| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
BMT Section, Transplantation Biology Research Center, Massachusetts General Hospital/Harvard Medical School, Boston, MA 02129;
Laboratory of Immunogenetics and Transplantation, Department of Medicine, Brigham and Women’s Hospital/Harvard Medical School, Boston, MA 02115; and
Department of Medicine, University of Pennsylvania, Philadelphia, PA 19104
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subunits), which
can undergo deletion only in the thymus at the
CD4+CD8+ double-positive
stage but not in the periphery (because they do not react efficiently
with superantigens presented by MHC class II), is not seen; and 3) the
early deletion of donor-reactive CD4+ PBL is also
observed in thymectomized recipients. Two broad categories of apoptotic cell death of lymphocytes are generally recognized (9). Activation-induced cell death (AICD) can be triggered by restimulating activated T cells with high concentrations of Ag (10). AICD is thought to play a role in the maintenance of self-tolerance by eliminating autoreactive lymphocytes (10, 11), is Fas (CD95)-dependent (10, 11, 12, 13), can be promoted by IL-2 (14), and can be inhibited by cyclosporin A (CyA) (15). AICD in most cases cannot be prevented by proteins of the Bcl family (11, 13). When expressed as transgenes, Bcl-2 and Bcl-xL in large part seem to function interchangeably (16). In contrast, passive cell death (PACD) is caused by the withdrawal of growth factors and activation stimuli (11), which can be a consequence of a lack of costimulatory signals (9, 17). PACD is thought to play a role in terminating immune responses to foreign Ags, and can be prevented by the overexpression of Bcl-2 or Bcl-xL (11, 18) but is considered to be independent of the Fas pathway (11, 13, 19). However, in contrast to these seemingly distinct characteristics of AICD and PACD, several in vitro studies have indicated that Bcl proteins could inhibit Fas-mediated cell death (17, 20, 21, 22, 23). Thus, considerable uncertainty persists regarding the precise relationship of the Fas-mediated and Bcl-inhibited cell death pathways. To better characterize the mechanism of peripheral deletion after costimulatory blockade and BMT, we examined this phenomenon using Fas-deficient (lpr) and Bcl-xL-transgenic recipients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female C57BL/6 (B6: I-E-), B6.lpr (lpr; Fas-deficient, I-E-), B10.A (B10.A: I-E+), and A.SW mice were purchased from Frederick Cancer Research Center (Frederick, MD) or from The Jackson Laboratory (Bar Harbor, ME). B6.Bcl-xL-transgenic mice (with transgenic expression of the long form of the Bcl-x gene (Bcl-xL) targeted to the T cell lineage (Ref. 24 ; B6.Bcl-xL, I-E-) and B6 nontransgenic littermates (B6, I-E-) were bred in the colony of Dr. L.A. Turka. Mice were maintained in a specific pathogen-free microisolator environment, as described (25).
BMT protocols
In the standard-dose BMT model, recipient mice were treated with
3 Gy total body irradiation (TBI) and were injected i.v. on the same
day (day 0) with 
20 x 106 unseparated BM
cells (BMC) harvested from fully MHC-mismatched B10.A donors. In the
high-dose BMT model, recipients were injected i.v. with 200 x
106 unseparated BMC harvested from B10.A donors.
In both BMT models, a hamster anti-mouse CD154 mAb (MR1; 0.5 or 2
mg) was injected i.p. on day 0, and murine CTLA4Ig (0.5 mg) was
injected i.p. on day +2. In the high-dose BMT experiment using
B6.Bcl-xL recipients, rat anti-mouse B7.1
(1G10) and rat anti-mouse B7.2 (2D10) mAbs (0.5 mg of each on day
+2) were used instead of CTLA4Ig. CyA was administered s.c. at a dose
of 20 mg/kg/d for the first 2 wk after BMT where indicated. CTLA4Ig was
a gift of Bristol-Myers Squibb (Seattle, WA); the MR1 hybridoma was
provided to us by Dr. Randolph J. Noelle; and the anti-B7.1 and
anti-B7.2 mAbs were purchased from Bioexpress (West Lebanon,
NH).
Flow cytometric (FCM) analysis of TCR V families
PBL were stained with anti-V5.1/2-FITC, V11-FITC, and
V8.1/2-FITC mAb vs PE-conjugated anti-CD4 (all purchased from
PharMingen, San Diego, CA). Two-color FCM analysis was performed on
gated CD4+ cells. Background staining (as
determined with nonreactive mAb HOPC-FITC) was subtracted from the
percentage of cells staining with each anti-V mAb. All
p values were calculated using a two-tailed Student’s
t test.
