| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Medical Biophysics, University of Toronto and Ontario Cancer Institute, Toronto, Canada; and
Division of Cancer Biology, Sunnybrook Research Institute, Sunnybrook Hospital, Toronto, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, it has been shown that some T cells die in a suicidal manner after activation, and the mechanisms of this activation-induced cell death (AICD) are being intensively studied (3). Evidence has been presented that some forms of AICD are controlled by interactions between members of the TNFR and TNF-ligand superfamilies, in particular the interaction between Fas and its ligand (4). These molecules clearly play a role in the deletion of superantigen-reactive T cells in vivo (5, 6) and in cell death resulting when the TCRs of previously activated T cell lines or clones are cross-linked (7, 8, 9). Less information is available regarding the control of AICD in primary activated T cells, which consist mainly of naive cells.
In this work, the role of perforin in the control of AICD was studied. The results indicate that, in addition to its role in killing by granule exocytosis, perforin is involved in an AICD pathway that occurs early after activation by Ag and is independent of pathways controlled by members of the TNFR family.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6J (B6) (H-2b, Thy-1.2), C56BL/6J-bg/bg, and C57BL/6J-gld/gld mice were obtained from The Jackson Laboratory (Bar Harbor, ME). pko mice on the B6 background (1, 2) and C.B-17 scid mice were bred and maintained in the defined flora animal colony at Ontario Cancer Institute (Toronto, Canada). The original pko breeding pair was generously provided by Dr. H. Hentgartner (Zurich, Switzerland). OVA mice (10), which carry a transgenic TCR specific for amino acids 323–339 of chicken OVA presented by I-Ad (11), were obtained originally from Dr. D. Y. Loh (St. Louis, MO) and have been backcrossed at least six times onto BALB/c in Ontario Cancer Institute animal facility. OVA-scid mice were bred by crossing with C.B-17-scid/scid mice. C.B-17 is congenic to BALB/c, differing only in the region of the Ig heavy chain allele. BALB/c are IgHa and C.B-17 are IgHb (12). F1 mice from the cross were interbred, and progeny was selected for the presence of the transgene and the scid phenotype. The TCR can be identified by the Id-specific Ab, KJ1-26 (13). A high percentage of CD4+KJ26+ PBL in the absence of circulating B or CD8+ T cells indicated the presence of both the transgene and the scid defect. Offspring were derived by Caesarian section and foster-mothered under specific pathogen-free defined flora conditions. All work in this study was performed with such specific pathogen-free animals. 2C mice (14), which carry a transgenic TCR specific for the class I MHC molecule Ld, were also originally obtained from Dr. D. Y. Loh. They have been backcrossed onto B6 more than 10 times, and the expression of the transgene was determined using the mAb, 1B2 (15), specific for the 2C TCR, in flow-cytometric analyses of peripheral blood.
Abs, reagents, and cell lines
Anti-CD3 (145-2C11) (14) was purified from culture supernatants
by protein G column chromatography. Phycoerythrin- or FITC-labeled CD4
and CD8 Abs were purchased from Sigma Chemical Co. (St. Louis, MO). The
anti-Fc
RIII Ab, 2.4G2 (16), was obtained from American Type
Culture Collection (ATCC, Rockville, MD) and used as a culture
supernatant. Phycoerythrin- and FITC-labeled Thy-1.2, Vß8, and
annexin-V Abs and biotinylated anti-Fas Abs were purchased from
PharMingen (San Francisco, CA). 7-AAD and streptavidin-phycoerythrin
were purchased from Sigma Chemical Co. The anti-IL-2 Ab, S4B6.2
(17), was a gift of Dr. T. Mosmann (Edmonton, Alberta, Canada); the
anti-TNF Ab, MP6-XT22 (18), was generously provided by Dr. J.
A. Abrams (DNAX, Palo Alto, CA); and ascites from both were prepared in
pristane-pretreated scid mice. Anti-perforin mAb ascites was
purchased from Kamiya Biomedical Co. (Seattle, WA). The 1B2 Ab, Id
specific for the 2C transgene, was initially obtained from Dr. H. Eisen
(Massachusetts Institute of Technology, Cambridge, MA) (15) and was
purified and biotinylated in our laboratory.
Mouse IL-2 cDNA-transfected X63Ag8-653 cells were a generous gift of H. Karasuyama (Basel Institute Basel, Switzerland) (19). Con A was purchased from Pharmacia (Uppsala, Sweden).
