positive. If no change was found, it
19


indicated that DTNB test was negative. If the color of the mixture changed to pale-
yellow, it indicated that only small amount thiol was in the sample. The thiol
recovery from the resin was usually stopped after DTNB test for filtrate was pale-
yellow.
5.3 Cell Uptake Experiment
The J82 cell line (human bladder carcinoma) was obtained from American
Type Culture Collection. Cells were cultured in a-Minimal Eagle Medium with
GlutaMax (a-MEM) containing 10% Fetal Bovine Serum and 1%
Penicillin/Streptomycin/Fungizone at 37 C in an atmosphere of 5% CO2. Cells were
grown to confluence in 25 cm2 culture flasks. These cells were trypsinized and
resuspended in the same medium, and then cultured again to confluence m a 75 cm
culture flask. These cells were trypsinized, resuspended in the same medium with
5% dimethyl sulfoxide (DMSO), and frozen in aliquots at -70 C.
For the cell uptake experiment, aliquots of cells were defrosted and plated
into 6-well coming culture plates (well diameter: 35mm) containing 3 mL medium
and grown to confluence.
There were three 6-well plates for the cell uptake experiment. The first (wells
1-6) as a medium control contained medium and compound only (no cells), to
determine the effect of medium on compounds over the time of the experiment. The
second (wells 7-12) was the experimental plate which had cells treated with either
control solution (PBS), GSH, or NAGME. The third plate (wells 13-18) was used for
20


cell counting. Cells were treated in the same way as in the experimental wells, but
were counted to determine the number of cells and the effect of compounds on cell
viability.
Just before the experiment began, 100 mM GSH (in PBS) and 100 mM
NAGME (in PBS) were adjusted to pH 7.0 with 10 N NaOH and sterilized by
filtration. The medium in each well was replaced with 2.7 mL a-MEM. 0.3 mL of
either GSH, NAGME, or PBS was added to the appropriate wells. The final thiol
concentration in the GSH or NAGME wells was therefore 10 mM. Wells were
incubated at 37 C in a 5% CO2 / 95% air atmosphere for 2 hr.
After incubation, 600 pL of medium from each of the medium control wells
and the experimental wells was removed and added to a microeppendorf tube
containing 300 pL of 12.9% SSA with 1.5 mM DTP A. The solutions were
derivatized for HPLC analysis (see 5.4.1 Derivatization for HPLC).
In experimental wells 7-12, the rest of the medium was aspirated. The cells
were rinsed with 5x5 mL PBS. The cells in each well were lysed by adding 0.5 mL
of ice-cold 4.31% SSA containing 0.5 mM DTP A. After 20 min on ice, the cells
were scraped from the bottom of each well. The entire cell suspension from each
well was transferred to a microeppendorf tube and centrifuged for 10 min at 15, 000
rpm. The supernatants were derivatized for HPLC analysis (see 5.4.1 Derivatization
for HPLC).
21


For counting wells 13-18, the medium was aspirated. The cells were rinsed
with 5x5 mL PBS and then each well was overlaid with 1 mL of ice-cold trypsin
(0.25%). After 30 seconds at room temperature, the cells became detached from the
well surface. To 0.7 mL of the cell suspension were added 0.1 mL of Fetal Bovine
Serum and 0.2 mL of Trypan Blue Stain (0.4%) to inactivate trypsin and stain the
cells. A drop of this mixture was used for cell counting (see 5.4.2 Cell Counting).
5.4 Analysis for Cell Uptake Experiment
5.4.1 Derivatization for HPLC
For each medium sample, 10 pL of the medium/SSA mixture was removed
and mixed with 4.31% SSA with 0.5 mM DTPA (350 pL), 4 mM DTPA (120 pL),
2 M Tris (pH 9, 100 pL) and 0.1 M mBBr (in CH3CN, 5 pL). This reaction mixture
was kept in the dark for 20 min. The reaction was then quenched with 15 pL of
glacial acetic acid.
For each cell lysate, 360 pL of the supernatant was removed and mixed with
4 mM DTPA (120 pL), 2 M Tris (pH 9, 100 pL) and 0.1 M mBBr (in CH3CN, 5
pL). This reaction mixture was kept in the dark for 20 min. The reaction was then
quenched with 15 pL of glacial acetic acid.
Standards of GSH and NAGME were also prepared tor HPLC analysis. 10
mM solutions of each compound were prepared in 4.31% SSA with 0.5 mM DTPA
and then diluted 10 times with 4.31% SSA with 0.5 mM DTPA. These 1 mM
22


solutions were diluted again as follows: [1 mM compound solution (pL) / 4.31%
SSA with 0.5 mM DTPA(pL): 0/360, 20/340, 40/320, 60/300, 80/280, and 100/260].
Each 360 pL new dilution was mixed with 4 mM DTPA (120 pL), 2 M Tris (pH 9,
100 pL) and 0.1 M mBBr (in CH3CN, 5 pL). The reactions were kept in the dark for
20 min. and then quenched with 15 pL of glacial acetic acid.
All derivatizations were analyzed by HPLC as described in 5.2.2 HPLC
Analysis of the Thiols.
5.4.2 Cell Counting
Cells were counted using a hemacytometer. Cells were counted in two grids
of 4 x 4. The count was averaged for the two grids (number of cells in one 4x4 grid
= number of cells in 104 mL).
5.4.3 Calculations
After all standard compounds were analyzed by HPLC, the peak area vs
concentration standard curve for each compound was generated using Excel 5.0. The
concentration of each compound in the medium or cell sample was then calculated
from the corresponding standard curve.
To express the medium contents of each well in mM, the calculation was
applied as follows:
(1) Moles of the thiol in 3 ml of medium (nmol) =
Moles of the thiol in derivatized sample (nmol) x 900 pL x 3000 pL /
(600 pL x 10 pL)
23


(2) Concentration of the thiol in 3 ml of medium (mM) =
Moles of the thiol in 3 ml of medium (nmol) x 10'6 / (3 x 10'3 L)
To express the cell contents of each well in nmol/106 cells, the calculation
was applied as follows:
(1) Moles of the thiol in entire 500 pi of cell lysate (nmol) =
Moles of the thiol in derivatized sample (nmol) x 500 pL / 360 pL
(2) Number of cells in each well =
Number of cells in one 4x4 grid x 1 mL x 1 mL / (10"4 mL x 0.7 mL)
(3) The cell content (nmol/106 cells) =
Moles of the thiol in entire 500 pL of cell lysate (nmol) / number
of cells in each well
24


6. Results and Discussion
6.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester
6.1.1 Reaction Mechanism
The synthesis of glutathione monoethyl ester follows the mechanism of
Fischer esterification reaction. This reaction is an acid-catalyzed condensation of an
alcohol and a carboxylic acid and then an ester and water are produced. In the
synthesis of glutathione monoethyl ester sulfuric acid was used as a catalyst and
excess absolute ethanol was used to shift the position of reaction equilibrium to favor
the formation of ester. The mechanism is depicted in Figure 6.1. At the first step, the
acid catalyst protonates the carboxylic acid group of the GSH glycine residue at its
carbonyl oxygen. Then ethanol as a nucleophile attacks the protonated carboxylic
acid to yield an intermediate. The intermediate is unstable under the acid-catalyzed
condition of its formation and undergoes dehydration to form the glutathione
mono(glycyl)ethyl ester. The product is precipitated with a large amount of diethyl
ether and recrystallized from a mixture of water and excess ethanol.
The resulting product still contains H2SO4 that interferes with the synthesis of
NAGME. To remove this impurity an anion exchanger DE 53 was used. This weakly
basic anion exchanger is based on the diethylaminoethyl tertiary amine functional
group [cellulose-0-(CH2)2N(C2H5)2]. In the water solution of the product, the
functional group of DE 53 works like a weak base and becomes cellulose-O-
25


(CH2)2N+H(C2H5)2. This group with a positive charge removes SO42' ions by binding
them in noncovalent interaction. GSH monoester remains in the solution because of
its zero overall charge. After DE 53 is filtered out, the SO4 '-free solution is dried.
The dry product is free base glutathione mono(glycyl)ethyl ester. The advantage of
using DE 53 other than AG 1X2 (OH' form) to yield salt-free GEE is that DE 53 is a
weakly basic anion exchanger. Even if excess DE 53 is added to the solution, this
weak base can not influence the pH of the solution greatly. The pH of the solution
was just changed from <4.5 to 6.0 during DE 53 treatment. By contrast, AG1X2 is a
strongly basic anion exchanger. Its active group is R-CH2N+ (CH3)3. As a strong base,
excess AG1X2 used for complete removal of SO42' can increase the pH of the
solution greatly. When AG1X2 (OH' form) was used to remove sulfate in this
experiment, it was found that the pH of the solution was changed from <4.5 to >10.
The resulting basic environment destroys the product because of deesterification and
oxidation.
Acid-catalysed esterification of GSH may lead to the formation of two forms
of monoester, the glycyl and y-glutamyl monoesters, and a diester (28). After 18-22
hr esterification of GSH, the glycyl monoester was a major product (over 80%). GSH
diester and GSH each accounted for about 10% of the total thiol compounds detected
by HPLC. The y-glutamyl monoester did not show up on HPLC spectrum. This is
probably related to the neighbouring amino group of the y-glutamyl residue.
Apparently glycine residue is more easily attacked by ethanol than glutamic acid
26


residue, perhaps due to steric reasons. The interaction between the un-ionized
carboxylate -COOH group and the ammonium -NH3+ group of the glutamic acid
residue may also influence the process of acid-catalyzed esterification. The different
pKa between two carboxylic acid is possible another reason to limit the esterification
on the glutamic acid. Because the pKa of glutamic acid is lower than that of glycine
acid, the protonation on it might be more difficult.
The diethyl ester and residual GSH were removed during the purification.
27


(1)R---C

O
H

OH
\a
- R---C
H
sOH

OH
OH
-H
(2)R-C w R---C----OH----> R--C---OH
S0H
t
c

'C&5

Â¥*>
oc2h5
Tetrahedral intermediate
OH
:OH
K
Cl */
H
(3)R---C----OH-----R----C-----O: ^ ~w R-
H
OC^5
OC^5
.OHT
C + Hp
^ocyis
T+
(4) R-----C
.OH
-H
//
O
- R--C
"OC^ ^OqHg
R = Remainder of GSH
Figure 6.1 The Mechanism of the Acid-catalyzed Esterification of GSH (29)
28


6.1.2 The Percent Yield
The synthesis of glutathione mono(glycyl)ethyl ester hemihydrosulfate
(GEET/2H2SO4) produced 58%-63% yields. When the reaction was stopped about
10% of GSH and 10% of DIGEE were observed in the reaction solution. That means
that not all of GSH had reacted with ethanol and some GEE had further been
esterified to DIGEE. GEE was purified from the crude product by recrystallization in
absolute ethanol. Because the product still contained the acid which was able to
increase the solubility of product in ethanol, recrystallization and excess ethanol
rinse resulted in loss of product in these acidic condition. Some product was also lost
during the recovery from the flask and filter.
The reaction temperature and the acid concentration also influence the
percent yield of product. High temperature or high concentrations of acid increase
the rate of the esterification, but result in a high yield of contaminating diester. Low
temperature or low concentrations of acid decrease the rate of esterification. The
optimal reaction conditions should include optimal reaction temperature and optimal
acid concentration that result in high yield of product and low yield of by-product.
The conditions as described resulted in relatively high yield of product and low yield
of by-product.
73-80% of free base GEE was recovered after removal of SO42' from
GEET/2H2SO4 with DE 53. After the removal was complete (BaCl2 test was
29


negative), the DE53 was filtered and washed with 7 x 25 mL Milli Q water. But
DTNB test showed that there was still some product adsorbed by DE 53
nonspecifically. To avoid generating a huge volume of solution which would make
lyophilization more difficult, washing DE 53 was stopped when most of the thiol
compound was recovered. This results in loss of product. Some product was also lost
during the process of transferring the dry product from the flask to a watchglass.
It was observed that the color of the free base GEE product was pale pink
other than white as expected. The color may be from metal contamination of DE 53.
The pale pink color could not be removed by washing with ethyl ether.
6.1.3 HPLC Analysis
HPLC was used to monitor the process of reactions, to determine the purity
of the product and to analyze the extent of cell uptake of NAGME.
All samples for HPLC were first derivatized by monobromobimane (mBBr).
Monobromobimane, a fluorescent labeling agent, is highly specific for reactive thiol
groups. The reaction of monobromobimane (nonfluorescent) with thiol groups
results in adducts which are highly fluorescent (30). The fluorescent label is stable to
air, to chemical and biochemical procedures, and resistant to fading under intense
irradiation (e.g. under examination with a fluorescence microscope). The reaction of
a thiol with mBBr is illustrated below (Figure 6.2). This is a nucleophilic substitution
reaction. RS group acts as a nucleophile to replace bromine of mBBr and form a
30


thiol-bimane adduct. The adducts can then be detected by a fluorometer after they
are separated and eluted by HPLC.
mBBr Thiol-bimane adduct
(Nonfluorescent) (Fluorescent)
R = Remainder of a thiol compound
Figure 6.2 The Reaction of a Thiol with mBBr
A Cl8 reverse-phase column was used to separate all GSH derivatives. This
hydrophobic column causes compounds which are more hydrophobic and neutral to
interact with the stationary phase more tightly. Compounds which are less
hydrophobic and have net charges interact less strongly with the stationary phase and
are eluted earlier from the column. In addition, the mobile phase also played a role in
the separation of GSH and GSH derivatives. In this experiment, a linear gradient
elution was applied. During the first 25 min of the elution, the solvent was linearly
changed from 100% buffer A to 65% buffer A / 35% buffer B. Because buffer A was
more polar than buffer B, the compounds which are less hydrophobic were eluted
during this period. The polarity of the solvent was further reduced by increasing the
31


concentration of buffer B from 35% to 100% over 10 min. The column was then
washed with 100% buffer B for an additional 9 min. The compounds which were
more hydrophobic or neutral were eluted during this 19 min. Under these conditions
the retention times of GSH and its derivatives were: GSH 14 min, GEE 29 min,
DIGEE 34 min, NAG 20 min, NAGME 30 min, NAGDI 38min.
A blank containing all the derivative reagents was required to determine
which peaks of each spectrum could be assigned to the reagents. After the reagent
peaks were excluded, the peaks of GSH and GSH derivatives were identified by their
specific retention times. Then the peak areas of all non-reagent peaks were totaled
and a percentage of each thiol compound was calculated.
For the synthesis of GEET/2H2SO4, HPLC analysis indicates about 10% of
DIGEE, 10% GSH and 80% of GEE in the reaction solution (Figure 6.3), but little
GSH or DIGEE in the final GEET/2H2SO4 product (Figure 6.4). Most of DIGEE
and unreacted GSH were removed during recrytallization and rinse. It was
determined by HPLC analysis that free base GEE was pure (containing <5% other
thiols) (Figure 6.5).
32


