| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-Glutamyl Transpeptidase Regulates Antiproliferative Effects of S-Nitrosoglutathione on Human T and B Lymphocytes1
,
*
Simmons Arthritis Research Center, University of Texas Southwestern Medical Center, Dallas, TX 75235; and Departments of
Medicine and
Immunology, University of Colorado Health Sciences Center, Denver, CO 80262
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-glutamyl transpeptidase (GGT) is
regulated on T lymphocytes. It is present at a low level on naive T
cells, at a high level on activated T cells, and at an intermediate
level on resting memory T cells. GGT cleaves the glutamyl group from
glutathione, which is the first step in the uptake of extracellular
glutathione. In vitro, purified GGT also metabolizes the naturally
occurring nitrosothiol, S-nitrosoglutathione (GSNO).
Because of this relationship, the effects of cellular GGT on the
metabolism of and cellular response to GSNO were tested. The
GGT-negative lymphoblasts Ramos and SupT1 were transfected with cDNA
for human GGT. In the presence of cells lacking GGT, GSNO is extremely
stable. In contrast, GGT-expressing cells rapidly metabolize GSNO
leading to nitric oxide release. The nitric oxide causes a rapid
(<2-h) inhibition of DNA synthesis. There is a concomitant decrease in
the concentration of intracellular deoxyribonucleotides, suggesting
that one effect of the nitric oxide generated from GSNO is the
previously described inactivation of the enzyme ribonucleotide
reductase. GSNO also caused a rapid, GGT-dependent cytostatic effect in
Hut-78, a human T cell lymphoma, as well as in activated peripheral
blood T cells. Although DNA synthesis was decreased to 16% of control
values in anti-CD3-stimulated Hut-78, the production of IL-2 was
unchanged by GSNO. These data show that GGT, a regulated ectoenzyme on
T cells, controls the rate of nitric oxide production from GSNO and
thus markedly affects the physiological response to this biologically
active nitrosothiol. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-glutamyl transpeptidase (GGT,3
EC 2.3.2.2) is capable of hydrolyzing
-glutamyl peptide bonds and transferring the glutamyl group to a
suitable acceptor (1). A major substrate for GGT is
glutathione (GSH). This tripeptide is the most abundant intracellular
nonprotein thiol and is thus the primary determinant of the cellular
redox state. The cleavage of GSH by GGT is the first step in the
glutamyl cycle, whereby extracellular GSH, which cannot directly enter
cells, is hydrolyzed into glutamate (or a glutamyl-amino acid),
cysteine, and glycine. These components are then transported back into
the cell for resynthesis of GSH (reviewed in 2). The functional
importance of GGT in the maintenance of intracellular glutathione
levels has recently been documented in mice in which the GGT gene has
been inactivated (3). These animals have intracellular
levels of GSH that are 
50% of wild-type levels whereas plasma and
urine GSH levels are markedly elevated. GGT is found on many cell types, primarily in secretory or absorptive tissues such as the proximal convoluted tubule of the kidney, the liver, seminal vesicles/prostate, small intestine, choroid plexus, and mammary gland. Resting and activated human mononuclear cells, as well as many transformed lymphoblastoid cell lines, have been shown to express GGT by enzymatic assays (4, 5, 6, 7). Recently, we have developed a novel mAb to human GGT (3A8). With this Ab, it was possible to demonstrate several unrecognized features of GGT expression (8).4 First, although the level of GGT protein is low on resting lymphocytes, it begins to be up-regulated 2–3 days after stimulation with mitogens or superantigens. Second, the level of GGT on resting memory T cells is significantly greater than that seen on naive T cells. The highest levels are seen on those memory T cells that have the capacity to migrate across an endothelial barrier. This population of memory T cells with high GGT levels is expanded in patients with rheumatoid arthritis. Third, GGT was found to associate on both T and B cell surfaces with a subset of tetraspan proteins. These proteins, with characteristic four-membrane-spanning domains, include CD81, CD82, and CD53. Associations between GGT and the B cell complement receptor/signaling molecules CD21 and CD19 were also demonstrated, likely mediated by the coassociation of these molecules and CD81. These molecular associations were demonstrated by coimmunoprecipitation of GGT by Abs to the tetraspan proteins and by cocapping of GGT with the other proteins on live cells. Finally, in a survey of common T and B lymphoblastoid cell lines, it was demonstrated that GGT expression is not uniform. Although most cell lines expressed the Ag, pre-B cells such as Nalm-6, mature B cells such as Daudi and Ramos, and T cells such as SupT1 and HSB-2 did not express immunoreactive or enzymatically functional protein. Together, these data show that the expression of GGT on lymphocytes is regulated not only by cellular activation but also by virtue of cellular differentiation. Therefore, cells that express high levels of GGT may respond differently in an environment containing substrates for this enzyme.