CTLL-2 and P815 tumor lines were obtained from ATCC. P815 was
maintained in exponential growth by serial passage in complete media
(CM: 
-MEM, 10% FCS, 5 x 10-5 M, 2-ME, 2 mM
L-glutamine) at 37°C in an atmosphere of 5%
CO2. CTLL-2 was maintained identically, except that the
medium was supplemented with 25 U/ml IL-2.
TCR complex religation of primary activated T cells
Our assay for T cell activation and induction of AICD in vitro has been previously described (20). Briefly, spleen cells from B6 and pko mice were cultured at 2 x 106 cells/ml with 1 µg/ml of soluble anti-CD3 Ab. After 48 h, these primary activated blasts were isolated using Lympholyte separation medium (Cedarlane Laboratories, Hornsby, Ontario, Canada), washed, and resuspended at 5 x 105 cells/ml in CM. Anti-CD3 Abs were bound to 96-well EIA plates (Corning Costar, Cambridge, MA) by incubation of 100 µl of 2 µg/ml protein G (Pharmacia)-purified Ab in PBS for 3 h at 37°C or overnight at 4°C. The plates were washed three times in PBS before use, and 100 µl of CM was added for 1 h at 37°C to block nonspecific binding. Then 100 µl of the primary activated blasts were added and incubated at 37°C. Control wells were simply blocked with CM. IL-2 (50 U/ml) was added to all cultures.
DNA analysis
The method of Telford et al. was followed (21). T cells (approximately 1 x 106) were washed and fixed in 70% ethanol at -20°C for several days at 106 cells/ml. The cells were then washed in Ca2+, Mg2+-free PBS and resuspended in 1 ml of Ca2+, Mg2+-free PBS with 0.1% Triton X-100, 0.1 mM EDTA, and 50 µg/ml RNase for 1 h at 37°C. This incubation period allows low m.w. DNA to escape through the permeabilized membranes (22). Cells were then washed and resuspended in staining buffer (Ca2+, Mg2+-free PBS, 0.1 mM EDTA, 0.1% Triton X-100, and 50 µg/ml of propidium iodide (Sigma Chemical Co.)) at room temperature in the dark for 4 to 12 h. Cells were then filtered through nylon mesh and analyzed on a FACscan flow cytometer (Becton Dickinson, Mountain View, CA) using LYSIS II software.
Mixed lymphocyte responses
Responder cells were spleen cells diluted to 5 x 106 cells/ml in CM. Irradiated (2000 cGy) spleen cells from BALB/c mice at 5 x 106 cells/ml were used as stimulators. Cultures (200 µl of a 1:1 mixture) were then incubated for 72 h, and 1 µCi of [3H]thymidine (sp. act., 6.7 Ci/mmol) (DuPont NEN, Boston, MA) was added to the cultures for a subsequent 18 h. The cells were then harvested, and the amount of incorporated thymidine was measured in a beta scintillation counter.
Determination of IL-2 secretion
Culture supernatant (100 µl) was transferred to a single well of a 96-well plate, and threefold serial dilutions were made. Each well then received 5000 CTLL-2 cells in 50 µl of complete medium. After 24 h, 1 µCi of [3H]thymidine was added to each well, and the thymidine uptake after a further 18 h of culture was determined by scintillation counting. A standard IL-2 titration curve was performed with each experiment so that the results could be reported as units per milliliter. The standard curve was obtained using supernatants from murine IL-2 cDNA-transfected X63Ag8-653 cells, as described above. The [3H]thymidine incorporation was converted to a percentage of maximum cpm and plotted against the dilutions of the control supernatant on semilog paper. The dilution that gave 50% activity was defined to contain 1 U of activity. The percentage of maximal cpm in the experimental wells was then used to determine the relative amount of IL-2 by comparison with the dilution that gave half-maximal activity.
Cytotoxicity assays
The cells to be tested as effectors were transferred from the
wells in which they had been activated to 96-well U-bottom plates
(Nunclon, Roskilde, Denmark). Target cells were P815 tumor targets in
exponential growth phase or spleen cells that had been activated
48 h previously at 2 x 106 cells/ml with 2
µg/ml of Con A. Con A blasts were harvested using Lympholyte
separation medium (Cedarlane) and washed with -MEM in the presence
of -methyl-mannoside (Sigma Chemical Co.). Tumor targets were
collected by centrifugation. Both pellets were resuspended in two drops
of 100% FCS and radiolabeled with 50 µl of sodium chromate (7.14
mCi/ml) (Dupont NEN). Labeling was allowed to proceed for 1 h in
the case of the tumor targets and 90 min for the Con A blasts.