Figure 6.3 The HPLC Spectrum for Synthesis of GEE-1/2H2S04
(The sample was from the reaction solution at 19 hr 40 min.)
(Peak at 13.150 min, GSH; Peak at 29.617 min, GEE;
Peak at 34.467 min, DIGEE; Other Peaks, reagents)


Figure 6.4 The HPLC Spectrum for GEET/2H2SO4 Product (recrystallized)
(Peak at 29.517 min, GEE; Other Peaks, reagents)


mV.
Figure 6.5 The HPLC Spectrum for Free Base GEE Product
(Peak at 29.567 min, GEE; Other Peaks, reagents)


6.1.4 BaCh Test
The result of BaCl2 test for SO42' on GEE free base was negative. This
indicates that DE 53 removed sulfate from GEET/2H2SO4 successfully.
6.1.5 DTNB Assay
DTNB (see figure 6.6), known as Ellmans Reagent (27), is used to determine
free thiol (-SH, reduced sulfhydryl) qualitatively or quantitatively. This is based on
the specific reaction between DTNB and free sulfhydryl group. The mechanism of
this reaction is depicted in figure 6.7. The deprotonated SH (S'), acting as a
nucleophile, attacks the disulfide bond of DTNB. This results in a cleavage of the
original disulfide and a formation of a new disulfide between the original thiol and
thionitrobenzoate. The other product is a new free thiol. This new free thiol, called
thionitrobenzoate (TNB anion), is bright orange and absorbs electromagnetic
radiation at 412 nm with an extinction coefficient of 13,600 M_1cm_1. The
concentration of TNB anion is easily measured using an UY-Visible
spectrophotometer. Since one mole of free thiol produces one mole of TNB anion,
the concentration of TNB anion reflects the amount of free thiols in the sample. This
is the theoretical basis for the DTNB assay.
36


HOOC
COOH
N02 N02
Figure 6.6 The Structure of 5,5Dithiobis(2-nitro)benzoic Acid (DTNB)
(1) R---SH-~f=- R-----S'+H+
R = Remainder of a thiol compound rT,>TT,
TNB anion
(bright orange)
Figure 6.7 The Mechanism of the Reaction Between DTNB and a Free Thiol
The reaction between DTNB and free thiol quickly takes place at a pH
around 7. At extremely acid pH, the sulfhydryl group does not deprotonate well
enough to be nucleophilic, and the TNB anion also does not deprotonate. At high
basic pH, the disulfide bonds of DTNB may be broken and more TNB anions are
formed (31). So low pH induces a false negative result, while a false positive result is
obtained because of high pH.
DTNB is light and temperature sensitive especially when it is in solution. The
37


adduct which absorbs light at 412 nm is also sensitive to light and temperature.
DTNB is therefore susceptible to decomposition which can cause inconsistent
results. To avoid these problems, DTNB was added directly to the cuvette and the
solutions were immediately analyzed by UV-Visible spectrophotometer.
The DTNB assay verified the percentage of free thiol in GEET/2H2SO4 as
92-98% and 88-95% in free base GEE.
38


6.2 Synthesis of N-acetylglutathione Mono(glycyI)ethyl Ester
6.2.1 Reaction Mechanism
:0 O:
HO: O:
(1) ch^cocch3
H
.+
ch^cocch3
t
CH-
Y
H CH
(2) R n C= OH+^
1 \ -
CH'i
R---N-
H CH-
-H,
OH^-RN-
-OH
H O
O
\
/
CHa
:0
\
/
C=o
ch3
Tetrahedral intermediate
CHo
tf1" I "\
(3)R-N--C--OH-^R---N--CO----B~^R-
CHq
O
\
/
CHq
c=o
(k
"\=OH+
CHq
H CH-
N--C=0
+
O
CHgC---OH
R = Remainder of GEE
Figure 6.8 The Mechanism of the Acetylation of GEE (29)
The synthesis of NAGME from free base GEE follows the mechanism for
39


acetylation of an amine with an anhydride. In this synthesis acetic acid was used as a
catalyst and solvent and excess acetic anhydride was used to shift the position of
reaction equilibrium to favor the formation of NAGME. The mechanism of this
reaction is depicted in Figure 6.8. First, acetic anhydride is activated toward
nucleophilic addition by protonation of one of its carbonyl groups. Then GEE, a
primary amine (a nucleophile), adds to the protonated carbonyl group to form a
tetrahedral intermediate. This tetrahedral intermediate is dissociated by the acid
catalyst and produces NAGME and acetic acid. NAGME is precipitated by ethyl
ether. Acetic acid, the only by-product (also the reaction solvent), is separated during
the precipitation because it is soluble in ethyl ether where NAGME is not.
The free base form of GEE was used to prepare NAGME. There are two
reasons: (1) acetic anhydride, one of reagents in the synthesis of NAGME, is
possibly hydrolyzed in the presence of the strong acid H2SO4 from GEET/2H2SO4.
(2) SO42' is a potential source of contamination for the NAGME product. It is more
difficult to remove SO4 from NAGME than from GEE by an anion exchanger
because NAGME also has a net negative charge.
6.2.2 The Percent Yield
The preparation of N-acetylglutathione mono(glycyl)ethyl ester (NAGME)
from free base GEE produced yields of 56%-57%. Reducing the volume of the
reaction mixture before the precipitation with ether resulted in over 10% increase in
the percent yield during the preparation of NAG from GSH (32). Reducing the
40


volume helps to remove excess solvent which may keep the product in the solution
when ethyl ether is added. During the preparation of NAGME, the volume was also
reduced to about 1/3 volume of the original reaction solution. This resulted in a
concentrated product solution. Since NAGME is insoluble in ethyl ether, without
excess solvent most of product may be precipitated by diethyl ether. However, in the
product solution there was still a high concentration of the acid, which might
increase the solubility of NAGME in diethyl ether. This may result in loss of
product.
It was observed that the color of NAGME was white as expected. The pale
pink color from free base GEE was removed after the preparation and purification of
NAGME.
6.2.3 HPLC Analysis
For the synthesis of NAGME, HPLC analysis indicates that there was 100%
of NAGME in the reaction solution after 20-22 hr reaction (Figure 6.9). Because the
retention time of NAGME (30 min) is close to that of GEE (29 min), a mixture of
GEE standard solution and reaction solution was also run on HPLC during the
experiment to confirm the synthesis of NAGME. The HPLC spectrum for this
mixture clearly shows that there were two peaks at 29 min and 30 min respectively
(Figure 6.10). Since there was just a peak at 30 min in the HPLC spectrum of the
reaction solution, the peak at 29 min in the mixture was from GEE standard and the
other peak at 30 min was NAGME. The peak at 30 min in the HPLC spectrum of the
41


reaction solution was also identified as the peak of NAGME by acid-catalyzed
hydrolysis. An aqueous solution of the product was mixed with concentrated HC1
and stirred in a water bath (35-40 C). The hydrolysis of the reaction product was
monitored by HPLC. It was observed that hydrolysis of the compound at 30 min
resulted in a peak at 15 min. To identify this peak, the hydrolytic reaction was mixed
with a GSH standard and analyzed by HPLC. A new peak, GSH, was observed at
12.5 min (data not shown). So the peak at 15 min was from NAG. This indicated that
the compound at 30 min was NAGME, not GEE. (Because of leakage in HPLC
injector, the retention times for NAG and GSH were shifted in this case.)
The HPLC analysis for NAGME product indicates that the product was pure
(containing <5% other thiols) (Figure 6.11).
42


Figure 6.9 The HPLC Spectrum for Synthesis of NAGME
(The sample was from the reaction solution at 21.5 hr)
(Peak at 30.700 min, NAGME; Other peaks, reagents)


000.00
600.00
^ 400.00
200.00
0.00
l
0.00
--- Prcnt D
20.00
Minutos
J\_____X
I
40.00
Figure 6.10 The HPLC Spectrum for Mixture of NAGME Reaction Solution
(at 21.5 lor) and GEE Standard Solution
(Peak at 29.750 min, GEE; Peak at 30.767, NAGME;
Other peaks, reagents)


Figure 6.11 The HPLC Spectrum for NAGME Product
(Peak at 30.433 min, NAGME; Other peaks, reagents)


6.2.4 DTNB Assay
The DTNB assay verified that the percentage of free thiol in NAGME
product was 84-89.5%. It is somewhat lower than that in reagent GEE (88-95%)
because of oxidation during the synthesis. GEE has no net charge. The pH of its
aqueous solution is around 7.0. Oxidation could happen in this neutral environment
after GEE was dissolved in water.
6.2.5 NMR Analysis
The proton magnetic resonance (]H NMR) spectrum for NAGME product is
shown in figure 6.12. It indicates that there was residual ethyl ether in the product.
The peak assignments for this spectrum are summarized in table 6.1. The assigned
structure for the product is shown in figure 6.13.
46


Chemical Shift (ppm) Splitting Pattern Number of Protons Character of Protons
1.2 to 1.3 Triplet 3 p-Ethyl
2.05 Singlet 3 Acetyl
1.9 to 2.3 Multiplet 2 P-Glutamyl
2.45 to 2.55 Multiplet 2 y-Glutamyl
2.9 to 2.95 Doublet 2 P-Cysteinyl
3.5 to 3.65 Multiplet N/A N/A
4.0 to 4.05 Singlet 2 Glycyl
4.15 to 4.25 Quadruplet 2 a-Ethyl
4.3 to 4.4 Multiplet 1 a-Glutamyl
4.5 to 4.6 Triplet 1 a-Cysteinyl
Table 6.1 NMR Data for NAGME Product
The protons bonded to oxygen, nitrogen and sulfur undergo exchange with
deuterium of solvent D2O. Because deuterium does not give a signal under the
conditions of proton nmr spectroscopy, exchanges of these protons by deuterium
lead to the disappearance of their nmr peaks. So only the protons bound to carbon
generate the peaks.
TMSPA as an internal reference is shown at 0 6. The triplet at 1.2 to 1.3 8,
which integrates for 3 protons, is caused by the methyl protons of the ester. The
singlet at 2.05 8, which integrates for 3 protons, arises from the resonance of the
protons of the acetyl group. The complex two-proton resonance at 1.9 to 2.3 arises
from P-glutamyl proton. The two protons bound to P-glutamyl carbon have different
chemical shifts as a result of their magnetic inequivalence. The peaks at 2.45 to 2.55
47


5, which integrate for 2 protons, are caused by y-glutamyl protons. The doublet at 2.9
to 2.95 5, which integrates for 2 protons, arises from (3-cysteinyl protons. The tiny
peaks at 3.5 to 3.65 6 are suspected as ethyl ether. The singlet at 4.0 to 4.05 6 which
integrates for 2 protons, is caused by glycyl protons. This peak is shifted downfield
to 4.0 5 due to the deshielding of the vicinal carbonyl and amide groups. The two-
proton quadruplet resonance at 4.15 to 4.25 5 arises from the methylene protons of
the ester. The peaks at 4.3 to 4.4 6, which integrate for 1 proton, are from the
resonance of the a-glutamyl proton. Since the proton is deshielded by the carbonyl
group and amide nitrogen, its signal is shifted downfield. The triplet at 4.5 to 4.6 8,
which integrates for 1 proton, is caused by the a-cysteinyl proton. This proton is also
deshielded by its vicinal carbonyl group and amide nitrogen and shifted downfield.
The sulfhydryl group of NAGME is possibly acetylated by acetic anhydride
during the reaction. The singlet at 2.39 8 is suspected from the resonance of the
protons of the s-acetyl group.
48


Figure 6.12 H1 NMR Spectrum at 200 MHz of NAGME Product in D20
6fr


Ui
o
CHi Acetyl
I X, -V
uu c#^ <#^ <#^
NH CH CHj- CHj- c
COOI-I
NHCHCNH-
?H2 ft-Cysteinyl
SH
^ O
$
CH;
c
0
d^^thyl
CH3P-hy.l.
Figure 6.13 The Structure of NAGME


6.3 Cell Uptake Experiment
6.3.1 HPLC Analysis for the Media
The results of HPLC analysis for the media are in Table 6.2 and Table 6.3.
Thiol applied (well #) Levels of thiols in media (mM)
GSH NAG NAGME
None(PBS) (1) N.D. N.D. N.D.
None(PBS) (2) N.D. N.D. N.D.
GSH (3) 12.3 N.D. N.D.
GSH (4) 11.1 N.D. N.D.
NAGME (5) N.D. N.D. 12.3
NAGME (6) N.D. N.D. 12.6
Table 6.2 The Levels of Thiols in Medium Control Wells (N.D. = not detected)
Thiol applied (well #) Levels of thiols in media (mM)
GSH NAG NAGME
None(PBS) (7) N.D. N.D. N.D.
None(PBS) (8) N.D. N.D. N.D.
GSH (9) 9.5 N.D. N.D.
GSH (10) 10.4 N.D. N.D.
NAGME (11) N.D. N.D. 12.4
NAGME (12) N.D. N.D. 12.4
Table 6.3 The Levels of External Thiols in Experimental Wells (N.D. = not detected)
51


The results in Table 6.2 indicate that GSH and NAGME were stable in the
medium. They were not degraded by the medium during the course of the
experiment.
The results in Table 6.3 show that the concentration of GSH or NAGME was
not substantially changed compared to the concentration of GSH or NAGME in
medium control. This indicated that nothing on the outer cell surface metabolized or
degraded the thiols during the incubation.
6.3.2 HPLC Analysis for Cell Lysates
The results of HPLC analysis for cell lysates are in Table 6.4.
Thiol applied (well #) Levels of thiols in cells (nmol/10o cells)
GSH NAG NAGME
None(PBS) (7) 54.6 N.D. N.D.
None(PBS) (8) 62.0 N.D. N.D.
GSH (9) 62.4 N.D. N.D.
GSH (10) 44.5 N.D. N.D.
NAGME (11) 44.8 N.D. N.D.
NAGME (12) 45.5 N.D. N.D.
Table 6.4 The Levels of Cellular Thiols in Experimental Wells (N.D. = not detected)
Few dead cells (<10%) were found during counting. The numbers of cells
found in each well ranged from 0.56 x 106 to 1.32 x 106. No correlation between cell
number and thiol treatment was found, so the cell counts in all 6 wells were averaged
in calculating the thiol contents in Table 6.4.
52


