| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Molecular Microbiology and Immunology, University of Southern California School of Medicine, Los Angeles, CA 90033
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 (H-2b), B6.C-H-2bm1 (Kbm1; I-Ab), and B6. C-H-2bm12 (Kb; I-Abm12) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). To isolate cortisone-resistant thymocytes, 6-wk-old mice were injected with 4 mg of hydrocortisone (Sigma, St. Louis, MO) in PBS i.p. and the thymi harvested 2 days later (6). To induce delayed-type hypersensitivity (DTH), mice were primed with 100 µg of keyhole limpet hemocyanin (KLH) in CFA (i.p.) on day 0. On day 5, mice were challenged into the footpad with 150 µg of KLH in saline and assayed for footpad swelling on day 6. Similarly, mice were primed with 109 pfu of a replication-deficient recombinant adenovirus (i.v.) on day 0 and challenged in the footpad on day 9 with the same dose of virus and assayed for footpad swelling on day 10. The defective adenovirus was obtained from Microbix Biosystems (Toronto, Ontario, Canada) and propagated as described by the supplier. Footpad swelling was measured with a gauge (Mitsutoyo, Tokyo, Japan). To assay humoral immune responses, mice were immunized i.p. with 100 µg of DNP-KLH (Calbiochem, San Diego, CA) in CFA on day 0 and bled on day 9. Ig subsets were determined by ELISA using DNP-BSA (Calbiochem)-coated plates and Ig class-specific Abs (Caltag, South San Francisco, CA).
Cell purification, fluorometric analysis, cell migration assays, and treatment of cells with phosphatidylinositol-specific phospholipase C (PIPLC)
To isolate peritoneal exudate cells (PEC), mice were injected
with 2 ml of 2x thioglycolate (Becton Dickinson, Mountain View, CA)
i.p. PEC were harvested 3 days later by peritoneal lavage. To deplete T
cells, spleen cells were treated with anti-Thy-1 (T24.31.7) for 45
min at 4°C, followed by 1:10 low tox rabbit complement (Accurate
Chemical, Westbury, NY) for 1 h at 37°C (7). Live cells were
isolated by Ficoll-Hypaque centrifugation (1, 7). Ig-positive cells
were purified by panning spleen cells with anti-IgA, -G, and -M Abs
(Jackson ImmunoResearch, West Grove, PA) (8). T cells were isolated by
passage over nylon wool columns (9), and CD4 or CD8 cells were
collected by panning. Nylon wool nonadherent (NWNA) cells were treated
with anti-CD4 (GK1.5) or anti-CD8 (AD4) for 45 min at 4°C;
washed and suspended in complete RPMI 1640 medium containing 5 x
10-5 M ß-mercaptoethanol, 0.2 mM glutamine, 1 mM
pyruvate, 0.1 mM nonessential amino acids, 10% FCS, and 20 mM HEPES
(pH 7.2); and then incubated for 45 min on ice-cold plastic dishes that
had been coated with anti-Ig Ab. Adherent cells were harvested (8, 10). To isolate splenic macrophages, spleen cells were incubated in
complete RPMI medium containing 20% FCS for 60 min at 37°C on
plastic dishes and adherent cells were harvested. The purity of cell
populations was analyzed on a FACStarplus (Becton
Dickinson) instrument by using FITC anti-Thy1.2 (53-2.1), FITC
anti-CD3
(145-2C11), phycoerythrin anti-CD4 (GK1.5), FITC
anti-CD8
(53-6.7) (PharMingen, San Diego, CA), and phycoerythrin
anti-mouse IgM mAbs (Caltag). Purified T cells contained more than
90% T cells and less than 2% B cells and purified B cells more than
97% B cells and less than 1% T cells.
Lymphocyte migration assays were performed as described (11). Briefly, 5 x 106 cells were labeled with 5 µCi/100 µl Na2[51Cr]O4 (ICN, Costa Mesa, CA) for 1 h at 37°C and cell aliquots of 100 µl injected into the tail vein of C57BL/6 mice. After 1 h, organs were harvested and counted in a gamma scintillation counter. Values are calculated as the percent of total radioactivity injected and plotted as % cell recovery. ADPRT was released from intact cells by incubating 1 to 2.5 x 108/ml in complete RPMI medium containing 5 U/ml PIPLC (Boehringer Mannheim, Indianapolis, IN) for 1 h at 37°C (1).
