| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
to CD8+ T Cell-Mediated Rejection of Pancreatic Islet Allografts1
Departments of
*
Immunology and
Medicine, Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center, Denver, CO 80262
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in triggering
graft destruction. To study CD8+ T cell effector pathways
independently of other lymphocyte populations, purified alloreactive
CD8+ T cells were adoptively transferred into severe
combined immune-deficient (SCID) recipients bearing established islet
allografts. Results indicate that to reject established islet
allografts, primed CD8+ T cells do not require the
individual action of the conventional cytotoxic effectors perforin and
Fas ligand. In contrast, the ability to produce IFN- is critical for
efficient CD8+ T cell-mediated rejection of established
islet allografts. Furthermore, alloreactive CD8+ TCR
transgenic T cells (2C) also show IFN- dependence for mediating
islet allograft rejection in vivo. We speculate from these results that
the production of IFN- by alloreactive CD8+ T cells is a
rate-limiting step in the process of islet allograft
rejection. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Despite these findings, the mechanism by which alloreactive CD8+ T cells destroy islet grafts is not known. Because activated CD8+ T cells are capable of contact-dependent cytotoxicity as well as abundant cytokine production, there are numerous effector processes to consider. The most extensively studied property of CD8+ T cells is their ability to kill allogeneic tumor cells and other targets in vitro. This cytotoxicity has been ascribed to two separate but complementary pathways, each requiring physical contact between the T cell and its target. The first involves the vectorial release of cytotoxic granules from the effector cell toward the target. The most important granule component is perforin, which forms multimeric pores in the target cell membrane. These pores cause osmotic lysis of the target and also allow the entry of other granule contents such as granzyme B and granulysin, which destroy intracellular structures and may trigger apoptotic pathways (6, 7). The second pathway involves an interaction between Fas ligand (FasL)3 (CD95L), a member of the TNF superfamily, and Fas (CD95), a member of the TNFR superfamily. When FasL on the surface of an activated CD8+ T cell binds to Fas on the surface of a susceptible target, the interaction is sufficient to initiate apoptosis of the target cell (6). While other TNF/TNFR family members such as TNF-related apoptosis-inducing ligand (TRAIL) and TNF itself are capable of mediating apoptotic signals (8), these appear to be less pronounced pathways of cytotoxicity in vitro. In fact, perforin and FasL are thought to account for nearly all of the in vitro killing activity exhibited by cytotoxic CD8+ T cells (CTLs) (9, 10).
In addition to their cytotoxic effector function, activated
CD8+ T cells also have the ability to produce
high levels of proinflammatory Th1-type cytokines including IFN-
(11). In an alloimmune response, IFN- up-regulates
expression of MHC molecules and enhances alloantigen presentation on
target tissues; IFN- also serves to enhance inflammation and
stimulate nonspecific effector cells such as macrophages and NK cells
(12). There is considerable evidence of increased
intragraft expression of IFN- during rejection of cardiac, renal,
and islet transplants (13), and treatment of animals with
anti-IFN- Abs has been shown to prolong graft survival in some
cases (14). Surprisingly, in other model systems IFN-
appears to have an immunoregulatory function, limiting T cell
proliferation and facilitating the induction of allograft tolerance
(15, 16).
Which, if any, of these effector functions are critical for CD8+ T cell-mediated islet allograft rejection is not clear. A predominant view is that immune-mediated islet destruction involves the actions of perforin and/or FasL. This hypothesis has received support from studies of the pathogenesis of autoimmune diabetes, in which islet destruction has been demonstrated to be either perforin dependent (17, 18) or Fas/FasL dependent (19, 20, 21) depending on the model system under scrutiny. However, in contrast to these reports, perforin and FasL do not appear to be essential mediators of islet allograft rejection, as mice lacking a functional perforin gene are capable of rejecting both normal as well as Fas-deficient (lpr/lpr) islet allografts (22, 23). Similar results have been obtained from studies of other types of allografts. For example, allogeneic tumor cells that are insensitive to lysis through Fas are promptly rejected in perforin-deficient mice (24), and Fas-deficient skin grafts are rejected by perforin-deficient CD4+ T cells used to reconstitute athymic recipients (25). Furthermore, wild-type heart allografts are rejected normally in perforin-deficient hosts (26) and lpr/lpr heart allografts are rejected by wild-type hosts (27).
