UNIVERSITY OF CINCINNATI
______, 20 _____
I,______, hereby submit this as part of the requirements for the degree of:
______in: ______It is entitled: ______
Approved by: ______
Role of Glutamate-Cysteine Ligase in Maintaining Glutathione
Homeostasis and Protecting against Oxidative Stress
A dissertation submitted to the
Division of Research and Advanced Studies University of Cincinnati
in partial fulfillment of the requirements for the degree of
DOCTORATE OF PHILOSOPHY (Ph.D.)
in the Department of Environmental Health of the College of Medicine
2003
by
Yi Yang
M.D., Sun Yat-sen University of Medical Sciences, 1995 M.S., Sun Yat-sen University of Medical Sciences, 1997
Committee Chair: Daniel W. Nebert, M.D. Professor Department of Environmental Health University of Cincinnati
Abstract
Glutamate-cysteine ligase (GCL) is the rate-limiting enzyme catalyzing the first step of
glutathione (GSH) biosynthesis. In higher eukaryotes, this enzyme is a heterodimer comprising a
catalytic subunit (GCLC) and a modifier subunit (GCLM); the latter changes the catalytic
characteristics of the holoenzyme. In the first part of this dissertation, the heterodimer formation
between GCLC and GCLM was investigated, using the yeast two-hybrid system and affinity
chromatography. A strong and specific interaction between GCLC and GCLM was observed in
both systems. Deletion analysis showed that most regions, except part of the C-terminal region in
GCLC and part of the N-terminal region in GCLM, are required for the interaction to occur.
Point mutations on selected amino acids were also tested for the binding activity. GCLC C248A-
C249A and P158L mutants had the same strength of binding to GCLM as the full-length GCLC,
yet the catalytic activity was dramatically decreased in these mutations. The functions of GCLC
and GCLM were further studied in knockout mouse lines with the targeted disruption of the gene
encoding either subunit. Deletion of the Gclc gene specifically in hepatocytes led to a ~98% loss
of GSH in liver and plasma. The affected livers showed significant oxidative damage,
mitochondrial degeneration, and hepatocyte death. These mice died of hepatic failure around post-partum day 30 (d30). Supplementing N-acetylcysteine in the drinking water, starting at d21, was able to prevent early death; however, these mice developed pathology characteristic of liver
cirrhosis. Gclm(-/-) mice, on the other hand, showed no overt phenotype. However, GSH levels
decreased to 9-16% of normal in tissues and plasma. Compared to the GCL holoenzyme partially
purified from livers of Gclm(+/+) mice, hepatic GCLC in Gclm(-/-) mice showed a ~2-fold
increase in Km for glutamate and a dramatically enhanced sensitivity to GSH inhibition. The major decrease in GSH, combined with diminished GCL activity, rendered Gclm(-/-) fetal fibroblasts more strikingly sensitive to chemical oxidants such as hydrogen peroxide. These data demonstrate that both GCLC and GCLM are critical in maintaining GSH homeostasis.
Disruption of either component is strikingly detrimental to the cell’s or animal’s protection system against oxidative stress.
Acknowledgments
I would like to thank members of my dissertation committee, Drs. Daniel W. Nebert, Timothy P.
Dalton, Howard G. Shertzer, and Anil Menon for all the suggestions and criticisms. I especially want to express my sincere gratitude to my advisor, Dr. Dan Nebert, and my lab mentor, Dr.
Timothy Dalton, for the wonderful opportunity to work on this project, for the advice, support, and encouragement over the past 6 years. You have made my graduate education a fruitful experience.
My gratitude also goes to Dr. Alvaro Puga and Dr. Mario Medvedovic for the advice and inspiration during my graduate study. I would also liker to thank Dr. Marian Miller for her technical assistance.
Special thanks to the former graduates from our lab, Dr. Matthew Z. Dieter and Dr. Willy A.
Solis, for sharing the good and bad times in our graduate student lives, for valuable discussions and suggestions, and for the friendship and encouragement. I would also like to thank for the rest of the members in my lab for various help. You have made the lab an enjoyable place to work.
My deepest gratitude goes to my father, Zhilin Yang, my mother, Jinzhong Yang, and my brother, Genghua Yang. Your unconditional love, your endless support, and your tremendous sacrifice have always been the greatest resources of my motivation. I also want to thank my husband, Feng Wang, for your love, patience, understanding, and encouragement. Nothing would have been possible without a strong support from all of you, my dear family. 1
Table of Contents
List of tables ………………………………………………………………………………...…. 3
List of figures …………………………………………………………………………………...4
Abbreviations ………………..……………………………………………………………..….. 5
Introduction ……………….…………………………………………………………….…...… 6
Chapter I
Characterization of the Interaction between Glutamate-Cysteine Ligase Subunits
Abstract ……….……………………………………………………………………………… 21
Introduction ……………….………………………………………………………………..… 22
Materials and Methods ………………………………………………………………..……… 23
Results ………………..………………………………………………………………….…… 26
Discussion ………………………………………………………………………………..…... 29
References ……………………………………………………………………………….…… 32
Chapter II
Hepatocyte-Specific Knockout of Glutamate-Cysteine Ligase Catalytic Subunit Gclc(-/-) in
Mouse: Early Death with Progressive Liver Degeneration and Rescue by N-acetylcysteine
Abstract …………………………………………………………………………….………. 47
Introduction ………………………………………………………………………………….. 49
Materials and Methods ………………...……………………………………………….……. 51
Results ………………………………………………………….…………………….……… 54
Discussion ………………………………………………………………..…………….……. 59 2
References ……………………………………………………………………….………….. 63
Chapter III
Initial Characterization of the Glutamate-Cysteine Ligase Modifier Subunit Gclm(-/-) Knockout
Mouse: Novel Model System for a Severely Compromised Oxidative Stress Response ….... 84
Discussion …………………………………..…………………………………….…………. 85
References for introduction and discussion sections …….…………………….……………. 93
3
List of Tables
Chapter I
Table 1 - Plasmids constructs used in this study ……………………………………………. 39
Table 2 - Enzymatic activity of wild-type and mutant GCLC ……………………………… 40
Chapter II
Table 1 - GSH levels in tissues of Gclc(f/f) mice and Gclc(h/h) mice ………………..…..... 74
Table 2 - Mitochondria GSH levels in livers from Gclc(f/f) mice and Gclc(h/h) mice …….. 75
Table 3 - Cysteine levels in tissues of Gclc(f/f) mice and Gclc(h/h) mice ……………..…... 76
Table 4 - Liver biochemical functions in Gclc(f/f) mice and Gclc(h/h) mice …..……..…… 77
4
List of Figures
Chapter I
Figure 1 - GCLC and GCLM interaction in yeast …………………………………………… 41
Figure 2 - Interaction of GCLC and GCLM on Ni-NTA resin ……………………………… 42
Figure 3 - GCLC deletion studies in yeast two-hybrid system ……………………….....….... 43
Figure 4 - GCLM deletion studies in yeast two-hybrid system ……………………………… 44
Figure 5 - Interaction of GCLC mutations with GCLM ……………………………….…….. 45
Chapter II
Figure 1 - Generation of Gclc floxed mice ……………………………………………..…….. 78
Figure 2 - Hepatocyte-specific conversion of Gclc(f) allele to Gclc(h) allele ……………...… 79
Figure 3 - Liver histology of Gclc(f/f) mice and Gclc(h/h) mice …………………………….. 80
Figure 4 - Liver morphology and histology of Gclc(f/f) mice and Gclc(h/h) mice after NAC supplement ………………………………………………………………………………….... 81
Figure 5 - Liver mRNA levels in Gclc(f/f) and Gclc(h/h) mice ……………………………… 82
Figure 6 - Lipid peroxidation in livers of Gclc(f/f) mice and Gclc(h/h) mice …...……..…….. 83
5
Abbreviations
ALT - alanine aminotransferase
AST- aspartate aminotransferase
BSO - buthionine sulfoximine
γ-GC - γ-glutamylcysteine
GCL - glutamate-cysteine ligase
GCLC – glutamate-cysteine ligase catalytic subunit (Gclc = mouse gene)
GCLM – glutamate-cysteine ligase modifier subunit (Gclm = mouse gene)
GD - gestational day
GGT - γ-glutamyltranspeptidase
GSH - reduced glutathione
GSH-EE – glutathione monoethyl ester
GSSG - glutathione disulfide
GPX - glutathione peroxidase
H2O2 - hydrogen peroxide
MFF – mouse fetal fibroblasts
NAC – N-acetylcysteine
. - O2 - superoxide anion
.OH – hydroxyl radical
ROS - reactive oxygen species
SOD - superoxide dismutase
TBARS - thiobarbituric acid-reactive substances
6
Introduction
Oxygen is essential for most forms of life. Its reduction to water provides the energy that
allows virtually all complex functions to proceed in aerobic organisms. However, oxygen can be
potentially dangerous if not completely reduced. The reactive oxygen species (ROS), such as
. - . superoxide anion ( O2 ), hydrogen peroxide (H2O2), and hydroxyl radical ( OH), can arise from
normal metabolic reactions as well as environmental insult. The accumulation of ROS may
perturb the cell’s natural antioxidant defense systems, resulting in damage to a variety of
biochemical and physiological processes. Oxidative stress, by definition, refers to a disturbance
in the prooxidant-antioxidant balance in favor of the former [1]. Oxidative stress has been
implicated in a variety of diseases and degenerative conditions such as aging, inflammation,
carcinogenesis, diabetes, cataract, and neurodegenerative diseases [2-5].
The potential toxicity of oxygen arises from its molecular properties. In fundamental
chemical terms, molecular oxygen is a highly reactive bi-radical because of its two unpaired
electrons. Addition of a single electron to O2 yields the superoxide anion. This radical is not
particularly reactive relative to other species. However, either by spontaneous action or through
enzymatic action of superoxide dismutases, it is converted into H2O2 and O2 [6,7]. H2O2 can also be formed directly from O2 through the two-electron transfer by a number of oxidases (e.g. urate
oxidase, monoamine oxidase, and xanthine oxidase). Compared with oxygen radicals, H2O2 is
relatively stable and can easily diffuse across cellular membranes. In the presence of reduced
transition metal ions such as copper or iron, H2O2 can undergo reduction to the hydroxyl anion via the Fenton reaction, yielding the hydroxyl radical [7]. This radical can also arise from the hemolytic fission of hydrogen peroxide by ultraviolet light, the hydrolysis of water by ionizing 7
radiation, or the metabolic reactions of certain chemicals [7]. The .OH is probably the most
potent oxygen radical and can react non-specifically with any biological molecules.
Sources of Reactive Oxygen Species (ROS)
There appear to be four key sources in aerobic organisms to generating ROS:
mitochondrial electron transport, peroxisomal metabolism, cytochrome P450 reactions, and
phagocytotic cells. In mitochondria, the respiration chain normally involves a coordinated four-
electron reduction of O2 to H2O. Starting from NADH or succinate, the electrons are first
transferred to complexes I and II. Ubiquinone, which accepts electrons, undergoes two sequential
one-electron reductions to ubisemiquinone and ubiquinol. Ultimately, the reducing equivalents
are transferred to the remainder of the electron transport chain: cytochrome b (complex III),
cytochrome c, cytochrome oxidase (complex IV), and finally, O2. However, the sequential
electron transport in mitochondria is imperfect, and one-electron reduction of O2 also occurs,
generating superoxide anion [8]. Even under tightly coupled conditions, as much as 2-4% of the reducing equivalents will escape the respiratory chain to generate superoxide. This free radical
can then form hydrogen peroxide and hydroxyl radical through the reactions discussed above.
Therefore, the mitochondria are a major source of endogenous ROS under normal cellular
conditions.
Peroxisomes are another source of oxidative metabolites in the cell. Peroxisomes contain
a number of H2O2-generating oxidases, including fatty acyl-CoA oxidase and urate oxidase.
These enzymes remove two electrons from substrates and ultimately transfer the electrons to O2 to form H2O2, which can be decomposed locally by catalase. It is unclear whether leakage of
H2O2 from peroxisomes contributes significantly to cytosolic oxidative stress under normal
circumstances. However, a class of nonmutagenic carcinogens––the peroxisome proliferators–– 8 may increase the number of peroxisomes and cause enhanced expression of several oxidases [9].
Such disproportionate increases in H2O2-generating enzymes and H2O2-degrading enzymes often lead to increased oxidative stress in exposed cells.
The cytochrome P450 enzymes also generate generous portions of ROS in the cell. These enzymes are primarily mixed-function oxidases found in the microsomes. The endogenous substrates of these enzymes include fatty acids, steroids, prostaglandins, leukotrienes, and thromboxanes [10,11]. In addition, the P450s are able to act on a variety of xenobiotic compounds. The catalytic mechanism of cytochrome P450s involves several steps of one- electron transfers among NADPH, the substrates, and O2. Such steps, if uncoupled, may generate superoxide anion and H2O2 as by-products [10]. Moreover, certain cytochrome P450 substrates, notably paraquat and doxorubicin, undergo redox cycling by accepting single electrons from cytochrome P450 and transferring them to oxygen [12,13]. This step leads to the generation of superoxide anion and simultaneously recovers the substrate, allowing subsequent rounds of ROS generation.
Lastly, neutrophils and macrophages are able to enhance the uptake of oxygen and use it to generate a mixture of free radicals, including superoxide anion, H2O2, and nitric oxide [14].
These oxidants are part of the defense mechanism to allow immune cells to kill invading pathogens. Under conditions of chronic inflammation, such “respiratory bursts” may constitute a major source of endogenous oxidants.
In addition to these four sources of ROS, there are numerous other enzymes capable of generating oxidants under normal or pathological conditions, often in a tissue-specific manner.
For example, deprivation of oxygen in vascular endothelium cells leads to the conversion of xanthine dehydrogenase to its oxidase form and increases the availability of hypoxanthine from 9
ATP degradation [15]. Hypoxanthine can be oxidized by xanthine oxidase when the tissue is re-
oxygenated, causing rapid generation of superoxide anion and H2O2, which might in turn
enhance the tissue damage.
Cellular and Subcellular Targets of ROS
ROS are highly dangerous in living organisms, because they can attack almost any
cellular structure or macromolecule. When the reaction is with DNA, ROS can cause DNA-
protein crosslinks, damage to the deoxyribose-phosphate backbone, and form chemically
modified purine and pyrimidine bases [16]. The oxidative base modifications may result in mutations, whereas oxidation of deoxyribose moieties may induce base release or DNA strand breaks. Oxidative DNA lesions are both cytotoxic and mutagenic. They contribute substantially to the process of aging, tumor generation, and cardiovascular diseases [17,18].
ROS also react with cellular membranes and cause changes in membrane fluidity and permeability [19]. In addition, the resulting lipid peroxides can easily decompose into several reactive species, including lipid alkoxyl radicals, aldehydes, alkanes, lipid epoxides, and alcohols.
These radicals are themselves toxic and may serve as second messengers for oxidative damage
[20].
Proteins represent another key target for ROS. The structural as well as functional groups of proteins can be modified through interactions with ROS or the lipid oxidation products. Such oxidative damage not only leads to the loss of protein functions, but also contributes to secondary damage to related cellular processes [21,22].
The “Oxidative Stress Response” by the Organism
At the cellular levels, ROS and the oxidative macromolecules interfere with the normal cellular processes––leading to mitochondria dysfunction, apoptosis, and necrosis [23,24]. 10
Ultimately, such damage may lead to a variety of clinical manifestations. To protect against
oxidative damage, cells have evolved a battery of defense mechanisms involving both enzymatic
and non-enzymatic strategies. The enzymatic scavengers include superoxide dismutase (SOD),
catalase, and several glutathione peroxidases (GPXs) [25]. SOD converts the superoxide radical
to hydrogen peroxide. In eukaryotic cells, SOD exits in three forms: one with manganese in
mitochondria (SOD2), one with copper and zinc (SOD1) in the cytoplasm, and one extra-cellular
Cu/Zn-SOD (SOD3) immunologically distinct from SOD1. Both catalase and GPXs can
decompose hydrogen peroxide to water. Catalase is compartmentalized in the peroxisome and is
particularly efficient in removing high concentrations of H2O2 because of the high Km for H2O2.
The normal low rate of cellular production of H2O2 seems to be mainly dealt with by GPXs,
which uses glutathione (GSH) as a proton donor. In the process, GSH is oxidized to GSSG.
Mitochondria lack catalase (except in heart) and depend solely on GPXs to remove hydrogen
peroxide [26].
The non-enzymatic antioxidants include: (1) hydrophilic radical scavengers such as GSH,
ascorbic acid, and urate; (2) lipophilic radical scavengers such as tocopherols, flavonoids,
carotenoids, and ubiquinol; (3) antioxidant minerals such as copper, zinc, and selenium. All
these antioxidants work in synergy with one other and against different types of ROS [26,27]. To
maintain the reduced form of these antioxidants, cells are also equipped with enzymes involved
in the regeneration of oxidized forms of small molecular antioxidants (GSH reductase,
dehydroascorbate reductase) or the reducing equivalent (glucose-6-phosphate dehydrogenase).
Among the sophisticated cellular defense systems, GSH is clearly one of the most
important antioxidants. It not only directly scavenges ROS, but also is responsible for
maintaining the essential thiol status of many proteins. In addition, GSH participates in 11 enzymatic antioxidant reactions by providing the reducing equivalents. The reminder of this dissertation will describe the function of GSH in more detail.
Reduced (GSH) and Oxidized (GSSG) Glutathione
Glutathione, a tripeptide of L-glutamate, L-cysteine, and glycine, is ubiquitously present in animals, plants, and microorganisms [28-31]. It contains an unusual γ-peptide bond between glutamate and cysteine, which prevents glutathione from being hydrolyzed by intracellular peptidases. The key functional element of the GSH molecule is the cysteinyl moiety that provides the reactive thiol group and allows it to participate in redox reactions or conjugate to electrophiles. In mammals, GSH is the most highly concentrated intracellular antioxidant, with cellular levels ranging between 1 and 10 mM. Almost 90% of cellular GSH is located in the cytosol, 10% in the mitochondria, and a small percentage in the endoplasmic reticulum [32]. In the mitochondria, GSH is particularly important, since it is the major source for breaking down hydrogen peroxide. The depletion of mitochondrial GSH has been suggested as a key step leading to cell death [26,33,34]. Correspondingly, mitochondria develop an effective import system that allows maintaining the mitochondrial pool even when cytosolic GSH levels are low
[32,35]. Glutathione exists in two forms: the reduced form, which is conventionally called GSH; and the oxidized disulfide form known as GSSG. The majority of glutathione resides in the cell as the reduced form with only a small amount oxidized to GSSG. In the endoplasmic reticulum, where GSH is implicated in protein disulfide bond formation, the GSH/GSSG ratio is 3:1 [36]. In the cytoplasm and mitochondria, this ratio exceeds 10:1 [28]. Under conditions of oxidative stress, such ratios can dramatically change as GSSG accumulates.
GSH is synthesized from its precursor amino acids in the cytosol [37,38] . It is a two-step reaction catalyzed by glutamate-cysteine ligase (GCL) and GSH synthetase. First, an amide 12
linkage is formed between cysteine and the γ-carboxyl of glutamate; then, glycine is added to the
cysteine carboxyl of γ-glutamylcysteine (γ-GC):
L-glutamate + L-cysteine + ATP → γ-glutamylcysteine + ADP + Pi (1)
γ-glutamylcysteine + L-glycine + ATP → GSH + ADP + Pi (2)
The first step of GSH biosynthesis, catalyzed by GCL, is rate-limiting and is feedback-inhibited
by GSH [39]. When GSH is consumed and the feedback inhibition is lost, availability of cysteine
as a precursor can become the limiting factor [38]. GSH synthetase apparently has no regulatory
role and is not subject to feedback inhibition by GSH [40]. Once synthesized, γ-GC is rapidly
converted to GSH. The over-expression of GSH synthetase fails to increase GSH levels, whereas
induction of GCL increases GSH levels [41-43], which is consistent with the fact that GCL is the
rate-limiting enzyme of GSH synthesis.
Functions of GSH
Intracellular GSH is utilized through three major reactions––including conjugation,
oxidation, and degradation. The conjugation of GSH is catalyzed by the GSH S-transferases, a family of cytosolic and membrane-bound homologous enzymes [29,44]. The substrates are mostly
exogenous electrophiles, which constitute major pathways of drug metabolism and detoxification.