After the 2 hr incubation, no other thiol except GSH was found in PBS, GSH
or NAGME treated cells. In comparison with the level of GSH in PBS treated cells
the level of GSH in GSH or NAGME treated cells did not show an increase. It
indicated that treatment with either GSH or NAGME failed to increase the level of
GSH in cells. The HPLC spectrum for the lysate of NAGME treated cells also did
not show any peaks which corresponded to NAGME or any products from degraded
NAGME, such as NAG. It indicated that NAGME failed to be taken up or
metabolized by cells. Because NAGME was stable in the medium during the time of
the experiment as described above, the failure of NAGME cell uptake was not
attributed to GSH formation through extracellular deacetylation and deesterification
[GSH is not effectively transported into most types of cells under a variety of
experimental conditions (9)]. The reason for the failure of NAGME cell uptake
perhaps is NAGME itself. NAGME has net charge of -1 at physiological pH. With
the negative charge NAGME might not diffuse through the cell membrane. The other
reasons might be: (1) 2 hr incubation was too short for NAGME to permeate through
cell membrane; (2) there is no carrier on the membrane of the J82 cell line which can
mediate NAGME transport into cells.
The NAGME-treated cells actually show a decrease in GSH level compared
to PBS-treated control. However, one of the two GSH-treated wells shows a similar
decrease. Therefore, no definite conclusions can be drawn without additional study.
53


7. Conclusion and Further Study
The cell uptake experiment showed that NAGME could not increase the level
of GSH or the level of other thiols in J82 cells. It indicated that NAGME was not
taken up by J82 cells.
Although J82 cells did not take up NAGME after 2 hr incubation, the
probability can not be excluded for other cell lines to have specific mechanism to
transport NAGME into cells. 2 hr incubation may also be too short for the transport
of NAGME. Other cell lines or longer incubation can thus be attempted to further
study the transport of NAGME. Other cell lines may include cisplatin-sensitive
cancer cell line, such as ovarian cancer cell line, or normal cell line.
Other derivatives of GSH can be designed to make them diffuse through cell
membrane and increase the cellular level of GSH or thiol in further study. The
hydrophobicity of GSH derivatives seems critical to their transport into cells.
Esterified GSH with higher alcohol might be a strategy to make derivatives
hydrophobic. Butyl alcohols could be candidates in further study to make GSH butyl
monoesters or butyl diesters. Uptake of them into cells would then be explored.
54


8. Failures
8.1 Synthesis of NAGME from NAGDI
Synthesis of NAGDI, the N-acetylated diethyl ester of GSH, is easy and
results in high yields. A synthesis of NAGME was attempted via hydrolysis of
NAGDI. Hydrochloric acid was used as catalyst. The process of hydrolysis was
monitored by HPLC. The hydrolysis was stopped when the reaction mixture
contained about 42% NAGDI, 45% NAGME and about 13% NAG. To separate
NAGME from this mixture, this mixture was first neutralized to pH 5.5 by 10 M
NaOH. At pH 5.5 NAGDI has no charge, NAGME has one negative charge while
NAG has two negative charges. The mixture was then loaded on a DE 53 column.
NAGDI was neutral and was first eluted by water. NAGME and NAG were anionic
and should be bound by DE 53. After elution of NAGDI was done NAGME and
NAG effluent would be collected respectively by switching the solvent from water to
0.1 M HC1. The fractions of elution were monitored by HPLC and DTNB. It was
found that all water effluents contained NAGDI, NAGME and NAG, and DTNB
tests for water effluent were brown-yellow. The acid effluents were also mixtures of
NAGDI, NAGME, and NAG, but DTNB tests for them were yellow. This indicated
that most of NAGDI, NAGME and NAG were eluted together by water. DE 53 did
not bind NAGME and NAG tightly as expected. The salt in the reaction mixture
might also influence the separation by interfering with the binding between DE 53
55


and NAGME or NAG. To avoid the interference of the salt, Bio-Gel P2 was tried to
remove them. Before loaded on the DE 53 column, the reaction mixture was first
loaded on a Bio-Gel P2 column. All fractions of elution with positive DTNB tests
were collected and monitored by HPLC. Unfortunately just NAG, no NAGME or
NAGDI was found in the effluent. NAGME and NAGDI might be hydrolyzed to
NAG during the elution.
8.2 Preparation of GEE Free Base with AG1X2 (Bicarbonate Form) Resin
To avoid the interference of the acid in GEE4/2H2SO4 when NAGME is
synthesized, SO42' has to be removed from GEET/2H2SO4 to make free base GEE.
AG1X2 (bicarbonate form) resin was first tried to remove SO42' from an aqueous
solution of GEET/2H2SO4. Resin was added to GEET/2H2SO4 solution and a
negative BaCl2 test was obtained. The BaCl2 test for the resulting product was
negative and the thiol percentage of the product was 96%, however, the yield was
low (about 20%). To remove all of SO4 ', excess AG1X2 has to be used. One reason
for the low yield may be the excess AG1X2 resin which still carried a lot of product
away after the resin was washed with water to recover the product.
8.3 Direct Preparation of GEE Free Base
Direct preparation of GEE free base has been reported by Campbell et al
(33). The method uses a higher concentration of H2SO4 and a larger excess of
ethanol. The esterification was therefore quickly completed within 4-6. When the
synthesis was complete, AG1X2 resin in the OH' form was added directly to the
56


reaction mixture for sulfate removal and acid neutralization. When at least two
consecutively negative BaCl2 tests for the supernatant were obtained, the resin was
filtered and rinsed with ethanol. The filtrate was kept in a freezer to crystallize GEE
free base. After several days, sticky (not fluffy as described by Campbell) precipitate
was observed. Ethyl ether was then applied to wash the product. Unfortunately, the
yield was very low (about 20% on average) and not reproducible. The product still
had sulfate and the thiol % of the product was below 80% which is not acceptable for
further synthesis.
To remove all sulfate from the product, excess AG 1X2 was used. But
monoester began to crystallize in the presence of excess resin. The yield of product
obtained from the filtrate was correspondingly decreased. Another problem might be
from the hydroxide form of AG1X2. A lot of OH' groups which were replaced by
SO4 groups possibly made the solution basic. Monoester in this basic environment
was more likely hydrolyzed or oxidized. That is why the yield and thiol % were low.
8.4 Reduction of NAGME with DTT
Because the degree of oxidation of the final NAGME product was generally
greater than 10 %, attempts were made to reduce the product.
In this experiment, dithiothreitol [DTT, HSCH2CH(OH)CH(OH)CH2SH] was
used to reduce batches of NAGME which had about 84% thiol percentage. First,
NAGME was dissolved in water, and then mixed with an aqueous solution of DTT.
In the reaction solution, the concentration of NAGME was 10 mM and the
57


concentration of DTT was 100 mM. After 15 min reaction in the dark, the reaction
solution was transferred to a separatory tunnel. Then ethyl acetate was added to
extract DTT. After extraction 5 times, the water layer that contained NAGME was
lyophilized. NAGME has a chain containing 12 carbon, 3 nitrogen and 1 oxygen
atom and has no charge at low pH. So it is likely hydrophobic and somewhat soluble
in ethyl acetate. This resulted in a low yield (about 25%). To increase the solubility
of NAGME in water, the pH of NAGME solution was adjusted to pH 5.5 with 10 M
NaOH. The pH adjusted NAGME solution was then mixed with DTT solution. After
15 min reaction in the dark, the extraction was performed as before. Due to pH
adjustment NAGME was not found in ethyl acetate layer and the yield went up to
100%. But the product became sticky after exposure to air. To solve this problem,
the product was dissolved in minimum absolute ethanol and precipitated, and then
washed with ethyl ether. Unfortunately, the resulting product still became sticky after
exposure to air and thiol percentage of this product was lower' than 80%. Thus the
method was modified again to solve these problems (dryness and oxidation). 1 mM
DTPA was added to NAGME water solution before neutralization to remove residual
metal that was possibly the oxidizing agent. NAGME solution was then neutralized
to pH 5.5 with 10 M NaOH and mixed with DTT solution. After 15 min reaction in
the dark, DTT was extracted by ethyl acetate. The resulting water layer (containing
NAGME and DTPA) was acidified to pH 2.0 with 0.1 M HC1 to make NAGME
hydrophobic. Then ethyl acetate was added to extract NAGME in this solution while
58


DTPA was expected to stay in water layer (but DTPA was also possibly hydrophobic
enough in the acid form to be extracted by ethyl acetate and contaminate the
product). NAGME was obtained by drying ethyl acetate layer under a flow of N2.
The yield was just 50% because even in the acid environment NAGME is still more
soluble in water than in ethyl acetate. But the product was dry even exposed to air
(the previous product became sticky after exposure to air). The problem of dryness
was overcome, but oxidation still took place. The thiol percentage of the product was
around 75% which was below 84% of starting material. Oxidation might happen
during the extraction in the absence of DTT or during the acidification of water
layer.
59


References
1. Petsko, G. A. (1995) Cancer Chemother. 377,580-581
2. Rosenberg, B. (1985) Cancer 55, 2303-2316
3. Bellon, S. F., Coleman, J. H., and Lippard, SJ. (1991) Biochemistry 30, 8026-
8035
4. Dedon, P. C. and Borch, R. F. (1987) Biochemical Pharmacol. 36,1955-1964
5. Chu, G., Mantin, R., Shen, Y.-M., Baskett, G., and Sussman, H. (1993) Cancer
72, 3707-3714
6. Chu, G. (1994) J. Bio. Chem. 269, 787-790
7. Andrews, P. A. and Howell, S. B. (1990) Cancer Cells 2, 35-43
8. Pinzani, V., Bressolle, F., Haug, I. J., Galtier, M., Blayac, J. B., and Balmes, P.
(1994) Cancer Chemother Pharmacol. 35,1-9
9. Meister, A. (1991) Pharmac. Ther. 51,155-194
10. Arrick, B. A. and Nathan, C. F. (1984) Cancer Res. 44,4224-4232
11. Ishikawa, T. and Ali-Osman, F. (1993) J. Biol. Chem. 268, 20116-20125
12. Godwin, A. K., Meister, A., ODwyer, P., Huang, C. S., Hamilton, T. C., and
Anderson, M. E. (1992) Proc. Natl. Acad. Sci. USA 89, 3070-3074
13. Meijer, C., Mulder, N. H., Hospers, G. A. P., Uges, D. R. A., and de Vries, E. G.
E. (1990) Br. J. Cancer 62, 72-77
60


14. Hormas, R. A., Andrews, P. A., Murphy, M. P., and Bums, C. P. (1987) Cancer
Lett. 34, 9-13
15. Richon, V. M., Schulte, N. A., and Eastman, A. (1987) Cancer Res. 47, 2056-
2061
16. Andrews, P. A., Murphy, M. P., and Howell, S. B. (1985) Cancer Res. 45, 6250-
6253
17. Andrews, P. A., Schiefer, M. A., Murphy, M. P., and Howell, S. B. (1988) Chem.
Biol. Interact. 65, 51-58
18. Eastman, A. (1987) Chem. Biol. Interact. 61, 241-248
19. Anderson, M. E. and Meister, A. (1989) Anal. Biochem. 183,16-20
20. Levy, E. J., Anderson, M. E., and Meister, A. (1993) Proc. Natl. Acad. Sci. USA
90,9171-9175
21. Levy, E. J., Anderson, M. E., and Meister, A. (1994) Methods Enzymol. 234,
499-505
22. Anderson, M. E., Naganuma, A., and Meister, A. (1990) FASEB J. 4, 3251-3254
23. Peters, R. H., Jollow, D. J., and Stuart, R. K. (1991) Cancer Res. 51,2536-2541
24. Peters, R. H., Ballard, K., Oatis, J. E., Jollow, D. J., and Stuart, R. K. (1990)
Cancer Chemother. Pharmacol. 26, 397-402
25. Minhas, H. S. and Thomalley, P. J. (1995) Biochem. Pharmacol. 49,1475-1482
26. Anderson, M. E., Levy, E. J., and Meister, A. (1994) Methods Enzymol. 234,
492-499
61


27. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70-77
28. Thomalley, P. K. (1991) Biochem. J. 275, 535-539
29. Carey, F. A. (1987) Organic Chemistry (Mcgraw-Hill, Inc), pp. 575-578
30. Gainer, H. and Kosower, N. S. (1980) Histochem. 68, 309-315
31. Danehy, J. P. and Hunter, W. E. (1967) J. Org. Chem. 32, 2047-2053
32. White, T. K. (1997) The synthesis of N-acetylglutathione and N-
acetylglutathione diethyl ester and cellular uptake study (Senior Honors Thesis),
PP-11
33. Campbell, E. B. and Griffith, O. W. (1989) Anal. Biochem. 183, 21-25
62


Full Text

PAGE 1

N-Acetylglutathione Mono(glycyl)ethyl Ester: Synthesis, Characterization and Transport into A Human Cancer Cell Line by MeiXu B.S., Sichuan University, 1986 M.S., Southwest Agriculture University, 1989 A thesis submitted to the University of Colorado at Denver in partial fulfillment ofthe requirements for the degree of Master of Science Chemistry 1998

PAGE 2

This thesis for the Master of Science degree by MeiXu has been approved by i..../ --------------------------------------------d' Ellen J. Le y F. Dyckes Teresa Audesirk __ s)_ljgr_ __ Date

PAGE 3

Xu, Mei (M.S., Chemistry) N-Acetylglutathione Mono(glycyl)ethyl Ester: Synthesis, Characterization and Transport into A Human Cancer Cell Line Thesis directed by Assistant Professor Ellen J. Levy ABSTRACT In this thesis N-acetylglutathione mono(glycyl)ethyl ester (NAGME) was synthesized and characterized by DTNB, HPLC and NMR. In addition, the uptake of this compound in J82 (human bladder carcinoma) cells was explored. Glutathione Mono(glycyl)ethyl Ester Hemihydrosulfate (GEE-1/2H2S04 ) was first synthesized through the acid-catalyzed esterification of GSH. The sulfate in this product was removed by an anion exchanger DE 53 to produce free base GEE. The overall percent yield for free base GEE synthesis was 42-50%. NAGME was then synthesized through the acetylation of free base GEE. The percent yield was 56-57%. The DTNB assay verified that the free thiol content ofNAGME was 84-89.5%. The HPLC spectrum of NAGME showed that there was no other thiol compounds present. The H1 NMR spectrum of NAGME indicated that there was little diethyl ether present. In NAGME cell uptake experiment, GSH level was not elevated in both GSH-treated and NAGME-treated cells. Besides GSH no other thiol was detected in lll

PAGE 4

NAGME-treated cells. These results indicated that J82 cells did not take up NAGME. NAGME could not increase the level of GSH in J82 cells under the experimental condition. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. iv

PAGE 5

ACKNOWLEDGMENTS I wish to express my sincere thanks to Dr. Levy for her advice, support, understanding, and patience throughout this project. I would like to also thank my committee members, Dr. Dyckes and Dr. Audesirk, for their guidance and support during the completion of my degree. Finally, I would like to thank all those who contributed in any way by their moral support, patience, and kind words, especially Dr. Kimbrough, Dr. Zapien, Dr. Damrauer, Dr. Anderson, Marc-Andrew Donsky and all of my wonderful friends.