Cell cultures, cytokine, and D-myo-inositol 1,4,5-trisphosphate (IP3) assays
MLR were set up in complete RPMI medium containing 10% FCS with
1.5 x 106 responder and 3 x
106 irradiated (1000 rad) stimulator cells per 1-ml well
(1). CTL activity was assayed on day 5 at different E:T ratios in a 4-h
assay and percent cytotoxicity expressed as percentage of releasable
counts after subtraction of spontaneous release (1). To induce CTL,
specific for adenovirus, mice were injected with 109 pfu
i.v. on day 0 and splenocytes harvested on day 9. Spleen cells were
restimulated in vitro by culture for 5 days with 1 pfu of adenovirus
per cultured cell. CTL activity was assayed in a 6-h Cr51
release assay using C57SV targets that had been infected with
adenovirus (50 pfu/cell) 24 h before (12). Spleen cells or
purified T cells were either cultured in 96-well flat-bottom tissue
plates (Becton Dickinson) (5 x 105 cells/well) coated
with 5 µg/ml of anti-CD3 mAb (500A2) or in the presence of PMA
(10 ng/ml) and calcium ionophore (A23187; 100 ng/ml), Con A (10
µg/ml), or LPS (10 µg/ml) (Sigma) for 2 days at 37°C.
Proliferative responses were assayed by pulsing cells for the last
18 h of the culture with 0.5 µCi/well of [3H]TdR
(ICN). Incorporated radioactivity was counted in a liquid scintillation
counter. To assay cytokine secretion, cells from draining lymph nodes
of mice that had been primed with 100 µg of KLH in CFA s.c. were
cultured at 8 x 105 cells per 200 µl with 50
µg/ml of KLH. Supernatants were harvested 24 h (IL-2) or 72
h (IFN-
, IL-4, IL-10) later and assayed by ELISA using a commercial
Ab kit purchased from PharMingen (13).
Generation of IP3 was assayed in NWNA cells after CD3 cross-linking. Cells (2 x 107) were suspended in 200 µl of 20 mM HEPES buffer (pH 7.4) containing 137 mM NaCl, 4.5 mM KCl, 1.2 mM MgCl2, 1 mM CaCl2, 10 mM glucose, 0.5% BSA, 5 mM LiCl, and 5 µg anti-CD3 mAb (500A2) for 1 min at 37°C followed by anti-hamster IgG (5 µg) (Jackson ImmunoResearch) for the times indicated. Ice-cold TCA was added to a final concentration of 7.5% and proteins removed by centrifugation. Supernatants were extracted, washed three times with 10 vol of diethylether, neutralized with NaHCO3, and IP3 assayed by a solid phase binding kit (Amersham, Arlington Heights, IL) (14).
Cell labeling with [32P]NAD, preparation of cell lysates, immunoprecipitations, gel electrophoresis, and enzyme assays
Cells (107 cells/ml) were labeled with
[32P]NAD by incubation in complete RPMI medium containing
100 µCi/ml of [32P]NAD (Amersham), 100 µM of NAD, and
1 mM of ADP-ribose (Sigma) at 37°C for 1 h (1). Cells were
excessively washed in large volumes of medium to remove radioactivity.
To prepare crude cell lysates, cells were suspended in lysis buffer
(PBS containing 1% Nonidet P-40 1 mM EDTA, 1 mM PMSF), incubated on
ice for 30 min, and then centrifuged at 12,000 x g at
4°C for 10 min and supernatants collected. For immunoprecipitations,
cell lysates (107 cells/200 µl) were precleared with 20
µl of 50% (v/v) protein G-Sepharose (Pharmacia, Uppsala, Sweden) for
1 h at 4°C under rotation. After removal of G-Sepharose beads by
centrifugation, 5 µg of anti-LFA-1 (M17/4), anti-LFA-2
(RM2-5), anti-4 integrin (R1-2), anti-L-selectin
(MEL-14), anti-CD3 (500A2), anti-CD4 (GK1.5),
anti-CD8 (53.6.7), anti-CD27 (LG.3A10), anti-CD28
(37.51), anti-CD43 (S7), anti-CD44 (IM7), anti-CD45 mAb
(30F11.1), or anti-CD48 (HM48-1) (PharMingen) were added, followed
by incubation for 2 h at 4°C with rotation. Immune complexes
were adsorbed to 40 µl of 50% (v/v) protein G-Sepharose for 1 h
at 4°C with rotation. After washing in lysis buffer,
immunoprecipitates were solubilized with 40 µl of 2x concentrated
SDS sample buffer (20% glycerol, 9% ß-mercaptoethanol, 4% SDS,
0.005% bromophenol blue, and 120 mM Tris-HCl, pH 6.8) at 100°C for 5
min and centrifuged at 12,000 x g for 15 s to
remove insoluble material. Proteins were separated on SDS-PAGE and
dried gels exposed to X-OMAT films (Eastman Kodak, Rochester, NY) at
-80°C with intensifying screens.