Chromium-labeled targets were washed three times with -MEM +
1% FCS, and 2000 targets were added to the U-bottom plates in 100 µl
of conditioned media. The plates were centrifuged at 600 rpm for 3 min
and then incubated at 37°C for 4 h. Plates were then centrifuged
at 800 rpm for 5 min, and 100 µl of the supernatant was transferred
to Fisherbrand flint glass tubes (Fisher Scientific, Pittsburgh, PA)
and counted in a gamma counter (CompuGamma Model 1282; LKB, Stockholm,
Sweden). Total release (TR) was measured by lysis of tumor targets with
1% acetic acid, and spontaneous release (SR) was measured in the
absence of effector cells. Percentage of cytotoxicity was determined by
the ratio (cpm - SR)/(TR - SR) x 100%.
For the redirected lysis assays, 1 µg of anti-CD3 Ab was added to the P815 targets before addition of effectors (23).
FITC labeling of cells
Aliquots of the cells to be labeled were washed in PBS and
resuspended at 20 to 50 x 106 cells/ml in PBS.
FITC (Sigma Chemical Co.) was added (final concentration 30 µg/ml),
and the cells were incubated at 37°C for 17 min. The reaction was
stopped by adding 9 ml of CM. Both the labeled and unlabeled cells were
centrifuged a second time using Lympholyte separation medium, and the
interface layer was washed in 50 ml of -MEM (24).
Immunofluorescence
Cell staining (in 100 µl vol) was performed after first blocking nonspecific binding by a 10-min incubation at room temperature with 10 µl of mouse serum (Cedarlane) and 40 µl of 2.4G2 culture supernatant. Cells (5 x 105) were then stained with the appropriate Abs for 20 min, washed, incubated with 7-AAD (Sigma) to identify dead cells, and then analyzed on a FACScan flow cytometer (Becton Dickinson) using LYSIS II software.
Western blot analysis of perforin
Cells were harvested at various times after TCR recross-linking, washed with PBS, and lysed in 50 to 100 µl of lysis buffer (50 mM HEPES, pH 6.8, 1% SDS, 150 mM NaCl, 50 mM NaF, 1 mM sodium vanadate, and 1 mM sodium pyrophosphate). The proteins in the lysate (50 µg/lane) were separated by SDS-PAGE and transferred onto nylon membranes using a semidry transfer apparatus (Hoefer Scientific Instruments, San Francisco, CA). The membranes were blocked in Tris-buffered saline Tween-20 (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.05% Tween-20) containing 5% skim milk, and incubated with antiperforin ascites (1/1000 dilution) in the same medium and 1% skim milk, according to the manufacturer’s instructions. After washing, the membranes were treated with secondary Abs conjugated to horseradish peroxidase, and the bands were visualized using an enhanced chemiluminescence kit (DuPont NEN).
Blockade of perforin expression by antisense oligonucleotides
Three sense oligonucleotides (5'-TGT GCC TGC AGC ATC ATG GCC, 5'-ACT ACC CTC CGC CGC TCC GGC, and 5'-CCA CGG AGG GTG TGG CCC GGG (sense 1, 2, and 3; positions 134–154, 296–316, and 483–503 on perforin mRNA, respectively)) and the corresponding antisense oligonucleotides (5'-GGC CAT GAT GCT GCA GGC ACA, 5'-GCC GGA GCG GCG GAG GGT AGT, and 5'-CCC GGG CCA CAC CCT CCG TGG (antisense 1, 2, and 3)) were synthesized as phosphothioates. The sequences were chosen from the mouse perforin cDNA sequence in GenBank (accession number J04148) (25). They were added at the indicated concentrations during both initial activation and restimulation of splenic cultures with apoptosis measured by eosin uptake 48 h after reactivation.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A series of studies from our laboratory has shown that T cell
blasts undergo programmed cell death when their TCR is reaggregated
48 h after initial activation (20, 26, 27). Activated wild-type
and pko T cells were replated onto anti-CD3 Abs with 50 U/ml IL-2.
Cell viability and DNA fragmentation were assessed 48 h later. As
shown in Table I
, the percentage of
specific death caused by CD3 religation after 48 h was reduced
significantly in cultures of activated pko T cells, while total cell
numbers were little different (Table I). The changes in cell viability
over time following CD3 religation in both cultures were also
determined (Fig. 1,
A and B).