PAGE 6

DEDICATION I dedicate this thesis to my husband William Tran, my parents Yuxun Xu, Hongmei Ou and my parents-in-law Anh V. Tran, Muoi Lam for their continual love, support and encouragement.

PAGE 7

CONTENTS Figures ........................................................................................ x Tables ......................................................................................... xi Abbreviations ................................................................................ xii Chapter 1 Introduction ............................................ .. ........... ... ................ ... 1 1.1 Cisplatin ................................ ................................................. 1 1.2 Glutathione ............................................................................... 3 1.3 Interaction between Glutathione and Cisplatin ...................................... .4 1.4 Glutathione Derivatives ................................................................. 6 2. Objective and Hypothesis ................................................................ 8 3. Materials ................................................................................... 11 4. Instrumentation ..................................................................... ...... 12 5. Methods .................................................................................... 13 5.1 Synthesis Procedure ..................................................................... 13 5.1.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester Hemihydrosulfate ......... 13 5 .1.2 Synthesis of Free Base Glutathione Mono(glycyl)ethyl Ester ................... 14 5.1.3 Synthesis ofN-acetylglutathione Mono(glycyl)ethyl Ester ...................... 15 5.2 Analysis for the Synthesis ............................................................... 16 Vll

PAGE 8

5.2.1 Derivatization for High-performance Liquid Chromatography (HPLC) ....... 16 5 .2.2 HPLC Analysis of the Thiols ......................................................... 17 5.2.3 DTNB Assay for Free -SH (Thiol) in Unknown Compounds ................... 17 5.2.4 NMR Analysis for N-acetylglutathione Mono(glycyl)ethyl Ester ............... 19 5.2.5 BaC12 Test. ............................................................................. 19 5.2.6 DTNB Test. ............................................................................. 19 5.3 Cell Uptake Experiment. ................................................................ 20 5.4 Analysis for Cell Uptake Experiment .................................................. 22 5.4.1 Derivatization for HPLC ............................................................... 22 5 .4.2 Cell Counting ............. ... ........................................................... 23 5.4.3 Calculations ............................................................................. 23 6. Results and Discussion .................................................................... 25 6.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester ................................... 24 6.1.1 Reaction Mechanism ................................................................... 24 6.1.2 The Percent Yield ....................................................................... 29 6.1.3 HPLC Analysis .......................................................................... 30 6.1.4 BaCl2 Test. ............................................................................... 36 6.1.5 DTNB Assay ............................................................................. 36 6.2 Synthesis ofN-acetylglutathione Mono(glycyl)ethyl Ester .......................... 39 6.2.1 Reaction Mechanism ................................................................... 39 viii

PAGE 9

6.2.2 The Percent Yield ....................................................................... 40 6.2.3 HPLC Analysis .......................................................................... 41 6.2.4 DTNB Assay ............................................................................ 46 6.2.5 NMR. Analysis ......................................................................... 46 6.3 Cell Uptake Experiment. .......... .. .................................................... 51 6.3.1 HPLC Analysis for the Media ........................................................ 51 6.3.2 HPLC Analysis for Cell Lysates .................................................... .52 7. Conclusion and Further Study ........................................................... 54 8. Failures ...................................................................................... 55 8.1 Synthesis ofNAGME from NAGDI. .................................................. 55 8.2 Preparation of GEE Free Base with AG 1X2 (Bicarbonate Form) Resin .......... 56 8.3 Direct Preparation of GEE Free Base .............................................. .... 56 8.4 Reduction ofNAGME with DTT ...................................................... 57 References ...................................................................................... 60 IX

PAGE 10

FIGURES Figure 1.1 The Structure of Cisplatin [Cis-diamminedichloroplatinum (IT)] .................. 1 1.2 The Structure of Glutathione (y-glutamylcysteinylglycine) .... ..................... 3 6.1 The Mechanism of the Acid-catalyzed Esterification of GSH .................... 28 6.2 The Reaction of A Thiol with mBBr. ................................................ 31 6.3 The HPLC Spectrum for Synthesis of GEEl/2H2S04 ................... ........ .33 6.4 The HPLC Spectrum for GEE/2H2S04 Product. ................................. 34 6.5 The HPLC Spectrum for Free Base GEE Product ................................... 35 6.6 The Structure of 5,5' -Dithiobis(2-nitro )benzoic Acid (DTNB) .................. 37 6.7 The Mechanism of the Reaction between DTNB and A Free Thiol. ............ 37 6.8 The Mechanism of the Acetylation ofGEE ..................... ........ .......... 39 6.9 The HPLC Spectrum for Synthesis ofNAGME .................................... .43 6.10 The HPLC Spectrum for Mixture ofNAGME Reaction Solution and GEE Standard Solution ........................................................................ 44 6.11 The HPLC Spectrum for NAG ME Product. ........................................ .45 6.12 H1 NMR Spectrum at 200 MHz ofNAGME Product in D20 ..................... .49 6.13 The Structure ofNAGME .............................................................. 50 X

PAGE 11

TABLES Table 6.1 NMRData forNAGME Product .................................................... .47 6.2 The Levels ofThiols in Medium Control Wells ..................................... 51 6.3 The Levels of External Thiols in Experimental Wells .............................. 51 6.4 The Levels of Cellular Thiols in Experimental Wells .............................. 52 Xl

PAGE 12

Abbreviation a-MEM ATP BSO DIGEE DTNB DTPA DTT GEE GSH GS-X HPLC mBBr NAG NAGDI NAG ME NMR PBS ABBREVIATIONS Name a-Minimal Eagle Medium with GlutaMax Adenosine Triphosphate Buthionine Sulfoximine Glutathione Diethyl Ester 5, 5' -Dithiobis(2-nitro )benzoic Acid Diethylenetriamine-pentaacetic Acid Dithiothreitol Glutathione Mono(glycyl)ethyl Ester Glutathione (y-glutamylcysteinylglycine) Glutathione S-conjugate Export High-performance Liquid Chromatography Monobromobimane N-acetylglutathione N-acetylglutathione Diethyl Ester N-acetylglutathione Mono(glycyl)ethyl Ester Nuclear Magnetic Resonance Dulbecco's Phosphate-buffered Saline xii

PAGE 13

SSA TMSPA TNB Tris 5-Sulfosalicylic Acid 3-(Trimethylsilyl)propionic-2,2,3,3-d4 Acid, Sodium Salt Thionitrobenzoate Tris(hydroxymethyl) Aminomethane xiii

PAGE 14

1. Introduction 1.1 Cisplatin Cis-diamminedichloroplatinum ( cisplatin), an inorganic compound (Figure 1.1), was first synthesized in 1845. Its biological activity was discovered serendipitously in 1965 when Barnett Rosenberg investigated the effect of an electric current on Escherichia coli (1 ). The cisplatin, which was synthesized in the electric field, was found to induce filamentous growth in E. coli and also to be active against implanted tumors in mice. Now cisplatin is one of the largest selling cancer chemotherapeutic agents and has been used successfully in treating many malignancies, including testicular, ovarian, head and neck, bladder, esophageal and small cell lung cancers (2). Figure 1.1 The Structure ofCisplatin [Cis-diamminedichloroplatinum (II)] The cytotoxicity of cisplatin is believed to be due to the formation of DNA adducts, which include DNA-protein cross-links, DNA monoadducts, and interstrand and intrastrand DNA cross-links. The intrastrand cisplatin cross-link produces a severe local distortion in the DNA double helix, leading to unwinding and kinking which eventually causes cancer cell death (3). Besides DNA, cisplatin also has

PAGE 15

affinity for sulfur in methionine and cysteine because of its platinum center. Thus all peptides and proteins containing methionine or cysteine are potential ligands of cisplatin in biological systems. The ability of cisplatin to bind to some these species, thus potentially affecting enzyme activity or protein conformation, may also play a role in systemic toxicity (4). Although cisplatin plays an important role in cancer therapy, unfortunately, it leads to severe multiorgan toxicities, such as nephrotoxicity (5). Additionally, after an initially favorable response, there can be rapid growth of a cisplatin-resistant cell population (6). The acquired resistance is a serious barrier to chemotherapeutic success. Numerous potential mechanisms of resistance to cisplatin have been elucidated. They include decreased drug accumulation, increased levels of intracellular thiols, increased DNA repair, and overexpression of metallothioneins (7). Actually there has been a continuous search for biological and pharmacological strategies to prevent cisplatin-induced nephrotoxicity and drug-resistance. These strategies include modification of administration modes to dilute cisplatin in the tubule and decrease systemic peak concentrations, the use of antidotes to antagonize the activity of cisplatin, and development of new galenic forms (a pharmaceutical preparation of a drug of animal or plant origin) (8). 2

PAGE 16

1.2 Glutathione R1NH-CH-CHz-CHz-CO-NH-CH-CO-NH-CHz-COOR3 I I CHzSH Figure 1.2 The Structure of Glutathione (y-glutamylcysteinylglycine) R 1=R2=R3=H Glutathione (GSH, Figure 1.2) Is a tripeptide thiol, y-glutamylcysteinylglycine. GSH is not required in the diets of animals. It is synthesized in many types of cells from glutamate (Glu), cysteine (Cys), and glycine (Gly), which are formed as products of cell metabolism and are also obtained from the diet. GSH is synthesized in a two-step pathway [(1) Glu + Cys y-Glu-Cys; (2) y-Glu-Cys + Gly y-Glu-Cys-Gly] involving the ATP-dependent enzymes y-glutamylcysteine synthetase and glutathione synthetase, respectively. The first step is rate-limiting and inhibited by glutathione itself and by buthionine sulfoximine (BSO), a transition-state analog (9). Glutathione has a variety of physiologically important functions, for example: (a) maintenance of the thiols of proteins (and other compounds) and of antioxidants, (b) reduction of ribonucleotides to form the deoxyribonucleotide precursors of DNA, (c) protection against oxidative damage, free radical damage, and other types of toxicity, and (d) synthesis and transport of biologically active, endogenous substances, such as cysteine moieties (9). 3

PAGE 17

Under normal steady-state conditions, the majority of GSH exists in the reduced form (0.5 to 10 mM) in mammalian cells. NADPH-dependent reduction of glutathione disulfide (GSSH) to GSH produces a GSH:GSSH ratio of over 100:1, which permits GSH to function as an intracellular reducing agent (1 0). In situations in which GSH is consumed by intracellular reactions, for example in detoxification reactions, therapeutic delivery of GSH to the cell might be sufficiently rapid to make up for the constant utilization of GSH and enhance the function of GSH. It was reported that elevation of GSH levels by addition of GSH monoethyl ester effectively protected human lymphoid cells against the lethal effects of irradiation, while depletion of GSH increased cell sensitivity to radiation under hypoxic conditions (9). This suggests that manipulation of GSH metabolism may be a rewarding therapeutic strategy. Since definite conclusions about "normal", "maximal", and "minimal" cellular GSH levels cannot yet be drawn, there must also be concern about the possibility that very high levels would lead to toxic effects. For example, high thiollevels in cell culture media may lead to toxicity due to peroxide formation (9). 1.3 Interaction between Glutathione and Cisplatin Glutathione interacts with a wide variety of drugs that include cisplatin. It is reported (11) that in human leukemia HL-60 cells, each GSH molecule acts as a bidentate chelating ligand, coordinating to platinum via cysteinyl sulfur and nitrogen atoms. After a GSHplatinum complex is formed in a 2:1 molar ratio, the complex is 4

PAGE 18

eliminated from the cell by an ATP-dependent glutathione S-conjugate export pump (GS-X pump). Although the structure of GSH-cisplatin complexes needs to be further explored, it has been reported by many sources that GSH plays a role in the resistance of normal and tumor cells to cisplatin toxicity Elevated levels of cellular GSH were observed in human ovarian cancer cell lines that have high resistance to cisplatin (12). Similarly, depleting GSH levels can enhance the toxicity of cisplatin in a lung cancer cell line (13) The exact mechanism by which GSH influences cisplatin cytotoxicity is not yet completely clear. There are several ways in which glutathione could affect cisplatin toxicity, for example, by preventing it from reacting with DNA, by quenching monofunctional adducts with DNA before they rearrange to the more toxic bifunctional lesions or by increasing the rate of cellular elimination of cisplatin by direct binding and efflux through a GS-X pump. The possible roles and means of glutathione in modifying the biological activity of cisplatin, and thus its roles in cisplatin resistance and toxicity, are clearly varied and complex. Elevated intracellular GSH is a component of the cisplatin-resistant phenotypes, but the significance of elevated glutathione in cisplatin resistance is unclear. If elevated glutathione is responsible for cisplatin resistance, then experimentally lowering the glutathione concentration should reverse resistance. Buthionine sufoximine (BSO) is the agent of choice for reducing glutathione concentration, because it is a potent and specific inhibitor of glutathione synthesis, 5

PAGE 19

specifically at the y-glutamylcysteine synthetase step. It has been shown that GSH depletion with BSO can sensitize some cisplatin-resistant cell lines to the drug by increasing both the amount ofPt bound to DNA and the Pt-GpG adduct (12, 14). In other studies, little or no sensitization was observed in other cisplatin-resistant cell lines after BSO treatment (15, 16). This suggests that there are other cellular mechanisms of resistance. Andrews et al. demonstrated that only after a prolonged decrease in the glutathione levels did human ovarian carcinoma cells become sensitized to cisplatin or more sensitized to cisplatin, and this was true for either cisplatin-resistant or cisplatin-sensitive cells (17). The intracellular concentration of glutathione is generally in the range of 0.510 mM. At 1 mM, only a 20% reduction in the platination of DNA was observed in an in vitro incubation (18) At 10 mM a 70% reduction in platination of DNA was observed. Hence, if the major role of glutathione is to reduce platination of DNA, it might be possible to modulate toxicity at the higher ranges of glutathione concentrations, but not at intracellular concentrations of less than 1 mM. This may explain the ambiguity in the results obtained after experimentally reducing glutathione concentration, and this suggests that clinical alteration of cellular glutathione levels would be one of means of influencing cisplatin activity. 1.4 Glutathione Derivatives Because elevated glutathione has been shown to be a component of drug resistant phenotypes in a number of cisplatin-resistant cells, manipulation of cellular 6