In addition, several endogenously formed compounds also follow similar metabolic pathways.
Some examples of endogenous compounds include leukotriene A, estrogens and prostaglandins
[44]. After conjugation, the compounds are excreted from the cell, and most are further
metabolized through the mercapturic acid pathway [29]. This pathway begins with the
conversion of GSH S-conjugates to the corresponding conjugates of cysteinylglycine by removal
of the γ-glutamyl moiety through γ-glutamyltranspeptidase. The resulting cysteinylglycine S-
conjugates are cleaved by dipeptidase to yield cysteine S-conjugates, which are in turn acetylated 13
to form mercapturic acid and excreted in the urine. As a result, the conjugation of GSH usually ends up with irreversible consumption of intracellular GSH. Such loss of GSH can be reversed only by de novo synthesis.
The oxidation of GSH occurs either directly with free radicals or through several enzymes. The antioxidant GPX enzymes use GSH to detoxify peroxides. In addition, GSH acts as an essential cofactor for GSH transhydrogeneases to convert dehydroascorbate to ascorbate, ribonucleotides to deoxyribonucleotides, and in a variety of disulfide-thiol conversions [45,46].
After GSH has been oxidized to GSSG, the recycling of GSSG to GSH is accomplished mainly by the enzyme glutathione reductase, which utilizes NADPH as the reducing power. Under normal circumstances there is essentially no net loss of GSH through oxidation. However, severe oxidative stress may overcome the capacity of GSSG reductase or the supply of NADPH, leading to accumulation of GSSG within the cytosol [44,47]. Because GSSG is not taken up intact by cells, but is rather degraded extracellularly, loss of GSSG from cells under conditions of oxidative stress leads to increases in the cell’s requirements for de novo GSH synthesis.
GSH Degradation
The breakdown of GSH occurs extracellularly and is catalyzed by two membrane-bound enzymes, γ-glutamyltranspeptidase (GGT) and dipeptidase. Once GSH is exported from the cell, the γ-glutamyl bond of GSH is cleaved by GGT. In the presence of amino acids, this reaction leads to the formation of cysteinylglycine and γ-glutamyl amino acids. Cysteinylglycine may be split at the cell’s membrane surface by dipeptidase to form cysteine and glycine, which are then taken back into the cell and reutilized for GSH synthesis. The γ-glutamyl amino acids formed by
GGT are transported into the cells and cleaved by the intracellular enzyme γ- glutamylcyclotransferase, yielding 5-oxoproline and the corresponding free amino acids. The 14
resulting 5-oxoproline is converted to glutamate in the ATP-dependent reaction catalyzed by 5-
oxoprolinase. The breakdown of GSH, together with the biosynthesis pathway mentioned earlier,
form the so-called γ-glutamyl cycle [29,48]. In addition, in vitro studies have shown that cystine
is among the active amino acid acceptors in the transpeptidase reaction catalyzed by GGT [49].
Thus, the corresponding product is γ-glutamylcystine, which may be transported into certain
cells and reduced to cysteine and γ-glutamylcysteine by GSH. Utilization of the latter by GSH synthetase yields GSH, by-passing the step catalyzed by GCL. These series of reactions appear to serve as an alternative or salvage pathway of GSH de novo synthesis. However, it is not clear
whether this pathway can make a real contribution in vivo.
The sequential process of GSH synthesis, secretion, degradation, amino acid uptake, and
GSH re-synthesis can operate locally, but is particularly important when it occurs between
tissues in the intact animal as an inter-organ mechanism of GSH and cysteine transport [50].
Most commonly, the process originates in liver, an organ rich in GCL and GSH synthetase.
Moreover, hepatocytes are able to derive cysteine from methionine and serine via the
transsulfuration pathway, whereas other cells can obtain cysteine only from diet or protein
breakdown. Such unique features of liver make it the major organ of GSH biosynthesis. Hepatic
GSH is then secreted into the plasma and supports GSH synthesis in peripheral tissues. Although
intact GSH is not taken up at a significant rate by most peripheral tissues, cells expressing GGT
can degrade GSH and take up the resulting cysteinylglycine and cysteine; adjacent cells and cells
“downstream” may also benefit by taking up these products. Such hepatic GSH-dependent
cysteine transport underscores the importance of liver in maintaining the inter-organ GSH
homeostasis. 15
GSH as an Important Antioxidant
One of the major functions of GSH involves its antioxidant properties. Reduced GSH is capable of directly scavenging radicals and peroxides by being oxidized to either GSSG or to a mixed disulfide [31]. GSH can also reduce hydrogen peroxide and lipid peroxide in the presence of GSH peroxidase [51]. In addition to scavenging free radicals, GSH is essential in maintaining the intracellular redox balance and the essential thiol status of proteins. To achieve this, GSH (or
GSSG) undergoes thiol-disulfide exchange in a reversible reaction catalyzed by thiol-transferase.
Meanwhile, GSH reduces (or GSSG oxidizes) the key sulfhydryl groups of the proteins. Whereas many proteins are active when the key sulfhydryls are in the thiol form, others require them to be in the oxidized disulfide form [52]. The equilibrium is determined by the redox state of the cell, which depends on the concentrations of GSH and GSSG.
GSH is also considered as the storage and transfer form of cysteine. Circulating GSH is stable in that it reacts very slowly with oxygen. Cysteine, on the other hand, is unstable extracellularly and rapidly auto-oxidized to cystine, producing potentially toxic oxygen free radicals [53]. To avoid the toxicity of autoxidation, most of the nonprotein cysteine is stored as
GSH. After GSH is exported from the cell, it can be quickly converted to cysteine through the γ- glutamyl cycle mentioned above. The resulting cysteine is then used elsewhere for protein synthesis and for the biosynthesis of taurine and other sulfur metabolites. The liver, which provides more than 80% of plasma GSH, plays a major role in the homeostasis of GSH and cysteine [50].
Role of GSH in Detoxification
GSH also plays a major role in detoxifying many reactive metabolites by either spontaneous conjugation or by a reaction catalyzed by the glutathione S-transferases [44]. 16
Although many chemicals in the environment are not toxic in the naturally occurring form, some
of those can be metabolized within the cell to unstable electrophilic metabolites, which can be
ultimate toxic or carcinogenic species. GSH can react with those electrophilic centers, and the
conjugates are then excreted from the cell via the mercapturic acid pathway.
Role of GSH in Critical Life Processes
In addition to the functions mentioned above, GSH is involved in a number of other
essential tasks including DNA synthesis and repair, protein synthesis, membrane transport, gene
regulation, enzyme activation, and immune function [30,54]. Due to such multiple roles of GSH,
there is considerable potential for alterations in GSH to be causally associated with diseases. In
fact, low GSH levels have been associated with the pathology of a number of diseases, such as
the progression of HIV and hepatitis C infections, liver cirrhosis, diabetes, chronic obstructive
pulmonary disease, atherosclerosis, Parkinson’s diseases, Alzheimer’s disease, and cataracts [55-
58]. For instance, in Parkinson’s disease patients, more than 40% decline of GSH levels has been
reported in the substantia nigra, the primary site of neuronal death, and the magnitude of GSH
depletion appears to parallel disease severity [59]. In pre-symptomatic patients, GSH depletion is
the earliest known indicator of oxidative stress, preceding lipid peroxidation, DNA damage, and
mitochondrial dysfunction [60]. The decreases of both plasma and liver GSH have also been documented in patients with hepatitis, alcoholic, and non-alcoholic cirrhosis, which are usually accompanied with increased levels of ROS and lipid peroxidation in affected livers [61-63]. In addition, an age-related decline in GSH has been observed in a number of studies [64,65]. This
decline in GSH levels has been suggested as a key factor underlying a number of changes during the normal aging process. 17
Glutamate-Cysteine Ligase (GCL) as the Rate-limiting Step in GSH Biosynthesis
Most of the diseases mentioned above are associated with signs of oxidative stress- mediated damage [18,55,63]. Given that GSH is a critical component of antioxidant defense, it follows that the decline of GSH may initiate or contribute to these disease manifestations through oxidative stress. However, since some disease conditions themselves are also accompanied by an increased generation of ROS, the observed GSH depletion may simply be the secondary effects of these pathological processes. To test the role of GSH deficiency experimentally, the best approach would be to study the modulation of GSH homeostasis through the rate-limiting biosynthesis enzyme, GCL.
GCL has been identified in a variety of living organism. The gene endoding the enzyme in E.coli has been cloned, and the gene product exists as a single polypeptide chain (58 kDa) [66].
Interestingly, in higher eukaryotes, GCL is a heterodimer comprising a catalytic subunit (GCLC,
72.8-kDa) and a modifier subunit (GCLM, 30.8-kDa) [47]. The enzymes from both bacteria and mammals have similar substrate specificity and are feedback-inhibited by GSH, but there is no significant sequence homology between the E. coli and the GCLC of the mammalian enzyme
[67]. The genes encoding the two subunits have been mapped in human and mouse. The human
GCLC gene is on chromosome 6, and the mouse Gclc gene is on chromosome 9. The human
GCLM gene is on human chromosome 1, and the mouse Gclm gene is on chromosome 3 [68].
Expression of the two subunits varies in a tissue-specific manner and in response to different inducers [69-71]. Such non-coordinated regulation may represent an intrinsic mechanism to maintain GSH homeostasis in a complex environment.
The catalytic subunit GCLC itself exhibits all the catalytic activity, as well as feedback inhibition by GSH [39,72]. GCLC is the key component in maintaining GSH homeostasis. In 18
GCLC-deficient patients, point mutations in the GCLC protein lead to 2% of normal GSH levels in red cells, although tissue levels of GSH are likely not as severely depleted. These patients showed symptoms of intermittent jaundice, hemolytic anemia, and neurological disorders [73-
75]. Experimentally, GCLC activity can be irreversibly inhibited by a compound called buthionine sulfoximine (BSO). For many years this chemical has been one of the major tools in studying the consequences of GSH depletion. In cell culture studies, no signs of cytotoxicity were observed when cellular GSH was depleted more than 80% by BSO, yet those cells are more vulnerable to toxicants [76-78]. When GSH levels dropped to 5%, cells subsequently showed increased levels of ROS, mitochondrial dysfunction, and eventually cell death [79]. Moreover, cell viability seemed to be correlated very well with the time course of mitochondria GSH depletion, which is more resistant to BSO treatment than are cytosolic GSH pools [80,81]. Such observations are consistent with that in animal studies [26,82]. Prolonged BSO treatment (2-3 weeks) did not lead to cellular damage in adult mice in the liver, heart, or kidney, when mitochondrial GSH levels were still greater than 40% as compared with control levels. There were significant mitochondria swelling, mitochondria degeneration, and vacuolization in skeletal muscle, lens, lung type II cells, and jejunal mucosal cells, when mitochondrial GSH levels dropped below 20% of that in controls. In newborn rats and mice, BSO treatment led to the development of cataracts, which occurred when the lens mitochondrial GSH becomes depleted.
Although BSO-induced GSH depletion provides important information, such studies are often subject to criticism because of the nonspecific effects of such chemicals. Furthermore, since
BSO is not sufficiently potent to deplete GSH in vivo and is rapidly excreted in the urine [83,84], it is hard to study the long-term effects of GSH depletion. To circumvent such problems, we 19
have developed a genetic approach for studying the function of GCLC, which will be discussed
in detail in Chapter II of this dissertation.
GCLC-GCLM Interactions
Despite the extensive studies of GCLC, our knowledge about the modifier subunit
GCLM is very limited. The GCLM gene has been sequenced and cloned from the human, rat, mouse, and drosophila [85]. To date, the gene product has not been associated with any enzymatic activities. However, comparison of enzyme kinetics between purified GCLC and the
holoenzyme has indicated that GCLM modifies GCLC catalytic properties by decreasing the Km
of glutamate and increasing the Ki for GSH inhibition [67,85,86]. Therefore, the interaction
between the two subunits seems to be able to regulate GSH homeostasis by changing GCL
activity. The purified holoenzyme cannot readily be separated by native gel filtration, and only
partially separated by native gel electrophoresis under reducing conditions, suggesting that the
association between GCLC and GCLM involves both disulfide bond formation as well as non- covalent interaction [87]. There are 14 cysteines in GCLC and 6 cysteines in GCLM. Mutation studies have shown that at least Cys-533 in GCLC is involved in disulfide formation between
GCLC and GCLM [88]. Reduction of GCL by GSH or mutation of Cys-533 to Gly-533 inhibits
the GCL activity. However, the holoenzyme––after either of these treatments––is still more
efficient than GCLC alone [88,89], suggesting that the non-covalent interaction is important to
achieve optimal enzyme activity. The mechanism of such interaction and the structure-function
relationship will be discussed in detail in Chapter I of this dissertation.
At physiological conditions, the intracellular levels of GSH are in the range of 1 to 10 mM
and glutamate levels are 1 to 3 mM. Given the in vitro kinetic data, the glutamate levels are typically lower than the Km for GCLC, and GSH levels are higher than the Ki for GCLC, 20
respectively. It has been suggested that GCLM is necessary for maintaining cellular GSH levels
at physiological conditions [67]. The experimental support for this hypothesis, however, is mixed.
Over-expression of GCLM increases cellular GSH levels by 2-fold, rendering cells resistant to
oxidative stress [43]. In agreement with this result, down-regulation of GCLM mRNA by
ribozyme expression leads to a decrease of GSH levels in cultured pancreatic islet cells [90]. On
the other hand, up-regulation of GCLC alone was also reported to support high levels of
intracellular GSH [42,91]. Furthermore, despite a decrease in GCLM following antisense RNA
inhibition of GCLM translation, no changes in GSH were noted in human hepatoblastoma
HepG2 cell cultures [38,69]. To define better the cellular function of GCLM, we have developed
a mouse model with a targeted disruption of the Gclm gene, which will be discussed in Chapter
III of this dissertation.
Conclusions of this Introduction
In summary, GSH is the most abundant non-protein thiol with multiple functions in the
intact animal. The homeostasis of GSH is very important in maintaining the redox balance of the
cellular environment. As the rate-limiting enzyme in GSH biosynthesis pathway, understanding
the function and regulation of GCL is critical in our understanding of the biological functions of
GSH and the consequence of GSH depletion in disease development. In an effort to characterize
the functions of GCL in maintaining GSH homeostasis and in protecting against oxidative stress,
this dissertation is divided into three parts to address the following questions: What is the
mechanism of interaction between the GCL subunits? What is the function of GCLC and GCLM
in vivo? What is the contribution of GCLC versus GCLM in protecting against oxidative stress?
21
Chapter I
Characterization of the Interaction between the Glutamate-Cysteine
Ligase Subunits
Abstract
Glutamate-cysteine ligase (GCL) is the rate-limiting enzyme in the glutathione (GSH)
biosynthesis pathway. This enzyme is a heterodimer comprising a catalytic subunit (GCLC) and
a modifier subunit (GCLM). Although GCLC alone can catalyze the formation of L-γ-glutamyl-
L-cysteine, its binding with GCLM enhances the enzyme activity by lowering the Km for
glutamate and increasing the Ki for GSH inhibition. To characterize the enzyme structure-
function relationship, the heterodimer formation between GCLC and GCLM was investigated in
vivo using the yeast two-hybrid system and in vitro using affinity chromatography. A strong and
specific interaction between GCLC and GCLM was observed in both systems. Deletion analysis
showed that most regions, except a portion of the C-terminal region in GCLC and part of the N-
terminal region in GCLM, are required for the interaction to occur. Point mutations on selected
amino acids were also tested for binding activity. GCLC C248A-C249A and P158L mutations
showed the same strength of binding of GCLC to GCLM as to the wild-type GCLC, yet the catalytic activity was dramatically decreased in these mutations. The results suggest that: [a] the heterodimer formation may not be dependent on amino acid sequence but instead might involve a complex formation of the tertiary structure; and [b] the catalytic activity of GCLC is dissociable from its binding with the GCLM subunit. 22
Introduction
Glutathione (GSH) is the most abundant non-protein thiol in mammalian cells. It plays a
major role in cellular defenses against oxidative stress and in cellular detoxification [1]. GSH is synthesized from the precursor amino acids by the sequential actions of glutamate-cysteine ligase
(GCL) and glutathione synthetase. The first step of GSH biosynthesis is rate-limiting and is feedback-inhibited by GSH [2]. Mammalian GCL is a heterodimer comprising a catalytic subunit
(GCLC; ~72.8 kDa) and a modifier subunit (GCLM; ~30.8 kDa), which are encoded by different genes on different chromosomes [3]. Studies with both native and recombinant GCL protein demonstrate that the isolated catalytic subunit exhibits all the catalytic activity, as well as all of the feedback inhibition by GSH [4,5].
Although GCLC is catalytically active, it has a higher Km value for glutamate (18.2 vs.
1.4 mM in rat kidney) and a lower Ki value for GSH (1.8 vs. 8.2 mM in rat kidney), compared with that of the holoenzyme [6]. Given that the intracellular levels of GSH and glutamate are
1~10 mM and 1~3 mM, respectively, the presence of GCLM is thought to modify the overall function of GCL in vivo, and the GCLC alone would have very low activity. This modification is especially important to the oxidative stress response and cellular detoxification of endogenous metabolites as well as foreign chemicals. Studies suggest that the survival of cells under such conditions depends more on their ability to synthesize additional GSH rapidly, rather than on the initial GSH concentration [7]. Therefore, the subunits’ interaction is believed, at least in part, to regulate GCL activity and thus GSH homeostasis.
Despite the importance of the holoenzyme, the exact mechanism of the subunits’ interaction remains poorly understood. Purified holoenzyme can be dissociated into two subunits by sodium dodecyl sulfate and/or reducing agents, indicating that there are both covalent and 23
non-covalent forces between the subunits [8,9]. Whereas reversible disulfide bond formation
could represent a mechanism of regulation under oxidative stress, the available in vitro data
suggest that the reduced form of holoenzyme contributes to most of the activities under
physiological conditions [10,11]. The reduced form of GCL is also a dominant form in drug-
resistant cells where both elevated GSH levels and GCL protein levels have been found [12,13].
Therefore, identification and characterization of the sites of non-covalent interaction are crucial
for understanding the enzyme structure-function relationship. Such sites of interaction may also
be suitable for the design of drugs that would function to disrupt protein-protein contacts
between GCLC and GCLM. In the present studies, we describe the characterization of the
subunits’ interaction in both in vivo and in vitro systems.
Materials and methods
Bacterial and yeast cells. All bacterial plasmids, except for histidine-tagged expression vectors,
were maintained in E. coli DH5α (Stratagene, La Jolla, CA). The histidine-tagged plasmids were
transformed into E.coli BL21(DE3)pLysS (Invitrogen, Carlsbad, CA). Saccharomyces cerevisiae
strain HF7c (Clontech, Palo Alto, CA) was used for the yeast two-hybrid analysis.
Plasmid constructs. The full-length mouse GCLC and GCLM cDNAs were obtained by PCR
amplification from C57/BL6 mice. The fragments were cloned into the appropriate vectors,
either by enzymatic digestion or by site-directed mutagenesis (Stratagene) (Table 1). pBluescript
II KS(+/-) (Stratagene) (pBS) was used in the TNT-T3-coupled wheat-germ extract system.
pRSET vectors (Invirtogen) were used to make histidine-tagged proteins. pAS2-1 and pACT2
vectors (Clontech) were used for the yeast two-hybrid system. All constructs were verified by restriction endonuclease digest and sequencing (data not shown). 24
Yeast two-hybrid analysis. Yeast two-hybrid analysis was performed using the Matchmaker
system (Clontech). Briefly, yeast strain HF7c was first transformed by electroporation with
plasmid pAS2-1 containing GCLM (or GCLC). The same colony was further transformed with
plasmid pACT2 containing GCLC (or GCLM). The double transformants were plated on yeast
minimal medium (1 M sorbital, 1.7 g of yeast nitrogen base/L, 1 g/L ammonium sulfate, 10 g/L
succinic acid, 6 g/L NaOH, 20 g/L dextrose, 2% (w/v) agar, and a mixture of essential amino acids minus leucine and tryptophan). For the β-galactosidase assays, minimal medium minus leucine and tryptophan was inoculated with an individual yeast colony and grown overnight at 30
°C to reach an OD600 of 0.8. Cells were harvested by centrifugation at 3000 g x 10 min at 4 °C and were resuspended in 300 µl of ice-cold Z buffer (0.06 M Na2HPO4⋅7H2O, 0.04 M
NaH2PO4⋅H2O, 0.01 M KCl, 0.001 M MgSO4⋅7H2O, pH 7.0). Cells were then lysed by two cycles of freezing in liquid nitrogen and thawing at 37 °C. Crude extract (100 µl) was added to
800 µl of buffer Z containing the substrate O-nitrophenyl-β-D-galactopyranoside (4 mg/ml). The reaction was incubated at 30 °C for various periods of time and stopped by adding 400 µl of 1 M
Na2CO3. β-galactosidase activity was assayed by measuring absorbance at 405 nm and
normalized relative to numbers of cells. In all cases, the amount of yeast-fusion proteins, whether
full-length or mutants, was evaluated by Western blot analysis using mouse anti-GAL4 AD and anti-GAL4 DNA-BD monoclonal antibodies (Clontech).