Arginine-specific ADPRT was assayed using agmatine as substrate, and the transfer of [32P]ADP-ribose from [32P]NAD to agmatine was determined (3, 15). After incubation in 50 mM sodium phosphate (pH 7.5), 1 mM EDTA, 1 mM DTT, 1 mM ADP-ribose, 100 µM NAD (5000 cpm/nmole of [32P]NAD) and 20 mM agmatine (Sigma) in a final volume of 100 µl for 90 min at 30°C, samples were applied to 1.8 ml of QAE-Sephadex (Pharmacia). Noncharged [32P]ADP-ribosylagmatine was eluted with 3 ml of water and quantified by liquid scintillation counting. Enzyme activity is expressed as picomoles of ADP-ribosylagmatine formed in 90 min.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We had previously reported that the presence of a
glycosylphosphatidylinositol-anchored ADPRT on CTL correlates with
regulatory effects of NAD on their in vitro functions (1). These
findings raised the question of whether other lymphocyte subsets are
also regulated by NAD and whether this regulatory pathway is
demonstrable in vivo. To examine this, mice were primed with KLH and
injected with 1 mg of NAD i.p., followed by an assay for cell-mediated
immunity, i.e., a DTH response. Results shown in Figure 1
A show that DTH is
significantly inhibited in mice that had received injections of NAD
starting on the day of KLH sensitization and the following 4 days
thereafter. Injection of NAD before sensitization with KLH had only
very small effects (data not shown). To examine whether inhibition of
DTH is due to effects on the sensitization or effector phase of the
response, mice were primed with KLH and again injected with NAD for 5
days. Mice were rested for 2 wk and then challenged with KLH in the
footpads. Figure 1B shows that NAD-treated mice express
considerably lower DTH, indicating that the sensitization step of the
DTH reaction is inhibited. When animals were primed with KLH on day 0,
received NAD on days 4 and 5, KLH injected on day 5, and DTH assayed on
day 6, DTH is found to be somewhat inhibited (Fig. 1C). Interestingly, an additional injection of NAD
6 h after the KLH challenge and the last NAD injection (day 5 plus
6 h) causes a more significant inhibition (Fig. 1C). Therefore, NAD appears to also inhibit the
effector step of the DTH reaction.
|
D). Treatment of mice with NAD,
starting with the day of immunization, causes significant suppression
of the DTH reaction, similar to effects seen with KLH. Spleen cells
from these mice were harvested, restimulated with virus in vitro, and
assayed for cytolytic activity 5 days later. Effector cells derived
from NAD-injected mice express much lower cytotoxicity (Fig. 1E). Therefore, NAD inhibits the in vivo priming
of CTL.
Both DTH and induction of CTL are Th1-type immune responses. Therefore,
the question is raised whether Ab production may similarly be affected
by injection of NAD. To find out, mice were immunized with DNP-KLH and
injected with NAD for 5 days beginning on the day of immunization. On
day 10, sera were harvested and assayed for the presence of IgM, IgG1,
IgG2a, and IgG3. The results shown in Figure 2 show that there is no difference in Ab
class production in NAD-injected mice. Therefore, NAD is predicted to
not significantly influence the relative production of Th1 vs Th2
cytokines, which was suggested from the Ab responses of B cells.
|
The selective in vivo effect of NAD on cell-mediated immune
responses suggests that not all lymphoid cells express ADPRT on their
cell surface. To examine this, spleen cells, lymph node cells, and
thymocytes were treated with PIPLC and supernatants assayed for
released ADPRT activity. The results shown in Figure 3A show that lymph node cells
release high enzyme activity, spleen cells release intermediate
activity, and thymocytes release almost no activity. Because this
points to the presence of ADPRT on mature T cells, mice were injected
with cortisone to eliminate immature thymocytes, and the remaining
cells were tested for ADPRT activity. Cortisone-resistant cells release
higher enzyme activity, consistent with the notion that ADPRT is
present on mature T cells (Fig. 3B). In agreement
with these results, treatment of spleen cells with anti-Thy-1 and
complement abolishes the ability to release ADPRT (Fig. 3A), raising the question of whether both CD4 and CD8
cells express the enzyme. Figure 3C shows that purified CD4
and CD8 cells release ADPRT, whereas B cells, splenic macrophages, and
PEC do not.
|
D). In contrast,
LPS-induced B cell proliferation is unaffected (Fig. 3D). Therefore, the sensitivity of lymphocytes to NAD
correlates with the expression of ADPRT on the cell surface and
is in agreement with the differential effects NAD has on cell-mediated
vs humoral immunity.
The presence of ADPRT on T cells predicts that incubation with
radioactive NAD should result in cell surface labeling. Figure 4A shows that lysates,
prepared from identical numbers of [32P]NAD-labeled
spleen cells, purified T cells, CD4 cells, and CD8 cells, but not
lysates from B cells or macrophages, contain radioactive proteins.