PAGE 20

glutathione levels is probably one of means of influencing cisplatin activity. But glutathione is not effectively transported into most types of cells under a variety of experimental conditions (9). So administration of glutathione does not result in substantial cellular elevation. Fortunately it has been shown that simple organic derivatives of GSH can be transported into cells more effectively than GSH itself and can be converted to GSH after transport into the cells and increase GSH levels. Two of these derivatives are glutathione mono(glycyl)ethyl ester (GEE, R 3 =CH2CH 3 in figure 1.2) and glutathione diethyl ester (DIGEE, Rz=R3= CH2CH 3 in figure 1.2). They can be transported into cells and intracellularly hydrolyzed to yield GSH and also result in large increase in total cellular thiols (including cysteine and the esters themselves) (19, 20). On the other hand, whether N-acetylglutathione derivatives are effectively transported into cells and yield similar increases in GSH has been only minimally explored. Studies on suspensions of human erythrocytes have shown evidence for uptake of N-acetyl-GSH diethyl ester (NAGDI, R 1 =CH 3 CO, Rz=R3=CH2CH3 in figure 1.2) and for partial deesterification of this compound intracellularly; no cleavage of the N-acetyl moiety was found. Initial studies on hamsters indicate that administration of N-acetylGSH diethyl ester leads to some increase in the level of GSH in the liver and kidney of buthionine sulfoximine treated animals (21). There is no report about the influence of the transport of N acetyl-GSH monoester (NAGME, R1=CH3CO, Rz=H, R3=CH2CH 3 in figure 1.2) into cells on GSH levels in cells. 7

PAGE 21

2. Objective and Hypothesis The mono (glycyl) ethyl ester of glutathione has been shown to offer protection against cisplatin toxicity to both cells and mice (22, 23). The effect of the diethyl ester on cisplatin activity is not known. However, the diester has been shown to protect K-562 cells (a kind of cancer cell line) against the cytotoxicity of the potential cancer drug, 4-hydroperoxycyclophosphamide (4-HC) (24). The ability of glutathione esters to elevate cellular glutathione levels is well-established (25). The ability ofN-acetyl-GSH monoester to influence cellular GSH levels is unexplored. The objectives of this study are: (1) to synthesize N-acetylglutathione mono(glycyl)ethyl ester (NAGME, R1=CH3CO and R3=CH2CH3 in figure 1.2) using a modification of the literature method and then characterize the product by NMR, HPLC, and the reaction with DTNB; (2) to investigate uptake of NAGME into a human cancer cell line to determine the influence of NAGME on the level of GSH and other thiols in cells. If NAGME could be transported into cells, the effect of NAGME on the cytotoxicity of cisplatin would be explored in further study. This study will therefore ultimately contribute to safer and more effective cancer chemotherapy. Although NAGME like GSH has a net charge of -1 at physiological pH, acetyl and ethyl groups in NAGME may make it different from GSH in hydrophobicity. The hypothesis for NAGME cell uptake was that NAGME could be 8

PAGE 22

transported into cells because it was more hydrophobic than GSH. After transported into the cells NAGME might be hydrolized to yield NAG or deacetylized to yield GEE or GSH and then increase the levels of intracellular thiols. Cislpatin is toxic to bladder cancer cells (2). In this study J82 cells (human bladder carcinoma) have therefore been chosen for cell uptake experiment. If NAGME could be taken up by J82 cells, it might then be like GSH, in which it could protect cancer cells against cisplatin toxicity. Although protection of cancer cells against cisplatin toxicity by elevation of thiol level is obviously not a goal, the selection of fast-growing cancer cells over normal cells is helpful for the initial study on NAGME uptake. In addition to the elevation of cellular glutathione, NAGME as a thiol may offer unique mechanisms for influencing the biochemistry of cisplatin. Four of these mechanisms, specific for cisplatin, might be: (a) direct binding of NAGME to cisplatin extracellularly, preventing transport of cisplatin across cell membranes. This mechanism may decrease the toxicity of cisplatin; (b) direct binding of NAGME to cisplatin intracellularly, yielding complexes which may interact with DNA or proteins in a manner different from that of glutathione-cisplatin complexes; (c) competition with GSH for intracellular cisplatin complexes, potentially decreasing the amount ofGSH-platinum and then decreasing cisplatin efflux through GSH-specific pumps; (d) competition of NAGME-platinum complexes with GSH platinum complexes for the GSH-specific pumps, potentially weakening the ability 9

PAGE 23

of GSH-specific pumps and then decreasing the amount of cisplatin efflux. The latter three mechanisms could potentially enhance toxicity The exploration of cisplatin interactions with glutathione derivatives and the effects of such interactions on cellular cisplatin activity will yield information critical to designing adjunct therapies for cisplatin cancer treatment. The ultimate purpose is to make chemotherapy safer and more effective, by lessening toxicity of cisplatin toward healthy cells and by overcoming the resistance of cancer cells to cisplatin. 10

PAGE 24

3. Materials Glutathione (98-1 00% reduced form) was from Sigma. Pre-swollen microgranular anion exchanger DE 53 [Diethylaminoethyl cellulose, cellulose-O (CH2)2N(C2H5)2] was from Whatman. AG1X2 resin (chloride form, 200-400 mesh) was from Bio-Rad Laboratories. 5, 5 '-Dithiobis(2-nitro )benzoic acid (DTNB), 3(Trimethylsilyl)propionic2,2,3,3-d4 acid sodium salt (TMSPA) and D20 (99.9 atom % D) were from Aldrich Chern. CO. Monobromobimane (mBBR) was from Molecular Probes. Absolute ethanol was from AAPER Alcohol and Chemical CO. Methanol (HPLC grade), Acetic anhydride and Sulfuric acid (trace metal grade, min 95.0%-max 98.0%) were from FisherChemical. Dulbecco's phosphate-buffered saline (PBS), a.-Minimal Eagle Medium with GlutaMax (a.-MEM), Trypsin (0.25%), Trypan Blue Stain (0.4%), antibiotic/antimycotic mixture and Fetal Bovine Serum were from Gibco/BRL. All other reagents were the highest grade available. 11

PAGE 25

4. Instrumentation The UV-Visible spectrophotometer (cary 1E) was from Varian Australia Pty. Ltd. The HPLC system consisted of a Beckman Ultrasphere 0.46x25 em C18 reverse-phase column, Waters 501 HPLC pump (Millipore), and a Waters U6K injector (Millipore). The detector was a model 121 fluorometer (Gilson). The Nuclear Magnetic Resonance spectrometer (200 MHz) was from Varian. 12

PAGE 26

5. Methods 5.1 Synthesis Procedure 5.1.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester Hemihydrosulfate Esterification conditions were essentially based on those described by Levy et al. (26). Reduced GSH (1 Og, 32.5 mmol) was added to an ice-cold solution of 100 mL absolute ethanol containing 2.74 mL concentrated H2S04 in a 1-liter flask. The reaction was started in an ice bath and stirred for 18 to 22 hr while warming to room temperature. Progress of the esterification was monitored by HPLC (see below). After 90% of GSH was gone, and over 80% of the glutathione monoester and 10% of diester appeared, diethyl ether was slowly added to the flask with swirling. When a white product formed, diethyl ether was poured up to the top of the flask. The flask was placed in an ice bath overnight. The supernatant was then decanted, and the syrup on the bottom of flask was dried under a flow of N2 to remove the residual ether. The syrup was dried overnight under vacuum over KOH. The fluffy dry white product (about 13 g) was dissolved in the minimal amount of warm Milli Q water (35-40 oc, about 3 mL), and absolute ethanol (75 mL) was slowly added with swirling. This solution deposited crystalline product after standing in the -20 C freezer overnight. The white crystals were filtered on a sintered glass funnel and washed with ice-cold absolute ethanol (3 x 80 mL) and diethyl ether (5 x 80 mL). The product was transferred to a large watchglass and air-dried in the hood to 13

PAGE 27

remove the last traces of ether. The product was weighed, analyzed by HPLC and DTNB, and then stored in a desiccator. The synthesis yielded 7.2-7.9 g product. The percent yield was 58%-63%. 5.1.2 Synthesis of Free Base Glutathione Mono(glycyl)ethyl Ester The process of removing acid from GEE/2H2S04 was carried out in a cold box. 60 g of DE 53 was washed with 5 x 100 mL Milli Q water and divided into portions of about 5 g. 4 g of GEEli2H2S04 was dissolved in 60 mL cold water. This solution was tested by 2% BaCh and 6 mM DTNB (in 100 mM phosphate buffer at pH 7.4) to check for the presence of sol and -SH (see 5.2.5 BaCh test and 5 2.6 DNTB test). The pH of this solution was checked by spotting 5 J.LL of the solution on a pH stick (pH range: 4.5-1 0). The BaCh test and the DTNB test for this initial solution were strong positive and weak positive respectively. The pH of this solution was <4.5. Two portions of clean DE53 were then added to the solution, and the mixture was stirred in a cold box to keep the resin suspended. After 5 min, stirring was stopped, and 0.5 mL of the supernatant was removed and centrifuged for 3 min at 15,000 rpm. The resulting clear supernatant was then tested by 2% BaCh, 6mM DTNB and a pH stick. If BaCh test was positive (white precipitate appeared), another portion ofDE53 was added to the solution to remove sulfate. This procedure was repeated until two consecutive negative BaCh tests were obtained (usually 6-7 portions were needed). During the DE 53 treatment all DTNB tests were positive and the pH of the solution was changed from <4.5 to 6.0. After DE 53 treatment was 14

PAGE 28

done, the DE 53 was immediately removed from the sulfate-free solution by filtration. The filtrate was tested with 2% BaCh, 6 mM DTNB and a pH stick. The resin was washed with 25 mL cold water, and then filtered. The filtrate was also tested with 2% BaCh, 6 mM DTNB and a pH stick. If DTNB test was positive (the solution was brown-yellow), another 25 mL of cold water was used to wash DE 53. This procedure was repeated until most of the -SH was washed out of the resin (DTNB test for the filtrate was pale-yellow). All filtrates with positive DTNB results were combined in a one-liter flask (if the filtrates were not clear, they were filtered again), shell frozen in a dry ice-ethanol bath and lyophilized overnight. The dried GEE free base was weighed and analyzed by HPLC, DTNB and BaCh. The purification yielded 2.5-2.8 g free base GEE. The percent yieid was 73%-80%. The overall percent yield for GEE synthesis was 42%-50%. 5.1.3 Synthesis of N-acetylglutathione Mono(glycyl)ethyl Ester NAGME was prepared from the free base form of GSH monoethyl ester using a method based on the method of Levy et al. (21) with modification. GEE free base (2g, 6 mmol) was dissolved in the minimal amount of warm water (35-40 ac, about 6 mL). While water was added dropwise to GEE, the solution was warmed in a water bath (35-40 C) to make GEE dissolve in water quickly. This solution was added over a period of at least 40 min to a 500 mL flask containing 20 mL of glacial acetic acid and 20 mL of acetic anhydride in an ice bath. The reaction solution was stirred during addition. After the addition was completed an additional 10 mL of 15

PAGE 29

acetic anhydride was added. The covered reaction flask was placed in an ice bath and allowed to come to room temperature overnight. The volume of the reaction solution was then reduced to about 15 mL under vacuum. Diethyl ether was slowly added to the mixture with swirling. When the white precipitate formed, diethyl ether was poured to the top of the flask. This solution, on standing overnight in an ice bath, gradually deposits the crystalline product. The product was filtered, washed with diethyl ether and air-dried (if needed, the wet product was kept drying in a vacuum desiccator over KOH). The dry product was weighed and characterized by HPLC, DTNB and NMR. The synthesis yielded 1.26-1.29 g NAGME. The percent yield was 56%-57%. 5.2 Analysis for the Synthesis 5.2.1 Derivatization for High-performance Liquid Chromatography (HPLC) HPLC was used for monitoring the progress of the esterification reaction, acetylation reaction and product purity (verifying if there was any other thiol compound). For monitoring esterification reaction progress, 5 J.LL of the reaction solution was diluted with 95 J.LL of3.33% 5-sulfosalicylic acid (SSA). To 5 J.LL ofthe dilution were added 3.33% SSA (65 J.LL), 4 mM diethylenetriamine-pentaacetic acid (DTPA, 192.5 J.LL), Tris-HCl (2M, pH 9.0, 20 J.LL), and monobromobimane (mBBr, 0.1 M in acetonitrile, 10 J.LL). This mixture was placed in the dark at room temperature. After 20 min, the reaction was quenched by addition of glacial acetic acid (7.5 J.LL). To monitor acetylation reaction progress, 5 J.LL of the reaction solution 16

PAGE 30

was directly derivatized as described above. For analysis of the final product, 1mM of the product solution (in 3.33% SSA) was made. To 40 J.LL of this solution were added 3.33% SSA (100 J.LL), 4 mM DTPA (385 J.LL), Tris-HCl (40 J.LL), and mBBr (20 J.LL) and then placed in the dark at room After 20 min, the reaction was quenched by addition of glacial acetic acid (15 J.LL). All derivatives were stored in a freezer ( -20 a c) until HPLC was available. 5.2.2 HPLC Analysis of the Thiols 50 J.LL of a mBBr-derivatized sample was chromatographed by usmg gradients of buffer A (12.8% methanol, 0.25% acetic acid, 86.95% water, pH 3.9 with 10M NaOH) and buffer B (90% methanol, 0.25% acetic acid, 9.75% water, not pH adjusted) with flow rate of 1mL/min. A 0.46x25 em fully capped C18 reverse phase column (Beckman) was used. After sample injection the solvent was changed from 100% buffer A to 65% buffer A I 35% buffer B by use of a linear gradient over 25 min. Then, over the next 10 min a linear gradient was run to give 100% buffer B. After 9 min in buffer B, the column was regenerated for another injection by running buffer A for 14 min. The retention times of the thiol compounds are: GSH 14 min, GEE 29 min, DIGEE 34 min, NAG 20 min, NAGME 30 min, NAGDI 38min. 5.2.3 DTNB Assay for Free -SH (Thiol) in Unknown Compounds The percentage of free thiols in the sample was determined using 5, 5'dithiobis(2-nitro )benzoic acid (DTNB) essentially as described by Ellman (27). 17

PAGE 31

(a) Preparation of GSH (reduced form, assuming 100% thiol) standard curve: 36.7 mM GSH solution was made in 10 mL ofDulbecco's Phosphate Buffered Saline (PBS) and then diluted 10 times with PBS. The dilution (3.67 mM) was diluted to a series of new dilutions [3.67mM GSH (j.!L) I PBS (1-!L): 0/200, 40/160, 80/120, 120/80, 160/40, and 200/0]. 30 j.!L of each new dilution was transferred to 1470 j.LL 6mM DTNB (in 0.1 M phosphate buffer, pH 7.4) in a 1 em-path polystyrene semi micro cuvette. The cuvette was covered with parafilm and turn-ed upside down gently twice to mix the reaction solution. The absorbances of these solutions at 412 nm were measured by UV-Visible spectrophotometer immediately. Then an absorbance vs concentration standard curve was generated using Excel5.0. (b) Preparation of unknown sample: 36.7 mM GSH derivative solution was made in 10 mL ofPBS and then diluted to 3.67 mM with PBS. 120 j.LL of3.67 mM GSH derivative solution was added to 80 j.!L of PBS. 30 j.LL of this dilution was mixed with 1470 j.!L of 6 mM DTNB in a cuvette and assayed by the UV-Visible spectrophotometer as described for the standard curve measurements. Using the standard curve generated with reduced glutathione, the actual concentration of thiol in the unknown sample was calculated The percentage of free thiols in the sample was obtained by dividing this actual concentration of thiol by the theoretical value. Each unknown sample was assayed by DTNB in triplicate. The variation among results was not greater than 5%. 18