In vitro transcription/translation. Plasmid pBS-GCLM or pBS-GCLC was used in an in vitro
transcription/translation reaction with the TNT-T3-coupled wheat-germ extract system (Promega,
Madison, WI). The full-length protein was radiolabeled with [S35] methionine (NEN, Boson, MA)
according to the manufacturer’s directions. Translated proteins were pre-cleared through Ni-
NTA agarose (Qiagen, Valencia, CA), and the supernatants were kept at –20 °C until use. 25
Purification of histidine-tagged protein. E.coli BL21(DE3)pLysS was transformed with
plasmid pRSET-GCLC or pRSET-GCLM using the CaCl2 method (Invitrogen). Bacterial
cultures (500 ml) were grown to an OD600 of 0.5-0.6, after which the synthesis of fusion protein was induced by the addition of isopropyl-thio-β-D-galactoside (IPTG) (1 mM final concentration). After 3 h, the bacteria were harvested by centrifugation and the pellet was stored at –70 °C until use.
The frozen bacterial cells were lysed by sonication in 30 ml of buffer A [20 mM Tris/HCl,
500 mM KCl, 10 mM imidazole, 10% (v/v) glycerol, 1 mM phenylmethylsulphonyl fluoride, 1%
(v/v) Nonidet P40, pH 7.9). After centrifugation, the supernatant was loaded onto a Ni-NTA column (0.5-ml bed volume) pre-equilibrated with buffer B [20 mM Tris/HCl, 500 mM KCl, 10 mM imidazole, 10% (v/v) glycerol, pH 7.9]. The column was then washed with 10 ml of buffer
B followed by 5 ml of buffer C [20 mM Tris/HCl, 150 mM KCl, 30 mM imidazole, 10% (v/v) glycerol, pH7.9]. The bound protein was eluted from the column with 1 ml of buffer D [20 mM
Tris/HCl, 150 mM KCl, 300 mM imidazole, 10% (v/v) glycerol, pH 7.9] and immediately dialyzed against either buffer E (10 mM Tris/HCl, 1 mM EDTA, 5 mM MgCl2, pH 8.4) for
activity assay purpose or buffer F (50 mM Tris/HCl, 5 mM MgCl2, 5 mM L-glutamate, pH 7.4)
for binding assay purpose. Proteins were quantified using the BCA protein assay kit (Pierce,
Rockford, IL) and stored at –20 °C by adding glycerol to a final concentration of 25%.
In vitro binding assay. A purified histidine-tagged GCLC or GCLC mutant (10 µg) was bound
to Ni-NTA agarose and washed three times with buffer G (20 mM Tris/HCl, 150 mM KCl, 20
mM immidazole, pH 7.4). As a control, 10 µg E. coli lysate containing pREST was used for the
assay. The protein-bound agarose was then incubated with an aliquot of radiolabeled GCLM at 4
°C for 4 h on a rocking platform. After washing four times with buffer H (20 mM Tris/HCl, 150 26
mM KCl, 20 mM immidazole, pH 7.4), bound proteins were eluted into buffer I (20 mM
Tris/HCl, 150 mM KCl, 300 mM immidazole, pH 7.4). All samples were denatured in an equal
volume of Laemmli buffer and separated by SDS-PAGE. [S35] methionine-labeled proteins were
detected by autoradiography and quantified with a PhosphoImager (Molecular Dynamics,
Sunnyvale, CA).
GCLC activity assay. GCLC activity was determined by directly measuring the rate of γ-
glutamylcysteine (γ-GC) formation under conditions described previously [14] . The reaction mixtures (200 µl) contained 100 mM Tris/HCl, pH 7.8, 50 mM KCl, 15 mM ATP, 20 mM
MgCl2, 1 mM Na2EDTA, 3 mM dithionite, 15 mM sodium L-glutamate, and 1 µg purified
GCLC or mutant GCLC protein. Reactions were allowed to proceed for 30 min at 37 °C upon adding 3 mM L-cysteine and then stopped by addition of trichloracetic acid. A fluorescent derivative of γ-GC was generated using o-phthalaldehyde, and the derivative was quantified using HPLC [15]. Activities are reported in nmol γ-GC formed per min per milligram protein.
Results
GCLC and GCLM can interact in yeast and in vitro. The full-length GCLC and GCLM cDNAs were cloned into the yeast two-hybrid vectors resulting in amino-terminal in-frame fusion of either the GAL4 DNA binding domain (BD) or the GAL4 activation domain (AD). The
plasmids were then transformed into Saccharomyces cerevisiae HF7c and analyzed for the
induction of β-galactosidase activity (Fig. 1). Coexpression of the empty vector with either
GCLC or GCLM resulted in no detectable β-galactosidase activity, whereas coexpression of
GCLC and GCLM led to β-galactosidase levels comparable to the interaction between p53 and 27
SV40TD (company’s positive control pairs pVA3-1 and pTD1-1). Neither GCLC nor GLCM
interacted with itself, indicating that the interaction of GCLC and GCLM is specific.
The results from the yeast system were confirmed in an in vitro binding assay. GCLC was tagged with 6 histidines at the amino-terminus and purified to homogeneity through a Ni-
NTA column. GCLM was translated in vitro using the TNT-T3-coupled wheat-germ extract system in the presence of [S35] methionine. The purified GCLC protein was first immobilized to
Ni-NTA agarose, followed by the addition of [S35] GCLM. As a control, a lysate of E.coli cells
hosted with empty vector was loaded onto the agarose. After 4 h of incubation at 4 °C, the
agarose was washed and eluted. Samples from the elution were analyzed by electrophoresis. The
control agarose did not retain any detectable amount of labeled GCLM, while a significant
fraction of the input [S35] GCLM was bound to the GCLC-attached agarose (Fig 2a). Both
GCLM and the heterodimer form were detected on the non-reducing SDS-PAGE gel; whereas
under reducing conditions, only GCLM was seen on the gel, suggesting that the subunits’
interaction involves both non-covalent bonds and disulfide bonds. The assay was also carried out
in the presence of the GCL substrate: L-glutamate, ATP, or the cysteine analog L-α- aminobutyrate. Quantitation of bound GCLM by PhosphorImager analysis indicated that none of the substrates affected the efficiency of subunit binding (Fig 2b). The converse experiment, in which GCLM was tagged with histidine and GCLC was radiolabeled, gave the similar results
(data not shown).
Interaction between GCLC and GCLM is mediated through multiple regions. In order to delineate regions within GCLC and GCLM that are responsible for the interaction, we designed a series of amino- and carboxyl-terminal deletions as well as several internal deletions based on the predicted protein secondary structure. These deletion constructs were fused to the GAL4 AD 28
domain, and the strength of interaction with the corresponding partner was measured as β-
galactosidase activity in the yeast two-hybrid system. All the fusion proteins were expressed at
similar levels, as revealed by Western blot analysis (data not shown).
The data in Fig. 3 summarizes the results from the nine GCLC deletions. The carboxy- terminal deletion of the last 146 residues retained 29% binding activity, compared with that of
the full-length GCLC protein. Internal deletion of residues 346 through 417 also gave a modest interaction with GCLM (23% of the full-length protein). Other constructs, including all the deletions in the amino-terminal region, showed no binding activity. Similar results were found in the GCLM deletion studies (Fig 4). Deletions that removed the 42 amino-terminal residues
(mutant ∆1-42) or 65 residues (mutant ∆1-65) gave a modest reduction in binding strength (35 and 15% of the full-length protein respectively), whereas the carboxy-terminal deletions showed no significant β-galactosidase activity compared with that of the controls. These data suggest that there is more than one region in GCLC and GCLM mediating the subunits’ interaction or promoting proper folding of the polypeptides. The amino-terminal region in GCLC and the carboxy-terminal region in GCLM are both required for the binding; however, neither of these regions is sufficient for the interaction.
The catalytic activity of GCLC is dissociable from its binding with GCLM. Since GCLC contains the capacity of binding with GCLM as well as the catalytic activity of forming γ- glutamylcysteine, we then asked if the two properties are dissociable. Protein sequence alignment shows that the two consecutive cysteines (Cys-248, Cys-249) are conserved in species as divergent as human and yeast. Mutations in either of the cysteines decreases the enzymatic activity to 10-15% of normal [16]. In addition, there is one point mutation (P158L) that is critical in human GCLC-deficient patients. The homozygote has only 2% GSH levels in erythrocytes 29
[17]. Based on this information, we cloned GCLC proteins containing these mutations (C248A,
C249A; and P158L) together with one of the deletions (∆346-417) into the histidine-tagged expression vector, and the enzymatic activities were then determined on the purified proteins.
Neither of the point mutations nor the deletion showed any detectable GCLC activity (Table 2).
Given the sensitivity of the assay, there was >50-fold decreases in the mutant protein activity compared with that of the full-length GCLC. We then tested the capacity of binding with GCLM in the yeast two-hybrid system and the in vitro assay (Fig. 5). Both of the point mutations were able to bind with GCLM at the same strength as the intact protein. These data suggest that the
GCLC catalytic activity can be divorced from its binding with GCLM.
Discussion
Mammalian GCL consists of two subunits, whereas only GCLC is found in E.coli and
yeast [18]. The mammalian GCLC and GCLM genes are located on different chromosomes, and the levels of expression appear to be tissue- and cell type-dependent [19,20]. The existence of
GCLM is believed to better maintain GSH homeostasis in response to various stimuli, in more complex biological systems. Compared with GCLC alone, the GCL holoenzyme is more optimized to the cellular environment. Available data suggest that the actual GCL activity is regulated by at least three mechanisms: the relative levels of GCLC and GCLM, the cellular
GSH concentration, and the covalent and non-covalent interactions between the subunits [21-23].
The first two mechanisms have been studied in depth during the past few years. Little is known, however, about the sequence requirements for the GCLC-GCLM subunits’ interaction.
Our study has shown that GCLC and GCLM can interact both in yeast and in an in vitro system. In vitro at 37 °C, this interaction was detectable in less than 5 min (data not shown), 30 indicating the non-covalent interaction may not involve enzymes. About 70% of disulfide-linked subunits have been reported in the isolated holoenzyme from the in vivo system [24]. However, under non-denaturing conditions, the holoenzyme form represents less than 20% of total input in our in vitro binding assay (data not shown). This portion was not increased in the presence of
GSSG or hydrogen peroxide. These results suggest that this reversible disulfide bond formation might require an enzyme system. The interaction of GCLC and GCLM was not affected by the substrates, indicating that the activity domain may not overlap with the binding region. These data also agree with our mutation studies.
The yeast two-hybrid system is a powerful tool for detecting protein-protein interactions.
This system has successfully detected the binding motifs in many of protein partners [25,26]. In this study we have attempted to delineate the interaction regions in GCLC and GCLM. To our surprise, only a few deletions were able to retain a modest interaction with the corresponding protein, whereas most constructs resulted in a complete loss of the binding capacity. These negative results are neither due to the expression level nor the protein stability, since Western blot analysis had detected a comparable level of expression in all yeast fusion proteins. The in vitro binding assay has further verified the results from our yeast two-hybrid studies. Therefore, the lack of binding may be attributed to either the removal of the protein binding-site or the improper folding of the truncated proteins. Computer-assisted analysis has indicated that both
GCLC and GCLM form compact globular proteins with no separable domains in either of the subunits [27,28]. Combined with our experimental data, we hypothesize that the heterodimerization of GCLC and GCLM is dependent on the tertiary structures formed by the three-dimensional folding of both proteins, rather than a sequence-specific recognition of an interacting domain. 31
Although the yeast two-hybrid system may not be a suitable approach in dissecting domains in this case, the β-galactosidase activity does reflect the strength of protein-protein interaction. As shown in Fig. 5, the data correlated fairly well with the in vitro binding assay.
Low GSH levels have been linked with decreased GCL activities in a couple of studies [29-31].
Whereas most researches have focused on amino acid changes that are associated with alterations in catalytic activities, the changes in binding capacity can also dramatically affect the efficiency of the enzyme. The yeast two-hybrid system described here will be an efficient approach to screen for the functional changes of amino acids critical for the subunits’ interaction.
Our data also suggest that the activity of GCLC is not necessarily associated with its binding with GCLM. The fact that the subunits’ binding can occur in the absence of GCL activity suggests that there might be mutations in which the enzymatic activity is lost, while the wild-type binding characteristics still remain. These mutations could serve as useful tools to create a dominant-negative model for studying the physical functions of GCLM.
32
Chapter I References
[1] Anderson, M. E. Glutathione: an overview of biosynthesis and modulation. Chem. Biol.
Interact. 111-112:1-14; 1998.
[2] Griffith, O. W.; Mulcahy, R. T. The enzymes of glutathione synthesis: gamma-
glutamylcysteine synthetase. Adv. Enzymol. Relat Areas Mol. Biol. 73:209-67, xii; 1999.
[3] Tsuchiya, K.; Mulcahy, R. T.; Reid, L. L.; Disteche, C. M.; Kavanagh, T. J. Mapping of the
glutamate-cysteine ligase catalytic subunit gene (GLCLC) to human chromosome 6p12 and
mouse chromosome 9D-E and of the regulatory subunit gene (GLCLR) to human
chromosome 1p21-p22 and mouse chromosome 3H1-3. Genomics 30:630-632; 1995.
[4] Yan, N.; Meister, A. Amino acid sequence of rat kidney gamma-glutamylcysteine
synthetase. J. Biol. Chem. 265:1588-1593; 1990.
[5] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[6] Huang, C. S.; Anderson, M. E.; Meister, A. Amino acid sequence and function of the light
subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol. Chem. 268:20578-20583;
1993.
[7] Meister, A. Glutathione deficiency produced by inhibition of its synthesis, and its reversal;
applications in research and therapy. Pharmacol. Ther. 51:155-194; 1991. 33
[8] Seelig, G. F.; Simondsen, R. P.; Meister, A. Reversible dissociation of gamma-
glutamylcysteine synthetase into two subunits. J. Biol. Chem. 259:9345-9347; 1984.
[9] Chang, L. S.; Chang, C. C. Biochemical regulation of the activity of gamma-
glutamylcysteine synthetase from rat liver and kidney by glutathione. Biochem. Mol. Biol.
Int. 32:697-703; 1994.
[10] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[11] Soltaninassab, S. R.; Sekhar, K. R.; Meredith, M. J.; Freeman, M. L. Multi-faceted
regulation of gamma-glutamylcysteine synthetase. J. Cell Physiol 182:163-170; 2000.
[12] Godwin, A. K.; Meister, A.; O'Dwyer, P. J.; Huang, C. S.; Hamilton, T. C.; Anderson, M. E.
High resistance to cisplatin in human ovarian cancer cell lines is associated with marked
increase of glutathione synthesis. Proc. Natl. Acad. Sci. U. S. A 89:3070-3074; 1992.
[13] Zhang, K.; Yang, E. B.; Wong, K. P.; Mack, P. GSH, GSH-related enzymes and GS-X
pump in relation to sensitivity of human tumor cell lines to chlorambucil and adriamycin.
Int. J. Oncol. 14:861-867; 1999.
[14] Seelig, G. F.; Meister, A. Glutathione biosynthesis; gamma-glutamylcysteine synthetase
from rat kidney. Methods Enzymol. 113:379-390; 1985.
[15] Senft, A. P.; Dalton, T. P.; Shertzer, H. G. Determining glutathione and glutathione
disulfide using the fluorescence probe o-phthalaldehyde. Anal. Biochem. 280:80-86; 2000. 34
[16] Tu, Z.; Anders, M. W. Identification of an important cysteine residue in human glutamate-
cysteine ligase catalytic subunit by site-directed mutagenesis. Biochem. J. 336 ( Pt 3):675-
680; 1998.
[17] Ristoff, E.; Augustson, C.; Geissler, J.; de Rijk, T.; Carlsson, K.; Luo, J. L.; Andersson, K.;
Weening, R. S.; van Zwieten, R.; Larsson, A.; Roos, D. A missense mutation in the heavy
subunit of gamma-glutamylcysteine synthetase gene causes hemolytic anemia. Blood
95:2193-2196; 2000.
[18] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[19] Gipp, J. J.; Bailey, H. H.; Mulcahy, R. T. Cloning and sequencing of the cDNA for the
light subunit of human liver gamma-glutamylcysteine synthetase and relative mRNA levels
for heavy and light subunits in human normal tissues. Biochem. Biophys. Res. Commun.
206:584-589; 1995.
[20] Diaz, D.; Krejsa, C. M.; White, C. C.; Keener, C. L.; Farin, F. M.; Kavanagh, T. J. Tissue
specific changes in the expression of glutamate-cysteine ligase mRNAs in mice exposed to
methylmercury. Toxicol. Lett. 122:119-129; 2001.
[21] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993. 35
[22] Soltaninassab, S. R.; Sekhar, K. R.; Meredith, M. J.; Freeman, M. L. Multi-faceted
regulation of gamma-glutamylcysteine synthetase. J. Cell Physiol 182:163-170; 2000.
[23] Wild, A. C.; Mulcahy, R. T. Regulation of gamma-glutamylcysteine synthetase subunit
gene expression: insights into transcriptional control of antioxidant defenses. Free Radic.
Res. 32:281-301; 2000.
[24] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[25] Vidal, M.; Legrain, P. Yeast forward and reverse 'n'-hybrid systems. Nucleic Acids Res.
27:919-929; 1999.
[26] McAlister-Henn, L.; Gibson, N.; Panisko, E. Applications of the yeast two-hybrid system.
Methods 19:330-337; 1999.
[27] Rost, B. PHD: predicting one-dimensional protein structure by profile-based neural
networks. Methods Enzymol. 266:525-539; 1996.
[28] Rost, B.; Sander, C.; Schneider, R. PHD--an automatic mail server for protein secondary
structure prediction. Comput. Appl. Biosci. 10:53-60; 1994.
[29] Walsh, A. C.; Feulner, J. A.; Reilly, A. Evidence for functionally significant polymorphism
of human glutamate cysteine ligase catalytic subunit: association with glutathione levels
and drug resistance in the National Cancer Institute tumor cell line panel. Toxicol. Sci.
61:218-223; 2001. 36
[30] Beutler, E.; Gelbart, T.; Kondo, T.; Matsunaga, A. T. The molecular basis of a case of
gamma-glutamylcysteine synthetase deficiency. Blood 94:2890-2894; 1999.
[31] Walsh, A. C.; Li, W.; Rosen, D. R.; Lawrence, D. A. Genetic mapping of GLCLC, the
human gene encoding the catalytic subunit of gamma-glutamyl-cysteine synthetase, to
chromosome band 6p12 and characterization of a polymorphic trinucleotide repeat within
its 5' untranslated region. Cytogenet. Cell Genet. 75:14-16; 1996.
37
Chapter I Figure Legends
Fig. 1. GCLC and GCLM protein interaction in yeast. Yeast strain HF7c was sequentially
transformed with the indicated plasmids, and positive transformants were selected. Strength of
interaction was measured as β-galactosidase activity. The values shown are the means ± S.D. of
three independent transformants.
Fig. 2. Interaction of GCLC and GCLM on Ni-NTA resin. [35S]-labeled GCLM from a
wheat-germ lysate was added to Ni-NTA resin, preincubated with either E. coli lysate or purified
His-6-GCLC. The resin was then washed and eluted with imidazole, as described in Materials and Methods. (A) Samples from each fraction were subjected to SDS-PAGE under reducing conditions (5% 2-mercaptoethanol) except lane 4. Lane 1, wheat-germ lysate containing [35S]- labeled GCLM. Lane 2, elute from the E. coli lysate-bound resin. Lane 3, elute from His-6-
GCLC-bound resin. Lane 4, elute from His-6-GCLC-bound resin without reducing agent. (B)
Effects of substrates on the subunits’ interaction. Binding of [35S]-labeled GCLM to His-6-
GCLC on Ni-NTA resin was performed in the presence of either 5 mM ATP, or 5 mM L- glutamate, or 5 mM L-aminobutyrate, or the combination of two or three. Elutes from each condition were loaded on SDS-PAGE under non-reducing conditions.