It has recently been shown that the nitrosothiol, S-nitrosoglutathione (GSNO) is a substrate for purified GGT in vitro (9, 10). Unlike free NO, which has a short half-life in aerobic solutions, nitrosothiols such as S-nitrosocysteine, GSNO, and S-nitrosoalbumin are more stable (11). For example, the half-life for the decomposition of S-nitrosocysteine is on the order of seconds to minutes, for S-nitroso-N-acetylpenicillamine it is minutes to hours, and for GSNO it has been estimated to be over 100 h. This makes nitrosothiols useful experimental tools for study of the effects of nitric oxide. For example, they cause relaxation of smooth muscle (12) and inhibit platelet activation (13). In addition, nitrosothiols have been detected in plasma and other biological fluids (reviewed in 14) where, for example, the reversible nitrosylation of hemoglobin controlled by oxygen tension regulates blood flow in peripheral tissues (15).
GSNO has been used to study the effects of nitric oxide on
mitogen-activated human peripheral blood T cells (16). It
inhibited proliferation at 72 h and caused up-regulation of
cellular cGMP levels. Because the spontaneous release of nitric oxide
from GSNO in vitro is greatly accelerated by hydrolysis of the
-glutamyl bond (10), we considered that the effect of
GSNO on activated T cells would be GGT dependent. GSNO has been
detected in culture supernatants of activated macrophages in vitro
(17). The increased expression of GGT on T cell subsets
would make these T cells particularly sensitive to nitric oxide in the
form of GSNO produced in an inflammatory microenvironment. Herein, we
show that GGT expression by lymphocytes is required for the rapid
decomposition of GSNO. This leads to the liberation of nitric oxide
from the resultant S-nitrosocysteinylglycine, which then
rapidly down-regulates de novo DNA synthesis of the GGT-expressing
cells. These results demonstrate a novel role for this tightly
regulated enzyme in controlling the cellular response to physiological
forms of NO.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SupT1 (human T cell lymphoma), Hut-78 (human T cell lymphoma), Ramos (human B cell lymphoma), and PA317 (murine amphipathic packaging cell line) were all obtained from the American Type Tissue Collection (Manassas, VA). The lymphoma cells were routinely cultured in RPMI 1640 supplemented with 10% FBS and antibiotics (RPMI/10). PA317 was cultured in DMEM supplemented with 5% FBS. Resting human T cells were purified from peripheral blood of healthy adult volunteers as previously described (18).
3A8, an IgG2a mAb to human GGT, was produced as described (8) and conjugated to FITC by standard techniques. 64.1 (anti-human CD3) was a kind gift from Dr. Ellen Vitetta (University of Texas Southwestern Medical Center).
Reagents
GSNO, spermine, and
1-N-[3-aminopropyl]-N-[4-(3-aminopropylammonio)butyl]-amino-diazen-1-ium-1,2-diolate
(spermine NONOate) were from Alexis Biochemicals (San Diego, CA).
Reduced glutathione and
L--glutamyl-p-nitroanilide were from
Calbiochem (San Diego, CA). G418 was from Life Technologies (Grand
Island, NY), and Polybrene and acivicin were from Sigma (St.
Louis, MO).
Transfection of SupT1 and Ramos with human GGT cDNA
The plasmid phGGT (19) was obtained from Dr. Henry
Pitot (University of Wisconsin, Madison, WI). A 2331-bp
EcoRI fragment containing the GGT coding region flanked by
492 bp of 5'- and 203 bp of 3'-untranslated region was cloned into the
EcoRI site of the amphipathic retroviral expression vector,
pLXSN (gift of Dr. A. Dusty Miller, Fred Hutchison Cancer Research
Center, Seattle, WA). The resulting pLGGTSN recombinant plasmid was
transfected into subconfluent PA317 cells by the calcium phosphate
technique. After transfection, the cells were selected in medium
containing 1 mg/ml G418. Drug-resistant cells were analyzed for human
GGT expression by staining with FITC-3A8 and analysis by flow
cytometry. Whereas G418-resistant cells transfected with the parental
pLXSN vector (control) were negative, the pLGGTSN-transfected packaging
cells were uniformly positive for expression of GGT (not shown). The
supernatants of confluent cultures of control and GGT-transfected PA317
were used as a source of retrovirus stock. They were diluted 1:10 with
RPMI 1640, and Polybrene (800 µg/ml) was added. SupT1 and Ramos cells
were cultured in the diluted viral stock at a concentration of
106/ml for 4 h at 37°C. They were then
washed and recultured in RPMI/10 overnight. The bulk cell cultures were
then diluted 1:1 with RPMI/10 containing 1 mg/ml G418. After 3 days,
the cells were again split 1:1 with G418-containing medium.