Interestingly, CD8 cells are labeled much stronger than the same number
of CD4 cells, which is consistent with the somewhat higher ADPRT
activity releasable from CD8 cells (Fig. 3C). In
comparing the labeling patterns of these cell populations, it can be
seen that they are quite similar, although differences in labeling
intensities of individual proteins are evident. For example, there is a
band in the 40-kDa range that is weakly labeled in spleen and CD4
cells, but very strongly labeled in purified T cells and CD8 cells. The
possibility that this protein constitutes the CD8 molecule itself was
examined by incubating T cell lysates with anti-CD8 Ab. Results
presented in Figure 4B show that this Ab precipitates a
40-kDa protein band, which comigrates with a band of similar molecular
mass in whole cell lysates. Figure 4B also shows that LFA-1,
previously identified on CTL as an ADP-ribosylated molecule (5), can be
precipitated as a labeled heterodimer. Because these as well as
previous results (2, 5) pointed to a preferential interaction of ADPRT
with adhesion molecules, labeling of other cell surface molecules was
examined. Data in Figure 4B show that four additional
proteins, i.e., CD27, CD43, CD44, and CD45, can be identified as
ADP-ribosylated proteins. Other molecules, which were found not to be
ADP-ribosylated, are CD4 (Fig. 4B), CD2 (LFA-2),
CD28, CD48, 4-integrins, and L-selectin (data not
shown). Attempts to precipitate TCR-associated, ADP-ribosylated
molecules were negative. Precipitation of Nonidet P-40 or digitonin
extracts from labeled cells with either anti-CD3- or
anti-
-chain Abs failed to reveal labeled protein bands (Fig. 4B and data not shown). Therefore, although a broad spectrum
of cell surface molecules can be ADP-ribosylated by action of the ADPRT
on naïve T cells, the TCR is not one of them.
|
The observation that adhesion molecules are ADP-ribosylated on T
cells raised the possibility that these modifications affect lymphocyte
receptor function, perhaps providing an explanation for the
immunoregulatory effects of NAD in vivo, especially the inhibition of
DTH effector function. Adhesion molecules are known to mediate multiple
functions, including the migration of sensitized T cells into the site
of a DTH reaction and the trafficking of lymphocytes through secondary
lymphoid organs. Lymphocyte homing has been reported to be a function
of some of the molecules shown here to be ADP-ribosylated, i.e., CD44,
CD43, and LFA-1 (16, 17, 18, 19). To examine whether modification of these
molecules correlates with changes in lymphocyte homing patterns, lymph
node cells were incubated with NAD for 3 h, labeled with
51Cr, and then injected into syngeneic recipients.
After 60 min, cell trafficking was assessed by determining
radioactivity in lymphoid and nonlymphoid organs. The data show (Fig. 5A) that homing of
NAD-treated cells into lymph nodes, spleen, Peyer’s patches, and lung
is strongly inhibited compared with controls. Titration of NAD
concentrations reveals a dose-dependent effect in which 100 µM and 1
mM of NAD induce maximal effects and 1 µM and 10 µM induce
intermediate effects (Fig. 5A).
|
B). However, the effects do not reach those obtained
when cells are incubated in vitro with 100 µM of NAD. This suggests
that the effective in vivo NAD concentration reached when 1 mg is
injected is likely below 100 µM.
Because T cells but not B cells express the cell surface ADPRT (Fig. 3C), a prediction is that the trafficking of T cells,
not B cells, should be regulated by NAD. To examine this, purified cell
populations were treated with NAD and injected into recipients. Results
shown in Figure 6A show that
the homing of T cells but not B cells is inhibited by NAD. Therefore,
the suppressive effect of NAD on lymphocyte migration correlates with
expression of ADPRT on the cell surface. Consistent with this, removal
of ADPRT from T cells by PIPLC, before incubation with NAD, results in
significant inhibition of its suppressive effect (Fig. 6B).
|
The effects of NAD on T cell trafficking could provide an explanation for the suppressive effects of NAD on the effector step of the DTH reaction. However, it is more difficult to understand why inhibition of lymphocyte trafficking should interfere with the sensitization phase of the DTH reaction, as well as, CTL responses. Moreover, inhibition of lymphocyte homing would not be able to explain why cell-mediated but not humoral immune responses are suppressed by NAD. The possibility was therefore entertained that inhibition of lymphocyte trafficking is due to a more general effect on cell functions, rather than due to a failure of homing receptors to bind to their ligands. It is well documented that lymphocyte migration and homing involve multiple reactions, triggered by receptor ligand binding and resulting in transmission of signals that in turn cause cytoskeletal movements, cell rolling, and cytokine gene activation (20, 21, 22). Therefore, inhibition of lymphocyte homing could be due to a block at one or another of these reactions. In agreement with this, the experiments presented above have shown already that NAD inhibits proliferative responses to Con A, consistent with a more general effect of NAD on T cell responses.