Fig. 3. GCLC deletion studies in the yeast two-hybrid system. Yeast strain HF7c was sequentially transformed with the pAS2-1 plasmid containing GCLM and the pACT2 plasmid containing GCLC (or its deletions). The strength of interaction between GCLC and GCLM in each transformant was represented by the percentage of β-galactosidase activity relative to the full-length GCLC. Data are expressed as means ± S.D. 38
Fig. 4. GCLM deletion studies in yeast two-hybrid system. Yeast strain HF7c was
sequentially transformed with the pAS2-1 plasmid containing GCLC and the pACT2 plasmid containing GCLM (or its deletions). The strength of interaction between GCLC and GCLM in each transformant was represented by the percentage of β-galactosidase activity relative to the full length GCLM. Data are expressed as means ± S.D.
Fig. 5. Interaction of GCLC mutants with GCLM. The interaction of GCLC mutants with
GCLM was tested in the yeast two-hybrid system (A) and on Ni-NTA resin (B), as described in
Materials and Methods. The amount of bound protein in the in vitro system was quantified with a PhosphoImager and was plotted against β-galactosidase activity in yeast (C).
39
Table 1. Plasmids constructs used in this study
Construct Description pBS-GCLC GCLC PCR fragment inserted into BamHΙ and NotΙ sites in pBS pBS-GCLM GCLM PCR fragment inserted into BamH1 and Not1 sites in pBS pACT2-GCLC GCLC PCR fragment inserted into BamH1 and Sac1 sites in pACT2 pACT2-GCLM pBS-GCLM digested with Sma1 and Sac1, fragment ligated into pACT2 pAS2-1-GCLC GCLC PCR fragment inserted into EcoR1 and BamH1 sites in pAS-2-1 pAS2-1-GCLM GCLM PCR fragment inserted into BamH1 site in pAS2-1 pACT2-GCLC∆1-41 GCLC∆1-41 PCR fragment inserted into BamH1 and EcoR1 sites in pACT2 pACT2-GCLC∆1-83 GCLC∆1-83 PCR fragment inserted into BamH1 and EcoR1 sites in pACT2 pACT2-GCLC∆1-175 GCLC∆1-175 PCR fragment inserted into BamH1 and EcoR1 sites in pACT2 pACT2-GCLC∆1-241 GCLC∆1-241 PCR fragment inserted into BamH1 and EcoR1 sites in pACT2 pACT2-GCLC∆491-637 pACT2-GCLC mutated to stop codon at GCLC491 pACT2-GCLC∆431-637 pACT2-GCLC mutated to stop codon at GCLC431 pACT2-GCLC∆256-281 pBS-GCLC281-256 PCR fragment digested with Sfo1 and religated, then digested with BamH1 and EcoR1, fragment ligated into pACT2 pACT2-GCLC∆281-340 pBS-GCLC340-281 PCR fragment digested with Sfo1 and religated, then digested with BamH1 and EcoR1, fragment ligated into pACT2 pACT2-GCLC∆346-417 pBS-GCLC417-346 PCR fragment digested with Sfo1 and religated, then digested with BamH1 and EcoR1, fragment ligated into pACT2 pACT2-GCLCP158L pACT2-GCLC mutated to proline at GCLC158 pACT2-GCLCC248AC249A pACT2-GCLC mutated to alaline at GCLC248249 pACT2-GCLM∆1-42 GCLM∆1-42 PCR fragment inserted into Sma1 and Sac1 sites in pACT2 pACT2-GCLM∆1-65 GCLM∆1-65 PCR fragment inserted into Sma1 and Sac1 sites in pACT2 pACT2-GCLM∆1-114 GCLM∆1-114 PCR fragment inserted into Sma1 and Sac1 sites in pACT2 pACT2-GCLM∆218-274 GCLM∆218-274 PCR fragment inserted into Sma1 and Sac1 sites in pACT2 pACT2-GCLM∆112-274 GCLM∆112-274 PCR fragment inserted into Sma1 and Sac1 sites in pACT2 pRSETA-GCLC pBS-GCLC digested with BamH1 and Not1, fragment ligated into BamH1 and PvuΙΙ sites in pRSETA pRSETB-GCLM pBS-GCLM digested with BamH1 and Not1, fragment ligated into BamH1 and PvuΙΙ sites in pRSETB pRSETC-GCLCP158L pACT2-GCLCP158L digested with BamH1 and Xho1, fragment ligated into pRSETC pRSETC-GCLCC248AC249A pACT2-GCLCC248AC249A digested with BamH1 and Xho1, fragment ligated into pRSETC pRSETC-GCLC∆1-41 pACT2-GCLC∆1-41 digested with BamH1 and Xho1, fragment ligated into pRSETC pRSETC-GCLC∆346-417 pACT2-GCLC∆346-417 digested with BamH1 and Xho1, fragment ligated into pRSETC
40
Table 2. Enzyme activity of wild-type and mutant GCLC
Protein GCLC activity (nmol/min/mg protein) GCLC 27 ± 2 GCLC P158L ND GCLC C248AC249A ND GCLC ∆346-417 ND Activity was measured as described in Materials and methods. Data are means ± S.D. from three measurements. ND, not detectable.
41
Fig 1.
6
5
4
3
2
1
beta-galactosidase activity 0
D D C C D L M M L /AD L L 0T D/A C M/A 4 B L L -GC -GC -GCLC -GCLM V D D D-GC D-GC -S /AD /AD D-GC D-GC D/A D/A C M/A C M/A B B B B L L L L 53/AD
D-GC D-GC D-GC D-GC D-p B B B B B
42
Fig. 2.
A.
B.
43
Fig. 3
β-gal activity (%)
1 637 GCLC N- -C 100 ± 8.5
∆1-41 N- -C 0 ± 0.06
∆1-83 N- -C 2 ± 0.08
∆1-175 N- -C 1 ± 0.07
∆1-241 N- -C 0 ± 0.07
∆491-637 N- -C 29 ± 6.1
∆431-637 N- -C 0 ± 0.06
∆256-281 N- -C 0 ± 0.07
∆281-340 N- -C 1 ± 0.07
∆346-417 N- -C 23 ± 4.6
44
Fig. 4
β-gal activity (%)
1 274 GCLM N- -C 100 ± 9.2
∆1-42 N- -C 35 ± 5.4
∆1-65 N- -C 15 ± 4.1
∆1-114 N- -C 2 ± 0.09
∆218-274 N- -C 1 ± 0.07
∆112-274 N- -C 1 ± 0.06
45
Fig. 5
A.
120
100
80
60
40
20 relative binding activity (%) 0
D -AD A D/AD A -B 417- CLC-AD 6- /G 4 D 8AC249 D3 GCLM -B 4 C L GC GCLM CLCC2 /G -BD/ D M GCLM-BD/GCLCP158L-AD L C G GCLM-B
46
B.
nput i
f
o 346-417 ¼ pRSET GCLC P158L C248AC249A ∆
C.
6 2 5 R = 0.9831 4 3 2 1 beta-gal activity beta-gal 0 0 2000 4000 6000 8000 10000 12000 arbitrary band intensity
47
Chapter II
Hepatocyte-specific knockout of glutamate-cysteine ligase catalytic
subunit in mouse: early death with progressive liver degeneration
and rescue by N-acetylcysteine
Abstract
GSH depletion has been associated with numerous liver diseases, but the cause-and-effect
relationship has not been established. To define a role for hepatic GSH in vivo, we generated a
hepatocyte-specific Cre recombinase-mediated disruption of the Gclc gene encoding the catalytic
subunit of glutamate-cysteine ligase, an essential enzyme in GSH biosynthesis. Mice that
expressed Cre recombinase driven by the hepatocyte-specific albumin promoter and in which
both Gclc alleles were flanked by loxP sites (floxed, f) became hepatocyte-specific knockouts,
Gclc(h/h), by 14 days after birth. These mice were seemingly normal, but had hepatic GSH
levels that were 1-2% of normal. By day 21, Gclc(h/h) mice failed to gain weight and died of
hepatic failure aound day 30. Plasma GSH levels dropped to 19%, 16%, and 2% of that in
Gclc(+/+) wild-type at days 14, 21 and 28, respectively, but GSH depletion (20-50%) in
nonhepatic tissues was not obvious until d21. The affected livers from d28 Gclc(h/h) mice showed significant oxidative damage, atypical mitochondria, and hepatocyte death. Remarkably,
N-acetylcysteine supplementation in the drinking water, starting at d21, was able to prevent early death; however, rescued Gclc(h/h) mice developed a pathological condition characteristic of 48 liver cirrhosis. Our data suggest that GCLC is essential for hepatocyte function, but some functions of hepatic GCLC, and perhaps GSH, may be compensated for by NAC.
49
Introduction
Glutathione (GSH) is the most abundant non-protein thiols found in most aerobic organisms.
It participates in many biological reactions related to cellular redox function, protection against oxidative damage, detoxification of endogenous and exogenous compounds, storage and transport cysteine, regulation of membrane transporters, as well as cell growth and DNA synthesis [1-5]. As the major organ for biosynthesis of macromolecules and detoxification of
toxic metabolites, liver has among the highest GSH concentrations (5-10 mM) in the body.
Besides the function mentioned above, hepatic GSH also plays a central role in inter-organ GSH
homeostasis by supplying plasma GSH through sinusoidal efflux [6]. A decrease in GSH has been found in association with a variety of oxidative stress-mediated liver diseases [7-9].
However, the essential role of GSH in liver function has not been established directly.
GSH is synthesized from its precursor amino acids in two sequential enzymatic reactions
[10,11]. Glutamate-cysteine ligase (GCL) catalyzes the formation of γ-glutamylcysteine (γ-GC) from glutamate and cysteine. Glutathione synthetase then couples glycine to γ-GC to form GSH.
The reaction catalyzed by GCL is rate-limiting in GSH biosynthesis, and the product GSH is a feedback inhibitor of GCL activity. Higher eukaryotes contain GCL as a heterodimer comprised of a 72.8-kDa catalytic subunit (GCLC) and a 30.8-kDa modifier subunit (GCLM) [12]. GCLC
exhibits all the catalytic activity, as well as feedback inhibition by GSH [13,14]. It is a key
component in maintaining GSH homeostasis. In GCLC deficient patients, point mutation of
GCLC protein leads to 2% of normal GSH levels in red cells. These patients showed symptoms
of intermittent jaundice, hemolytic anemia, and neurological disorders [15-17]. Experimentally,
GCLC activity can be irreversibly inhibited by a compound called buthionine sulfoximine (BSO).
Depending on the extent of GSH depletion, inactivation of GCLC is associated with increased 50
sensitivity of toxicant insults, mitochondria damage, and cell death [18-22]. Furthermore, mice
with the targeted disruption of the Gclc gene died between gestational day 7.5 to 8.5; yet permanent cell lines derived from the mutant blastocysts can grow indefinitely in GSH-free medium supplemented with N-acetylcysteine (NAC) [23,24]. These data suggest that GCLC, and most likely GSH, is essential for embryonic development but not for survival of cells in culture.
Much of our understanding about the function of GCLC, and GSH homeostasis, in vivo has been the result of studies in chemical-induced GSH depletion in rats and mice and, recently, in the Gclc knockout mouse. Analysis of these mouse models has been limited, however, by several issues. First, the non-specificity of chemicals often leads to toxicities that cannot be
ascribed exclusively to GSH depletion. Second, the long-term effect of GSH depletion in vivo is
hard to achieve using chemicals. Third, the embryonic lethality of Gclc knockout mice makes it
impossible to study in adult animals. Moreover, the role of the hepatic GSH in the pathogenesis
of liver diseases cannot be addressed directly using those animal models.
Characterization of the cell-specific roles of GCLC in GSH homeostasis requires an
animal model in which the protein can be selectively eliminated in selected sites without
affecting expression in other sites. To attain this, we have made use of the Cre/loxP gene-
targeting strategy, because it enables us to develop mouse lines containing cell-specific gene deletions [25,26]. We describe here the generation of mice bearing the conditional (“floxed”)
Gclc allele. By intercrossing cell-specific Cre transgenes with the floxed Gclc allele, we generated mice with a hepatic-specific defect in GCLC expression. Analysis of these tissue-
specific knockouts has allowed us to directly define the function of hepatic GSH in adult livers and in maintaining the inter-organ GSH homeostasis.
51
Materials and Methods
Materials. N-acetylcysteine (NAC) was purchased from Sigma (St Louis, MO).
Preparation of targeting construct and generation of Gclc floxed mice. Figure 1A shows the
cloning scheme used to create the Gclc floxed mouse. The key features of the targeting vector
are a neomycin resistance gene (neoR), a herpes simplex virus thymidine kinase mini-cassette
(HSV-TK), and three 34-bp loxP sequences. Two of the loxP sites flank neoR in intron 3, and the
third loxP site is located in intron 6 of the Gclc gene. The targeting construct was electroporated
into embryonic stem (ES) cells derived from 129/SvJ mice. ES clones, resistant to both geneticin
(G418) and gancyclovir, were picked and expanded. ES cells harboring the homologous
recombined construct were determined by Southern blot analysis. The correctly targeted ES cell
clones were microinjected into C57BL/6J blastocysts and transferred into pseudo-pregnant
mothers. Male chimeric mice were bred against female C57BL/6J mice, and germline
transmission was identified by both Southern blot and PCR analysis. All animal studies were
approved by the University of Cincinnati Medical Center (UCMC) Institutional Animal Care and
Use Committee (IACUC). For abbreviations from now on, the four different Gclc alleles
discussed in this paper are denoted as: “+” for the wild-type allele; “f ” for the floxed Gclc allele;
“-” for the ubiquitously deleted Gclc allele by Cre recombinase; “h” for the hepatocyte-specific
Gclc deletion.
Cre transgenic mice. CAGGS-Cre transgenic mice were provided from Japan. The mice express
Cre recombinase under the control of the chicken β-actin promoter. After a genetic cross with
these mice, Cre-mediated recombination events occur ubiquitously during early development
[27,28]. Alb-Cre transgenic mice [C57BL6-TgN(AlbCe)21Mgn] were purchased from the
Jackson Laboratory (Bar Harbor, ME). These mice contain the rat albumin enhancer/promoter 52 gene, and have been shown to be very efficient at performing hepatocyte-specific gene knockouts using the Cre/loxP system [29].
Southern blotting and PCR analysis. For the differentiation of Gclc(f) vs. Gclc(+) allele, genomic DNA isolated from ES cells or mouse tissues was digested overnight with Sph I and processed for Southern blot analysis using a 32P-labeled 5' probe outside the region encompassed by the targeting construct (Fig. 1B). For the differentiation of Gclc(f) vs. Gclc(h) allele, genomic
DNA was digested with Sac I and blotted against a 3' probe (Fig. 1A). The band intensity was quantified using a Storm 860 Phosphorimager (Molecular Dynamics; Sunnyvale, CA) and
ImageQuant 5.0 software.
Genotyping was also confirmed by PCR analysis. The Gclc(f) allele was detected using primers A (5'- CGGGTGTTGGGTCGTTTGT -3') within the NEO gene, and B (5'-
CTATAATGTCCTGCACTGGG -3') near the second loxP site within intron 3. The Gclc(-) or
Gclc(h) allele was detected using primers C (5'- TAGTGAACGGTGTTAAAGG -3') near the first loxP site within intron 3, and D (5'- TCACTGGATTCTCTCACC -3') near the third loxP site within intron 6. The Gclc(+) allele was detected using primers B and C. PCR primers for
Cre transgene were purchased from the Jackson Laboratory (Bar Harbor, ME).
Northern blot and Western immunoblot analysis. Mouse tissues were harvested at appropriate stages, frozen immediately on dry ice, and stored at –70 °C until use. For Northern blot analysis, total RNA was isolated and analyzed for the presence of heme oxygenase-1 and metallothionein-
1 mRNA as described [30]. Equal loading was determined by rehybridizing blots with a probe for β-actin mRNA. For detection of GCLC and GCLM protein, the cytosolic fractions were resolved by 12% SDS-PAGE and blotted with the specific antibodies as described previously
[31]. 53
Histological Analysis. Mouse livers were fixed in an iso-osmolar paraformaldehyde/ glutaldehyde-phosphate-buffered solution, post-fixed in 1% phosphate-buffered osmium tetroxide, dehydrated in graded ethanol solutions ranging from 30-100%, treated with propylene oxide, and embedded in Spurr’s resin. One-micron-thick sections were stained with toluidine blue for determining the extent of hepatocyte apoptosis. The sections for electron microscopy were placed on naked copper grids and stained with uranyl acetate and lead citrate.
GSH, GSSG, and cysteine measurements. GSH, oxidized glutathione (GSSG), and cysteine levels were determined as described previously [32,33].
Plasma biochemical analysis. Mouse blood was collected by cardiac puncture using syringes coated with lithium heparin. The plasma layer was collected after centrifugation of the blood at
2,000g x 1 min at 4 °C. Plasma alanine aminotransferase (ALT), aspartate aminotransferase
(AST), and bilirubin levels were measured in the Clinical Laboratories at Cincinnati Children's
Hospital Medical Center following standard procedures.
Lipid peroxidation. Liver samples were homogenized immediately in 10 volumes of
homogenization buffer (154 mM KCl, 5 mM diethylenetriaminepentaacetic acid, 0.1 M
potassium phosphate buffer, and 10 mM MgCl2, pH 6.8) using a Teflon homogenizer. Lipid oxidation was determined by the presence of thiobarbituric acid-reactive substances (TBARS) as
described previously [34].
Statistical analysis. All data are expressed as the means ± S.E.M. (standard error of the mean).
Group means were compared by one-way analysis of variance (ANOVA) using the SAS
program (Windows version 8.0). P values <0.05 were considered statistically significant. When
the overall test of significance led to rejection of the null hypothesis, a multiple comparison was
performed to determine the source of the effect. 54
Results
Generation and characterization of Gclc(f/f) mice. To investigate the tissue-specific function of GCLC, we generated a Gclc inducible knockout mouse line. The inducible Gclc(f) allele that contains three loxP sites and a NEO mini-gene cassette (Fig. 1A) was created by gene targeting into ES cells. The positive clones were used for injecting into the blastocyst to generate chimeric mice. Such mice were mated with C57BL/6J female mice, and agouti animals were tested for germline transmission of the targeted Gclc(f) allele. The heterozygous Gclc(+/f) mice were then intercrossed, and the resultant offspring were genotyped by Southern blot (Fig. 1B) and PCR analysis (data not shown).
All three genotypes, Gclc(f/f), Gclc(f/+), and Gclc(+/+), were observed in the offspring from Gclc(+/f) intercross. The ratio of the three genotypes followed Mendelian inheritance. We also surveyed the GCLC protein content and GSH levels in a number of tissues including liver, kidney, pancreas, lung, heart, and spleen (data not shown). No differences were found among the three genotypes, suggesting that Gclc(f) allele is functionally indistinguishable from the Gclc(+) allele.
To determine whether the Cre recombinase is able to delete the floxed Gclc fragment in
Gclc(f) allele, homozygous Gclc(f/f) mice were bred with CAGGS-Cre transgenic mice, which express the Cre recombinase in a ubiquitous manner during early development. Male mice heterozygous for both Cre and Gclc(f) allele were then bred with female mice homologous for
Gclc(f) allele. Embryos from gestational day-15 (GD15) were genotyped for the presence of the
Gclc(-) allele. Out of 25 embryos we genotyped, no Gclc(-/-)/Cre(+/-) mice were found; this finding is consistent with the embryonic lethality of Gclc knockout mice from previous studies 55
[35,36]. Such results suggested that the Gclc(f) allele is converted into a null allele following
Cre-mediated recombination.