Drug-resistant cells were allowed to grow to near confluence and
then tested for GGT expression by flow cytometry. The GGT expressing
cells (50–75% of the G418-resistant cells) were then sorted on the
FACStarPlus for a uniform population of
GGT-expressing cells, termed SupT1/GGT and Ramos/GGT. G418-resistant
cells that grew from SupT1 or Ramos transfected with empty retroviral
vector were used without further manipulation and termed SupT1/X and
Ramos/X.
Spectrophotometric assays
GGT enzyme activity on intact cells was assayed by resuspending
106 washed, GGT-expressing Ramos or SupT1 cells
in 1 ml PBS containing 60 mM glycylglycine, pH 7.4, and 2.5 mM
L--glutamyl-p-nitroanilide (Calbiochem). All
solutions were treated with Chelex resin (Bio-Rad, Richmond, CA) to
remove trace metal contamination, because this has been shown to
accelerate the decomposition of nitrosothiols. In addition,
106 GGT-expressing cells were treated with a 200
µM concentration of the irreversible GGT inhibitor acivicin
(20) for 1 h at 37°C and washed before assay. After
90 min at 37°C, the reaction was stopped by the addition of 2 ml 1.5
M acetic acid. The cells were pelleted by centrifugation, and the
absorbance of the supernatants was read at 410 nm.
The intact S-nitroso bond of GSNO was measured by spectrophotometry (10). intact Ramos or SupT1 cells (106) were resuspended in 1 ml PBS plus glycylglycine containing 500 µM GSNO at 37°C in the dark. GGT-expressing cells were also tested after acivicin treatment as described above. At the indicated intervals, the absorbance of the supernatant at 336 nm was measured. Control incubations contained PBS, glycylglycine, and GSNO, but no cells.
The production of nitrite after the incubation of cells with GSNO in vitro was used as a surrogate for NO production. Nitrite was determined by the Greiss reaction. Supernatants (50 µl) from the incubation of control and GGT-expressing Ramos and SupT1 cells with GSNO in PBS plus glycylglycine, as above, or from control or GGT-expressing cells incubated in RPMI 1640 with 10% FCS containing 500 µM GSNO were incubated at room temperature with 100 µl of 1% sulfanilamide in 2.5% phosphoric acid for 5 min in flat-bottom 96-well plates. Naphthylethylenediamine, 0.5% (100 µl) in 2.5% phosphoric acid were added for an additional 5 min. The absorbance at 550 nm was measured, and the nitrite concentration was determined by comparison to sodium nitrite standards. In some experiments, 1 mM EDTA, pH 7.4, or 4 µM CuSO4 was added to the GSNO solution during incubation with the Ramos and SupT1 cells before nitrite determination. All enzyme assays were done at least twice.
T cell activation assays
The effect of GSNO on lymphoblast proliferation was determined by thymidine incorporation. Control or GGT-expressing Ramos or SupT1 cells (105) were washed and resuspended in 50 µl serum-free DMEM lacking cysteine and methionine (to prevent trans-nitrosylation by GSNO), and plated in 96-well U-bottom plates. Glutathione, GSNO, spermine, and spermine NONOate were added at 0, 125, 250, and 500 µM. After 2 h at 37°C, 1 µCi [methyl-3H]thymidine (New England Nuclear, Boston, MA, 6.7 Ci/mmol) was added for an additional 4 h. The cells were harvested onto glass microfiber filters, and incorporation of thymidine into DNA determined by liquid scintillation counting. All measurements were done in triplicate. Similar assays were performed on human peripheral blood T lymphocytes. T cells were purified from venous blood as described and incubated with 10 ng/ml PMA and 0.5 µM ionomycin for 72 h. The cells were then incubated with GSNO as described and assayed for thymidine incorporation.
Hut-78 cells were incubated with GSNO as above for 2 h and then transferred in the GSNO-containing medium to flat-bottom 96-well microtiter plates that had been coated with 1 µg/ml anti-CD3 (64.1). The cultures were pulsed with [3H]thymidine, and incorporation into DNA was determined after a 6-h incubation. The supernatants of replicate cultures were collected for determination of IL-2 content by ELISA (PharMingen, San Diego, CA).