To examine which functions might be impaired by NAD, in vitro responses
of CD4 and CD8 cells to antigenic stimulation were tested. To assay
activation of CD4 cells, spleen responder cells were incubated with MHC
class II discordant stimulator cells and tested for cell proliferation.
Results shown in Figure 7A
show that the proliferation of C57BL/6 spleen cells, stimulated with
MHC class II discordant bm12 cells, is completely inhibited by 100 or
1000 µM of NAD. Intermediate effects are induced with 10 µM of NAD.
Quite similar results are seen when C57BL/6 responder cells are
incubated with MHC class I discordant bm1 stimulator cells (Fig. 7A). In these cultures, CD8 cells are the principal
responders. Not only is cell proliferation inhibited by 100 or 1000
µM of NAD, the induction of cytolytic activity is also inhibited
(Fig. 7B). Therefore, T cells responding to either
MHC class I or MHC class II Ags, CD8 and CD4 cells, respectively, are
inhibited by NAD in their response to allogeneic stimulator cells.
|
). It
was possible, however, that NAD exerts subtle effects on the secretion
of cytokines that are important for induction of cell-mediated immune
responses. To investigate this, lymph node cells from KLH-primed mice
were restimulated in vitro with KLH in the presence of NAD, and the
levels of secreted IL-2, IFN-, IL-4, and IL-10 were assayed. Results
shown in Figure 8 show that 100 or 1000
µM of NAD suppresses secretion of all four cytokines. At 10 µM NAD,
only weak suppressive effects are seen on the secretion of IFN-,
IL-4, and IL-10. In contrast, at this NAD concentration, secretion of
IL-2 is completely inhibited. Lower but significant effects are also
seen at 1 µM of NAD. Therefore, IL-2 secretion appears to be
particularly sensitive to the suppressive effects of NAD. From the
lymphocyte homing assays it had been concluded that injection of NAD
leads to effective NAD concentrations in vivo that are below 100 µM.
Therefore, concentrations of NAD in the 1- to 10-µM range are likely
to be responsible for the in vivo effects of NAD on cell-mediated
immunity.
|
The finding that NAD inhibits Ag-stimulated T cell responses in
vitro points to the inhibition of important cell activation pathways.
The possibility that treatment of cells with NAD leads to apoptotic
cell death was investigated. In T cells incubated with NAD, no
induction of DNA fragmentation or loss of cell survival were
demonstrable (data not shown). Rather, it appeared that in the presence
of NAD or after NAD treatment, cells persist in an unresponsive state.
They ultimately die at the same rate as untreated T cells in the
absence of Ag stimulation. NAD does not penetrate cell membranes and
ADPRT is a cell surface enzyme. Therefore, it is most likely that the
inhibitory effects are caused by ADP-ribosylation of cell surface
molecules that participate in the early steps of cell activation
processes. Because Ag-induced responses are inhibited by NAD, the
ability of the TCR to induce cell activation was assayed. Purified T
cells were incubated on anti-CD3-coated plates or with anti-CD3
and a second cross-linking Ab added to accomplish optimal receptor
ligation. Two responses were assayed, the generation of
IP3 and the induction of cell proliferation. Figure 9, A and B, shows
that both the generation of IP3 and the induction of cell
proliferation are inhibited by NAD. That this effect is not due to the
failure of the anti-CD3 Ab to bind to NAD-treated cells is shown in
Figure 9C. There is no decrease of anti-CD3 binding
demonstrable by fluorometric analysis.
|
B shows that NAD has no effect on PMA plus
ionomycin-induced cell proliferation. These results show that the
suppressive effect of NAD on T cell responses is due to effects on the
level of transmembrane signaling. Moreover, the finding that NAD has no
effects on PMA and ionomycin-induced cell proliferation is consistent
with the finding that incubation of cells with NAD does not induce
apoptosis. The finding that signaling through CD3 is inhibited by NAD,
despite the fact that neither CD3 nor any of its associated peptides
are ADP-ribosylated (Fig. 4) suggests that inhibition of TCR signaling
by NAD may be due to an indirect effect involving ADP-ribosylation of
cell surface molecules that modulate TCR signaling. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These molecules represent major protein bands in cell extracts from
NAD-labeled cells, but they likely do not constitute all the proteins
that can be modified. Labeling of these molecules raises the question
of whether ADPRT specifically interacts with and recognizes these
proteins. It had been reported that ADPRT on skeletal muscle myotubes
ADP-ribosylates integrin 7 (2), prompting the
hypothesis that cell surface ADPRT selectively modifies adhesion
molecules. While this remains to be proven, experiments with soluble
enzyme preparations have shown that not only the arginine analogue
agmatine, but also histone and other proteins, may provide a substrate
for T cell ADPRTs (1, 23). The alternative possibility, therefore, that
modification of cell surface molecules is a function of their
preferential association with the ADPRT cannot be excluded.