Generation of hepatocyte-specific Gclc knockout mice. Using CAGGS-Cre transgenic mice, we experimentally validated the Cre/loxP system in Gclc(f/f) mice. To generate hepatocyte- specific Gclc knockout mice, Gclc(f/f) homozygous mice were first bred with Alb-Cre transgenic mice, which express the Cre transgene under the control of the hepatocyte-specific albumin promoter. The resulting Gclc(f/+)/AlbCre(+/-) offspring were mated with Gclc(f/f) mice to obtain Gclc(h/h)/AlbCre(+/-) mice and littermate control groups of the following genotype:
Gclc(f/f)/AlbCre(-/-); Gclc(h/+)/AlbCre(+/-); Gclc(f/+)/AlbCre(-/-). Gclc(h/h)/AlbCre(+/-) mice were born at expected Mendelian frequency and seemingly without phenotype. For simplicity, the Gclc(h/h)/AlbCre(+/-) mice are denoted as Gclc(h/h), and the Gclc(f/f)/AlbCre(-/-) mice are denoted as Gclc(f/f) in the remainder of this paper.
To monitor Cre-mediated gene recombination of the Gclc(f) allele, genomic DNA was isolated from Gclc(h/h) mice at different time points ranging from GD16 to postnatal day 28
(d28). The presence of the Gclc(h) allele was detected exclusively in liver by PCR analysis as early as GD16, but not in any other tissues (Fig. 2A). Southern blot analysis showed that the deletion of Gclc(f) allele reached ~65% by 1 week after birth and increased to ~80% at d14 (Fig.
2B). Since liver tissue is composed of various cells other than hepatocytes, the residual undeleted portions most likely reflects the Gclc(f) allele still present in the non-parenchymal cells. Taken together, these data suggest that Gclc(f) allele is efficiently and specifically deleted in hepatocytes following Cre-mediated recombination, and the deletion was completed by postnatal day 14. Deletion of the Gclc(f) allele was also accompanied by the loss of GCLC protein. As shown in Fig. 2C, GCLC protein was below the level of dectection in the liver homogenate 56 obtained from Gclc(h/h) mice since d14, whereas the changes of the GCLM protein levels were unremarkable in the Gclc(f/f) mice.
GSH and cysteine levels in hepatocyte-specific Gclc knockout mice. GCL is the rate-limiting enzyme in GSH de novo synthesis pathway. If the deletion of the Gclc(f) allele results in a null allele, we would expect the complete loss of GSH in the knockout hepatocytes. Indeed, liver
GSH levels in Gclc(h/h) mice dropped to 3% of that seen in the Gclc(f/f) mice by d14 (Table 1), and the residual amounts of GSH are most likely produced by non-parenchymal cells in the liver.
Cellular GSH is compartmentalized in a cytosolic and a mitochondrial pool. The latter is thought to serve as the last reservoir of cellular GSH and plays a crucial role in cell survival
[37,38]. Most of GSH-depleting reagents can rapidly decrease cytosolic GSH levels, but are less effective in depleting GSH from the mitochondrial pool. Would deletion of Gclc allele affect sub-cellular GSH pools differently? We therefore measured the mitochondrial GSH levels in our
Gclc(h/h) mice. Compared to the Gclc(f/f) mice, liver mitochondrial GSH levels in Gclc(h/h) mice decreased dramatically, following a pattern similar to that for cellular GSH (Table 2).
Hepatic GSH is thought to play a central role in inter-organ GSH homeostasis by providing cysteine to extra-hepatic tissues. Following sinusoidal efflux from liver into the blood stream, GSH is hydrolyzed at tissue sites by two membrane-bound enzymes: γ-glutamyl- transpeptidase (GGT) and dipeptidase. Hydrolysis of GSH results in the liberation of cysteine, which is then taken up into the cells to synthesize GSH and/or proteins. To evaluate the role of liver in inter-organ homeostasis of GSH and cysteine, we also measured the thiol levels in extra- hepatic tissues from Gclc(h/h) mice. Following GSH depletion in liver, GSH levels from other tissues––e.g. kidney, pancreas, and lung––remained unchanged at d14, but dropped to 50-70% of the Gclc(f/f) mouse levels around d21 (Table 1). Plasma GSH levels dropped to 20% when the 57 liver GSH became depleted. The decline of plasma GSH levels reached 2% when GSH depletion occurred in non-hepatic tissues, indicating that most of the plasma GSH is derived from liver. On the other hand, there was no statistical difference in tissue cysteine levels between Gclc(h/h) mice and Gclc(f/f) mice (Table 3).
Growth retardation and liver failure in hepatocyte-specific Gclc knockout mice. At birth,
Gclc(h/h) mice were indistinguishable from control littermates in appearance or body weight.
However, they failed to gain weight around d19 and died between d29 to d33. The process coincided with increased levels of plasma AST, ALT, and bilirubin, indicating a progressive liver dysfunction in Gclc(h/h) mice (Table 4). Liver abnormalities were also evident by light microscopy from d21, but were most significant at d28. Liver tissue from these mice showed areas of necrosis, apoptosis, and inflammation (Fig. 3B). Many hepatocytes also showed enlarged nucleoli, DNA condensation and fragmentation, representing serious stress conditions.
The necrotic or apoptotic hepatocytes comprised about 2% of the overall population. However, by electron microscopic analysis, most of the hepatocytes showed significant ultra-structural alterations, including swollen mitochondria, compact tubular cristae formation, enrichment of small cytoplasmic vesicles, lack of rough endoplasmic reticulum, and segregation of nuclear
RNA and DNA components (Fig. 3D, F, G).
Rescue of the hepatocyte-specific Gclc knockout mouse by NAC. NAC is an antioxidant that has been in clinical use primarily to reduce hepatic injury following acetaminophen overdose.
NAC is transported across the plasma membrane, where it serves as a source for cysteine during
GSH biosynthesis [39]. In addition, NAC is believed to have direct antioxidant properties [40,41].
In an attempt to rescue Gclc(h/h) mice, NAC was added to the drinking water (10 mg/ml) starting at d21. Although NAC supplementation had little effects on liver GSH levels (Table 1), 58
these mice started to gain weight around d30, and appeared to be normal by 2 months of age. We
are currently observing some rescued mice alive 6 months after birth. Livers from these mice,
however, exhibited a nodular-like morphology with a hardened texture (Fig. 4B). Both liver and
spleen were twice as big as those from Gclc(f/f) mice. Biochemical function of the liver showed
elevated plasma ALT, AST, conjugated bilirubin, and unconjugated bilirubin levels, indicating
serious hepatocyte damage (Table 4). The abnormal liver architecture was also confirmed by
histology (Fig. 4D), which was characterized as irregular hepatic lobules, extensive fibrosis, increased inflammation, and enlargement of hepatocytes. Compared with that seen in the d28
Gclc(h/h) mice, the abnormalities of the endoplasmic reticulum were persistent in rescued mice.
However, similar mitochondria and nucleolus changes were not evident after NAC treatment.
Instead, most of the hepatic nuclei were unusually large, accompanied with nuclear inclusions.
Taken together, our data suggest that the rescued mice may suffer a pathological condition characteristic of liver cirrhosis.
Oxidative stress response in livers from hepatocyte-specific Gclc knockout mice. One of the
major functions of GSH is to protect against ROS. With the deletion of Gclc, livers from
Gclc(h/h) mice might represent a vulnerable state for oxidative injury. To measure the oxidative
stress responses in livers of these mice, we first chose to evaluate the levels of several oxidative
stress-inducible genes by Northern blot analysis. Heme oxygenase-1 catalyzes the conversion of
heme to bilirubin, and metallothionein-1 is a heavy-metal-binding protein with antioxidant
properties. Both of these genes are well established as sensitive oxidative stress indicators [42].
As shown in Fig. 5, both of these mRNAs were elevated in livers of Gclc(h/h) mice as early as
d14, compared to the Gclc(f/f) mice. There were even greater inductions by d28, indicating a
persistent oxidative stress response in livers of the Gclc(h/h) mice. In the absence of 59
environmental insults, such stress is most likely generated by endogenous ROS via the
mitochondrial respiratory chain. If not scavenged by reducing equivalents, excessive ROS can
bind to cellular maromolecules, such as lipid, DNA and protein. To evaluate the oxidative
damage in Gclc(h/h) mice, we measured lipid peroxidation by the TBARS assay. Compared to
the Gclc(f/f) mice, lipid peroxidation increased 3-fold in livers from Gclc(h/h) mice (Fig. 6).
In NAC-rescued mice, the patterns of mRNA expression and the lipid peroxidation were very similar to that seen in the non-NAC-treated Gclc(h/h) mice. Our data suggest that, although
NAC might be able to replace some critical function(s) of GSH, it appears not to be able to combat against the vast majority of endogenous oxidative stress. Furthermore, the accumulated oxidative damage might play a major role in causing the phenotype seen in the Gclc(h/h) mouse.
Discussion
In this study, we successfully generated a hepatocyte-specific Gclc knockout mouse line using the Cre/loxP system. Following Cre expression in hepatocytes, Gclc was completely deleted around d14. These mice died of liver failure within one month, apparently as a result of
GSH depletion. Our study provides the first direct evidence with regard to the GSH depletion causing liver failure. The exact mechanism of hepatocyte death deserves further study. GSH depletion has been associated with the activation of apoptosis through mitochondrial dysfunction, production of sphingomyelin metabolites, and the increased degradation of BCL2 [43-46]. In contrast, low GSH levels can also lead to necrosis by inhibiting caspase activity [47]. Although both mitochondrial changes and different stages of apoptotic cells were evident in livers from
Gclc(h/h) mice, we could not detect any significant cytochrome c release nor caspase-3 activation at any stage of liver dysfunction (data not shown). In addition, there was a good 60
proportion of necrotic cells in livers from Gclc(h/h) mice. The distribution of apoptotic and
necrotic cells seemed to be well correlated with the anatomic substructure of liver. In zone 3,
where blood to hepatocytes is supplied by central veins that are relatively depleted of oxygen,
most cells underwent necrosis; apoptotic cells were more concentrated in zone 1, on the other
hand, at which hepatocytes are supplied by portal veins. Such unsynchronized events suggest
that cellular responses to GSH depletion are complicated and may depend on their own
metabolic activities as well. Alternatively, this may reflect the differential susceptibility of cells
to GSH depletion due to their requirements for reducing equivalents.
One major function of GSH is to serve as a reservoir for cysteine. Unlike GSH, cysteine
is extremely unstable outside the cell and rapidly auto-oxidizes to cystine, generating potentially
toxic oxygen radicals. To maintain a constant flow of cysteine, extracellullar GSH is hydrolyzed
to release cysteine, by way of membrane-bound enzyme GGT and dipeptidase. Since liver is the
major provider of plasma GSH, it was thought to maintain cysteine homeostasis as well.
However, although hepatic GSH depletion greatly decreased GSH levels in non-hepatic tissues, no significant differences of cysteine levels were found in tissues from Gclc(h/h) mice. Relative
to our hepatocyte knockout model, Ggt(-/-) knockout mice show both cysteine and GSH
deficiency in various tissues due to the inability of the liver to recover cysteine from exported
GSH [48]. In addition to the similar symptoms of GSH depletion observed here, these mice also
developed the loss of agouti coat color as the result of cysteine deficiency. On the other hand,
Gclm(-/-) mice, which have functional GGT but have decreased capacity to synthesize GSH,
showed severe GSH depletion in various tissues and cysteine depletion in tissues rich in GGT
[49]. Taken together, these data suggest that cysteine homeostasis is maintained by the normal
functions of GSH biosynthesis and catabolism, rather than the hepatic GSH synthesis alone. 61
Besides its function as a cysteine transporter, GSH has been associated with a number of
critical cellular processes such as DNA and protein synthesis. However, its essential functions
have been challenged by the fact that cultured cells derived from Gclc(-/-) blastocysts can grow
normally in the absence of GSH [50]. In our study, the Gclc(h/h) mouse dies of liver dysfunction
at an early age as the result of GSH depletion, suggesting that––at least in the intact animal––
GSH is still required for normal adult liver function. However, the early death and mitochondrial
damage can be prevented with NAC supplement. One possible explanation is that NAC may
replenish GSH by providing its substrate cysteine. Analysis of GCLC and GSH levels in rescued
mice revealed that it was not the case. NAC itself is also nucleophilic and can directly scavenge
ROS. As a result, it is involved in a number of cellular protective effects associated with DNA
repair, mitochondrial function, signal transduction, cell survival and apoptosis [40,51,52].
Furthermore, some studies in vitro have shown that NAC can serve as an effective thiol substrate
in the reaction catalyzed by ebselen, a GPXs mimic, resulting in the reduction of hydrogen
peroxide [53]. It is intriguing to consider that such a reaction may actually occur in the intact
animal, so that the role of GSH in mitochondria may be replaced by NAC.
In our study, although NAC supplement was able to rescue Gclc(h/h) mice to adulthood,
these mice developed a pathological condition characteristic of liver cirrhosis. Clinically, liver
cirrhosis represents the end stages of chronic liver cell injury. It is characterized by recursive
destruction of liver cells, chronic inflammation and fibrogenic responses, deposition of collagen
and extracellular matrix, and eventually the pseudo-lobular formation in the organ. Although the
initiation of liver cirrhosis may vary upon inducers, most studies have suggested that ROS and its
metabolites may play a pathogenetic role. In cultured human or rat stellate cells, ROS may directly stimulate stellate cell proliferation and collagen synthesis [54,55]. It may also serve as 62
pro-inflammatory mediators to activate Kupffer cells [56]. In addition, oxidation products, such as lipid peroxidation, are able to up-regulate the expression of procollagen as well as fibrogenic cytokines in a number of studies [57-59]. Given the extent of GSH depletion and the significant oxidative damage in liver, such ROS-mediated fibrogenic events––in all likelihood––have occurred in NAC-rescued mice as well. It wil be interesting to test if this process can be alleviated or reversed by the use of antioxidants.
63
Chapter II References
[1] Meister, A.; Anderson, M. E. Glutathione. Annu. Rev. Biochem. 52:711-760; 1983.
[2] Meister, A. Glutathione metabolism. Methods Enzymol. 251:3-7; 1995.
[3] Hammond, C. L.; Lee, T. K.; Ballatori, N. Novel roles for glutathione in gene expression,
cell death, and membrane transport of organic solutes. J. Hepatol. 34:946-954; 2001.
[4] Sies, H. Glutathione and its role in cellular functions. Free Radic. Biol. Med. 27:916-921;
1999.
[5] Manciu, L.; Chang, X. B.; Buyse, F.; Hou, Y. X.; Gustot, A.; Riordan, J. R.; Ruysschaert, J.
M. Intermediate structural states involved in MRP1-mediated drug transport. Role of
glutathione. J. Biol. Chem. 278:3347-3356; 2003.
[6] Ookhtens, M.; Kaplowitz, N. Role of the liver in interorgan homeostasis of glutathione and
cyst(e)ine. Semin. Liver Dis. 18:313-329; 1998.
[7] Shigesawa, T.; Sato, C.; Marumo, F. Significance of plasma glutathione determination in
patients with alcoholic and non-alcoholic liver disease. J. Gastroenterol. Hepatol. 7:7-11;
1992.
[8] Purucker, E.; Winograd, R.; Roeb, E.; Matern, S. Glutathione status in liver and plasma
during development of biliary cirrhosis after bile duct ligation. Res. Exp. Med. (Berl)
198:167-174; 1998.
[9] Ljubuncic, P.; Tanne, Z.; Bomzon, A. Evidence of a systemic phenomenon for oxidative
stress in cholestatic liver disease. Gut 47:710-716; 2000. 64
[10] Anderson, M. E. Glutathione: an overview of biosynthesis and modulation. Chem. Biol.
Interact. 111-112:1-14; 1998.
[11] Lu, S. C. Regulation of hepatic glutathione synthesis: current concepts and controversies.
FASEB J. 13:1169-1183; 1999.
[12] Griffith, O. W.; Mulcahy, R. T. The enzymes of glutathione synthesis: gamma-
glutamylcysteine synthetase. Adv. Enzymol. Relat Areas Mol. Biol. 73:209-67, xii; 1999.
[13] Yan, N.; Meister, A. Amino acid sequence of rat kidney gamma-glutamylcysteine
synthetase. J. Biol. Chem. 265:1588-1593; 1990.
[14] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[15] Beutler, E.; Gelbart, T.; Kondo, T.; Matsunaga, A. T. The molecular basis of a case of
gamma-glutamylcysteine synthetase deficiency. Blood 94:2890-2894; 1999.
[16] Ristoff, E.; Augustson, C.; Geissler, J.; de Rijk, T.; Carlsson, K.; Luo, J. L.; Andersson, K.;
Weening, R. S.; van Zwieten, R.; Larsson, A.; Roos, D. A missense mutation in the heavy
subunit of gamma-glutamylcysteine synthetase gene causes hemolytic anemia. Blood
95:2193-2196; 2000.
[17] Ristoff, E.; Larsson, A. Patients with genetic defects in the gamma-glutamyl cycle. Chem.
Biol. Interact. 111-112:113-121; 1998. 65
[18] Beutler, E.; Gelbart, T.; Kondo, T.; Matsunaga, A. T. The molecular basis of a case of
gamma-glutamylcysteine synthetase deficiency. Blood 94:2890-2894; 1999.
[19] Ristoff, E.; Augustson, C.; Geissler, J.; de Rijk, T.; Carlsson, K.; Luo, J. L.; Andersson, K.;
Weening, R. S.; van Zwieten, R.; Larsson, A.; Roos, D. A missense mutation in the heavy
subunit of gamma-glutamylcysteine synthetase gene causes hemolytic anemia. Blood
95:2193-2196; 2000.
[20] Merad-Boudia, M.; Nicole, A.; Santiard-Baron, D.; Saille, C.; Ceballos-Picot, I.
Mitochondrial impairment as an early event in the process of apoptosis induced by
glutathione depletion in neuronal cells: relevance to Parkinson's disease. Biochem.
Pharmacol. 56:645-655; 1998.
[21] Meister, A. Glutathione deficiency produced by inhibition of its synthesis, and its reversal;
applications in research and therapy. Pharmacol. Ther. 51:155-194; 1991.
[22] Meister, A. Mitochondrial changes associated with glutathione deficiency. Biochim.
Biophys. Acta 1271:35-42; 1995.
[23] Shi, Z. Z.; Osei-Frimpong, J.; Kala, G.; Kala, S. V.; Barrios, R. J.; Habib, G. M.; Lukin, D.
J.; Danney, C. M.; Matzuk, M. M.; Lieberman, M. W. Glutathione synthesis is essential for
mouse development but not for cell growth in culture. Proc. Natl. Acad. Sci. U. S. A
97:5101-5106; 2000.
[24] Dalton, T. P.; Dieter, M. Z.; Yang, Y.; Shertzer, H. G.; Nebert, D. W. Knockout of the
mouse glutamate cysteine ligase catalytic subunit (Gclc) gene: embryonic lethal when 66
homozygous, and proposed model for moderate glutathione deficiency when heterozygous.
Biochem. Biophys. Res. Commun. 279:324-329; 2000.
[25] Kuhn, R.; Torres, R. M. Cre/loxP recombination system and gene targeting. Methods Mol.
Biol. 180:175-204; 2002.
[26] Ray, M. K.; Fagan, S. P.; Brunicardi, F. C. The Cre-loxP system: a versatile tool for
targeting genes in a cell- and stage-specific manner. Cell Transplant. 9:805-815; 2000.
[27] Araki, K.; Araki, M.; Miyazaki, J.; Vassalli, P. Site-specific recombination of a transgene
in fertilized eggs by transient expression of Cre recombinase. Proc. Natl. Acad. Sci. U. S. A
92:160-164; 1995.
[28] Sunaga, S.; Maki, K.; Komagata, Y.; Ikuta, K.; Miyazaki, J. I. Efficient removal of loxP-
flanked DNA sequences in a gene-targeted locus by transient expression of Cre
recombinase in fertilized eggs. Mol. Reprod. Dev. 46:109-113; 1997.
[29] Postic, C.; Shiota, M.; Niswender, K. D.; Jetton, T. L.; Chen, Y.; Moates, J. M.; Shelton, K.
D.; Lindner, J.; Cherrington, A. D.; Magnuson, M. A. Dual roles for glucokinase in glucose
homeostasis as determined by liver and pancreatic beta cell-specific gene knock-outs using
Cre recombinase. J. Biol. Chem. 274:305-315; 1999.
[30] Maier, A.; Dalton, T. P.; Puga, A. Disruption of dioxin-inducible phase I and phase II gene
expression patterns by cadmium, chromium, and arsenic. Mol. Carcinog. 28:225-235; 2000.