In both cases, control cells treated with IL-2 alone did not exhibit a
significant drop in cell viability. However, in cultures of wild-type
cells treated with anti-CD3, cell viability dropped to less than
60% by 48 h (Fig. 1A). pko cells exhibited only
a slight decrease in cell viability over the same time period. When the
cultures were incubated further (for up to 60 h), protection from
cell death was still observed in the pko cultures (data not shown),
suggesting that a simple delay in cell death was not the cause.
|
|
C showed that the number of T cells with hypodiploid
DNA content 48 h after reactivation was decreased in the absence
of perforin. Hypodiploid peaks are considered to be a quantitative
measure of the number of cells undergoing apoptosis in a cell
population (22). Therefore, primary activated pko T cells appeared to
have a defect in their ability to undergo AICD. The death pathway controlled by perforin affects mainly CD8+ T cells in unfractionated spleen cell cultures
To determine whether reactivated CD4+ and
CD8+ T cell blasts were affected similarly by the absence
of perforin, triple staining for 7-AAD, annexin-V, and CD4 or CD8 was
performed 21 h after reactivation. Phosphatidylserine molecules on
the inner membrane of cells undergoing apoptosis flip to the outer
membrane (28, 29), where they can bind to annexin-V in a
calcium-dependent manner. This change takes place early in cells
undergoing apoptosis. 7-AAD is a nuclear dye taken up by apoptotic
cells when their outer membrane is not intact. Cells that stain with
annexin-V, but exclude 7-AAD, have been shown to be in the early stages
of apoptosis (28, 29). As shown in Figure 2, more CD8+ T cells than
CD4+ T cells died, by these criteria, following
reactivation of unfractionated spleen cell cultures. Protection from
apoptosis in the absence of perforin was significant only for
CD8+ T cells (Fig. 2).
|
Expression of perforin protein after initial activation was
determined by Western blotting. As shown in Figure 3, perforin was expressed 24 h after
reactivation and was present throughout the time period in which cells
underwent apoptosis (lanes 2 and
5).
|
Although the development of T cells in pko mice is assumed to be
normal (1, 2, 30), perhaps the molecular pathways that actually mediate
AICD fail to develop when perforin is absent during ontogeny,
accounting for the observed survival of pko T cells. To directly test
the role of perforin in AICD of wild-type cells, the effect of
antisense oligonucleotides that interfered with the translation of
perforin mRNA was studied. Viability of reactivated B6 spleen cells was
increased with antisense 3, but not the sense oligonucleotide (Table II). The experiments were repeated using
spleen cells from OVA-scid mice as a monoclonal source of
naive T cells. Despite the fact that these cells were CD4+,
they expressed large quantities of perforin in response to activation
by anti-CD3 Abs (data not shown), in agreement with recent studies
showing that activation of CD4+ T cells in the absence of
CD8+ T cells results in perforin expression (31, 32). As
shown in Table II, two different antisense oligonucleotides, added at
both the time of initial activation and recross-linking, increased the
viability of the cells in a concentration-dependent manner. The
corresponding sense oligonucleotides did not prevent cell death. These
results implicated perforin directly in the apoptosis observed after
reactivation.
|
Nonspecific killing of bystander cells by perforin released after
reaggregation of the TCR on activated T cells would be the most
conventional explanation for the involvement of perforin in AICD. We
first asked whether recross-linked cells could kill nonspecifically
activated syngeneic T cells. As shown in Table I (lines 1 and 2,
columns 7 and 8), control and recross-linked T cells were highly
cytotoxic to P815 tumor cells in a redirected lysis assay. That this
nonspecific cytotoxicity was due to perforin was revealed by its
absence when pko T cells were evaluated in the same assay (Table I,
lines 3 and 4, columns 7 and 8). Despite the presence of strong
nonspecific cytotoxicity, 48-h activated Con A blasts were at best
minimally killed by the recross-linked cells (Table I, line 2, column
9). These results suggested that the observed AICD was not just caused
by the TCR recross-linked cells killing each other through nonspecific
perforin release.
In another attempt to address the issue of whether perforin mediated
suicide or fratricide, T cells from B6 or pko mice were activated for
48 h, labeled with fluorescein, and mixed with equal numbers of
unlabeled cells of the opposite perforin-expressing phenotype. They
were then replated on anti-CD3-bound plates. Forty-eight hours
later, the recross-linked cells, along with control mixes grown only in
IL-2, were harvested, and the number of viable cells of each type
remaining in the cultures was enumerated by flow cytometry. If perforin
controlled an autonomous death pathway, more pko cells than B6 cells
should survive the TCR recross-linking step. If there was fratricidal
killing of pko cells by B6, the number of surviving cells of each type
should be approximately the same. As shown in Table III, pko cells had a survival advantage
compared with B6 T cells following reactivation that was not observed
in the control cultures and was quantitatively compatible with an
autocidal functioning of the perforin-dependent death pathway. For
example, in experiment 1 (Table III), the percentages of wild-type and
pko cells in the control culture were 60 and 40%, respectively.