PAGE 32

5.2.4 NMR Analysis for N-acetylglutathione Mono(glycyl)ethyl Ester 1 mL of 150 mM NAGME in D20 was made. A trace amount of 3( trimethylsilyl)propionic2,2,3 ,3-c4 acid, sodium salt [TMSPA, (CH3)SiCD2CD2C02Na] as a standard was dissolved in NAGME solution. 700 1-f.L of this mixture was transferred to a clean dry NMR tube. An H 1 spectrum for the product was obtained. 5.2.5 BaCh Test BaC}z was used to test the presence of so/by forming white precipitate (BaS04). For the removal of so/from GEE/2H2S04 75 1-f.L of the solution was added in a small clean glass tube containing 300 1-f.L of2% BaC}z and mixed. For the free base GEE product, 5 mg of product was dissolved in 5 drops of water in a clean glass tube and then 5 drops of 2% BaCh was added in this solution and mixed. Appearance of white precipitate indicates BaC}z test positive. Absence of white precipitate indicates BaC}z test negative. 5.2.6 DTNB Test DTNB was used to test the presence of any free thiol by the change of solution color from colorless to yellow. For the reaction solution or filtrate, 75 1-f.L of the sample was added in a test tube containing 75 1-f.L of 6 mM DTNB (in 100 mM phosphate buffer at pH 7.4) and mixed. If the color of the mixture was changed to brown-yellow, it indicated DTNB test was positive. If no change was found, it 19

PAGE 33

indicated that DTNB test was negative. If the color of the mixture changed to pale yellow, it indicated that only small amount thiol was in the sample. The thiol recovery from the resin was usually stopped after DTNB test for filtrate was pale yellow. 5.3 Cell Uptake Experiment The J82 cell line (human bladder carcinoma) was obtained from American Type Culture Collection. Cells were cultured in a-Minimal Eagle Medium with GlutaMax (a.-MEM) containing 10% Fetal Bovine Serum and 1% Penicillin/Streptomycin!Fungizone at 37 oc in an atmosphere of 5% C02 Cells were grown to confluence in 25 cm2 culture flasks. These cells were trypsinized and resuspended in the same medium, and then cultured again to confluence in a 75 cm2 culture flask. These cells were trypsinized, resuspended in the same medium with 5% dimethyl sulfoxide (DMSO), and frozen in aliquots at -70 C. For the cell uptake experiment, aliquots of cells were defrosted and plated into 6-well corning culture plates (well diameter: 35mm) containing 3 mL medium and grown to confluence. There were three 6-well plates for the cell uptake experiment. The first (wells 1-6) as a medium control contained medium and compound only (no cells), to determine the effect of medium on compounds over the time of the experiment. The second (wells 7-12) was the experimental plate which had cells treated with either control solution (PBS), GSH, or NAGME. The third plate (wells 13-18) was used for 20

PAGE 34

cell counting. Cells were treated in the same way as in the experimental wells, but were counted to determine the number of cells and the effect of compounds on cell viability. Just before the experiment began, 100 mM GSH (in PBS) and 100 mM NAGME (in PBS) were adjusted to pH 7.0 with 10 N NaOH and sterilized by filtration. The medium in each well was replaced with 2.7 mL a-MEM. 0.3 mL of either GSH, NAGME, or PBS was added to the appropriate wells. The fmal thiol concentration in the GSH or NAG ME wells was therefore 10 mM. Wells were incubated at 37 C in a 5% C02 I 95% air atmosphere for 2 hr. After incubation, 600 J..lL of medium from each of the medium control wells and the experimental wells was removed and added to a microeppendorf tube containing 300 J..lL of 12.9% SSA with 1.5 mM DTP A. The solutions were derivatized for HPLC analysis (see 5.4.1 Derivatization for HPLC). In experimental wells 7-12, the rest of the medium was aspirated. The cells were rinsed with 5 x 5 mL PBS. The cells in each well were lysed by adding 0.5 mL of ice-cold 4.31% SSA containing 0.5 mM DTP A. After 20 min on ice, the cells were scraped from the bottom of each well. The entire cell suspension from each well was transferred to a microeppendorftube and centrifuged for 10 min at 15, 000 rpm. The supernatants were derivatized for HPLC analysis (see 5.4.1 Derivatization forHPLC). 21

PAGE 35

For counting wells 13-18, the medium was aspirated. The cells were rinsed with 5 x 5 mL PBS and then each well was overlaid with 1 mL of ice-cold trypsin (0.25%). After 30 seconds at room temperature, the cells became detached from the well surface. To 0.7 mL of the cell suspension were added 0 1 mL ofFetal Bovine Serum and 0.2 mL of Trypan Blue Stain (0.4%) to inactivate trypsin and stain the cells. A drop of this mixture was used for cell counting (see 5.4.2 Cell Counting). 5.4 Analysis for Cell Uptake Experiment 5.4.1 Derivatization for HPLC For each medium sample, 10 I-lL of the medium/SSA mixture was removed and mixed with 4.31% SSA with 0.5 mM DTPA (350 !-LL), 4 mM DTPA (120 !-LL), 2M Tris (pH 9, 100 !-lL) and O.l M mBBr (in CH3CN, 5 !-lL). This reaction mixture was kept in the dark for 20 min. The reaction was then quenched with 15 I-lL of glacial acetic acid. For each cell lysate, 360 I-lL of the supernatant was removed and mixed with 4 mM DTPA (120 !-LL), 2M Tris (pH 9, 100 !-LL) and 0.1 M mBBr (in CH3CN, 5 !-lL). This reaction mixture was kept in the dark for 20 min. The reaction was then quenched with 15 I-lL of glacial acetic acid. Standards of GSH and NAG ME were also prepared for HPLC analysis. 10 mM solutions of each compound were prepared in 4.31% SSA with 0.5 mM DTP A and then diluted 10 times with 4.31% SSA with 0.5 mM DTPA. These 1 mM 22

PAGE 36

solutions were diluted again as follows: [1 mM compound solution (J.tL) I 4.31% SSA with 0.5 mM DTPA(J.tL): 0/360, 20/340, 40/320, 60/300, 80/280, and 100/260]. Each 360 JlL new dilution was mixed with 4 mM DTPA (120 J.tL), 2M Tris (pH 9, 100 J.tL) and 0.1 M mBBr (in CH3CN, 5 J.tL). The reactions were kept in the dark for 20 min. and then quenched with 15 JlL of glacial acetic acid. All derivatizations were analyzed by HPLC as described in 5.2.2 HPLC Analysis of the Thiols. 5.4.2 Cell Counting Cells were counted using a hemacytometer. Cells were counted in two grids of 4 x 4. The count was averaged for the two grids (number of cells in one 4 x 4 grid =number of cells in 10--4 mL). 5.4.3 Calculations After all standard compounds were analyzed by HPLC, the peak area vs concentration standard curve for each compound was generated using Excel 5.0. The concentration of each compound in the medium or cell sample was then calculated from the corresponding standard curve. To express the medium contents of each well in mM, the calculation was applied as follows: (1) Moles of the thiol in 3 m1 of medium (nmol) = Moles of the thiol in derivatized sample (nmol) x 900 JlL x 3000 JlL I (600 JlL X 10 JlL) 23

PAGE 37

(2) Concentration of the thiol in 3 ml of medium (mM) = Moles ofthe thiol in 3 ml of medium (nmol) x 10-6 I (3 x 10-3 L) To express the cell contents of each well in nmol/1 06 cells, the calculation was applied as follows: (1) Moles of the thiol in entire 500 J.Ll of cell lysate (nmol) = Moles of the thiol in derivatized sample (nmol) x 500 J.LL I 360 J.LL (2) Number of cells in each well= Number of cells in one 4 x 4 grid x 1 mL x 1 mL I (10-4 mL x 0.7 mL) (3) The cell content (nmol/106 cells)= Moles of the thiol in entire 500 J.LL of cell lysate (nmol) I number of cells in each well 24

PAGE 38

6. Results and Discussion 6.1 Synthesis of Glutathione Mono(glycyl)ethyl Ester 6.1.1 Reaction Mechanism The synthesis of glutathione monoethyl ester follows the mechanism of Fischer esterification reaction. This reaction is an acid-catalyzed condensation of an alcohol and a carboxylic acid and then an ester and water are produced. In the synthesis of glutathione monoethyl ester sulfuric acid was used as a catalyst and excess absolute ethanol was used to shift the position of reaction equilibrium to favor the formation of ester. The mechanism is depicted in Figure 6.1. At the first step, the acid catalyst protonates the carboxylic acid group of the GSH glycine residue at its carbonyl oxygen. Then ethanol as a nucleophile attacks the protonated carboxylic acid to yield an intermediate The intermediate is unstable under the acid-catalyzed condition of its formation and undergoes dehydration to form the glutathione mono(glycyl)ethyl ester. The product is precipitated with a large amount of diethyl ether and recrystallized from a mixture of water and excess ethanol. The resulting product still contains HzS04 that interferes with the synthesis of NAGME. To remove this impurity an anion exchanger DE 53 was used. This weakly basic anion exchanger is based on the diethylaminoethyl tertiary amine functional group [ cellulose-O-(CHz)2N(C2Hs)z]. In the water solution of the product the functional group of DE 53 works like a weak base and becomes cellulose-0-25

PAGE 39

This group with a positive charge removes solions by binding them in noncovalent interaction. GSH monoester remains in the solution because of its zero overall charge. After DE 53 is filtered out, the sol-free solution is dried. The dry product is free base glutathione mono(glycyl)ethyl ester. The advantage of using DE 53 other than AG1X2 (OHform) to yield salt-free GEE is that DE 53 is a weakly basic anion exchanger. Even if excess DE 53 is added to the solution, this weak base can not influence the pH of the solution greatly. The pH of the solution was just changed from <4.5 to 6.0 during DE 53 treatment. By contrast, AG1X2 is a strongly basic anion exchanger. Its active group is R-CH 2N\CH3 ) 3 As a strong base, excess AG1X2 used for complete removal of solcan increase the pH of the solution greatly When AG1X2 (OHform) was used to remove sulfate in this experiment, it was found that the pH of the solution was changed from <4.5 to > 10. The resulting basic environment destroys the product because of deesterification and oxidation. Acid-catalysed esterification of GSH may lead to the formation of two forms of monoester, the glycyl and y-glutamyl monoesters, and a diester (28). After 18-22 hr esterification of GSH, the glycyl monoester was a major product (over 80% ). GSH diester and GSH each accounted for about 10% of the total thiol compounds detected by HPLC. The y-glutamyl monoester did not show up on HPLC spectrum. This is probably related to the neighbouring amino group of the y-glutamyl residue. Apparently glycine residue is more easily attacked by ethanol than glutamic acid 26

PAGE 40

residue, perhaps due to steric reasons. The interaction between the un-ionized carboxylate -COOH group and the ammonium -NH3 + group of the glutamic acid residue may also influence the process of acid-catalyzed esterification. The different pKa between two carboxylic acid is possible another reason to limit the esterification on the glutamic acid. Because the pKa of glutamic acid is lower than that of glycine acid, the protonation on it might be more difficult. The diethyl ester and residual GSH were removed during the purification. 27

PAGE 41

.. 0 IV'"""Cils R = Remainder of GSH Tetrahedral intermediate Figure 6.1 The Mechanism of the Acid-catalyzed Esterification of GSH (29) 28

PAGE 42

6.1.2 The Percent Yield The synthesis of glutathione mono(glycyl)ethyl ester hemihydrosulfate (GEEli2H2S04 ) produced 58%-63% yields. When the reaction was stopped about 10% of GSH and 10% of DIGEE were observed in the reaction solution. That means that not all of GSH had reacted with ethanol and some GEE had further been esterified to DIGEE. GEE was purified from the crude product by recrystallization in absolute ethanol. Because the product still contained the acid which was able to increase the solubility of product in ethanol, recrystallization and excess ethanol rinse resulted in loss of product in these acidic condition. Some product was also lost during the recovery from the flask and filter. The reaction temperature and the acid concentration also influence the percent yield of product. High temperature or high concentrations of acid increase the rate of the esterification, but result in a high yield of contaminating diester. Low temperature or low concentrations of acid decrease the rate of esterification. The optimal reaction conditions should include optimal reaction temperature and optimal acid concentration that result in high yield of product and low yield of by-product. The conditions as described resulted in relatively high yield of product and low yield of by-product. 73-80% of free base GEE was recovered after removal of solfrom GEEl/ZH2S04 with DE 53. After the removal was complete (BaCh test was 29

PAGE 43

negative), the DE53 was filtered and washed with 7 x 25 mL Milli Q water. But DTNB test showed that there was still some product adsorbed by DE 53 nonspecifically. To avoid generating a huge volume of solution which would make lyophilization more difficult, washing DE 53 was stopped when most of the thiol compound was recovered This results in loss of product. Some product was also lost during the process of transferring the dry product from the flask to a watchglass. It was observed that the color of the free base GEE product was pale pink other than white as expected. The color may be from metal contamination of DE 53. The pale pink color could not be removed by washing with ethyl ether. 6.1.3 HPLC Analysis HPLC was used to monitor the process of reactions, to determine the purity of the product and to analyze the extent of cell uptake ofNAGME. All samples for HPLC were first derivatized by monobromobimane (mBBr). Monobromobimane, a fluorescent labeling agent, is highly specific for reactive thiol groups. The reaction of monobromobimane (nonfluorescent) with thiol groups results in adducts which are highly fluorescent (30) The fluorescent label is stable to air, to chemical and biochemical procedures, and resistant to fading under intense irradiation (e.g. under examination with a fluorescence microscope). The reaction of a thiol with mBBr is illustrated below (Figure 6.2). This is a nucleophilic substitution reaction. RSgroup acts as a nucleophile to replace bromine of mBBr and form a 30