[31] Yang, Y.; Dieter, M. Z.; Chen, Y.; Shertzer, H. G.; Nebert, D. W.; Dalton, T. P. Initial
characterization of the glutamate-cysteine ligase modifier subunit Gclm(-/-) knockout 67
mouse. Novel model system for a severely compromised oxidative stress response. J. Biol.
Chem. 277:49446-49452; 2002.
[32] Senft, A. P.; Dalton, T. P.; Shertzer, H. G. Determining glutathione and glutathione
disulfide using the fluorescence probe o-phthalaldehyde. Anal. Biochem. 280:80-86; 2000.
[33] Shertzer, H. G.; Vasiliou, V.; Liu, R. M.; Tabor, M. W.; Nebert, D. W. Enzyme induction
by L-buthionine (S,R)-sulfoximine in cultured mouse hepatoma cells. Chem. Res. Toxicol.
8:431-436; 1995.
[34] Shertzer, H. G.; Berger, M. L.; Tabor, M. W. Intervention in free radical mediated
hepatotoxicity and lipid peroxidation by indole-3-carbinol. Biochem. Pharmacol. 37:333-
338; 1988.
[35] Shi, Z. Z.; Osei-Frimpong, J.; Kala, G.; Kala, S. V.; Barrios, R. J.; Habib, G. M.; Lukin, D.
J.; Danney, C. M.; Matzuk, M. M.; Lieberman, M. W. Glutathione synthesis is essential for
mouse development but not for cell growth in culture. Proc. Natl. Acad. Sci. U. S. A
97:5101-5106; 2000.
[36] Dalton, T. P.; Dieter, M. Z.; Yang, Y.; Shertzer, H. G.; Nebert, D. W. Knockout of the
mouse glutamate cysteine ligase catalytic subunit (Gclc) gene: embryonic lethal when
homozygous, and proposed model for moderate glutathione deficiency when heterozygous.
Biochem. Biophys. Res. Commun. 279:324-329; 2000.
[37] Smith, C. V.; Jones, D. P.; Guenthner, T. M.; Lash, L. H.; Lauterburg, B. H.
Compartmentation of glutathione: implications for the study of toxicity and disease.
Toxicol. Appl. Pharmacol. 140:1-12; 1996. 68
[38] Kaplowitz, N.; Fernandez-Checa, J. C.; Kannan, R.; Garcia-Ruiz, C.; Ookhtens, M.; Yi, J.
R. GSH transporters: molecular characterization and role in GSH homeostasis. Biol. Chem.
Hoppe Seyler 377:267-273; 1996.
[39] Sjodin, K.; Nilsson, E.; Hallberg, A.; Tunek, A. Metabolism of N-acetyl-L-cysteine. Some
structural requirements for the deacetylation and consequences for the oral bioavailability.
Biochem. Pharmacol. 38:3981-3985; 1989.
[40] Moldeus, P.; Cotgreave, I. A. N-acetylcysteine. Methods Enzymol. 234:482-492; 1994.
[41] Labib, R.; Abdel-Rahman, M. S.; Turkall, R. N-acetylcysteine pretreatment decreases
cocaine and endotoxin-induced hepatotoxicity. J. Toxicol. Environ. Health A 66:223-239;
2003.
[42] Dalton, T. P.; Shertzer, H. G.; Puga, A. Regulation of gene expression by reactive oxygen.
Annu. Rev. Pharmacol. Toxicol. 39:67-101; 1999.
[43] Merad-Boudia, M.; Nicole, A.; Santiard-Baron, D.; Saille, C.; Ceballos-Picot, I.
Mitochondrial impairment as an early event in the process of apoptosis induced by
glutathione depletion in neuronal cells: relevance to Parkinson's disease. Biochem.
Pharmacol. 56:645-655; 1998.
[44] Celli, A.; Que, F. G.; Gores, G. J.; LaRusso, N. F. Glutathione depletion is associated with
decreased Bcl-2 expression and increased apoptosis in cholangiocytes. Am. J. Physiol
275:G749-G757; 1998. 69
[45] Liu, B.; Andrieu-Abadie, N.; Levade, T.; Zhang, P.; Obeid, L. M.; Hannun, Y. A.
Glutathione regulation of neutral sphingomyelinase in tumor necrosis factor-alpha-induced
cell death. J. Biol. Chem. 273:11313-11320; 1998.
[46] Okamoto, Y.; Obeid, L. M.; Hannun, Y. A. Bcl-xL interrupts oxidative activation of neutral
sphingomyelinase. FEBS Lett. 530:104-108; 2002.
[47] Hentze, H.; Kunstle, G.; Volbracht, C.; Ertel, W.; Wendel, A. CD95-Mediated murine
hepatic apoptosis requires an intact glutathione status. Hepatology 30:177-185; 1999.
[48] Lieberman, M. W.; Wiseman, A. L.; Shi, Z. Z.; Carter, B. Z.; Barrios, R.; Ou, C. N.;
Chevez-Barrios, P.; Wang, Y.; Habib, G. M.; Goodman, J. C.; Huang, S. L.; Lebovitz, R.
M.; Matzuk, M. M. Growth retardation and cysteine deficiency in gamma-glutamyl
transpeptidase-deficient mice. Proc. Natl. Acad. Sci. U. S. A 93:7923-7926; 1996.
[49] Yang, Y.; Dieter, M. Z.; Chen, Y.; Shertzer, H. G.; Nebert, D. W.; Dalton, T. P. Initial
characterization of the glutamate-cysteine ligase modifier subunit Gclm(-/-) knockout
mouse. Novel model system for a severely compromised oxidative stress response. J. Biol.
Chem. 277:49446-49452; 2002.
[50] Shi, Z. Z.; Osei-Frimpong, J.; Kala, G.; Kala, S. V.; Barrios, R. J.; Habib, G. M.; Lukin, D.
J.; Danney, C. M.; Matzuk, M. M.; Lieberman, M. W. Glutathione synthesis is essential for
mouse development but not for cell growth in culture. Proc. Natl. Acad. Sci. U. S. A
97:5101-5106; 2000. 70
[51] Droge, W.; Gross, A.; Hack, V.; Kinscherf, R.; Schykowski, M.; Bockstette, M.; Mihm, S.;
Galter, D. Role of cysteine and glutathione in HIV infection and cancer cachexia:
therapeutic intervention with N-acetylcysteine. Adv. Pharmacol. 38:581-600; 1997.
[52] De Flora, S.; Izzotti, A.; D'Agostini, F.; Balansky, R. M. Mechanisms of N-acetylcysteine
in the prevention of DNA damage and cancer, with special reference to smoking-related
end-points. Carcinogenesis 22:999-1013; 2001.
[53] Sies, H. Ebselen, a selenoorganic compound as glutathione peroxidase mimic. Free Radic.
Biol. Med. 14:313-323; 1993.
[54] Parola, M.; Pinzani, M.; Casini, A.; Albano, E.; Poli, G.; Gentilini, A.; Gentilini, P.;
Dianzani, M. U. Stimulation of lipid peroxidation or 4-hydroxynonenal treatment increases
procollagen alpha 1 (I) gene expression in human liver fat-storing cells. Biochem. Biophys.
Res. Commun. 194:1044-1050; 1993.
[55] Nieto, N.; Friedman, S. L.; Cederbaum, A. I. Stimulation and proliferation of primary rat
hepatic stellate cells by cytochrome P450 2E1-derived reactive oxygen species. Hepatology
35:62-73; 2002.
[56] Tsukamoto, H.; Rippe, R.; Niemela, O.; Lin, M. Roles of oxidative stress in activation of
Kupffer and Ito cells in liver fibrogenesis. J. Gastroenterol. Hepatol. 10 Suppl 1:S50-S53;
1995.
[57] Parola, M.; Robino, G.; Marra, F.; Pinzani, M.; Bellomo, G.; Leonarduzzi, G.; Chiarugi, P.;
Camandola, S.; Poli, G.; Waeg, G.; Gentilini, P.; Dianzani, M. U. HNE interacts directly
with JNK isoforms in human hepatic stellate cells. J. Clin. Invest 102:1942-1950; 1998. 71
[58] Houglum, K.; Bedossa, P.; Chojkier, M. TGF-beta and collagen-alpha 1 (I) gene expression
are increased in hepatic acinar zone 1 of rats with iron overload. Am. J. Physiol 267:G908-
G913; 1994.
[59] Armendariz-Borunda, J.; Katai, H.; Jones, C. M.; Seyer, J. M.; Kang, A. H.; Raghow, R.
Transforming growth factor beta gene expression is transiently enhanced at a critical stage
during liver regeneration after CCl4 treatment. Lab Invest 69:283-294; 1993.
72
Chapter II Figure Legends
Fig. 1. Generation of Gclc floxed mice. (A) Schematic of the wild-type Gclc allele, targeting construct, targeted allele, and the deleted allele after Cre-mediated recombination. NEO, neomycin-resistance mini-cassette, and HSV-TK, herpes simplex virus thymidine kinase mini- cassette, represent genes used as selection markers. The positions of primers for PCR analysis are shown as arrows. (B) Southern blot analysis. The genomic DNA from mouse spleen was digested with Sph I and hybridized with the probe shown in part A. The 7.5-kb and 3.0-kb bands represent the Gclc(+) and Gclc(f) allele, respectively.
Fig. 2. Hepatocyte-specific conversion of Gclc(f) allele to Gclc(h) allele. (A) PCR analysis of genomic DNA from Gclc(f/f) (left lane) or Gclc(h/h) mice (the other lanes). Liver tissues were harvested at different time points as indicated. All other tissues were taken from Gclc(h/h) mice at postnatal day 28. K, kidney; P, pancreas; H, heart; L, lung; S, spleen; T, tail. (B) Southern blot analysis of liver DNA isolated from Gclc(f/f) or Gclc(h/h) mice at different time points. The 3.6 kb and 1.2 kb bands represent the Gclc(f) and Gclc(h) allele. (C). Western blot analysis for
GCLC and GCLM protein, respectively, from livers of Gclc(f/f) or Gclc(h/h) mice at day 14, 21,
28 (without NAC supplement), or day 120 (with NAC supplement since day 21). Results shown are representative of samples taken from 3 mice per group.
Fig. 3. Liver histology of Gclc(f/f) mice and Gclc(h/h) mice at d28. Light microscopy of livers from Gclc(f/f) mice (A) demonstrate normal looking hepatocytes. In contrast, Gclc(h/h) mice (B) showed areas of apoptosis, necrosis, and inflammation. Also evident are big nucleoli and condensation of DNA. Under electron microscopy, hepatocytes from Gclc(f/f) mice showed 73 stacks of rough endoplasmic reticulum (C), regular mitochondria (E), and small nucleoli (G). On the other hand, Gclc(h/h) mice were rich of small cytoplasmic vesicles (D) and had very little rough endoplasmic reticulum in hepatic cytoplasm. Mitochondria from these mice were swollen in some areas and compacted with tubular cristae in other areas (F). Nucleoli were also large with clumps of DNA, RNA, and ribonuclear protein constituents (G). RER, rough endoplasmic reticulum; V, cytoplasmic vesicles; M, mitochondria; No, nucleolus.
Fig. 4. Liver morphology and histology of Gclc(f/f) mice and Gclc(h/h) mice at d120 after
NAC supplement. Liver from Gclc(h/h) mice (B) were morphologically distinct from that of
Gclc(f/f) mice (A). By light microscopy, the NAC-rescued Gclc(h/h) mice showed areas of fibrosis (long arrow) and increases in both cytoplasmic and nuclear areas (short arrow) in hepatocytes.
Fig. 5. Liver mRNA levels in Gclc(f/f) and Gclc(h/h) mice. Liver samples were taken from days 14, 21, 28 (without NAC supplement), or day 120 (with NAC supplement since d21).
Results shown are representative of samples taken from 3-4 mice per group. HMOX1, heme oxygenase-1; MT1, metallothionein-1.
Fig. 6. Lipid peroxidation in livers of Gclc(f/f) mice and Gclc(h/h) mice. Livers were obtained from mice at d28 (without NAC supplement) or d120 (with NAC supplement since d21). After measurement, the absorbance was converted to MDA equivalents according to the malondialdehyde standard curve.
74
Table 1. GSH levels in tissues of Gclc(f/f) mice and Gclc(h/h) mice.
Age Gclc Liver Kidney Pancreas Lung Plasma
(f/f) 2.9 ± 0.2 0.9 ± 0.1 0.7 ± 0.1 0.7 ± 0.0 26.0 ± 2.0 Day † † 14 (h/h) 0.1 ± 0.0 0.9 ± 0.0 0.7 ± 0.0 0.6 ± 0.0 5.2 ± 0.5 (3%) (20%)
(f/f) 4.1 ± 0.1 1.5 ± 0.1 0.9 ± 0.2 0.7 ± 0.0 33.0 ± 2.6 Day † † † † † 21 (h/h) 0.1 ± 0.0 1.0 ± 0.1 0.4 ± 0.0 0.5 ± 0.0 5.4 ± 0.6 (2%) (67%) (44%) (71%) (16%)
(f/f) 4.4 ± 0.2 1.9 ± 0.1 0.9 ± 0.0 0.8 ± 0.0 59.0 ± 2.2 Day † † † † † 28 (h/h) 0.1 ± 0.0 1.2 ± 0.1 0.5 ± 0.0 0.5 ± 0.0 1.0 ± 0.1 (2%) (63%) (55%) (63%) (2%)
(f/f) 7.1 ± 0.2 2.1 ± 0.2 1.1 ± 0.0 0.9 ± 0.0 59.6 ± 5.9 Day † † 120* (h/h) 0.5 ± 0.0 1.8 ± 0.2 1.2 ± 0.2 0.9 ± 0.0 4.8 ± 0.6 (7%) (8%)
GSH levels are expressed as following: µmol/g tissues in liver, kidney, pancreas, and lung; µM in plasma. Values are means ± S.E.M. of 3 or 4 mice. Numbers in parenthesis are % of the Gclc(f/f). †p < 0.01 when comparing (f/f) and (h/h) mice. *Day 120: Sample was taken from mice supplemented with NAC since day 21.
75
Table 2. Mitochondria GSH levels in livers from Gclc(f/f) mice and Gclc(h/h) mice.
Gclc Day 21 Day 28 Day 120* (f/f) 1.53 ± 0.09 1.83 ± 0.13 4.15 ± 0.28 † † † (h/h) 0.04 ± 0.02 0.26 ± 0.02 0.77 ± 0.08
GSH levels are expressed as nmol /mg protein. Values are means ± S.E.M. of 3 or 4 mice. †p < 0.01 when comparing (f/f) and (h/h) mice. *Day 120: Sample was taken from mice supplemented with NAC since day 21.
76
Table 3. Cysteine levels in tissues of Gclc(f/f) mice and Gclc(h/h) mice.
Age Gclc Liver Kidney Pancreas Lung Day (f/f) 0.24 ± 0.01 0.53 ± 0.02 0.34 ± 0.02 0.17 ± 0.00 14 (h/h) 0.16 ± 0.01 0.45 ± 0.07 0.28 ± 0.03 0.16 ± 0.01
Day (f/f) 0.28 ± 0.01 0.55 ± 0.05 0.87 ± 0.13 0.14 ± 0.00 21 (h/h) 0.29 ± 0.04 0.51 ± 0.01 0.66 ± 0.05 0.14 ± 0.01
Day (f/f) 0.28 ± 0.07 1.34 ± 0.38 0.40 ± 0.00 0.19 ± 0.01 28 (h/h) 0.26 ± 0.03 1.09 ± 0.25 0.35 ± 0.03 0.15 ± 0.00
Cysteine levels are expressed as µmol/g tissues. Values are means ± S.E.M. of 3 or 4 mice.
77
Table 4. Liver biochemical functions in Gclc(f/f) mice and Gclc(h/h) mice.
conjugated unconjugated Age Gclc AST (IU/L) ALT (IU/L) bilirubin (mg/dl) Bilirubin (mg/dl) Day (f/f) 0.0 ± 0.0 0.0 ± 0.0 100 ± 5 30 ±3 21 (h/h) 0.4 ± 0.0 0.0 ± 0.0 681 ± 25 165 ± 10
Day (f/f) 0.0 ± 0.0 0.0 ± 0.0 122 ± 8 37 ± 2 28 (h/h) 9.2 ± 0.6 0.2 ± 0.0 3400 ± 261 1722 ± 124
Day 0.0 ± 0.0 0.0 ± 0.0 107 ± 8 28 ± 6 (f/f) 120* (h/h) 11.3 ± 1.0 0.6 ± 0.2 853 ± 31 428 ± 41
Values are means ± S.E.M. of 3 or 4 mice. *Day 120: Sample was taken from mice supplemented with NAC since day 21.
78
Fig. 1.
A.
B.
79
Fig. 2.
A.
B.
C.
80
Fig. 3.
81
Fig. 4.
82
Fig. 5.
83
Fig. 6.
84
Chapter III
Initial characterization of the glutamate-cysteine ligase modifier
subunit Gclm(-/-) knockout mouse: novel model system for a
severely compromised oxidative stress response
Published in Journal of Biological Chemistry 277 (51), December 2002
85
Discussion of the Entire Thesis
The intrinsic balance between life and death can be influenced by several environmental
stresses. Reactive oxygen species (ROS) are among the most potent threats faced by any living
organism. The accumulation of ROS can arise from normal metabolic reactions and/or toxic
insults. ROS are highly damaging, because they can attack almost any cellular structure or
macromolecules. To protect themselves against oxidative damage, cells have evolved a battery of
defense mechanisms involving both enzymatic and non-enzymatic strategies. However, some of
the antioxidants may not be essential because of the functional redundancy in biological systems.
For instance, mice deficient of SOD1, the mitochondrial form of superoxide dismutase, die
within 10 days of birth with dilated cardiomyopathy, metabolic acidosis, and lipid accumulation
in liver and skeletal muscle [92]. However, knockouts of either the cytosolic or the extracellular
form of SOD causes no overt phenotype under physiological conditions [93,94], suggesting that
their antioxidant functions can be substituted by other systems. Similarly, mice deficient of
GPX1, the major form of GPXs, are apparently healthy and show no increased sensitivity to
hyperoxia [95]. The results suggest that the contribution from a single antioxidant to the overall well being of the intact animal, in some cases, could be very limited under normal physiological conditions.
As the most abundant water-soluble antioxidant molecule in aerobic organisms, GSH has attracted tremendous attention as to its potential role as a cellular protection system. Strategies to increase GSH levels have been applied to protect against compound toxicity in a number of
studies [96-98]. In order to fully characterize the protective role of GSH, it is also equally
important to ask if GSH is essential for maintaining the cellular redox balance. In addition to its 86
antioxidant properties, GSH also participates in many other functions, including DNA synthesis, immune function, and membrane transport [30,54]. Is GSH absolutely required in these reactions?
In a first attempt to answer these questions, a Gclc conventional knockout mouse line was
generated by our lab and another group [99,100]. GCLC is the catalytic subunit of GCL, a rate- limiting enzyme in GSH biosynthesis pathway. Therefore, a complete depletion of GSH is expected when GCLC is genetically removed. Indeed, no GSH was detected in either the Gclc(-/-) knockout embryos or their derived cell lines. Compared to the chemical-induced GSH depletion approach, this model has the advantage of directly demonstrating the physiological function of
GSH without introducing any non-specific effects due to the addition of one or another toxic
chemical that most likely has nonspecific effects. At GD6.5, the Gclc(-/-) knockout embryos
could not be distinguished morphologically from their wild-type littermates; however, high rates of apoptosis began to be detected in the distal region of the mutant embryos. Theses mice died between GD7.5 and 8.5, indicating that GSH is essential for early mammalian development.
Attempts to rescue the mutant embryos failed––either by administration of N-acetylcysteine
(NAC) or the cell-permeable GSH monoethyl ester (GSH-EE) to pregnant dams. In contrast,
permanent cell lines derived from the mutant blastocysts can grow indefinitely in GSH-free
medium supplemented with NAC, indicating that GSH may not be required for cell growth under
culture conditions.