Assuming that the viability after 48 h was 75% for reactivated
pko blasts and 50% for reactivated wild-type blasts (Fig. 1A), then the viability of reactivated pko blasts in
the mixture should be 50%, which is very close to the value of 52%
observed experimentally.
|
The granule exocytosis pathway is defective in mice homozygous for the
beige (bg) mutation (33), in this case due to
a defect in granule formation (34) rather than the absence of perforin
as in pko mice. If the observed cell death in the AICD assay used in
this study was fratricidal and mediated by the granule exocytosis
pathway, then reactivated T cells from bg/bg mice should
also be protected from death. Figure 4A confirmed that redirected
lysis of P815 cells by activated bg/bg T cells was
approximately 10-fold lower than by activated wild-type B6 T cells.
However, the number of apoptotic bg/bg T cells was even
greater than wild-type cells 48 h after reactivation (Fig. 4B). This result indicates that AICD occurs in
bg/bg T cells and is consistent with our hypothesis of an
autocidal role for perforin.
|
It is known that IL-2 can program activated T cells to undergo
AICD (35, 36, 37). We wondered whether defective IL-2 production, rather
than the absence of perforin itself, might account for the decreased
AICD observed with pko T cells. Accordingly, IL-2 production from pko
and B6 wild-type T cells activated in an MLR was determined (Table IV). The pko T cells secreted
approximately two- to threefold more IL-2 than wild-type T cells in the
first 3 days of an MLR. Addition of wild-type T cells did not prevent
the increased production of IL-2 (Table IV). Since pko T cells produce
more IL-2 at short times after activation than wild-type T cells, it
seems unlikely that the decreased death of pko cells is a direct effect
of IL-2 levels.
|
), consistent with previous
results (35). IL-2 blockade of pko cultures did not increase further
the viability of the remaining cells, suggesting that other apoptotic
pathways were not blocked. Figure 3 confirmed the known dependence of
perforin protein expression on IL-2, as perforin could not be
demonstrated, or was markedly decreased, when anti-IL-2 ascites was
added throughout the culture period (Fig. 3, lanes 2 and
5). These results suggested that changes in perforin
may help to explain the pro-apoptotic effects of IL-2, at least in this
short-term cell death assay.
|
There is much experimental evidence showing that Fas and its
ligand cause the death of activated T cells in vivo and in vitro.
Despite previously published results demonstrating normal Fas signaling
in pko mice (2) and that the AICD assay used in these studies is Fas
independent (5), the possibility that Fas defects could account for the
behavior of pko cells described above was further examined. It is known
that cell death after recross-linking the TCR of short-term T cell
lines is mediated by Fas (3). Accordingly, short-term polyclonal lines
were obtained by soluble anti-CD3 stimulation of pko and B6 spleen
cells, followed by expansion in IL-2 for 5 days. As shown in Figure 5A, the level of Fas
expression on both types of lines at the end of the culture period was
equivalent. Cells were then replated onto anti-CD3 Ab-coated plates
in the presence of IL-2, and the number of surviving cells was
determined. As shown in Figure 5B, cell death occurred
rapidly and was the same for B6 wild-type and pko blasts. As a positive
control for the role of Fas in this form of cell death, T cell lines
from B6-gld mice were protected from death (Fig. 5B).
|
TNF- has recently been shown to cause the death of
reactivated CD8+ T cells by signals through the p75 TNFR
(39, 40). These studies were performed using short-term T cell lines.
Since the death pathway controlled by perforin in the AICD assay used
in our current study seemed to mainly involve CD8+ T cells,
the role of TNF- was studied. Reactivation of pko T cells was
performed in the presence of an anti-TNF- mAb, and the degree of
apoptosis was determined by eosin uptake 48 h following TCR
religation. The logic is similar to that of gene complementation
experiments. If TNF- blockade prevented AICD in both pko and
wild-type T blasts, it acts independently of the perforin-dependent
death pathway. If TNF- blockade protected wild-type cells, but not
pko T cells from AICD, then the protection afforded by the perforin
deficiency acts through TNF-. As shown in Table VI, TNF blockade increased the survival
of reactivated wild-type T cells, suggesting that TNF- played a role
in the apoptosis observed in the AICD assay used in this study. TNF-
blockade also increased the survival of reactivated pko blasts, and
their previously demonstrated survival advantage was maintained
compared with similarly treated wild-type cells. This result suggested
that the death pathway mediated by perforin was independent of that
mediated by TNF-.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1) Reactivation of previously activated pko T cells caused less cell
death, as measured by reduced eosin staining, hypodiploid DNA peaks,
and annexin-V positivity (Table I and Figs. 1 and 2).