Enzymatic measurement of dNTP pools
The concentrations of intracellular deoxyribonucleotides in cells treated with GSNO were determined by modification of the primer extension method of Sherman and Fyfe (21). SupT1/X or SupT1/GGT cells were incubated for 4 h in medium containing various concentrations of GSNO, washed, and extracted overnight in 60% methanol at -20°C. After the methanol-insoluble material was pelleted, the extracts were dried in a Speed-Vac (Savant Instruments, Farmingdale, NY) and resuspended in water at a concentration of 106 cell equivalents per 20 µl. Oligonucleotides specific for the measurement of dATP, dCTP, dGTP, and TTP were synthesized on an ABI 394 DNA/RNA synthesizer and annealed to form double-stranded primers as described (21). Reactions were set up in 25 µl containing 32 mM Tris-HCl, pH 7.5, 16 mM MgCl2, 40 mM NaCl, 4 mM DTT, 2 mM unlabeled dATP, 1.25 µCi [35S]dATP (1250 Ci/mmol, DuPont/NEN, Boston, MA) 100 pmol primer, 0.4 unit Sequenase T7 DNA polymerase (Amersham, Cleveland, OH) and 2.5 µl of either deoxyribonucleotide standards or cell extract. For the measurement of dATP, TTP and [35S]TTP were substituted for the dATP in the reaction mixtures. The reactions were incubated at 37°C for 1 h, and then aliquots were spotted in duplicate on Whatman DE81 filter paper (Fisher Scientific, Pittsburgh, PA). The filters were washed three times in 5% Na2HPO4, once in water, and once in 100% ethanol and dried. The radioactivity on the dried filters was then quantified by liquid scintillation counting.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To facilitate the study of GGT expression on cell physiology,
human GGT cDNA was expressed in two GGT-
lymphoblastoid cell lines. Many T and B cell lines tested express high
levels of GGT (8), but the B cell lymphoma, Ramos, and the
T cell lymphoma, SupT1, do not. A defective retrovirus capable of
expressing GGT from the viral LTR was constructed and used to infect
these cells. Cells infected with virus created from the empty vector
served as controls. The expression of GGT on the control (/X) and
transfected (/GGT) cell lines is shown in Fig. 1
. The levels of GGT expressed normally
on the surface of the T lymphoblastoid cell line, Hut-78, and on
PMA-stimulated peripheral blood T cells are shown for comparison. To
avoid artifacts arising from clonal selection, the cell lines were not
subcloned but were selected for high levels of GGT expression by
fluorescence activated cell sorting. Despite this, the SupT1/GGT line
continued to have heterogeneous expression. The level of GGT enzyme
activity correlated with the level of antigenic expression (Fig. 2). Acivicin (AT-125), an irreversible
inhibitor of GGT in vitro (20) was able to completely
prevent the cleavage of the synthetic GGT substrate, -glutamyl
paranitroanilide.
|
|
Although glutathione is the major substrate for GGT, other
glutamyl-containing molecules are cleaved by purified enzyme in vitro.
One of these is GSNO, a natural adduct of nitric oxide and glutathione.
The stability of GSNO in the presence of live cells was tested. The
control and GGT-expressing Ramos and SupT1 cells were incubated with
GSNO. The presence of intact nitrosothiol was measured by monitoring
the absorbance of the solution at 336 nm (10). The cell
lines containing the control vector did not degrade GSNO (Fig. 3). When GGT was expressed on the cell
surface, there was loss of the nitrosothiol absorbance. The rate of
disappearance of the nitrosothiol was correlated with the level of GGT
on the cell surface. That is, the SupT1/GGT cells degraded GSNO more
rapidly that the Ramos/GGT cells. Inclusion of acivicin prevented
cleavage of GSNO in the incubations with Ramos/GGT or SupT1/GGT cells.