Does this scenario hold true in adult animals? To answer this question, we developed the
Gclc inducible knockout mouse line, in order to circumvent the embryonic lethality of the
conventional knockout mouse. Since liver is the major organ of GSH biosynthesis, the function
of GCLC was first tested in hepatocytes by breeding the Gclc inducible knockout mice with the
Cre transgenic mice under the control of an albumin promoter. Following Cre-mediated 87
recombination, Gclc was completely deleted by post-partum day 14 (d14). The homozygous hepatic knockout mice, Gclc(h/h), died of liver failure within one month, apparently as the result
of hepatic GSH depletion. Surprisingly, NAC treatment was able to rescue the Gclc(h/h) mice
into adulthood. However, unlike the normal phenotype seen in the cultured Gclc(-/-) cell line,
these mice developed a pathological condition characteristic of liver cirrhosis––despite the NAC rescue. This discrepancy suggests that, at least in the intact animal, some unique properties of
GSH cannot be replaced to maintain the normal cellular functions, whereas the cells in culture
might have developed some compensatory mechanisms through genetic changes after many cell passages. Given the persistence of severe oxidative damages in spite of NAC treatment, it seems that one of the essential functions of GSH is to maintain the cellular redox balance under normal physiological conditions. Additional studies are necessary to examine other potential targets of
GSH depletion.
Under normal cellular conditions, mitochondria are the major source of endogenous ROS.
Because mitochondria have no catalase, its removal of hydrogen peroxide is thought to depend solely on GSH and GPXs. Therefore, mitochondria might be the most sensitive organelle in response to GSH depletion. Indeed, a major effect of GSH deficiency in BSO-treated animal models has been shown to be mitochondria damage, which seemed well correlated with the extent of mitochondrial GSH depletion. Consistent with those findings, Gclc(h/h) mice also showed significant mitochondria swelling and degeneration in hepatocytes. Interestingly, such morphological changes were reversed by NAC treatment in both the Gclc(h/h) mice and the cultured Gclc(-/-) cell line. NAC is the acetylated form of cysteine and can serve as a source of cysteine for GSH biosynthesis. However, this explanation was ruled out by the complete absence of GCLC protein and GSH levels in these experimental models. Rather, the data suggest 88 that the contribution of GSH in mitochondria is not due to the GSH molecule itself, but some common properties shared by NAC. Besides its activity as GSH precursor, NAC is responsible for a number of protective functions––including direct free radical scavenger, nucleophilic detoxification, thiol-disulfide exchange, immunomodulation, and DNA replication and repair
[58,101,102]. It is not clear exactly on which of these properties NAC acts to compensate for the loss of GSH in the hepatocyte-specific knockout mice. Some studies in vitro have shown that
NAC can serve as an effective thiol substrate in the reaction catalyzed by the GPX-mimic ebselen, resulting in the reduction of hydrogen peroxide [103]. It is intriguing to imagine that such a reaction may actually occur in the intact animal, so that the role of GSH in mitochondria may be replaced by NAC supplement.
GSH depletion has also been implicated as a mediator of cell death in chemical-induced models. Depending on the cell types and inducers, the mode of cell death can be either apoptosis or necrosis. Low GSH levels have been associated with a number of pro-apoptotic signals–– including mitochondrial dysfunction, production of sphingomyelin metabolites, and the increased degradation of BCL2 [79,104-106]. In contrast, GSH depletion has also been shown to prevent caspase activation, leading to either cell survival or necrosis [107-110]. Similarly, the results from mouse Gclc knockout models are also mixed. In Gclc(-/-) conventional knockout embryos, depletion of GSH was associated with apoptosis; however, in cultured Gclc(-/-) cell lines, cell death after NAC or GSH withdrawal was not the result of apoptosis. Furthermore, both hepatic necrosis and apoptosis were observed in livers from the hepatocyte-specific knockout mice. These data demonstrate that the role of GSH depletion in cell death appears to be very complex. Cell death depends on the rate of GSH withdrawal, the extent of GSH depletion, as well as the intrinsic metabolic activities of the cells. 89
In E. coli, GCL is a single peptide; whereas it is a heterodimer composed of GCLC and
GCLM in mammalians. Based on the previous discussion, GCLC is certainly needed for animal
survival. Then the next question is: What is GCLM needed for? To answer this question, we first
generated a mouse model with the targeted disruption of the Gclm gene. Homozygous mutation
of Gclm results in an 85-90% depletion of GSH levels in mice tissues and plasma. Comparing
the GCL holoenzyme partially purified from livers of the wild-type mice, the GCLC apoenzyme
derived from the Gclm(-/-) mice had a ~2-fold increase in Km for glutamate and a dramatically
enhanced sensitivity to GSH inhibition. These data support the previous in vitro kinetic studies
that GCLM is able to maintain normal GSH levels by optimizing GCL activities. Despite the
severe GSH depletion, Gclm(-/-) mice are viable, fertile, and have no overt phenotype, at least
until 6 months of age, suggesting that GCLM is not essential for viability. However, preliminary
experiments demonstrated that Gclm(-/-) mice were more sensitive to the heavy metal cadmium than their wild-type littermates. A strikingly increased sensitivity to H2O2 was also shown in the
mouse fetal fibroblasts (MFFs) derived from these knockout animals. Furthermore, increases in
the initial GSH levels by GSH-EE offered little protection in Gclm(-/-) MFFs. These findings
suggest that the sensitivity to oxidant insults in Gclm(-/-) is more likely due to the diminished
GCL activity rather than to the lowered levels of GSH. Therefore, the major function of GCLM
may be to improve GSH synthetic capacity as a defense mechanism against oxidative insults. It
also follows that at physiological conditions, GCLM may have a limited contribution to maintain
cellular redox balance; however, GCLM becomes critical when such balance is disturbed under
pathological condition or facing environmental insults. In addition, the process of aging has been
associated with enhanced ROS generation and decreased antioxidant activities [111,112]. It is
very likely that the Gclm (-/-) mice may eventually develop symptoms as they reach senescence. 90
These mouse genetic models have demonstrated that both GCLC and GCLM are responsible for maintaining GSH homeostasis in the intact animal. It is important to understand that GSH levels are also tightly regulated at the level of these two subunits. Once synthesized,
GCLC and GCLM may form heterodimers through non-covalent interaction. Deletion of a small number of amino acids in either subunit resulted a dramatic loss of heterodimer formation, suggesting that such interaction is mediated through multiple and separate regions of both subunits. In addition, the interaction between GCLC and GCLM also involves the formation of inter-subunit disulfide bonds, which can be easily reduced by low levels of GSH [87]. The available data suggest that both covalent and non-covalent interactions are required for the improved kinetic properties of the GCL holoenzyme [113,114]. Therefore, any factors that interfere with the physical interaction between the subunits would have significant effects on
GCL activity, leading to changes in GSH levels. For instance, GCLC undergoes caspase-3- dependent cleavage during apoptotic cell death [115]. Such cleavage does not seem to abolish
GCL activity when assayed under the standard conditions (i.e. with saturating glutamate and no
GSH) [116]. However, compared with the full-length protein, the truncated form of GCLC,
∆491-637, retained only 29% binding activity to GCLM in experiments using the yeast two- hybrid system. Since GCL activity is highly dependent on heterodimer formation, such lack of interaction may partially explain the observed GSH depletion in apoptotic cells. In addition, cellular redox status may also regulate GCL activity through the reversible disulfide bonds between GCLC and GCLM. The oxidizing conditions that result in GSH depletion may promote
GSH recovery by the formation of disulfide bonds. Physiological concentrations of GSH, on the other hand, reduce the intermolecular disulfide, producing conformational changes that favor
GSH feedback inhibition. 91
Besides the physical interactions between GCLC and GCLM, GCL activity is also regulated at the level of GCLC and GCLM gene expression. Studies have shown that the genes encoding both subunits can be up-regulated by a variety of chemicals and stress conditions, leading to increased levels of GSH [69,70]. However, in most cases, the regulation of the two subunits is not coordinated. Furthermore, even under normal physiological conditions, GCLC and GCLM are expressed at different rates. For instance, only free GCLC protein was detected in
our gel filtration study when liver homogenate was applied, indicating that GCLM may be limiting in liver. On the other hand, there seems to be an excess amount of GCLM in mouse fetal fibroblasts. Such tissue-specific expression may explain why some cells are responsive to up- or down-regulation of GCLM, and others are sensitive to GCLC levels.
In terms of the regulation of GCL activity, it is also important to consider the contribution of human genetic polymorphisms on GCL activity. The occurrence of such polymorphisms, which affect either gene expression or enzyme activity, may play an important role in the inter- individual variabilities in response to oxidative stress. At least two SNPs in the coding regions of the GCLC gene have been documented in some GCLC-deficient patients. In both cases, the
altered amino acid led to a significant loss of enzyme activity [74]. In addition, there exists a
polymorphic trinucleotide repeat in the human GCLC promoter region [117]. Some variants, as
defined by the number of repeats, are associated with GSH levels and/or drug sensitivity in the
screening of 60 tumor cell lines. In a similar manner, a functional polymorphism in the 5'-flanking
region of the human GCLM gene has been identified, and it is associated with an increased risk of myocardial infarction [118]. Because of our findings in Chapter I, it is also possible to envision that some polymorphic sequences may change GCL activity by affecting the interaction 92
between the two subunits. If so, the yeast two-hybrid system described in this dissertation will be
an efficient approach to screening for such variants.
The role of GSH in oxidative stress-mediated processes has been a very active field of
research in recent years. At the molecular level, GSH is suggested to be involved in signal
transduction, gene expression, and apoptosis through oxidative stress [119,120]. In population
studies, decreased GSH content has been found in a variety of aged tissues and in different
pathologic states, such as neurodegenerative disorders, cataracts, atherosclerosis, and
inflammatory diseases [121-123]. GSH deficiency has also been considered as an important
factor in assessing risk from environmental exposure to oxidants [124]. The study presented here
clearly shows that GCL, the rate-limiting enzyme of GSH biosynthesis, is critical in maintaining cellular GSH homeostasis. The existence of both GCLC and GCLM subunits, and their interaction, represents an important mechanism in defending against oxidative stress. Removal of any component of this rate-limiting step inevitably leads to decreases in GSH levels in the intact animal, as a result of decreased capacity for GSH biosynthesis. Depending on the extent of GSH depletion, the Gclc(-/-) or Gclm(-/-) knockout mice show either cellular damage as a result of endogenous oxidative injury, increased sensitivity to exogenous insults, or both. These genetically compromised mice provide excellent model systems to study the role of GSH in numerous pathophysiological conditions, as well as in chemical-mixture toxicities associated with oxidant insults. Furthermore, these model systems will allow researchers to evaluate potential therapeutic drugs to combat against oxidative stress and to promote better health conditions. 93
References
[1] Sies, H. Oxidative stress II. Oxidants and Antioxidants.: London Academic Press; 1991.
[2] Halliwell, B.; Cross, C. E. Oxygen-derived species: their relation to human disease and
environmental stress. Environ. Health Perspect. 102 Suppl 10:5-12; 1994.
[3] Aliev, G.; Smith, M. A.; Seyidov, D.; Neal, M. L.; Lamb, B. T.; Nunomura, A.; Gasimov,
E. K.; Vinters, H. V.; Perry, G.; LaManna, J. C.; Friedland, R. P. The role of oxidative
stress in the pathophysiology of cerebrovascular lesions in Alzheimer's disease. Brain
Pathol. 12:21-35; 2002.
[4] Penckofer, S.; Schwertz, D.; Florczak, K. Oxidative stress and cardiovascular disease in
type 2 diabetes: the role of antioxidants and pro-oxidants. J. Cardiovasc. Nurs. 16:68-85;
2002.
[5] Southorn, P. A.; Powis, G. Free radicals in medicine. II. Involvement in human disease.
Mayo Clin. Proc. 63:390-408; 1988.
[6] Fridovich, I. Superoxide radical and superoxide dismutases. Annu. Rev. Biochem. 64:97-
112; 1995.
[7] Halliwell, B.; Gutteridge, J. The chemistry of free radicals and related 'reactive species'.
Free radicals in Biology and Medicine. Oxford University press; 1999:36-104.
[8] Cadenas, E.; Davies, K. J. Mitochondrial free radical generation, oxidative stress, and
aging. Free Radic. Biol. Med. 29:222-230; 2000. 94
[9] Yeldandi, A. V.; Rao, M. S.; Reddy, J. K. Hydrogen peroxide generation in peroxisome
proliferator-induced oncogenesis. Mutat. Res. 448:159-177; 2000.
[10] Puntarulo, S.; Cederbaum, A. I. Production of reactive oxygen species by microsomes
enriched in specific human cytochrome P450 enzymes. Free Radic. Biol. Med. 24:1324-
1330; 1998.
[11] Nebert, D. W.; Roe, A. L.; Dieter, M. Z.; Solis, W. A.; Yang, Y.; Dalton, T. P. Role of
the aromatic hydrocarbon receptor and [Ah] gene battery in the oxidative stress response,
cell cycle control, and apoptosis. Biochem. Pharmacol. 59:65-85; 2000.
[12] Saito, M.; Thomas, C. E.; Aust, S. D. Paraquat and ferritin-dependent lipid peroxidation.
J. Free Radic. Biol. Med. 1:179-185; 1985.
[13] Goeptar, A. R.; Te Koppele, J. M.; Lamme, E. K.; Pique, J. M.; Vermeulen, N. P.
Cytochrome P450 2B1-mediated one-electron reduction of adriamycin: a study with rat
liver microsomes and purified enzymes. Mol. Pharmacol. 44:1267-1277; 1993.
[14] Robinson, J. M.; Badwey, J. A. Production of active oxygen species by phagocytic
leukocytes. Immunol. Ser. 60:159-178; 1994.
[15] Hille, R.; Nishino, T. Flavoprotein structure and mechanism. 4. Xanthine oxidase and
xanthine dehydrogenase. FASEB J. 9:995-1003; 1995.
[16] Dizdaroglu, M.; Jaruga, P.; Birincioglu, M.; Rodriguez, H. Free radical-induced damage
to DNA: mechanisms and measurement. Free Radic. Biol. Med. 32:1102-1115; 2002. 95
[17] Ames, B. N.; Gold, L. S. Endogenous mutagens and the causes of aging and cancer.
Mutat. Res. 250:3-16; 1991.
[18] Lee, S. H.; Blair, I. A. Oxidative DNA damage and cardiovascular disease. Trends
Cardiovasc. Med. 11:148-155; 2001.
[19] Richter, C. Biophysical consequences of lipid peroxidation in membranes. Chem. Phys.
Lipids 44:175-189; 1987.
[20] Romero, F. J.; Bosch-Morell, F.; Romero, M. J.; Jareno, E. J.; Romero, B.; Marin, N.;
Roma, J. Lipid peroxidation products and antioxidants in human disease. Environ. Health
Perspect. 106 Suppl 5:1229-1234; 1998.
[21] Linton, S.; Davies, M. J.; Dean, R. T. Protein oxidation and ageing. Exp. Gerontol.
36:1503-1518; 2001.
[22] Squier, T. C. Oxidative stress and protein aggregation during biological aging. Exp.
Gerontol. 36:1539-1550; 2001.
[23] de Zwart, L. L.; Meerman, J. H.; Commandeur, J. N.; Vermeulen, N. P. Biomarkers of
free radical damage applications in experimental animals and in humans. Free Radic.
Biol. Med. 26:202-226; 1999.
[24] Chandra, J.; Samali, A.; Orrenius, S. Triggering and modulation of apoptosis by oxidative
stress. Free Radic. Biol. Med. 29:323-333; 2000.
[25] Mates, M. Effects of antioxidant enzymes in the molecular control of reactive oxygen
species toxicology. Toxicology 153:83-104; 2000. 96
[26] Meister, A. Mitochondrial changes associated with glutathione deficiency. Biochim.
Biophys. Acta 1271:35-42; 1995.
[27] Chen, H.; Tappel, A. L. Protection by vitamin E selenium, trolox C, ascorbic acid
palmitate, acetylcysteine, coenzyme Q, beta-carotene, canthaxanthin, and (+)-catechin
against oxidative damage to liver slices measured by oxidized heme proteins. Free Radic.
Biol. Med. 16:437-444; 1994.
[28] Meister, A.; Anderson, M. E. Glutathione. Annu. Rev. Biochem. 52:711-760; 1983.
[29] Meister, A. Glutathione metabolism. Methods Enzymol. 251:3-7; 1995.
[30] Hammond, C. L.; Lee, T. K.; Ballatori, N. Novel roles for glutathione in gene expression,
cell death, and membrane transport of organic solutes. J. Hepatol. 34:946-954; 2001.
[31] Sies, H. Glutathione and its role in cellular functions. Free Radic. Biol. Med. 27:916-921;
1999.
[32] Smith, C. V.; Jones, D. P.; Guenthner, T. M.; Lash, L. H.; Lauterburg, B. H.
Compartmentation of glutathione: implications for the study of toxicity and disease.
Toxicol. Appl. Pharmacol. 140:1-12; 1996.
[33] Colell, A.; Garcia-Ruiz, C.; Miranda, M.; Ardite, E.; Mari, M.; Morales, A.; Corrales, F.;
Kaplowitz, N.; Fernandez-Checa, J. C. Selective glutathione depletion of mitochondria
by ethanol sensitizes hepatocytes to tumor necrosis factor. Gastroenterology 115:1541-
1551; 1998. 97
[34] Roychowdhury, S.; Wolf, G.; Keilhoff, G.; Horn, T. F. Cytosolic and mitochondrial
glutathione in microglial cells are differentially affected by oxidative/nitrosative stress.
Nitric. Oxide. 8:39-47; 2003.
[35] Kaplowitz, N.; Fernandez-Checa, J. C.; Kannan, R.; Garcia-Ruiz, C.; Ookhtens, M.; Yi, J.
R. GSH transporters: molecular characterization and role in GSH homeostasis. Biol.
Chem. Hoppe Seyler 377:267-273; 1996.
[36] Hwang, C.; Sinskey, A. J.; Lodish, H. F. Oxidized redox state of glutathione in the
endoplasmic reticulum. Science 257:1496-1502; 1992.
[37] Anderson, M. E. Glutathione: an overview of biosynthesis and modulation. Chem. Biol.
Interact. 111-112:1-14; 1998.
[38] Lu, S. C. Regulation of hepatic glutathione synthesis: current concepts and controversies.
FASEB J. 13:1169-1183; 1999.
[39] Yan, N.; Meister, A. Amino acid sequence of rat kidney gamma-glutamylcysteine
synthetase. J. Biol. Chem. 265:1588-1593; 1990.
[40] Oppenheimer, L.; Wellner, V. P.; Griffith, O. W.; Meister, A. Glutathione synthetase.
Purification from rat kidney and mapping of the substrate binding sites. J. Biol. Chem.
254:5184-5190; 1979.
[41] Grant, C. M.; MacIver, F. H.; Dawes, I. W. Glutathione synthetase is dispensable for
growth under both normal and oxidative stress conditions in the yeast Saccharomyces 98
cerevisiae due to an accumulation of the dipeptide gamma-glutamylcysteine. Mol. Biol.
Cell 8:1699-1707; 1997.
[42] Mulcahy, R. T.; Bailey, H. H.; Gipp, J. J. Transfection of complementary DNAs for the
heavy and light subunits of human gamma-glutamylcysteine synthetase results in an
elevation of intracellular glutathione and resistance to melphalan. Cancer Res. 55:4771-
4775; 1995.
[43] Tipnis, S. R.; Blake, D. G.; Shepherd, A. G.; McLellan, L. I. Overexpression of the
regulatory subunit of gamma-glutamylcysteine synthetase in HeLa cells increases
gamma-glutamylcysteine synthetase activity and confers drug resistance. Biochem. J. 337
( Pt 3):559-566; 1999.
[44] DeLeve, L. D.; Kaplowitz, N. Glutathione metabolism and its role in hepatotoxicity.
Pharmacol. Ther. 52:287-305; 1991.
[45] Holmgren, A. Antioxidant function of thioredoxin and glutaredoxin systems. Antioxid.
Redox. Signal. 2:811-820; 2000.
[46] Meister, A. Glutathione-ascorbic acid antioxidant system in animals. J. Biol. Chem.
269:9397-9400; 1994.
[47] Griffith, O. W.; Mulcahy, R. T. The enzymes of glutathione synthesis: gamma-
glutamylcysteine synthetase. Adv. Enzymol. Relat Areas Mol. Biol. 73:209-67, xii; 1999.