2) Death was not fratricidal since marked pko T cells behaved
independently when mixed with wild-type T cells in vitro (Table III).
Moreover, killing of chromium-labeled T cell blasts by reactivated T
cells could not be detected (Table I).
3) Antisense blockade of perforin expression reduced the death of
reactivated wild-type T cells (Table II).
4) The disproportionate number of CD4+ and
CD8+ T cells that stained positively with annexin-V FITC
(Fig. 2) after recross-linking also supported an autonomous role for
perforin. Fratricidal killing should not have discriminated between
CD4+ and CD8+ T cells in cultures that involved
both types of cells.
5) Reactivated T cells from bg/bg mice, in which perforin
expression is normal despite significant defects in the granule
exocytosis killing pathway, were not protected from apoptosis
(Fig. 4).
Increased numbers of surviving cells could potentially also be
explained by a faster doubling time of pko T cells. However, as shown
in Figure 1B, the numbers of both B6 and pko T cells after
reactivation were essentially the same for the first 30 h,
suggesting that their doubling times were the same. Differences began
to show up after this time, presumably reflecting the role of perforin
in increasing the number of apoptotic cells.
The cell death pathway controlled by perforin mainly seems to involve
CD8+ T cells when unfractionated spleen cells are
activated. This result fits intuitively with the known distribution of
perforin expression, which is mainly confined to CD8+
T cells and NK cells. However, some types of human and murine
CD4+ T cells, especially Th2 cells, express mRNA for
perforin (38). Recent studies have also shown that the context of
initial activation may determine the expression of perforin by
CD4+ cells. Alloreactive CD4+ T cell lines
derived by stimulation of whole spleen cell suspensions did not express
perforin compared with similarly reactive lines derived by first
purifying CD4+ T cells (31, 32). Certainly the
CD4+ OVA-scid cells expressed perforin after
activation by anti-CD3 Abs (Table II) perhaps because of the
absence of CD8+ T cells in this system.
The increased amount of IL-2 production noted in MLRs with pko T cell
responders can be explained in terms of a defect in AICD. Sarin et al.
(41) have shown previously that protease inhibitors can block AICD and
also increase IL-2 production by T cell lines. Their explanation is
that cells that would otherwise die continue to produce IL-2 in
response to Ag. This explanation may also apply to the finding of
increased IL-2 production from pko mice in vitro (Table III). Sad and
Mossman (42) have also noted increased production of cytokines by pko T
cells.
The in vitro assay used in these experiments may have emphasized the
role played by perforin in AICD. Recently, both Fas-Fas ligand and
TNF-TNFR (p75) interactions have been implicated in the control of AICD
of T cell hybridomas and primary activated T cells expanded in IL-2 for
more than 2 days (3, 40). The paradigm that has emerged suggests that
Fas-Fas ligand interactions cause the death of CD4+ T
cells, and TNF-p75TNFR interactions kill CD8+ T cells.
However, the assay used in this study involved direct reactivation of
primary activated T cells without prior expansion in IL-2. The period
of expansion in IL-2 may be an important step that allows for the full
development of the death machinery controlled by members of the TNFR
family (4). There are important quantitative and kinetic differences
between the two types of assays. As illustrated in Figure 5B, the death of short-term T cell lines mediated by
TNFR-family members is more destructive (higher percentage of dead
cells) and much faster (by 24 h) compared with the death of
primary activated T cells.
Fas-Fas ligand interactions do not seem to be important in the type of AICD involving primary activated T cells without expansion in IL-2. T cells from young (6- to 8-wk-old) lpr mice are not protected from cell death in this assay (5), and susceptibility to the Fas death signal requires 4 days of expansion after initial activation and occurs later than the AICD studied in this investigation (4, 43).
In a similar fashion, blockade of TNF signaling throughout the
culture period increased the viability of B6 blasts, but never to the
same degree as pko blasts (Table VI).
This result suggests that TNFR signaling is involved in the death
of primary activated T cells not expanded in IL-2, but the death
pathway involving perforin is independent of TNF-. The roles of
other members of the TNFR superfamily (such as NGFR (nerve growth
factor receptor), 4-1BB, Ox40, CD27, CD30, the TRAIL (tumor necrosis
factor-related apoptosis-inducing ligand) receptor, and CD40) (44) in
this death pathway remain to be determined.