FCM analysis of thymocyte subpopulations
Thymocytes were stained with anti-CD4-PE vs anti-CD8-CyChrome, and the percentages of CD4 or CD8 single-positive, CD4 CD8 double-positive, and double-negative thymocytes were determined by two-color FCM analysis.
CFU assay
CFUs were determined using a complete methylcellulose medium with recombinant cytokines (MethoCult GF M3434; Stem Cell Technologies, Vancouver, British Columbia, Canada) plus 1x penicillin/streptomycin plus 1x glutamine. BM from one tibia and one femur was flushed into IMDM with 2% FBS. RBCs were lysed in 2% acetic acid, and nucleated cells were counted. Cells were diluted to 1.33 x 105 cells/ml, and 1.5 ml of medium was plated into 35-mm petri dishes. After incubation for 7–8 days at 37°C with 5% CO2, the total number of colonies was counted on an inverted microscope.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Wild-type B6, Fas-deficient B6.lpr,
Bcl-xL-transgenic B6 mice, and nontransgenic
littermate controls were treated with 3 Gy TBI, 20 x
106 fully MHC-mismatched B10.A BMC, and
costimulatory blockade consisting of one injection each of
anti-CD154 plus CTLA4Ig. Seven days after BMT, the percentages of
CD4 cells bearing certain V subunits on their TCR were determined by
two-color FCM analysis. The donor strain B10.A expresses I-E, which is
required to present superantigens derived from mammary tumor virus
(Mtv)-8 and -9 endogenous retroviruses encoded in the B6 background
genome. Developing thymocytes whose TCR contain V11 or V5.1/2,
which bind to these superantigens, are deleted in the thymus of
I-E-positive B10.A mice, but not in B6 mice, because they do not
express I-E (26, 27, 28, 29). V5+ and
V11+ CD4 cells can also be subject to deletion
in the periphery when they recognize superantigen plus donor MHC class
II (I-E) under specific circumstances. T cells whose TCR contain
V8.1/2 do not bind these superantigens, are therefore not deleted,
and thus served as an irrelevant control to assure specificity of the
deletion.
In the first experiment, we sought to examine the role of the Fas
pathway in early deletion of donor-reactive CD4+
cells. As reported previously (6), wild-type B6 BMT
recipients demonstrated a substantial decline in the percentages of
V5+ and V11+
CD4+ PBL, by 74 and 59%, respectively, compared
with B6 controls receiving costimulatory blockade and TBI without BM
(p < 0.00005) (Fig. 1
A). The percentage of
V8+ CD4+ PBL was not
reduced in any group (data not shown), indicating that the observed
deletion was specific for endogenous superantigens presented by the
donor. In marked contrast, B6.lpr recipients showed a
reduction of only 39 and 22%, respectively, in the percentages of
V5+ and V11+
CD4+ PBL, compared with B6.lpr
controls treated with costimulatory blockade alone
(p = 0.1 for V11, p = 0.003
for V5). A significant difference was seen between wild-type B6 and
Fas-deficient BMT recipients in the extent of deletion
(p < 0.001 for V5 and V11).
|
5+ and
V11+ CD4+ PBL was seen 1
wk after BMT in B6 hosts (65 and 68% reduction, respectively, compared
with controls receiving costimulatory blockade alone) (Fig. 1
B). (As expected, nontransgenic littermate controls behaved
similarly to wild-type B6 mice; therefore, these mice are presented
together.) In contrast, Bcl-xL-transgenic BMT
recipients showed reductions of only 20 and 1% in
V5+ and V11+
CD4+ PBL, respectively (compared with
Bcl-xL-transgenic mice treated with costimulatory
blockade alone, p > 0.05). The difference in the
extent of the deletion between wild-type B6 and
Bcl-xL-transgenic BMT recipients was highly
significant (p < 0.002).