These data emphasize the stability of GSNO in aqueous solutions even
when exposed to living cells which could potentially destroy the
nitrosothiol through disulfide interchange with cell surface
sulfhydryl-containing proteins or low m.w. thiols secreted from the
cells.
|
It has been proposed that the release of nitric oxide from
nitrosothiols is catalyzed by metals, particularly
Cu+ ions (10, 22). Both enzymatic
and nonenzymatic mechanisms have been postulated. The Ramos and SupT1
transfectants were incubated with GSNO, and the appearance of nitrite
was determined as an indicator of nitric oxide production (Fig. 4, top). In the incubations
with the control transfectants, there was little or no accumulation of
nitrite in the supernatant. In the presence of GGT-expressing cells,
there was prompt production of nitrite that increased during the 2-h
incubation. The presence of serum had only a small effect on the
stability of GSNO in the absence of GGT. Therefore,
trans-nitrosylation of BSA or other proteins is not a
significant mechanism for the release of nitric oxide from GSNO. When 1
mM EDTA was included in the incubation (Fig. 4, bottom),
there was an almost total abrogation of nitrite production. Conversely,
when copper was added to the incubation, there was an increase in the
level of nitrite produced. The buffers used in these experiments were
treated with ion-exchange resins to remove trace metals. In
vitro, therefore, the metal needed to effect the conversion of GSNO to
nitric oxide presumably comes from the cells themselves. In vivo, the
metal ions are likely to be present in the extracellular fluid as well
as secreted by the cells.
|
, open bars). There was an 7-fold
increase in GGT activity per cell after activation that was inhibited
by >90% after a 1-h exposure to 200 µM acivicin. Nitrite production
from GSNO in the presence of resting T cells was only slightly over
background. In the presence of activated T cells, there was a
significant amount of nitrite produced from GSNO that was inhibited by
acivicin. Thus, the increased expression of GGT on human T cells after
activation facilitates the delivery of nitric oxide from this
physiological nitrosothiol.
|
Nitrite accumulation is a surrogate marker for nitric oxide
production. To demonstrate a physiological effect of the NO produced
after hydrolysis of GSNO by GGT, the transfected cell lines were
exposed to different concentrations of GSNO and then tested for the
incorporation of [3H]thymidine into DNA. After
a 2-h pretreatment with GSNO in tissue culture medium, there was a
concentration-dependent inhibition of DNA synthesis only in the
GGT-expressing cell lines (Fig. 6). This effect was not caused by GSNO
inhibition of initial thymidine uptake into the cells, in that no
difference was seen in the accumulation of radiolabeled thymidine by
mitomycin C-treated cells (data not shown).
|
Although the transfected cell lines provide the opportunity to test the
effect of GGT on GSNO degradation directly, they are limited in their
functional capacity. For example, SupT1 cells have low levels of TCR
expression and do not respond to anti-CD3 stimulation (D. Karp,
unpublished observations). To obsrve the GGT-dependent effects of GSNO
metabolism on T cell activation, other cells were analyzed. The
inhibition of DNA synthesis by short term exposure to GSNO was tested
in activated peripheral blood T cells and in the Hut-78 lymphoma line
(Table I). T cells were activated by a
72-h exposure to phorbol ester-ionomycin and then incubated with 250,
500, and 1000 µM GSNO, resulting in 23, 70, and 91% inhibition of
thymidine incorporation, respectively. Hut-78 cells express moderate
levels of GGT quantified by flow cytometry using 3A8 (Fig. 1
) and
secrete IL-2 in response to stimulation with anti-CD3. When exposed
to GSNO for 4 h, there is almost total inhibition of DNA synthesis
as expected from the experiments with GGT-transfected cell lines.
Addition of 200 µM acivicin to either the peripheral blood or Hut-78
cultures 1 h before the incubation with GSNO prevented the
decrease in thymidine incorporation. This acivicin treatment
irreversibly inhibits >90% of the GGT enzyme activity without
altering the expression of GGT on the cell surface determined by 3A8
staining (data not shown). Therefore, the inhibition of DNA synthesis
caused by exposure to GSNO depends on its metabolism by cell
surface GGT.
|
Various nitric oxide donors have been shown to cause a decrease in
the activity of ribonucleotide reductase (23, 24, 25). This
enzyme catalyzes the rate-limiting step in the pathway leading to DNA
synthesis. The catalytic reduction of ribonucleotides involves a
tyrosyl free radical that is subject to reversible nitrosylation by
nitric oxide (26, 27). The intracellular pools of the four
deoxyribonucleotides were measured in cells expressing GGT and exposed
to GSNO as an indicator of ribonucleotide reductase activity (Table II). Methanol extracts were prepared from
Hut-78 cells after a 4-h exposure to GSNO. The concentrations of the
individual deoxyribonucleotides in the extracts were determined by
separate primer-extension reactions, each dependent on the presence of
a single dNTP (21). Exposure of the Hut-78 cells to GSNO
was associated with concentration-dependent decreases in all four
deoxyribonucleotides. At the highest concentration tested (1 mM), dATP
was decreased to 67% of control, dCTP to 40% of control, dGTP to 30%
of control, and TTP to 56% of control.