One of the principal reasons for identifying the proteins that are ADP-ribosylated was to elucidate the mechanism by which NAD regulates T cell functions. This is made difficult, however, by the fact that these molecules have multiple functions. Consequently, it is impossible to predict which modification may be responsible for the observed effects. Therefore, the alternative approach, investigating at which step cell functions are inhibited by NAD, is a more promising one. It is shown here that cell activation induced by TCR cross-linking is inhibited in NAD-treated cells and that this is caused by a block on the level of transmembrane signaling. Activation of T cells by PMA and calcium ionophore, circumventing transmembrane signaling, is not affected by NAD.
This result has important implications. Localization of the inhibitory action to the cell membrane eliminates the albeit remote possibility of NAD action inside the cell. More important is that signaling through a structure, i.e., the TCR, shown not to be ADP-ribosylated, is inhibited. Therefore, NAD effects are likely due to an indirect action of ADP-ribosylated molecules on TCR signaling. The hypothesis is therefore proposed that ADP-ribosylation of cell surface molecules indirectly regulates TCR signaling. Several important molecules were found to be ADP-ribosylated, including CD8 and CD45. Modification of CD8 could be responsible for the inhibition of CTL, both in the induction and the effector phases. This is a particularly attractive hypothesis because the CD8 molecule binds the src-related kinase p56lck, which is important in TCR-mediated signaling (24, 25). In fact, down-regulation of this kinase in NAD-treated CTL has already been demonstrated (4). The observation that CD45 is ADP-ribosylated may provide an additional clue, perhaps explaining why CD4 cells are inhibited by NAD. CD45 is known to be necessary for activation of p56lck, which is important in TCR-mediated signaling (26, 27, 28, 29, 30). Modification of CD45 could lead to an inability to associate with cell surface molecules that have bound p56lck to their cytoplasmic tail, such as CD4 and CD8, and thereby could inhibit p56lck activation by CD45 (31, 32, 33, 34).
The failure of the TCR to transmit a transmembrane signal leading to the generation of second messengers and gene activation could explain the failure of NAD-incubated T cells to respond to Con A or allogeneic stimulation in MLR. It may, however, also provide an explanation for NAD’s inhibition of migration and homing of lymphocytes in vivo. Binding to and penetration of high endothelial venules by lymphocytes require receptor ligand engagement as well as transmission of transmembrane signals (20). Therefore, the failure of NAD-treated cells to efficiently home into lymphoid organs may be due to the observed block in transmembrane signaling. However, more direct effects of ADP-ribosylation on adhesion molecules, affecting their ability to bind to ligands, cannot be excluded. An interesting finding in this respect is that NAD not only inhibits trafficking of T cells to lymph nodes and Peyer’s patches, but also homing to the spleen. CD43 is a homing receptor for spleen-seeking lymphocytes (19). Therefore, it is intriguing that CD43 is ADP-ribosylated. Whether the modification of CD43 is directly responsible for the inhibition of lymphocyte homing to the spleen is not known; therefore, conclusions from the finding that adhesion molecules are ADP-ribosylated await further experimentation.
Our results in tissue culture, demonstrating inhibition of T cell responses to antigenic stimulation, as well as the in vivo results on lymphocyte trafficking predict that NAD could have effects when injected into animals. We show that not only are the induction and effector steps of DTH reactions inhibited, but also the sensitization of CTL. All these reactions require Ag recognition by TCRs and cell proliferation. Therefore, inhibition by NAD may not be unexpected. It is not known, however, whether suppression of the effector phase of the DTH reaction by NAD may be at least in part due to an inhibition of sensitized cells from migrating into the site of the DTH reaction. This would be consistent with the effects NAD has on the trafficking of normal T cells.
An interesting observation is that cell-mediated, but not humoral
immunity to KLH is inhibited by NAD. Given the finding that B cells
lack ADPRT and are insensitive to the action of NAD, this points to an
action of NAD on Th cells. The finding that Ig classes are not changed
in NAD-injected mice suggests that there are no major alterations in Th
cytokine secretion. Cytokine assays of spleen cells from primed mice,
restimulated with Ag in the presence of NAD, revealed that high
concentrations of NAD inhibit secretion of IL-2, IFN-, IL-4, and
IL-10, whereas low concentrations exert preferential effects on IL-2
secretion. It should be noted that in mice in which the IL-2 gene had
been inactivated by homologous recombination, the production of IgM is
severely suppressed (35). The fact that this is not seen in NAD-treated
mice suggests that production of this cytokine is not completely
inhibited. It is interesting that in IL-2-deficient mice, the
differentiation of CTL does not take place, indicating that IL-2 is
necessary for CTL generation (36). Whether this requirement for IL-2 in
the induction of CTL explains the preferential effect of NAD on
cell-mediated immunity remains to be shown. Therefore, presently it is
not clear why cell-mediated but not humoral immunity is inhibited in
NAD-injected animals.