The above results were surprising because they implicated both AICD
(Fas-dependent) and PACD (blocked by Bcl-xL
expression) in the deletion of donor-reactive CD4 cells in BMT
recipients. CyA has been shown to inhibit AICD (15), but
should not block PACD because it reduces IL-2 production. Therefore, we
were interested in determining the effect of CyA on deletion after BMT
with costimulatory blockade. We treated wild-type B6 mice with 3 Gy
TBI, 20 x 106 BMC, and costimulatory
blockade (n = 12) and compared
V5+ and V11+ deletion
1 wk after BMT with that in a group receiving the same protocol plus
daily CyA treatment beginning on the day of BMT (n =
13). Recipients of BMT with costimulatory blockade again showed the
expected reduction of V5+ and
V11+ CD4+ PBL (73 and
67% reduction, respectively, compared with normal B6 controls). In
mice receiving CyA in addition to these treatments, there was still
substantial deletion, but it was diminished to some degree (55 and 40%
reduction, respectively; p = 0.09 for V5,
p < 0.01 for V11 compared with the group without
CyA) (Table I). Control mice receiving
TBI, BMC, and CyA (but no costimulatory blockade, n =
5) did not show significant deletion (p >
0.05, compared with normal B6).
|
TBI is an essential component of the BMT model described above.
Therefore, it was important to rule out the possibility that Fas
deficiency or Bcl-xL overexpression indirectly
influenced the peripheral deletion of CD4 cells by abrogating a
participatory role for TBI. This becomes especially relevant in view of
evidence that Fas/Fas ligand interactions play a role in
radiation-induced apoptosis (30). Likewise,
Bcl-2-transgenic mice are more radioresistant than wild-type mice
(31, 32). However, much less is known about the
radiosensitivity of Bcl-xL-transgenic mice.
Therefore, we compared irradiated
Bcl-xL-transgenic and wild-type B6 mice. As shown
in Table II, lymphocytes in
Bcl-xL-transgenic mice were radiosensitive, but
overall less so than those in wild-type mice, as demonstrated by the
lesser decline in viable cells in thymus and spleen after irradiation
with 3 Gy (p < 0.01 for comparison of the
percent reduction in cell counts in these tissues in irradiated
wild-type vs irradiated Bcl-xL-transgenic mice).
BM and lymph node cell counts did not follow this trend, and CFUs were
also comparable between the two groups. Of note, double-positive
thymocytes were markedly less radiosensitive in
Bcl-xL-transgenic mice than in wild-type mice
(p < 0.05 comparing the change in percent
double-positive cells after irradiation in wild-type vs
Bcl-xL-transgenic mice). Thus, although the
radiosensitivity of some cell populations differed between wild-type
mice and Bcl-xL-transgenic mice, other cell types
seemed to react similarly.
|
Because both lpr and
Bcl-xL-transgenic mice react differently to
radiation than wild-type mice, we next used a radiation-free model to
evaluate mechanisms of peripheral T cell deletion after BMT with
costimulatory blockade. We have recently demonstrated that early
peripheral deletion of donor-reactive CD4+ T
cells also occurs after a protocol involving high-dose BMT with
costimulatory blockade without cytoreductive host conditioning
(8). Wild-type B6, Fas-deficient B6.lpr,
Bcl-xL-transgenic B6 mice, and B6-nontransgenic
littermate controls were treated with 200 x
106 fully MHC-mismatched BMC and anti-CD154
plus CTLA4Ig (or anti-CD154 plus anti-B7.1 and anti-B7.2;
Ref. 8). As in the experiments described above, the
percentages of V5+,
V11+, and V8+
CD4+ PBL were determined 7 days after BMT.
Similar to previous results, in wild-type B6 recipients the percentages
of V5+ and V11+ (but
not V8+) CD4+ PBL
declined by 81 and 79%, respectively, after high-dose BMT with
costimulatory blockade compared with controls receiving BMT only
without costimulatory blockade (p < 0.05)
(Fig. 2A) (8). In
contrast, Fas-deficient B6.lpr recipients showed no
significant reduction (p > 0.05) in either
V5+ or V11+
CD4+ PBL compared with normal B6.lpr
mice (p < 0.005 for comparison of B6 and
B6.lpr BMT recipients).
|
5+ and V11+
CD4+ PBL (58 and 59% reductions, respectively,
compared with naive nontransgenic littermates; p <
0.05) (Fig. 2B). In contrast, transplanted
Bcl-xL-transgenic recipients did not show a
significant reduction in V5+ or
V11+ CD4+ PBL (24 and
12% reductions, respectively, compared with naive nontransgenic
littermates; p = 0.05 for V5, p =
0.3 for V11). The difference in the extent of deletion between B6
and Bcl-xL-transgenic high-dose BMT recipients
reached statistical significance for V11 (p
< 0.05; p = 0.1 for V5). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In theory, either AICD or PACD could be responsible for the deletion
after BMT with costimulatory blockade. The injection of a large amount
of Ag in the form of fully allogeneic BMC could lead to repeated
activation of host T cells and subsequent AICD. In addition, several
cell types with veto-like activity contained in the unseparated BM
inoculum might induce Ag-specific apoptosis of host cells. Indeed, in
the high-dose BMT studies, a slight reduction of donor-reactive
CD4+ cells was observed in control mice receiving
only the BM without costimulatory blockade, but this effect was not
observed in a Fas-deficient control mouse (Fig. 2A and Ref.