|
Hut-78 cells were used to test the effect of GSNO on T cell
activation (Table III). After stimulation
with anti-CD3, these cells secrete IL-2. In the presence of
increasing concentrations of GSNO without exposure to
plastic-immobilized anti-CD3, there was a slight increase in
baseline IL-2 production that was statistically significant
(p < 0.01 for 0 vs 500 µM GSNO, Student’s
t test). Exposure of the cells to anti-CD3 caused
significant IL-2 to be secreted during the 6-h culture that was not
altered by exposure to GSNO. This indicates that in contrast to the
profound inhibition of DNA synthesis caused by GSNO, signaling from the
TCR and IL-2 gene transcription is not decreased by the NO produced
from GSNO. This provides further evidence that GSNO exposure does not
globally suppress cell functions but, like NO itself, results in
specific alterations in cell capacities.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
160 h
(11) in metal-free aqueous solution) and is subject to
enzymatic modification. In vitro, the removal of the -glutamyl group
from GSNO markedly increases the rate of release of nitric oxide into
the medium (9, 10). The enzyme that catalyzes this
reaction is GGT, which initiates the cleavage of extracellular GSH for
transport of the component amino acids into the cell and resynthesis as
intracellular GSH. Here we demonstrate that GGT naturally expressed at
the surface of lymphocytes is required for the rapid breakdown of GSNO.
This GGT-dependent delivery of NO causes cytostasis without leading to
a change in T cell function, as determined by IL-2 secretion. The level of GGT expression differs on subpopulations of human T cells. Both the enzymatic activity (28) and cell surface expression4 of GGT increase markedly with T cell activation. Moreover, the expression of GGT remains elevated on T cells that express CD45RO and intermediate levels of CD45RB. These isoforms of the CD45 protein phosphatase are markers for memory T cells. A subset of memory T cells has the highest capacity for transendothelial migration into sites of tissue inflammation such as rheumatoid synovium (29, 30). Moreover, the relative proportion of memory T cells is enhanced in the peripheral blood of patients with rheumatoid arthritis (31). We have recently documented that both the CD4+ T cells that exhibit rapid transendothelial migration and the expanded memory T cell population in patients with rheumatoid arthritis have very high levels of GGT compared with resting, naïve T cells.4 Therefore, the population of T cells that is found outside the vasculature, in inflamed tissue, will be able to metabolize GSNO and be subject to the effects of the released nitric oxide. The fact that GGT is expressed at the highest level on activated T cells explains the previous observation that GSNO is cytostatic for human peripheral blood T cells 3 to 5 days after exposure to mitogens (16). This corresponds to the time required for the expression of high levels of GGT on the cell surface.
GGT metabolism of GSNO is generally thought of as a mechanism for cells to recapture GSH as outlined above. However, there is also experimental evidence that under certain conditions, the cysteinylglycine produced from GSH by GGT has a prooxidant effect. This has been seen in cultured hepatocytes as well human monoblastic leukemia cell lines. In the latter system, an decrease in peroxide released from cells was seen when GGT was inhibited by acivicin. Also, an increase in exogenous lipid peroxidation could be seen when the cells were incubated with chelated iron and GSH. These effects were thought to be caused by the facilitated generation of reactive oxygen intermediates by the presence of the more basic thiolate anion of cysteinylglycine. It is unknown whether such a prooxidant effect of GGT occurs in lymphocytes as well. In the short term assays that have been performed in this study, we have been unable to see any effect of either cysteinylglycine at concentrations of up to 500 µM, or of exogenous bovine heart catalase on the proliferation of lymphocytes (data not shown). It is unlikely that the generation of reactive oxygen intermediates from the metabolism of either GSH or GSNO accounts for the cytostasis observed here.
GSNO has been documented in vivo in a number of systems. Low m.w. nitrosothiols including GSNO have been demonstrated in human (14) and murine (32) bronchoalveolar lavage fluid. The concentration of GSNO is highest in fluid from inflamed lungs, as a result of either infection or orthotopic lung transplantation. It is also present in human erythrocytes (15). Finally, GSNO has been produced by murine macrophages activated by LPS in the presence of GSH (17). The concentrations of GSNO found in these different fluids range from several hundred nanomolar to several micromolar (33). Although these are below the levels studied here, it is likely that higher concentrations are obtained in tissue than in plasma or other body fluids, where the GSNO is produced in response to induction of nitric oxide synthase. The expression of GGT on activated T cells at sites of inflammation would therefore increase the local concentration of NO leading to control of lymphocyte proliferation.