Further evidence that the death pathway controlled by perforin is distinct from that controlled by Fas is provided by the phenotype of mutant mice deficient for both perforin and functional Fas ligand. These mice develop more severe lymphoproliferative disease than Fas ligand-deficient gld/gld mice (45). This observation is more consistent with a second pathway of AICD controlled by perforin than the possibility that the observed defects in AICD in pko mice act through Fas.
IL-2 is another mediator of AICD of T cells. Again, many of its known
in vitro effects on T cell death have been studied in short-term cell
lines (35) rather than in primary cultures, as we have done in this
study. IL-2 most likely is involved in multiple processes that cause
AICD. The results of Table V suggest that one of these processes is its
control of perforin expression.
The findings in this study suggest that a hitherto unknown mechanism of AICD has been revealed by pko mice. The mechanism has not been determined in this work, but is currently under investigation in our laboratory. Our current hypothesis is that perforin acts internally on intracellular membranes, including mitochondria, the nucleus, or the endoplasmic reticulum, causing damage that activates the effector mechanisms of apoptosis. Perforin may act internally in concert with granzyme family members as when killing CTL targets. Consistent with this idea is that serine proteases of granular origin have been found in the nuclei of lymphokine-activated killer cells (46). These hypotheses strongly predict that perforin can be found at multiple sites outside of the granules after recross-linking and should be demonstrable by cell fractionation experiments and immunofluorescence or confocal microscopy. They also predict that constitutive overexpression of perforin in lymphocytes of transgenic mice may result in profound lymphopenia and immunodeficiency.
Perforin has recently been shown to have two routes of secretion (47). In addition to routing to the lytic granules, perforin can be secreted directly. Since this second route is intact in bg/bg mice, perforin in the secretory pathway may mediate the AICD studied in this investigation.
The results in this study suggest that there are multiple levels of control on AICD in T cells. One level may be controlled by perforin, mainly in CD8+ T cells, and fairly early after activation. At later times after activation, pathways controlled by members of the TNFR superfamily, such as Fas and p75 TNFR, may become dominant in the control of immune responses. This later dominance of the TNFR-controlled pathways may explain the fact that pko mice do not develop a lymphoproliferative syndrome, as seen in lpr and gld strains of mice. However, the importance of the perforin-controlled death pathway in vivo may be illustrated by the accelerated disease seen in gld-pko mice described above (45) and the hyperexpansion of (mainly) CD8+ cells in pko mice infected with noncytopathic viruses (30), or when T cells from pko mice are injected into sublethally irradiated C.B-17 scid mice (D. Spaner, manuscript in preparation). The fact that pko T cells are not completely protected from death in the assay used in the studies described in this work, plus the demonstrated independence from Fas and TNF mechanisms, suggests that there are other mediators of early cell death still to be described. Possible candidates include other members of the TNFR superfamily.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. David Spaner, Division of Cancer Biology, Sunnybrook Research Institute, S-116A, S-Wing, Research Building, Sunnybrook Hospital, 2075 Bayview Avenue, North York, Ontario, Canada M4N 3 M5. E-mail address:
3 Abbreviations used in this paper: pko, perforin knockout; 7-AAD, 7-aminoactinomycin D; AICD, activation-induced cell death; CM, complete media.
Received for publication March 14, 1997. Accepted for publication November 20, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß T lymphocytes for apoptosis. Nature 353:858.[Medline]
This article has been cited by other articles:
|
|
 |
|