8). Early deletion after BMT with costimulatory blockade
was also observed to a similar extent when T cell-depleted BM was used
(J.K. and M.S., unpublished data), suggesting that donor T cells are
not required for the deletion and that they are not required as
mediators of putative veto activity. Therefore, it is possible that the
effect of anti-CD154 and CTLA4Ig is primarily to prevent the
rejection of the injected BM (possibly by inducing anergy or
suppression), and that the prolonged presence of donor Ag is enough to
lead to continuous extrathymic deletion of donor-reactive T cells
through AICD or other BM-mediated pathways. Additional support for a
partial role of AICD in the deletion after BMT with costimulatory
blockade comes from the observation that CyA, which has been shown to
prevent AICD, seemed to inhibit deletion to some degree (but by far not
completely). A key feature of AICD is its dependence on Fas
(10, 11, 12, 13). Consistent with this characteristic, the lack of
functional Fas in lpr mice prevented 50% or more of the
peripheral deletion observed in wild-type control mice. However,
proteins of the Bcl-family are thought to be unable to prevent AICD
(4, 11, 13, 19). An additional argument against AICD as
the main mechanism of deletion in these BMT models is that we have not
found evidence for expansion of donor-reactive T cells (J.K. and M.S.,
unpublished observation), which preceded AICD at an early time point in
several studies (33, 34). However, in another system,
superantigen was able to induce deletion of
CD28-/- T cells without evidence of preceding T
cell expansion (35). Thus it seems possible that AICD can
occur without preceding T cell expansion in the absence of an intact
CD28 pathway, which would be the case in our model by day 2 when
CTLA4Ig is administered.
PACD can be induced as a consequence of the absence of growth factors
and activation stimuli due to a lack of costimulation signals (9, 17, 36). The CD28 pathway, which can be blocked by CTLA4Ig, is
considered to provide the most important costimulatory signal for
mature T cells by inducing IL-2 and other essential growth and
activation signals (37). A hallmark of PACD is that it can
be prevented by the overexpression of survival genes of the Bcl-2
family (4, 11, 18). In our experiments, peripheral
deletion of V5+ and
V11+ CD4+ cells was
effectively prevented in Bcl-xL-transgenic
recipients. Because the Bcl-xL transgene does not
augment proliferation of T cells with or without costimulation (but in
fact seems to delay cell cycle kinetics somewhat (Ref. 4 ,
and A.D.W. and L.A.T., unpublished observation), it is unlikely that
the powerful net effect on deletion observed in
Bcl-xL-transgenic BMT recipients reflects
increased proliferation rather than reduced deletion. Also, the lack of
evidence for expansion before the deletion and the inability of CyA to
prevent deletion completely is consistent with a major role for PACD.
However, PACD is generally considered not to be mediated by
Fas.
How can the observations that the extrathymic deletion in these BMT models is partly Fas-dependent but preventable by Bcl-xL expression be reconciled? First, Fas deficiency did not completely inhibit deletion in our studies, suggesting that there is a component of apoptosis that is Fas independent. This component of apoptosis could be due to PACD. However, because Bcl-xL is able to prevent deletion almost entirely, there seems to be a component that is simultaneously Fas-dependent and preventable by Bcl-xL. This form of apoptosis could be either AICD that can be overcome by Bcl-xL expression, or PACD that is Fas-dependent. An alternative explanation is that costimulatory blockade with anti-CD154 plus CTLA4Ig could lead to a novel form of apoptosis that is distinct from classical AICD and PACD. Some of our observations resemble results from in vitro studies describing apoptosis that occurs after TCR ligation without costimulation, is Fas-dependent, and is impeded by CyA (38). A role for T cell apoptosis by PACD and AICD, respectively, has been recently suggested in two transplant models involving peripheral tolerance induction (4, 5). However, considerable uncertainty still exists about the exact interrelationship between the Fas pathway and the Bcl-2/Bcl-xL pathways. Some evidence suggests that they are separate (9, 11, 13, 19, 39), whereas other studies argue that the Fas pathway intersects with the Bcl-2/Bcl-xL pathways. Fas-mediated apoptosis was partially inhibited by Bcl-2 and Bcl-xL, respectively, in in vitro studies using murine and human cell lines (20, 21). To our knowledge, the studies presented here provide the first direct in vivo evidence that Bcl-xL expression can prevent Fas-mediated T cell death and that these two pathways do "overlap."