The observations that nitric oxide release from GSNO is accelerated by
cellular GGT and is dependent on the presence of metals in the
extracellular environment support a biochemical mechanism derived from
in vitro studies. The homolytic cleavage of the S-nitroso
bond is catalyzed by transition metals (22). In
particular, reduced metals such as Cu+ are more
efficient than oxidized metals such as Cu2+. The
products of this cleavage are nitric oxide and thiyl radicals, which
then form disulfides. Different nitrosothiols are more susceptible to
metal-dependent cleavage than others (22). The product of
GGT metabolism of GSNO is S-nitrosocysteinylglycine, which
can bind metals at both the sulfur and free amino nitrogen (which is
not present in GSNO). S-Nitrosocysteinylglycine rapidly
decomposes to nitric oxide and the disulfide of cysteinylglycine
(10). The overall scheme for the release of nitric oxide
from GSNO is shown in Fig. 6. In
phosphate buffers or tissue culture medium, GSNO is
stable before cleavage by GGT. Once the nitrosocysteinylglycine is
formed, there is sufficient metal contamination from the buffer
solution, or from the cells themselves, to cause homolytic cleavage.
Chelation of the metals in solution by EDTA does not affect the degree
of cleavage of GSNO by GGT (10), whereas the resulting
nitrosothiol is still stable even in the presence of cells. Lastly, it
is possible that the products of GGT hydrolysis of GSNO have direct
effects on cells, unrelated to their release of NO or the transfer of
NO to another molecule. Certain tissues have been shown to have
specific and stereoselective receptors for intact nitrosothiols, such
as L-s-nitrosocysteine
(34). The similar effects of spermine NONOate and GSNO in
GGT-expressing cells argue against a direct effect in this system.
|
for 24
h activated endogenous nitric oxide synthesis and caused a 45%
decrease in ribonucleotide reductase activity that was associated with
an 86% inhibition of DNA synthesis (23). In the
experiments described here, DNA synthesis in Hut-78 cells was shown to
be inhibited by exposure to GSNO. The cytostatic effect was reversed by
inhibition of GGT. The different dNTP pools were decreased 32–70%.
These results are similar to level of dNTP pool depletion resulting
from inhibition of ribonucleotide reductase caused by other cytostatic
agents (40). The role of nitric oxide in T cell signaling is not clear. When examined in isolation, the MAP kinase pathways are up-regulated by nitric oxide. Nitrosylation of a specific cysteine residue of p21ras enhances GDP/GTP exchange (41). This is reflected by the 2.5- to 3.5-fold increase in ERK1/2 activity seen in Jurkat cells exposed to nitric oxide donors (42). Interestingly, the kinetics of nitric oxide-induced activation of the other MAP kinases, JNK, and p38 was different from that seen for ERK1/2. In resting Hut-78 T cells, an increase in ERK1/2 phosphorylation is seen after exposure to GSNO (D. Karp, unpublished observations). Whether nitric oxide has other effects on the intermediate stages of this pathway is unknown. The precise role of nitric oxide on T cell-signaling pathways in the presence or absence of T cell activation is currently under investigation. Finally, it is possible that the liberation of NO from GSNO by GGT expressed on memory/effector T cells may act in trans on neighboring cells, perhaps those that are producing the GSNO.
In conclusion, the data presented here show that GSNO is stable and does not decompose to form nitric oxide species even in the presence of B or T lymphocytes unless those cells express GGT. GGT is present in serum and, presumably, other extracellular fluids. However, the low level of the enzyme in serum is insufficient to degrade GSNO in short term assays (data not shown). The activity of GGT on the cell surface is both necessary and, in the presence of metal ions, sufficient to cause release of nitric oxide from GSNO. The nitric oxide released from GSNO by cell surface GGT causes rapid cytostasis without preventing T cell activation. This pathway could represent a mechanism for GSNO produced at sites of inflammation to regulate the proliferation of previously activated T cells that have up-regulated their expression of GGT. For example, synovial T cells from patients with rheumatoid arthritis proliferate poorly. These T cells also express very high levels of GGT.4 The synovium also contains cells with up-regulated nitric oxide synthase. Although it remains to be proved that these cells produce GSNO, it is possible that this represents a mechanism whereby the regulated expression of an ectoenzyme by T cells controls their function.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. David R. Karp, Harold C. Simmons Arthritis Research Center, 5323 Harry Hines Boulevard, Dallas, TX 75235-8884. E-mail address:
3 Abbreviations used in this paper: GGT, -glutamyl transpeptidase; GSH, glutathione; spermine NONOate, 1-N-[3-aminopropyl]-N-[4-(3-aminopropylammonio)butyl]-amino-diazen-1-ium-1,2-diolate; GSNO, S-nitrosoglutathione.