  V. Mateo, M. Menager, G. de Saint-Basile, M.-C. Stolzenberg, B. Roquelaure, N. Andre, B. Florkin, F. le Deist, C. Picard, A. Fischer, et al. Perforin-dependent apoptosis functionally compensates Fas deficiency in activation-induced cell death of human T lymphocytes Blood, December 15, 2007; 110(13): 4285 - 4292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Laforge, N. Bidere, S. Carmona, A. Devocelle, B. Charpentier, and A. Senik Apoptotic Death Concurrent with CD3 Stimulation in Primary Human CD8+ T Lymphocytes: A Role for Endogenous Granzyme B J. Immunol., April 1, 2006; 176(7): 3966 - 3977. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kawano, C. L. Efferson, G. E. Peoples, D. Carter, N. Tsuda, J. L. Murray, and C. G. Ioannides Sensitivity of Undifferentiated, High-TCR Density CD8+ Cells to Methylene Groups Appended to Tumor Antigen Determines Their Differentiation or Death Cancer Res., April 1, 2005; 65(7): 2930 - 2937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. X. Wu, J. La Rose, L. Chen, C. Neale, T. Mak, K. Okkenhaug, R. Wange, and R. Rottapel CD28 Regulates the Translation of Bcl-xL via the Phosphatidylinositol 3-Kinase/Mammalian Target of Rapamycin Pathway J. Immunol., January 1, 2005; 174(1): 180 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Astsaturov, T. Petrella, E. U. Bagriacik, M. de Benedette, R. Uger, G. Lumber, N. Berinstein, I. Elias, N. Iscoe, C. Hammond, et al. Amplification of Virus-Induced Antimelanoma T-Cell Reactivity by High-Dose Interferon-{alpha}2b: Implications for Cancer Vaccines Clin. Cancer Res., October 1, 2003; 9(12): 4347 - 4355. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bose, Y. Inoue, K. E. Kokko, and F. G. Lakkis Cutting Edge: Perforin Down-Regulates CD4 and CD8 T Cell-Mediated Immune Responses to a Transplanted Organ J. Immunol., February 15, 2003; 170(4): 1611 - 1614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. F. Davidson, C. Haudenschild, J. Kwon, and M. S. Williams T Cell Receptor Ligation Triggers Novel Nonapoptotic Cell Death Pathways That Are Fas-Independent or Fas-Dependent J. Immunol., December 1, 2002; 169(11): 6218 - 6230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Spaner, X. Sheng-Tanner, and A. C. Schuh Aberrant regulation of superantigen responses during T-cell reconstitution and graft-versus-host disease in immunodeficient mice Blood, August 28, 2002; 100(6): 2216 - 2224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. N. Balaji, N. Schaschke, W. Machleidt, M. Catalfamo, and P. A. Henkart Surface Cathepsin B Protects Cytotoxic Lymphocytes from Self-destruction after Degranulation J. Exp. Med., August 19, 2002; 196(4): 493 - 503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Raju, B. A. Rabinovich, L. G. Radvanyi, D. Spaner, and R. G. Miller A central role for IL-2 in fate determination of mature T cells--I: role in determining the Th1/Th2 profile in primary T cell cultures Int. Immunol., December 1, 2001; 13(12): 1453 - 1459. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Gorak-Stolinska, J.-P. Truman, D. M. Kemeny, and A. Noble Activation-induced cell death of human T-cell subsets is mediated by Fas and granzyme B but is independent of TNF-{alpha} J. Leukoc. Biol., November 1, 2001; 70(5): 756 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Quinn, M. Melo, D. Ethell, and Eli. E. Sercarz Relative resistance to nasally induced tolerance in non-obese diabetic mice but not other I-Ag7-expressing mouse strains Int. Immunol., October 1, 2001; 13(10): 1321 - 1333. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Opferman, B. T. Ober, R. Narayanan, and P. G. Ashton-Rickardt Suicide induced by cytolytic activity controls the differentiation of memory CD8+ T lymphocytes Int. Immunol., April 1, 2001; 13(4): 411 - 419. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. ARNOLD, S. BALKOW, R. STAATS, H. MATTHYS, W. LUTTMANN, and J. C. VIRCHOW Jr. Increase in Perforin-positive Peripheral Blood Lymphocytes in Extrinsic and Intrinsic Asthma Am. J. Respir. Crit. Care Med., January 1, 2000; 161(1): 182 - 186. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. E. Stepp, R. Dufourcq-Lagelouse, F. L. Deist, S. Bhawan, S. Certain, P. A. Mathew, J. Henter, M. Bennett, A. Fischer, G. d. S. Basile, et al. Perforin Gene Defects in Familial Hemophagocytic Lymphohistiocytosis Science, December 3, 1999; 286(5446): 1957 - 1959. [Abstract] [Full Text]
|
|
|
|
|
|
D. Spaner, X. Sheng-Tanner, K. Raju, B. Rabinovich, H. Messner, and R. G. Miller Long-term persistence of IL-2-unresponsive allogeneic T cells in sublethally irradiated SCID mice Int. Immunol., October 1, 1999; 11(10): 1601 - 1614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Spaner, K. Raju, B. Rabinovich, and R. G. Miller A Role for Perforin in Activation-Induced T Cell Death In Vivo: Increased Expansion of Allogeneic Perforin-Deficient T Cells in SCID mice J. Immunol., January 15, 1999; 162(2): 1192 - 1199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Deas, C. Dumont, M. MacFarlane, M. Rouleau, C. Hebib, F. Harper, F. Hirsch, B. Charpentier, G. M. Cohen, and A. Senik Caspase-Independent Cell Death Induced by Anti-CD2 or Staurosporine in Activated Human Peripheral T Lymphocytes J. Immunol., October 1, 1998; 161(7): 3375 - 3383. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