PAGE 44

thiol-bimane adduct. The adducts can then be detected by a fluorometer after they are separated and eluted by HPLC. N I N mBBr (N onfluorescent) R == Remainder of a thiol compound N I N Thiol-bimane adduct (Fluorescent) Figure 6.2 The Reaction of a Thiol with mBBr A C18 reverse-phase column was used to separate all GSH derivatives. This hydrophobic column causes compounds which are more hydrophobic and neutral to interact with the stationary phase more tightly. Compounds which are less hydrophobic and have net charges interact less strongly with the stationary phase and are eluted earlier from the column. In addition, the mobile phase also played a role in the separation of GSH and GSH derivatives. In this experiment, a linear gradient elution was applied. During the first 25 min of the elution, the solvent was linearly changed from 100% buffer A to 65% buffer A I 35% buffer B. Because buffer A was more polar than buffer B, the compounds which are less hydrophobic were eluted during this period. The polarity of the solvent was further reduced by increasing the 31

PAGE 45

concentration of buffer B from 35% to 100% over 10 min. The column was then washed with 100% buffer B for an additional 9 min. The compounds which were more hydrophobic or neutral were eluted during this 19 min. Under these conditions the retention times of GSH and its derivatives were: GSH 14 min, GEE 29 min, DIGEE 34 min, NAG 20 min, NAGME 30 min, NAGDI 38min. A blank containing all the derivative reagents was required to determine which peaks of each spectrum could be assigned to the reagents. After the reagent peaks were excluded, the peaks of GSH and GSH derivatives were identified by their specific retention times. Then the peak areas of all non-reagent peaks were totaled and a percentage of each thiol compound was calculated. For the synthesis of GEEl/2H2S0 4 HPLC analysis indicates about 10% of DIGEE, 10% GSH and 80% of GEE in the reaction solution (Figure 6.3), but little GSH or DIGEE in the fmal GEEl/2H2S04 product (Figure 6.4). Most of DIGEE and unreacted GSH were removed during recrytallization and rinse. It was determined by HPLC analysis that free base GEE was pure (containing <5% other thiols) (Figure 6.5). 32

PAGE 46

IN "IN ', 2000.00 1500.00 1000.00 l M I 500.00 -1 II ------II 0 II 0 ,.. N ,.. "' I I I I '---------I i ... ... A A A I I 0.00 o.oo Minutas Parcant 5 Figure 6.3 The HPLC Spectrum for Synthesis ofGEE/2H2S04 (The sample was from the reaction solution at 19 hr 40 min.) (Peak at 13.150 min, GSH; Peak at 29.617 min, GEE; Peak at 34.467 min, DIGEE; Other Peaks, reagents)

PAGE 47

I.IJ .,.. ..... 1200.00 1000.00 900.00 600.00 400.00 200.00 0.00 o.oo "' .. r1 "' 20.00 / .... ... r1 Minutes 40.00 Pe:ccent B Figure 6.4 The HPLC Spectrum for GEEl/2H2S04 Product (recrystallized) (Peak at 29 517 min, GEE; Other Peaks, reagents) i

PAGE 48

I.>) VI 1000.00 800.00 >-600. 00 E 400.00 200.00 o.oo o.oo Prcent B .... Minuto!r 40.00 Figure 6.5 The HPLC Spectrum for Free Base GEE Product (Peak at 29.567 min, GEE; Other Peaks, reagents)

PAGE 49

6.1.4 BaCh Test The result of BaCh test for solon GEE free base was negative. This indicates that DE 53 removed sulfate from GEEl/2H2S04 successfully. 6.1.5 DTNB Assay DTNB (see figure 6.6), known as Ellman's Reagent (27), is used to determine free thiol (-SH, reduced sulfhydryl) qualitatively or quantitatively. This is based on the specific reaction between DTNB and free sulfhydryl group The mechanism of this reaction is depicted in figure 6.7. The deprotonated SH (S-), acting as a nucleophile, attacks the disulfide bond of DTNB This results in a cleavage of the original disulfide and a formation of a new disulfide between the original thiol and thionitrobenzoate The other product is a new free thiol. This new free thiol, called thionitrobenzoate (TNB anion), is bright orange and absorbs electromagnetic radiation at 412 nm with an extinction coefficient of 13,600 M-1cm-1 The concentration of TNB anion is easily measured using an UVVisible spectrophotometer Since one mole of free thiol produces one mole of TNB anion the concentration of TNB anion reflects the amount of free thiols in the sample. This is the theoretical basis for the DTNB assay. 36

PAGE 50

HOOCS-S --COOH NO N02 2 Figure 6.6 The Structure of 5,5'Dithiobis(2-nitro)benzoic Acid (DTNB) HOOC COOH HOOC: Q -s-R w NO l NO NO Mixed disulfide 2 -R 2 2 s-R = Remainder of a thiol compound + TNB anion (bright orange) Figure 6.7 The Mechanism of the Reaction Between DTNB and a Free Thiol The reaction between DTNB and free thiol quickly takes place at a pH around 7. At extremely acid pH, the sulfhydryl group does not deprotonate well enough to be nucleophilic, and the TNB anion also does not deprotonate. At high basic pH, the disulfide bonds of DTNB may be broken and more TNB anions are formed (31). So low pH induces a false negative result, while a false positive result is obtained because of high pH. DTNB is light and temperature sensitive especially when it is in solution. The 37

PAGE 51

adduct which absorbs light at 412 run is also sensitive to light and temperature. DTNB is therefore susceptible to decomposition which can cause inconsistent results. To avoid these problems, DTNB was added directly to the cuvette and the solutions were immediately analyzed by lNVisible spectrophotometer. The DTNB assay verified the percentage of free thiol in GEEl/2H2S04 as 92-98% and 88-95% in free base GEE. 38

PAGE 52

6.2 Synthesis of N-acetylglutathione Mono(glycyl)ethyl Ester 6.2.1 Reaction Mechanism + .. :o o: (I) + HIT H CH:fOCCH 3 1 \3 r r3 + 1 r3 HO\ HO 0\ .. \ c==:o: c===o c==o I I I CH3 CH 3 CH3 Tetnihedral intermediate HI yH3 HI ciH 3 H I H+ I I (3) R--N-C-1 r1 \ c==o c===o:H I I'---'. CH3 CH 3 + 0 II Cfi.3C-OH R = Remainder of GEE Figure 6.8 The Mechanism of the Acetylation of GEE (29) The synthesis of NAGME from free base GEE follows the mechanism for 39

PAGE 53

acetylation of an amine with an anhydride. In this synthesis acetic acid was used as a catalyst and solvent and excess acetic anhydride was used to shift the position of reaction equilibrium to favor the formation of NAGME. The mechanism of this reaction is depicted in Figure 6.8. First, acetic anhydride is activated toward nucleophilic addition by protonation of one of its carbonyl groups. Then GEE, a primary amine (a nucleophile), adds to the protonated carbonyl group to form a tetrahedral intermediate. This tetrahedral intermediate is dissociated by the acid catalyst and produces NAGME and acetic acid. NAGME is precipitated by ethyl ether. Acetic acid, the only by-product (also the reaction solvent), is separated during the precipitation because it is soluble in ethyl ether where NAGME is not. The free base form of GEE was used to prepare NAGME. There are two reasons: (1) acetic anhydride, one of reagents in the synthesis of NAGME, is possibly hydrolyzed in the presence of the strong acid H2S04 from GEE/2H2S04 (2) solis a potential source of contamination for the NAGME product. It is more difficult to remove solfrom NAGME than from GEE by an anion exchanger because NAGME also has a net negative charge. 6.2.2 The Percent Yield The preparation of N-acetylglutathione mono(glycyl)ethyl ester (NAGME) from free base GEE produced yields of 56%-57%. Reducing the volume of the reaction mixture before the precipitation with ether resulted in over 10% increase in the percent yield during the preparation of NAG from GSH (32). Reducing the 40

PAGE 54

volume helps to remove excess solvent which may keep the product in the solution when ethyl ether is added. During the preparation of NAG ME, the volume was also reduced to about 1/3 volume of the original reaction solution. This resulted in a concentrated product solution. Since NAGME is insoluble in ethyl ether, without excess solvent most of product may be precipitated by diethyl ether. However, in the product solution there was still a high concentration of the acid, which might increase the solubility of NAGME in diethyl ether. This may result in loss of product. It was observed that the color of NAGME was white as expected. The pale pink color from free base GEE was removed after the preparation and purification of NAGME. 6.2.3 HPLC Analysis For the synthesis ofNAGME, HPLC analysis indicates that there was 100% ofNAGME in the reaction solution after 20-22 hr reaction (Figure 6.9). Because the retention time of NAGME (30 min) is close to that of GEE (29 min), a mixture of GEE standard solution and reaction solution was also run on HPLC during the experiment to confirm the synthesis of NAGME. The HPLC spectrum for this mixture clearly shows that there were two peaks at 29 min and 30 min respectively (Figure 6.10). Since there was just a peak at 30 min in the HPLC spectrum of the reaction solution, the peak at 29 min in the mixture was from GEE standard and the other peak at 30 min was NAGME. The peak at 30 min in the HPLC spectrum of the 41

PAGE 55

reaction solution was also identified as the peak of NAGME by acid-catalyzed hydrolysis. An aqueous solution of the product was mixed with concentrated HCl and stirred in a water bath (35-40 C). The hydrolysis of the reaction product was monitored by HPLC. It was observed that hydrolysis of the compound at 30 min resulted in a peak at 15 min. To identify this peak, the hydrolytic reaction was mixed with a GSH standard and analyzed by HPLC. A new peak, GSH, was observed at 12.5 min (data not shown). So the peak at 15 min was from NAG. This indicated that the compound at 30 min was NAGME, not GEE. (Because of leakage in HPLC injector, the retention times for NAG and GSH were shifted in this case.) The HPLC analysis for NAGME product indicates that the product was pure (containing <5% other thiols) (Figure 6.11). 42

PAGE 56

w . 800.00 600.00 200.00 0,00 o.oo 20.00 40.00 MinuteD l?ercont B Figure 6.9 The HPLC Spectrum for Synthesis ofNAGME (The sample was from the reaction solution at 21.5 hr) (Peak at 30.700 min, NAGME; Other peaks, reagents)

PAGE 57

t 900.00 600.00 400.00 200.00 0.00 0.00 20.00 ,.. ... ,.. 0 "' Minuto11 '---'-------1 40.00 Porcent D Figure 6.10 The HPLC Spect111m for Mixture ofNAGME Reaction Solution (at 21.5 hr) and GEE Standard Solution (Peak at 29.750 min, GEE; Peak at 30.767, NAGME; Other peaks, reagents)

PAGE 58

1600.00 1400.00 1200.0. 0 j 0 ..., "' 1000.00 -, .., :1! !; BOO. 00 600.00 400.00 .200.00 0.00 0.00 20.00 40.00 Minutos P:r:cant 8 Figure 6.11 The HPLC Spectrum for NAGME Product at 30.433 min, NAGME; Other peaks, reagents)

PAGE 59

6.2.4 DTNB Assay The DTNB assay verified that the percentage of free thiol in NAGME product was 84-89.5%. It is somewhat lower than that in reagent GEE (88-95%) because of oxidation during the synthesis. GEE has no net charge. The pH of its aqueous solution is around 7.0. Oxidation could happen in this neutral environment after GEE was dissolved in water. 6.2.5 NMR Analysis The proton magnetic resonance eH NMR) spectrum for NAGME product is shown in figure 6.12. It indicates that there was residual ethyl ether in the product. The peak assignments for this spectrum are summarized in table 6.1. The assigned structure for the product is shown in figure 6.13. 46

PAGE 60

Chemical Shift (ppm) Splitting Pattern Number of Character of Protons Protons 1.2 to1.3 Triplet 3 P-Ethyl 2.05 Singlet 3 Acetyl 1.9 to 2.3 Multiplet 2 f3-Giutamyl 2.45 to 2.55 Multiplet 2 y-Giutamyl 2.9 to 2.95 Doublet 2 f3-Cysteinyl 3.5 to 3.65 Multiplet N/A N/A 4.0 to 4.05 Singlet 2 Glycyl 4.15 to 4.25 Quadruplet 2 a-Ethyl 4.3 to 4.4 Multiplet 1 a-Giutamyl 4.5 to 4.6 Triplet 1 a-Cysteinyl Table 6.1 NMR Data for NAGME Product The protons bonded to oxygen, nitrogen and sulfur undergo exchange with deuterium of solvent D20. Because deuterium does not give a signal under the conditions of proton nmr spectroscopy, exchanges of these protons by deuterium lead to the disappearance of their nmr peaks. So only the protons bound to carbon generate the peaks. TMSPA as an internal reference is shown at 0 8. The triplet at 1.2 to 1.3 8, which integrates for 3 protons, is caused by the methyl protons of the ester. The singlet at 2.05 8, which integrates for 3 protons, arises from the resonance of the protons of the acetyl group. The complex two-proton resonance at 1.9 to 2.3 arises from p-glutamyl proton. The two protons bound to f3-glutamyl carbon have different chemical shifts as a result of their magnetic inequivalence. The peaks at 2.45 to 2.55 47

PAGE 61

8, which integrate for 2 protons, are caused by y-glutamyl protons. The doublet at 2.9 to 2.95 8, which integrates for 2 protons, arises from P-cysteinyl protons. The tiny peaks at 3.5 to 3.65 8 are suspected as ethyl ether. The singlet at 4.0 to 4.05 8 which integrates for 2 protons, is caused by glycyl protons. This peak is shifted downfield to 4.0 8 due to the deshielding of the vicinal carbonyl and amide groups. The two proton quadruplet resonance at 4.15 to 4.25 8 arises from the methylene protons of the ester. The peaks at 4.3 to 4.4 8, which integrate for 1 proton, are from the resonance of the a-glutamyl proton. Since the proton is deshielded by the carbonyl group and amide nitrogen, its signal is shifted downfield. The triplet at 4.5 to 4.6 8, which integrates for 1 proton, is caused by the a-cysteinyl proton. This proton is also deshielded by its vicinal carbonyl group and amide nitrogen and shifted downfield. The sulfhydryl group of NAG ME is possibly acetylated by acetic anhydride during the reaction. The singlet at 2.39 8 is suspected from the resonance of the protons of the s-acetyl group. 48

PAGE 62

'Tl .... ?' ..... N :I: z a: Cl:l (1) (') 2 8 a N 0 0 a: :I: N g, z > 0 a: trl "'0 a 0.. s:: (') .... .... ::s 0 N 0 .... "' f" ... w "' w ... N "' N ... ... "' ... .. .. "' C> .. 6P a-Cysteinyl (12 I a-Glutamyl (I) a-ethyl (22 Glycyl (22 i I \ \ 1 i 13-Cysteinyl {22 y-Glutamyl {22 f3-ethyl{32 I I l I I