4 D. R. Karp, T. C. Nichols, N. Oppenheimer-Marks, R. I. Brezinschek, and V. M. Holers. Submitted for publication.
Received for publication February 8, 1999. Accepted for publication June 4, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-glutamyl transpeptidase. Methods Enzymol. 77:237.[Medline]
-Glutamyl transpeptidase. Am. J. Pathol. 147:1175.[Abstract]
-glutamyl transpeptidase-deficient mice. Proc. Natl. Acad. Sci. USA 93:7923.-Glutamyl transpeptidase, a lymphoid cell-surface marker: relationship to blastogenesis, differentiation, and neoplasia. Proc. Natl. Acad. Sci. USA 73:2414.-glutamyl transpeptidase (-GT) in populations of mononuclear cells from human peripheral blood. Biol. Chem. Hoppe Seyler 374:287.[Medline]
-glutamyl transpeptidase in lymphoid cells. FEBS Lett. 115:273.[Medline]
-Glutamyl transpeptidase: cellular expression in populations of normal human mononuclear cells and patients suffering from leukemias. Ann. Hematol. 70:237.[Medline]
-Glutamyl transpeptidase, an ectoenzyme regulator of intracelluar redox potential, is a component of TM4 signal transduction complexes. Eur. J. Immunol. 28:4123.[Medline]
-glutamyl transpeptidase. Biochem. J. 323:477.
-glutamyl transpeptidase cDNA: comparison of hepatoma and kidney mRNA in the human and rat. Gene 76:1.[Medline]
-glutamyl transpeptidase with acivicin. J. Biol. Chem. 269:21435.-glutamyl transpeptidase, a lymphoid cell-surface marker: relationship to blastogenesis, differentiation, and neoplasia. Proc. Natl. Acad. Sci. USA 73:2414.
This article has been cited by other articles:
|
|
 |
|
  J. B. King, M. B. West, P. F. Cook, and M. H. Hanigan A Novel, Species-specific Class of Uncompetitive Inhibitors of {gamma}-Glutamyl Transpeptidase J. Biol. Chem., April 3, 2009; 284(14): 9059 - 9065. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Dietlin, F. M. Hofman, B. T. Lund, W. Gilmore, S. A. Stohlman, and R. C. van der Veen Mycobacteria-induced Gr-1+ subsets from distinct myeloid lineages have opposite effects on T cell expansion J. Leukoc. Biol., May 1, 2007; 81(5): 1205 - 1212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. R. Reyes, S. Danese, M. Sans, C. Fiocchi, and A. D. Levine Redox Equilibrium in Mucosal T Cells Tunes the Intestinal TCR Signaling Threshold J. Immunol., August 15, 2005; 175(4): 2158 - 2166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Macphail, C. A. Gibney, B. M. Brooks, C. G. Booth, B. F. Flanagan, and J. W. Coleman Nitric Oxide Regulation of Human Peripheral Blood Mononuclear Cells: Critical Time Dependence and Selectivity for Cytokine versus Chemokine Expression J. Immunol., November 1, 2003; 171(9): 4809 - 4815. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Roozendaal, E. Vellenga, M. A. de Jong, K. F. Traanberg, D. S. Postma, J. G. R. de Monchy, and H. F. Kauffman Resistance of activated human Th2 cells to NO-induced apoptosis is mediated by {{gamma}}-glutamyltranspeptidase Int. Immunol., April 1, 2001; 13(4): 519 - 528. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Karp, K. Shimooku, and P. E. Lipsky Expression of gamma -Glutamyl Transpeptidase Protects Ramos B Cells from Oxidation-induced Cell Death J. Biol. Chem., February 2, 2001; 276(6): 3798 - 3804. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), Ramos/GGT (•), Ramos/GGT treated with 200 µM acivicin for
1 h (
). Right, SupT1/X (
), Ramos/GGT (•),
Ramos/GGT treated with 200 µM acivicin for 1 h (
, addition of 1 mM EDTA.
Data are means ± SEM of triplicate determinations and are
representative of two different experiments.