In summary, the peripheral deletion of donor-reactive T cells in the early period after BMT with costimulatory blockade using anti-CD154 and CTLA4Ig is mediated in part by Fas and can be overcome by the constitutive expression of Bcl-xL. These data show for the first time: 1) that both Fas-dependent and Bcl-xL-reversible cell death play a role in peripheral deletion after BMT with costimulatory blockade, and 2) that the Fas and Bcl-xL pathways are not entirely separate but do intersect functionally in vivo. Thus these observations do not readily fit the criteria believed to distinguish AICD and PACD, but confirm in vivo the existence of a form of apoptosis in which the Fas-mediated and Bcl-xL-inhibited pathways intersect. This intersecting pathway may be important in the maintenance of self-tolerance by peripheral deletion of T cells encountering self-Ags in the absence of costimulation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Surgery, Vienna General Hospital, University of Vienna, Währingergürtel 18, A-1090 Vienna, Austria.
3 Address correspondence and reprint requests to Dr. Megan Sykes, Bone Marrow Transplantation Section, Transplantation Biology Research Center, Massachusetts General Hospital, MGH East, Building 149-5102, 13th Street, Boston, MA 02129.
4 Abbreviations used in this paper: BM, bone marrow; BMT, BM transplantation; AICD, activation-induced cell death; BMC, BM cell; CyA, cyclosporin A; FCM, flow cytometric; PACD, passive cell death; TBI, total body irradiation.
Received for publication September 14, 2000. Accepted for publication November 27, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T lymphocytes for apoptosis. Nature 353:858.[Medline]
11 T cells cosegregate with mammary tumour virus genomes. Nature 349:531.[Medline]
8+ CD4+ T cells in mice tolerant to Staphylococcus aureus enterotoxin B. Nature 349:245.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Manicassamy, D. Yin, Z. Zhang, L. L. Molinero, M.-L. Alegre, and Z. Sun A Critical Role for Protein Kinase C-{theta}-Mediated T Cell Survival in Cardiac Allograft Rejection J. Immunol., July 1, 2008; 181(1): 513 - 520. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Orbach, J. Rachmilewitz, M. Parnas, J.-H. Huang, M. L. Tykocinski, and M. Dranitzki-Elhalel CTLA-4 {middle dot} FasL Induces Early Apoptosis of Activated T Cells by Interfering with Anti-Apoptotic Signals J. Immunol., December 1, 2007; 179(11): 7287 - 7294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Najafian, M. J. Albin, and K. A. Newell How Can We Measure Immunologic Tolerance in Humans? J. Am. Soc. Nephrol., October 1, 2006; 17(10): 2652 - 2663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kurtz, J. Shaffer, A. Lie, N. Anosova, G. Benichou, and M. Sykes Mechanisms of early peripheral CD4 T-cell tolerance induction by anti-CD154 monoclonal antibody and allogeneic bone marrow transplantation: evidence for anergy and deletion but not regulatory cells Blood, June 1, 2004; 103(11): 4336 - 4343. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sandalova, C.-H. Wei, M. G. Masucci, and V. Levitsky Regulation of expression of Bcl-2 protein family member Bim by T cell receptor triggering PNAS, March 2, 2004; 101(9): 3011 - 3016. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Xu, P. M. Chilton, Y. Huang, C. L. Schanie, and S. T. Ildstad Production of Donor T Cells Is Critical for Induction of Donor-Specific Tolerance and Maintenance of Chimerism J. Immunol., February 1, 2004; 172(3): 1463 - 1471. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Blaha, S. Bigenzahn, Z. Koporc, M. Schmid, F. Langer, E. Selzer, H. Bergmeister, F. Wrba, J. Kurtz, C. Kiss, et al. The influence of immunosuppressive drugs on tolerance induction through bone marrow transplantation with costimulation blockade Blood, April 1, 2003; 101(7): 2886 - 2893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Chiffoleau, G. Beriou, P. Dutartre, C. Usal, J.-P. Soulillou, and M. C. Cuturi Role for Thymic and Splenic Regulatory CD4+ T Cells Induced by Donor Dendritic Cells in Allograft Tolerance by LF15-0195 Treatment J. Immunol., May 15, 2002; 168(10): 5058 - 5069. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