PAGE 63

CI-I3 Acetyl I o/'-' 0 !'-..
PAGE 64

6.3 Cell Uptake Experiment 6.3.1 HPLC Analysis for the Media The results ofHPLC analysis for the media are in Table 6.2 and Table 6.3. Thiol applied Levels ofthiols in media (mM) (well#) GSH NAG NAG ME None(PBS) (1) N.D. N.D. N.D. None(PBS) (2) N.D. N.D. N.D. GSH (3) 12.3 N.D. N.D. GSH (4) 11.1 N.D. N.D. NAGME(5) N.D. N.D. 12.3 NAGME(6) N.D. N.D. 12.6 Table 6.2 The Levels ofThiols in Medium Control Wells (N.D.= not detected) Thiol applied Levels ofthiols in media (mM) (well#) GSH NAG NAG ME None(PBS) (7) N.D. N.D. N.D. None(PBS) (8) N.D. N.D. N.D. GSH (9) 9.5 N.D. N.D. GSH (10) 10.4 N.D. N.D. NAGME (11) N.D. N.D. 12.4 NAGME(12) N.D. N.D. 12.4 Table 6.3 The Levels ofExtemal Thiols in Experimental Wells (N.D.= not detected) 51

PAGE 65

The results in Table 6.2 indicate that GSH and NAGME were stable in the medium. They were not degraded by the medium during the course of the experiment. The results in Table 6.3 show that the concentration of GSH or NAGME was not substantially changed compared to the concentration of GSH or NAGME in medium control. This indicated that nothing on the outer cell surface metabolized or degraded the thiols during the incubation. 6.3.2 HPLC Analysis for Cell Lysates The results ofHPLC analysis for celllysates are in Table 6.4. Thiol applied Levels ofthiols in cells (nmol/10 cells) (well#) GSH NAG NAG ME None(PBS) (7) 54.6 N.D. N.D. None(PBS) (8) 62.0 N.D. N .D. GSH (9) 62.4 N.D. N.D. GSH (10) 44.5 N.D. N.D. NAGME(11) 44.8 N.D. N.D. NAGME(12) 45.5 N.D. N.D. Table 6.4 The Levels of Cellular Thiols in Experimental Wells (N.D.= not detected) Few dead cells (<10%) were found during counting. The numbers of cells found in each well ranged from 0.56 x 106 to 1.32 x 106 No correlation between cell number and thiol treatment was found, so the cell counts in all 6 wells were averaged in calculating the thiol contents in Table 6.4. 52

PAGE 66

After the 2 hr incubation, no other thiol except GSH was found in PBS, GSH or NAGME treated cells. In comparison with the level of GSH in PBS treated cells the level of GSH in GSH or NAGME treated cells did not show an increase. It indicated that treatment with either GSH or NAGME failed to increase the level of GSH in cells. The HPLC spectrum for the lysate of NAGME treated cells also did not show any peaks which corresponded to NAGME or any products from degraded NAGME, such as NAG. It indicated that NAGME failed to be taken up or metabolized by cells. Because NAGME was stable in the medium during the time of the experiment as described above, the failure of NAGME cell uptake was not attributed to GSH formation through extracellular deacetylation and deesterification [GSH is not effectively transported into most types of cells under a variety of experimental conditions (9)]. The reason for the failure of NAGME cell uptake perhaps is NAGME itself. NAGME has net charge of -1 at physiological pH. With the negative charge NAGME might not diffuse through the cell membrane. The other reasons might be: (1) 2 hr incubation was too short for NAG ME to permeate through cell membrane; (2) there is no carrier on the membrane of the J82 cell line which can mediate NAGME transport into cells. The NAGME-treated cells actually show a decrease in GSH level compared to PBS-treated control. However, one of the two GSH-treated wells shows a similar decrease. Therefore, no definite conclusions can be drawn without additional study. 53

PAGE 67

7. Conclusion and Further Study The cell uptake experiment showed that NAGME could not increase the level of GSH or the level of other thiols in J82 cells. It indicated that NAGME was not taken up by J82 cells. Although J82 cells did not take up NAGME after 2 hr incubation, the probability can not be excluded for other cell lines to have specific mechanism to transport NAGME into cells. 2 hr incubation may also be too short for the transport of NAGME. Other cell lines or longer incubation can thus be attempted to further study the transport of NAGME. Other cell lines may include cisplatin-sensitive cancer cell line, such as ovarian cancer cell line, or normal cell line. Other derivatives of GSH can be designed to make them diffuse through cell membrane and increase the cellular level of GSH or thiol in further study. The hydrophobicity of GSH derivatives seems critical to their transport into cells. Esterified GSH with higher alcohol might be a strategy to make derivatives hydrophobic. Butyl alcohols could be candidates in further study to make GSH butyl monoesters or butyl diesters. Uptake of them into cells would then be explored. 54

PAGE 68

8. Failures 8.1 Synthesis of NAG ME from NAGDI Synthesis of NAGDI, the N-acetylated diethyl ester of GSH, is easy and results in high yields. A synthesis of NAGME was attempted via hydrolysis of NAGDI. Hydrochloric acid was used as catalyst. The process of hydrolysis was monitored by HPLC. The hydrolysis was stopped when the reaction mixture contained about 42% NAGDI, 45% NAGME and about 13% NAG. To separate NAGME from this mixture, this mixture was first neutralized to pH 5.5 by 10M NaOH. At pH 5 5 NAGDI has no charge, NAGME has one negative charge while NAG has two negative charges The mixture was then loaded on a DE 53 column. NAGDI was neutral and was first eluted by water. NAGME and NAG were anionic and should be bound by DE 53. After elution of NAGDI was done NAGME and NAG effluent would be collected respectively by switching the solvent from water to 0.1 M HCI. The fractions of elution were monitored by HPLC and DTNB. It was found that all water effluents contained NAGDI, NAGME and NAG, and DTNB tests for water effluent were brown-yellow. The acid effluents were also mixtures of NAGDI, NAGME, and NAG, but DTNB tests for them were yellow This indicated that most of NAGDI, NAGME and NAG were eluted together by water. DE 53 did not bind NAGME and NAG tightly as expected. The salt in the reaction mixture might also influence the separation by interfering with the binding between DE 53 55

PAGE 69

and NAGME or NAG. To avoid the interference of the salt, Bio-Gel Pz was tried to remove them. Before loaded on the DE 53 column, the reaction mixture was first loaded on a Bio-Gel P2 column. All fractions of elution with positive DTNB tests were collected and monitored by HPLC. Unfortunately just NAG, no NAGME or NAGDI was found in the effluent. NAGME and NAGDI might be hydrolyzed to NAG during the elution. 8.2 Preparation of GEE Free Base with AG1X2 (Bicarbonate Form) Resin To avoid the interference of the acid in GEEli2HzS0 4 when NAGME is synthesized, solhas to be removed from GEEli2H2S04 to make free base GEE. AG 1X2 (bicarbonate form) resin was first tried to remove solfrom an aqueous solution of GEEl/2HzS04. Resin was added to GEE-112H2S04 solution and a negative BaC}z test was obtained. The BaC}z test for the resulting product was negative and the thiol percentage of the product was 96%, however, the yield was low (about 20%). To remove all of SO/-, excess AG1X2 has to be used. One reason for the low yield may be the excess AG1X2 resin which still carried a lot of product away after the resin was washed with water to recover the product. 8.3 Direct Preparation of GEE Free Base Direct preparation of GEE free base has been reported by Campbell et al (33). The method uses a higher concentration of HzS04 and a larger excess of ethanol. The esterification was therefore quickly completed within 4-6. When the synthesis was complete, AG1X2 resin in the OHform was added directly to the 56

PAGE 70

reaction mixture for sulfate removal and acid neutralization. When at least two consecutively negative BaCh tests for the supernatant were obtained, the resin was filtered and rinsed with ethanol. The filtrate was kept in a freezer to crystallize GEE free base After several days, sticky (not fluffy as described by Campbell) precipitate was observed. Ethyl ether was then applied to wash the product. Unfortunately, the yield was very low (about 20% on average) and not reproducible. The product still had sulfate and the thiol% of the product was below 80% which is not acceptable for further synthesis. To remove all sulfate from the product excess AG1X2 was used. But monoester began to crystallize in the presence of excess resin. The yield of product obtained from the filtrate was correspondingly decreased. Another problem might be from the hydroxide form of AG1X2. A lot of OHgroups which were replaced by solgroups possibly made the solution basic. Monoester in this basic environment was more likely hydrolyzed or oxidized. That is why the yield and thiol % were low. 8.4 Reduction of NAG ME with DTT Because the degree of oxidation of the final NAGME product was generally greater than 10 %, attempts were made to reduce the product. In this experiment, dithiothreitol [DTT, HSCH2CH(OH)CH(OH)CH2SH] was used to reduce batches of NAGME which had about 84% thiol percentage. First, NAGME was dissolved in water, and then mixed with an aqueous solution of DTT. In the reaction solution, the concentration of NAGME was 10 mM and the 57

PAGE 71

concentration ofDTT was 100 mM. After 15 min reaction in the dark, the reaction solution was transferred to a separatory funnel. Then ethyl acetate was added to extract DTT. After extraction 5 times, the water layer that contained NAGME was lyophilized. NAGME has a chain containing 12 carbon, 3 nitrogen and 1 oxygen atom and has no charge at low pH. So it is likely hydrophobic and somewhat soluble in ethyl acetate. This resulted in a low yield (about 25%). To increase the solubility ofNAGME in water, the pH ofNAGME solution was adjusted to pH 5.5 with 10M NaOH. The pH adjusted NAGME solution was then mixed with DTT solution. After 15 min reaction in the dark, the extraction was performed as before. Due to pH adjustment NAGME was not found in ethyl acetate layer and the yield went up to 100%. But the product became sticky after exposure to air. To solve this problem, the product was dissolved in minimum absolute ethanol and precipitated, and then washed with ethyl ether. Unfortunately, the resulting product still became sticky after exposure to air and thiol percentage of this product was lower than 80%. Thus the method was modified again to solve these problems (dryness and oxidation). 1 mM DTP A was added to NAGME water solution before neutralization to remove residual metal that was possibly the oxidizing agent. NAGME solution was then neutralized to pH 5.5 with 10M NaOH and mixed with DTT solution. After 15 min reaction in the dark, DTT was extracted by ethyl acetate. The resulting water layer (containing NAGME and DTP A) was acidified to pH 2.0 with 0.1 M HCl to make NAGME hydrophobic. Then ethyl acetate was added to extract NAGME in this solution while 58

PAGE 72

DTP A was expected to stay in water layer (but DTP A was also possibly hydrophobic enough in the acid form to be extracted by ethyl acetate and contaminate the product). NAGME was obtained by drying ethyl acetate layer under a flow of N2. The yield was just 50% because even in the acid environment NAGME is still more soluble in water than in ethyl acetate. But the product was dry even exposed to air (the previous product became sticky after exposure to air). The problem of dryness was overcome, but oxidation still took place. The thiol percentage of the product was around 75% which was below 84% of starting material. Oxidation might happen during the extraction in the absence of DTT or during the acidification of water layer. 59

PAGE 73

References 1. Petsko, G. A. (1995) Cancer Chemother. 377, 580-581 2. Rosenberg, B. (1985) Cancer 55,2303-2316 3. Bellon, S. F., Coleman, J. H., and Lippard, S.J. (1991) Biochemistry 30, 80268035 4. Dedon, P. C. and Borch, R. F. (1987) Biochemical Pharmacol. 36, 1955-1964 5. Chu, G., Mantin, R., Shen, Y.-M., Baskett, G., and Sussman, H. (1993) Cancer 72, 3707-3714 6. Chu, G. (1994) J. Bio. Chern. 269,787-790 7. Andrews, P. A. and Howell, S. B. (1990) Cancer Cells 2, 35-43 8. Pinzani, V., Bressolle, F., Haug, I. J., Galtier, M., Blayac, J. B., and Balmes, P. (1994) Cancer Chemother Pharmacol. 35, 1-9 9. Meister, A. (1991) Pharmac. Ther. 51, 155-194 10. Arrick, B. A. and Nathan, C. F. (1984) Cancer Res. 44,4224-4232 11. Ishikawa, T. and Ali-Osman, F. (1993) J. Biol. Chern. 268, 20116-20125 12. Godwin, A. K., Meister, A., O'Dwyer, P., Huang, C. S., Hamilton, T. C., and Anderson, M. E. (1992) Proc. Natl. Acad. Sci. USA 89,3070-3074 13. Meijer, C., Mulder, N.H., Hospers, G. A. P., Uges, D. R. A., and de Vries, E. G. E. (1990) Br. J. Cancer 62, 72-77 60

PAGE 74

14. Hormas, R. A., Andrews, P. A., Murphy, M.P., and Burns, C. P. (1987) Cancer Lett. 34, 9-13 15. Richon, V. M., Schulte, N. A., and Eastman, A. (1987) Cancer Res. 47, 20562061 16. Andrews, P. A., Murphy, M.P., and Howell, S. B. (1985) Cancer Res. 45, 62506253 17. Andrews, P. A., Schiefer, M.A., Murphy, M.P., and Howell, S. B. (1988) Chern. Bioi. Interact. 65, 51-58 18. Eastman, A. (1987) Chern. Bioi. Interact. 61, 241-248 19. Anderson, M. E. and Meister, A. (1989) Anal. Biochern. 183, 16-20 20. Levy, E. J., Anderson, M. E., and Meister, A. (1993) Proc. Natl. Acad. Sci. USA 90,9171-9175 21. Levy, E. J., Anderson, M. E., and Meister, A. (1994) Methods Enzymol. 234, 499-505 22. Anderson, M. E., Naganuma, A., and Meister, A. (1990) FASEB J. 4, 3251-3254 23. Peters, R. H., Jollow, D. J., and Stuart, R. K. (1991) Cancer Res. 51, 2536-2541 24. Peters, R. H., Ballard, K., Oatis, J. E., Jollow, D. J., and Stuart, R. K. (1990) Cancer Chernother. Pharmacol. 26,397-402 25. Minhas, H. S. and Thomalley, P. J. (1995) Biochem. Pharmacol. 49, 1475-1482 26. Anderson, M. E., Levy, E. J., and Meister, A. (1994) Methods Enzymol. 234, 492-499 61

PAGE 75

27. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70-77 28. Thomalley, P. K. (1991) Biochem. J. 275, 535-539 29.Carey, F. A. (1987) Organic Chemistry (Mcgraw-Hill, Inc), pp. 575-578 30. Gainer, H. and Kosower, N. S. (1980) Histochem. 68, 309-315 31. Danehy, J.P. and Hunter, W. E. (1967) J. Org. Chern. 32,2047-2053 32. White, T. K. (1997) The synthesis of N-acetylglutathione and N acetylglutathione diethyl ester and cellular uptake study (Senior Honors Thesis), pp.ll 33. Campbell, E. B. and Griffith, 0. W. (1989) Anal. Biochem. 183, 21-25 62