However, the finding that cell-mediated immunity is suppressed documents that effective in vivo concentrations of NAD can be reached by injection. This observation raises the exciting possibility of influencing Th1 responses by the injection of NAD for therapeutic purposes in which suppression of cell-mediated immunity is desirable, for example, in certain autoimmune diseases. More difficult to answer is the question of how this regulatory mechanism participates in the regulation of normal immune responses. Extracellular NAD concentrations had been reported in mice to be in the range of 0.15 µM (37), and in vitro effects are demonstrable at concentrations as low as 1 µM of NAD in the experiments presented here. Therefore, NAD concentrations necessary to induce effects have to reach local NAD concentrations of 1 µM and higher, raising the question of the origin of extracellular NAD during T cell responses. One possibility is that NAD is released from cells, as a consequence of massive cell lysis during inflammatory immune reactions that could be sufficient to cause ADP-ribosylation of cell surface molecules and functional suppression of T cells. The local release of NAD during an inflammatory reaction could therefore serve to down-modulate the extent of a cell-mediated immune reaction. However, presently direct evidence for release of NAD from lysing cells is lacking.
The results presented here describe a novel and potentially important immunoregulatory mechanism in T cells. That this mechanism may be operative in vivo is supported by the observation that certain mouse and rat strains lacking one of the T cell ADPRTs, i.e., RT6, develop autoimmunity (23, 38, 39, 40, 41). Therefore, ADP-ribosylation of adhesion molecules may constitute a self-limiting autoregulatory mechanism that suppresses immunity to self-Ags.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Gunther Dennert, University of Southern California/Norris Comprehensive Cancer Center, NOR 621, M/S 73, P.O. Box 33800, 1441 Eastlake Avenue, Los Angeles, CA 90033-0800.
3 Abbreviations used in this paper: ADPRT, ADP-ribosyltransferase; NAD, nicotinamide adenine dinucleotide; PIPLC, phosphatidylinositol-specific phospholipase C; NWNA, nylon wool nonadherent; KLH, keyhole limpet hemocyanin; DTH, delayed-type hypersensitivity; PEC, peritoneal exudate cell; IP3, D-myo-inositol 1,4,5-triphosphate.
Received for publication October 10, 1997. Accepted for publication December 29, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4ß7 and LFA-1 in lymphocyte homing to Peyer’s patch-HEV in situ: the multistep model confirmed and refined. Immunity 3:99.[Medline]
This article has been cited by other articles:
|
|
 |
|
  F. Scheuplein, N. Schwarz, S. Adriouch, C. Krebs, P. Bannas, B. Rissiek, M. Seman, F. Haag, and F. Koch-Nolte NAD+ and ATP Released from Injured Cells Induce P2X7-Dependent Shedding of CD62L and Externalization of Phosphatidylserine by Murine T Cells J. Immunol., March 1, 2009; 182(5): 2898 - 2908. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hong, A. Brass, M. Seman, F. Haag, F. Koch-Nolte, and G. R. Dubyak Lipopolysaccharide, IFN-{gamma}, and IFN-beta Induce Expression of the Thiol-Sensitive ART2.1 Ecto-ADP-Ribosyltransferase in Murine Macrophages J. Immunol., November 1, 2007; 179(9): 6215 - 6227. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Adriouch, S. Hubert, S. Pechberty, F. Koch-Nolte, F. Haag, and M. Seman NAD+ Released during Inflammation Participates in T Cell Homeostasis by Inducing ART2-Mediated Death of Naive T Cells In Vivo J. Immunol., July 1, 2007; 179(1): 186 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Chen, Y.-G. Chen, P. C. Reifsnyder, W. H. Schott, C.-H. Lee, M. Osborne, F. Scheuplein, F. Haag, F. Koch-Nolte, D. V. Serreze, et al. Targeted Disruption of CD38 Accelerates Autoimmune Diabetes in NOD/Lt Mice by Enhancing Autoimmunity in an ADP-Ribosyltransferase 2-Dependent Fashion. J. Immunol., April 15, 2006; 176(8): 4590 - 4599. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Zolkiewska Ecto-ADP-ribose Transferases: Cell-Surface Response to Local Tissue Injury Physiology, December 1, 2005; 20(6): 374 - 381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Krebs, S. Adriouch, F. Braasch, W. Koestner, E. H. Leiter, M. Seman, F. E. Lund, N. Oppenheimer, F. Haag, and F. Koch-Nolte CD38 Controls ADP-Ribosyltransferase-2-Catalyzed ADP-Ribosylation of T Cell Surface Proteins J. Immunol., March 15, 2005; 174(6): 3298 - 3305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kawamura, F. Aswad, M. Minagawa, K. Malone, H. Kaslow, F. Koch-Nolte, W. H. Schott, E. H. Leiter, and G. Dennert P2X7 Receptor-Dependent and -Independent T Cell Death Is Induced by Nicotinamide Adenine Dinucleotide J. Immunol., February 15, 2005; 174(4): 1971 - 1979. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Ohlrogge, F. Haag, J. Lohler, M. Seman, D. R. Littman, N. Killeen, and F. Koch-Nolte Generation and Characterization of Ecto-ADP-Ribosyltransferase ART2.1/ART2.2-Deficient Mice Mol. Cell. Biol., November 1, 2002; 22(21): 7535 - 7542. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z.-X. Liu, O. Azhipa, S. Okamoto, S. Govindarajan, and G. Dennert Extracellular Nicotinamide Adenine Dinucleotide Induces T Cell Apoptosis In Vivo and In Vitro J. Immunol., November 1, 2001; 167(9): 4942 - 4947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Bortell, J. Moss, R. C. McKenna, M. R. Rigby, D. Niedzwiecki, L. A. Stevens, W. A. Patton, J. P. Mordes, D. L. Greiner, and A. A. Rossini Nicotinamide Adenine Dinucleotide (NAD) and Its Metabolites Inhibit T Lymphocyte Proliferation: Role of Cell Surface NAD Glycohydrolase and Pyrophosphatase Activities J. Immunol., August 15, 2001; 167(4): 2049 - 2059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Adriouch, W. Ohlrogge, F. Haag, F. Koch-Nolte, and M. Seman Rapid Induction of Naive T Cell Apoptosis by Ecto-Nicotinamide Adenine Dinucleotide: Requirement for Mono(ADP-Ribosyl)Transferase 2 and a Downstream Effector J. Immunol., July 1, 2001; 167(1): 196 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kahl, M. Nissen, R. Girisch, T. Duffy, E. H. Leiter, F. Haag, and F. Koch-Nolte Metalloprotease-Mediated Shedding of Enzymatically Active Mouse ecto-ADP-ribosyltransferase ART2.2 Upon T Cell Activation J. Immunol., October 15, 2000; 165(8): 4463 - 4469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z.-X. Liu, S. Govindarajan, S. Okamoto, and G. Dennert NK Cells Cause Liver Injury and Facilitate the Induction of T Cell-Mediated Immunity to a Viral Liver Infection J. Immunol., June 15, 2000; 164(12): 6480 - 6486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Koch-Nolte, T. Duffy, M. Nissen, S. Kahl, N. Killeen, V. Ablamunits, F. Haag, and E. H. Leiter A New Monoclonal Antibody Detects a Developmentally Regulated Mouse Ecto-ADP-Ribosyltransferase on T Cells: Subset Distribution, Inbred Strain Variation, and Modulation Upon T Cell Activation J. Immunol., December 1, 1999; 163(11): 6014 - 6022. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Weng, W. C. Thompson, H.-J. Kim, R. L. Levine, and J. Moss Modification of the ADP-ribosyltransferase and NAD Glycohydrolase Activities of a Mammalian Transferase (ADP-ribosyltransferase 5) by Auto-ADP-ribosylation J. Biol. Chem., November 5, 1999; 274(45): 31797 - 31803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z.-X. Liu, Y. Yu, and G. Dennert A Cell Surface ADP-ribosyltransferase Modulates T Cell Receptor Association and Signaling J. Biol. Chem., June 18, 1999; 274(25): 17399 - 17401. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-K. Han, Y.-S. Cho, Y. S. Kim, C.-Y. Yim, and U.-H. Kim Interaction of Two Classes of ADP-ribose Transfer Reactions in Immune Signaling J. Biol. Chem., June 30, 2000; 275(27): 20799 - 20805. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), PEC, Ig+ B cells
(Ig+), purified T cells (NWNA), CD4+ cells, and
CD8+ cells were labeled with [32P]NAD. Whole
cell lysates from 106 cells per lane were analyzed by
SDS-PAGE and autoradiography. B, T cell lysates from
107 cells per lane were immunoprecipitated with the Ab
specificities indicated and immunoprecipitates analyzed as in
A. The results shown were repeated in three independent
experiments.