[48] Orlowski, M.; Meister, A. The gamma-glutamyl cycle: a possible transport system for
amino acids. Proc. Natl. Acad. Sci. U. S. A 67:1248-1255; 1970. 99
[49] Thompson, G. A.; Meister, A. Utilization of L-cystine by the gamma-glutamyl
transpeptidase-gamma-glutamyl cyclotransferase pathway. Proc. Natl. Acad. Sci. U. S. A
72:1985-1988; 1975.
[50] Ookhtens, M.; Kaplowitz, N. Role of the liver in interorgan homeostasis of glutathione
and cyst(e)ine. Semin. Liver Dis. 18:313-329; 1998.
[51] Zhang, L. P.; Maiorino, M.; Roveri, A.; Ursini, F. Phospholipid hydroperoxide
glutathione peroxidase: specific activity in tissues of rats of different age and comparison
with other glutathione peroxidases. Biochim. Biophys. Acta 1006:140-143; 1989.
[52] Cotgreave, I. A.; Gerdes, R. G. Recent trends in glutathione biochemistry--glutathione-
protein interactions: a molecular link between oxidative stress and cell proliferation?
Biochem. Biophys. Res. Commun. 242:1-9; 1998.
[53] Olney, J. W.; Zorumski, C.; Price, M. T.; Labruyere, J. L-cysteine, a bicarbonate-
sensitive endogenous excitotoxin. Science 248:596-599; 1990.
[54] Manciu, L.; Chang, X. B.; Buyse, F.; Hou, Y. X.; Gustot, A.; Riordan, J. R.; Ruysschaert,
J. M. Intermediate structural states involved in MRP1-mediated drug transport. Role of
glutathione. J. Biol. Chem. 278:3347-3356; 2003.
[55] Bains, J. S.; Shaw, C. A. Neurodegenerative disorders in humans: the role of glutathione
in oxidative stress-mediated neuronal death. Brain Res. Brain Res. Rev. 25:335-358; 1997.
[56] Giblin, F. J. Glutathione: a vital lens antioxidant. J. Ocul. Pharmacol. Ther. 16:121-135;
2000. 100
[57] Khamaisi, M.; Kavel, O.; Rosenstock, M.; Porat, M.; Yuli, M.; Kaiser, N.; Rudich, A.
Effect of inhibition of glutathione synthesis on insulin action: in vivo and in vitro studies
using buthionine sulfoximine. Biochem. J. 349:579-586; 2000.
[58] Droge, W.; Gross, A.; Hack, V.; Kinscherf, R.; Schykowski, M.; Bockstette, M.; Mihm,
S.; Galter, D. Role of cysteine and glutathione in HIV infection and cancer cachexia:
therapeutic intervention with N-acetylcysteine. Adv. Pharmacol. 38:581-600; 1997.
[59] Benzi, G.; Moretti, A. Age- and peroxidative stress-related modifications of the cerebral
enzymatic activities linked to mitochondria and the glutathione system. Free Radic. Biol.
Med. 19:77-101; 1995.
[60] Jenner, P. Altered mitochondrial function, iron metabolism and glutathione levels in
Parkinson's disease. Acta Neurol. Scand. Suppl 146:6-13; 1993.
[61] Shigesawa, T.; Sato, C.; Marumo, F. Significance of plasma glutathione determination in
patients with alcoholic and non-alcoholic liver disease. J. Gastroenterol. Hepatol. 7:7-11;
1992.
[62] Purucker, E.; Winograd, R.; Roeb, E.; Matern, S. Glutathione status in liver and plasma
during development of biliary cirrhosis after bile duct ligation. Res. Exp. Med. (Berl)
198:167-174; 1998.
[63] Poli, G.; Parola, M. Oxidative damage and fibrogenesis. Free Radic. Biol. Med. 22:287-
305; 1997. 101
[64] Lang, C. A.; Naryshkin, S.; Schneider, D. L.; Mills, B. J.; Lindeman, R. D. Low blood
glutathione levels in healthy aging adults. J. Lab Clin. Med. 120:720-725; 1992.
[65] Julius, M.; Lang, C. A.; Gleiberman, L.; Harburg, E.; DiFranceisco, W.; Schork, A.
Glutathione and morbidity in a community-based sample of elderly. J. Clin. Epidemiol.
47:1021-1026; 1994.
[66] Watanabe, K.; Yamano, Y.; Murata, K.; Kimura, A. The nucleotide sequence of the gene
for gamma-glutamylcysteine synthetase of Escherichia coli. Nucleic Acids Res. 14:4393-
4400; 1986.
[67] Huang, C. S.; Anderson, M. E.; Meister, A. Amino acid sequence and function of the
light subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol. Chem.
268:20578-20583; 1993.
[68] Tsuchiya, K.; Mulcahy, R. T.; Reid, L. L.; Disteche, C. M.; Kavanagh, T. J. Mapping of
the glutamate-cysteine ligase catalytic subunit gene (GLCLC) to human chromosome
6p12 and mouse chromosome 9D-E and of the regulatory subunit gene (GLCLR) to
human chromosome 1p21-p22 and mouse chromosome 3H1-3. Genomics 30:630-632;
1995.
[69] Soltaninassab, S. R.; Sekhar, K. R.; Meredith, M. J.; Freeman, M. L. Multi-faceted
regulation of gamma-glutamylcysteine synthetase. J. Cell Physiol 182:163-170; 2000.
[70] Wild, A. C.; Mulcahy, R. T. Regulation of gamma-glutamylcysteine synthetase subunit
gene expression: insights into transcriptional control of antioxidant defenses. Free Radic.
Res. 32:281-301; 2000. 102
[71] Gipp, J. J.; Bailey, H. H.; Mulcahy, R. T. Cloning and sequencing of the cDNA for the
light subunit of human liver gamma-glutamylcysteine synthetase and relative mRNA
levels for heavy and light subunits in human normal tissues. Biochem. Biophys. Res.
Commun. 206:584-589; 1995.
[72] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[73] Beutler, E.; Gelbart, T.; Kondo, T.; Matsunaga, A. T. The molecular basis of a case of
gamma-glutamylcysteine synthetase deficiency. Blood 94:2890-2894; 1999.
[74] Ristoff, E.; Augustson, C.; Geissler, J.; de Rijk, T.; Carlsson, K.; Luo, J. L.; Andersson,
K.; Weening, R. S.; van Zwieten, R.; Larsson, A.; Roos, D. A missense mutation in the
heavy subunit of gamma-glutamylcysteine synthetase gene causes hemolytic anemia.
Blood 95:2193-2196; 2000.
[75] Ristoff, E.; Larsson, A. Patients with genetic defects in the gamma-glutamyl cycle. Chem.
Biol. Interact. 111-112:113-121; 1998.
[76] Sugimoto, C.; Matsukawa, S.; Fujieda, S.; Noda, I.; Tanaka, N.; Tsuzuki, H.; Saito, H.
Involvement of intracellular glutathione in induction of apoptosis by cisplatin in a human
pharyngeal carcinoma cell line. Anticancer Res. 16:675-680; 1996.
[77] Vahrmeijer, A. L.; van Dierendonck, J. H.; Schutrups, J.; van de Velde, C. J.; Mulder, G.
J. Effect of glutathione depletion on inhibition of cell cycle progression and induction of 103
apoptosis by melphalan (L-phenylalanine mustard) in human colorectal cancer cells.
Biochem. Pharmacol. 58:655-664; 1999.
[78] Shen, H.; Yang, C.; Liu, J.; Ong, C. Dual role of glutathione in selenite-induced oxidative
stress and apoptosis in human hepatoma cells. Free Radic. Biol. Med. 28:1115-1124;
2000.
[79] Merad-Boudia, M.; Nicole, A.; Santiard-Baron, D.; Saille, C.; Ceballos-Picot, I.
Mitochondrial impairment as an early event in the process of apoptosis induced by
glutathione depletion in neuronal cells: relevance to Parkinson's disease. Biochem.
Pharmacol. 56:645-655; 1998.
[80] Seyfried, J.; Soldner, F.; Schulz, J. B.; Klockgether, T.; Kovar, K. A.; Wullner, U.
Differential effects of L-buthionine sulfoximine and ethacrynic acid on glutathione levels
and mitochondrial function in PC12 cells. Neurosci. Lett. 264:1-4; 1999.
[81] Wullner, U.; Seyfried, J.; Groscurth, P.; Beinroth, S.; Winter, S.; Gleichmann, M.;
Heneka, M.; Loschmann, P.; Schulz, J. B.; Weller, M.; Klockgether, T. Glutathione
depletion and neuronal cell death: the role of reactive oxygen intermediates and
mitochondrial function. Brain Res. 826:53-62; 1999.
[82] Meister, A. Glutathione deficiency produced by inhibition of its synthesis, and its
reversal; applications in research and therapy. Pharmacol. Ther. 51:155-194; 1991.
[83] Bailey, H. H.; Ripple, G.; Tutsch, K. D.; Arzoomanian, R. Z.; Alberti, D.; Feierabend, C.;
Mahvi, D.; Schink, J.; Pomplun, M.; Mulcahy, R. T.; Wilding, G. Phase I study of 104
continuous-infusion L-S,R-buthionine sulfoximine with intravenous melphalan. J. Natl.
Cancer Inst. 89:1789-1796; 1997.
[84] Malaker, K.; Hurwitz, S. J.; Bump, E. A.; Griffith, O. W.; Lai, L. L.; Riese, N.; Coleman,
C. N. Pharmacodynamics of prolonged treatment with L,S-buthionine sulfoximine. Int. J.
Radiat. Oncol. Biol. Phys. 29:407-412; 1994.
[85] Fraser, J. A.; Saunders, R. D.; McLellan, L. I. Drosophila melanogaster glutamate-
cysteine ligase activity is regulated by a modifier subunit with a mechanism of action
similar to that of the mammalian form. J. Biol. Chem. 277:1158-1165; 2002.
[86] Misra, I.; Griffith, O. W. Expression and purification of human gamma-glutamylcysteine
synthetase. Protein Expr. Purif. 13:268-276; 1998.
[87] Seelig, G. F.; Simondsen, R. P.; Meister, A. Reversible dissociation of gamma-
glutamylcysteine synthetase into two subunits. J. Biol. Chem. 259:9345-9347; 1984.
[88] Tu, Z.; Anders, M. W. Identification of an important cysteine residue in human
glutamate-cysteine ligase catalytic subunit by site-directed mutagenesis. Biochem. J. 336
( Pt 3):675-680; 1998.
[89] Chang, L. S.; Chang, C. C. Biochemical regulation of the activity of gamma-
glutamylcysteine synthetase from rat liver and kidney by glutathione. Biochem. Mol. Biol.
Int. 32:697-703; 1994.
[90] Kijima, H.; Tsuchida, T.; Kondo, H.; Iida, T.; Oshika, Y.; Nakamura, M.; Scanlon, K. J.;
Kondo, T.; Tamaoki, N. Hammerhead ribozymes against gamma-glutamylcysteine 105
synthetase mRNA down-regulate intracellular glutathione concentration of mouse islet
cells. Biochem. Biophys. Res. Commun. 247:697-703; 1998.
[91] Yao, K. S.; Godwin, A. K.; Johnson, S. W.; Ozols, R. F.; O'Dwyer, P. J.; Hamilton, T. C.
Evidence for altered regulation of gamma-glutamylcysteine synthetase gene expression
among cisplatin-sensitive and cisplatin-resistant human ovarian cancer cell lines. Cancer
Res. 55:4367-4374; 1995.
[92] Li, Y.; Huang, T. T.; Carlson, E. J.; Melov, S.; Ursell, P. C.; Olson, J. L.; Noble, L. J.;
Yoshimura, M. P.; Berger, C.; Chan, P. H.; . Dilated cardiomyopathy and neonatal
lethality in mutant mice lacking manganese superoxide dismutase. Nat. Genet. 11:376-
381; 1995.
[93] Reaume, A. G.; Elliott, J. L.; Hoffman, E. K.; Kowall, N. W.; Ferrante, R. J.; Siwek, D.
F.; Wilcox, H. M.; Flood, D. G.; Beal, M. F.; Brown, R. H., Jr.; Scott, R. W.; Snider, W.
D. Motor neurons in Cu/Zn superoxide dismutase-deficient mice develop normally but
exhibit enhanced cell death after axonal injury. Nat. Genet. 13:43-47; 1996.
[94] Carlsson, L. M.; Jonsson, J.; Edlund, T.; Marklund, S. L. Mice lacking extracellular
superoxide dismutase are more sensitive to hyperoxia. Proc. Natl. Acad. Sci. U. S. A
92:6264-6268; 1995.
[95] Ho, Y. S.; Magnenat, J. L.; Bronson, R. T.; Cao, J.; Gargano, M.; Sugawara, M.; Funk, C.
D. Mice deficient in cellular glutathione peroxidase develop normally and show no
increased sensitivity to hyperoxia. J. Biol. Chem. 272:16644-16651; 1997. 106
[96] Sakamaki, H.; Akazawa, S.; Ishibashi, M.; Izumino, K.; Takino, H.; Yamasaki, H.;
Yamaguchi, Y.; Goto, S.; Urata, Y.; Kondo, T.; Nagataki, S. Significance of glutathione-
dependent antioxidant system in diabetes- induced embryonic malformations. Diabetes
48:1138-1144; 1999.
[97] Obrador, E.; Navarro, J.; Mompo, J.; Asensi, M.; Pellicer, J. A.; Estrela, J. M. Regulation
of tumour cell sensitivity to TNF-induced oxidative stress and cytotoxicity: role of
glutathione. Biofactors 8:23-26; 1998.
[98] Yamaguchi, N.; Yajima, N.; Ishida, M.; Shimada, T.; Hiraishi, H. Protection of cultured
gastric cells against tert-butyl hydroperoxide by glutathione isopropyl ester. Eur. J.
Pharmacol. 351:363-369; 1998.
[99] Dalton, T. P.; Dieter, M. Z.; Yang, Y.; Shertzer, H. G.; Nebert, D. W. Knockout of the
mouse glutamate cysteine ligase catalytic subunit (Gclc) gene: embryonic lethal when
homozygous, and proposed model for moderate glutathione deficiency when
heterozygous. Biochem. Biophys. Res. Commun. 279:324-329; 2000.
[100] Shi, Z. Z.; Osei-Frimpong, J.; Kala, G.; Kala, S. V.; Barrios, R. J.; Habib, G. M.; Lukin,
D. J.; Danney, C. M.; Matzuk, M. M.; Lieberman, M. W. Glutathione synthesis is
essential for mouse development but not for cell growth in culture. Proc. Natl. Acad. Sci.
U. S. A 97:5101-5106; 2000.
[101] Moldeus, P.; Cotgreave, I. A. N-acetylcysteine. Methods Enzymol. 234:482-492; 1994. 107
[102] De Flora, S.; Izzotti, A.; D'Agostini, F.; Balansky, R. M. Mechanisms of N-acetylcysteine
in the prevention of DNA damage and cancer, with special reference to smoking-related
end-points. Carcinogenesis 22:999-1013; 2001.
[103] Sies, H. Ebselen, a selenoorganic compound as glutathione peroxidase mimic. Free Radic.
Biol. Med. 14:313-323; 1993.
[104] Celli, A.; Que, F. G.; Gores, G. J.; LaRusso, N. F. Glutathione depletion is associated
with decreased Bcl-2 expression and increased apoptosis in cholangiocytes. Am. J.
Physiol 275:G749-G757; 1998.
[105] Liu, B.; Andrieu-Abadie, N.; Levade, T.; Zhang, P.; Obeid, L. M.; Hannun, Y. A.
Glutathione regulation of neutral sphingomyelinase in tumor necrosis factor-alpha-
induced cell death. J. Biol. Chem. 273:11313-11320; 1998.
[106] Okamoto, Y.; Obeid, L. M.; Hannun, Y. A. Bcl-xL interrupts oxidative activation of
neutral sphingomyelinase. FEBS Lett. 530:104-108; 2002.
[107] Hentze, H.; Kunstle, G.; Volbracht, C.; Ertel, W.; Wendel, A. CD95-Mediated murine
hepatic apoptosis requires an intact glutathione status. Hepatology 30:177-185; 1999.
[108] Hentze, H.; Gantner, F.; Kolb, S. A.; Wendel, A. Depletion of hepatic glutathione
prevents death receptor-dependent apoptotic and necrotic liver injury in mice. Am. J.
Pathol. 156:2045-2056; 2000.
[109] Galan, A.; Garcia-Bermejo, L.; Troyano, A.; Vilaboa, N. E.; Fernandez, C.; De Blas, E.;
Aller, P. The role of intracellular oxidation in death induction (apoptosis and necrosis) in 108
human promonocytic cells treated with stress inducers (cadmium, heat, X-rays). Eur. J.
Cell Biol. 80:312-320; 2001.
[110] Hampton, M. B.; Fadeel, B.; Orrenius, S. Redox regulation of the caspases during
apoptosis. Ann. N. Y. Acad. Sci. 854:328-335; 1998.
[111] Sohal, R. S.; Agarwal, S.; Sohal, B. H. Oxidative stress and aging in the Mongolian
gerbil (Meriones unguiculatus). Mech. Ageing Dev. 81:15-25; 1995.
[112] Mo, J. Q.; Hom, D. G.; Andersen, J. K. Decreases in protective enzymes correlates with
increased oxidative damage in the aging mouse brain. Mech. Ageing Dev. 81:73-82; 1995.
[113] Huang, C. S.; Chang, L. S.; Anderson, M. E.; Meister, A. Catalytic and regulatory
properties of the heavy subunit of rat kidney gamma-glutamylcysteine synthetase. J. Biol.
Chem. 268:19675-19680; 1993.
[114] Tu, Z.; Anders, M. W. Expression and characterization of human glutamate-cysteine
ligase. Arch. Biochem. Biophys. 354:247-254; 1998.
[115] Franklin, C. C.; Krejsa, C. M.; Pierce, R. H.; White, C. C.; Fausto, N.; Kavanagh, T. J.
Caspase-3-Dependent Cleavage of the Glutamate-L-Cysteine Ligase Catalytic Subunit
during Apoptotic Cell Death. Am. J. Pathol. 160:1887-1894; 2002.
[116] Siitonen, T.; Alaruikka, P.; Mantymaa, P.; Savolainen, E. R.; Kavanagh, T. J.; Krejsa, C.
M.; Franklin, C. C.; Kinnula, V.; Koistinen, P. Protection of acute myeloblastic leukemia
cells against apoptotic cell death by high glutathione and gamma-glutamylcysteine 109
synthetase levels during etoposide-induced oxidative stress. Ann. Oncol. 10:1361-1367;
1999.
[117] Walsh, A. C.; Feulner, J. A.; Reilly, A. Evidence for functionally significant
polymorphism of human glutamate cysteine ligase catalytic subunit: association with
glutathione levels and drug resistance in the National Cancer Institute tumor cell line
panel. Toxicol. Sci. 61:218-223; 2001.
[118] Nakamura, S.; Kugiyama, K.; Sugiyama, S.; Miyamoto, S.; Koide, S.; Fukushima, H.;
Honda, O.; Yoshimura, M.; Ogawa, H. Polymorphism in the 5'-flanking region of human
glutamate-cysteine ligase modifier subunit gene is associated with myocardial infarction.
Circulation 105:2968-2973; 2002.
[119] Arrigo, A. P. Gene expression and the thiol redox state. Free Radic. Biol. Med. 27:936-
944; 1999.
[120] Voehringer, D. W. BCL-2 and glutathione: alterations in cellular redox state that regulate
apoptosis sensitivity. Free Radic. Biol. Med. 27:945-950; 1999.
[121] Rahman, I.; MacNee, W. Oxidative stress and regulation of glutathione in lung
inflammation. Eur. Respir. J. 16:534-554; 2000.
[122] Bains, J. S.; Shaw, C. A. Neurodegenerative disorders in humans: the role of glutathione
in oxidative stress-mediated neuronal death. Brain Res. Brain Res. Rev. 25:335-358; 1997.
[123] Giugliano, D.; Ceriello, A.; Paolisso, G. Diabetes mellitus, hypertension, and
cardiovascular disease: which role for oxidative stress? Metabolism 44:363-368; 1995. 110
[124] Rahman, Q.; Abidi, P.; Afaq, F.; Schiffmann, D.; Mossman, B. T.; Kamp, D. W.; Athar,
M. Glutathione redox system in oxidative lung injury. Crit Rev. Toxicol. 29:543-568;
1999.