Equally complex data have emerged with regard to the role of IFN- in
islet allograft destruction. As with perforin and FasL, there is
compelling evidence that IFN- is required for the islet cell damage
that occurs in autoimmune diabetes. For example, IFN- has been shown
to be essential for onset of disease in a virally induced model of
diabetes (28), and disease progression in the nonobese
diabetic mouse appears to require production of IFN-
(29) as well as signaling through the IFN- receptor
(30). There is also a large body of evidence to support a
role for IFN- in the rejection of tissue allografts. Hao et al.
(31) showed that treating islet allograft recipients with
cyclosporine A, which inhibits cytokine mRNA production by T cells,
also abolishes the islet graft-destructive capacity of
CD8+ T cells in vivo. Subsequently, several other
investigators demonstrated a strong correlation between elevated
intragraft levels of Th1/Tc1-type cytokines, particularly IFN-, and
the rejection of various types of allografts (13). In
accordance with these observations, Ring et al. (32)
recently showed that host-derived IFN- is required for the efficient
rejection of MHC class II-disparate skin allografts by
CD4+ T cells. However, rejection of fully
allogeneic heart (33) and skin (16, 32)
grafts has been clearly shown to occur in the absence of host-derived
IFN-, suggesting that there must also exist IFN--independent
mechanisms of allograft destruction. In contrast, IFN- also appears
to be required for the induction of allograft tolerance following
costimulatory blockade, perhaps due to the inhibitory effect of this
cytokine on T cell proliferation (15, 16, 34). Thus,
IFN- may have multiple functions in vivo, serving as either a
negative regulator or as a pathogenic mediator of the allograft
response, depending on the circumstances surrounding its
expression.
Importantly, none of the studies cited above specifically examined the
requirements of the CD8+ effector T cell subset
in allograft rejection; rather, they addressed the ability of the
overall allograft response to proceed in the absence of the pathway(s)
being examined. We sought to clarify the mechanism of
CD8+ T cell-mediated islet allograft rejection by
studying CD8+ T cells independently of other
cells of the adaptive immune system. To do this, we employed an
adoptive transfer system in which SCID mice served as islet allograft
recipients. These animals, which lack functional B and T cells, were
rendered diabetic with streptozotocin, grafted with allogeneic islets,
and then reconstituted with highly purified, in vitro-primed
alloreactive CD8+ T cells obtained from a variety
of donors. In this way, the reconstituting CD8+ T
cells could be assayed for their ability to mediate allograft rejection
in the absence of B cells and CD4+ T cells. We
found that primed CD8+ T cells do not require
perforin and are only partially dependent on FasL to cause rejection of
islet allografts. Furthermore, we demonstrated that while the overall
islet allograft rejection response is IFN- independent, the
rejection of established islet allografts by CD8+
T cells is clearly an IFN--dependent process.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male C57BL/6J (H-2b), BALB/c ByJ
(H-2d), C57BL/6-Pfptm1 (B6
perforin-deficient), B6Smn.C3H-Faslgld (B6
FasL-deficient), C57BL/6J-Ifgtm1@ (B6
IFN--deficient), BALB/c-Ifgtm1@ (BALB/c
IFN--deficient), and C57BL/6J-scid/SzJ (C57BL/6J
scid/scid) mice were obtained from The Jackson Laboratory
(Bar Harbor, ME). Mice carrying the 2C TCR transgene (35)
were a generous gift from Richard Miller (Toronto, Canada). These mice
were backcrossed to C57BL6 mice for nine generations in our center
before use in these experiments.
Isolation of CD8+ T cells
In non-TCR transgenic experiments, four strains of mice were
used as CD8+ T cell donors: 1) wild-type control
B6 mice, 2) B6 perforin-deficient
(pfp-/-) mice, 3) B6
FasL-deficient (gld/gld) mice, and 4) B6 IFN--deficient
(IFN--/-) mice.
Following sacrifice by cervical dislocation, lymph nodes and spleens
were removed, ground to a single-cell suspension, and enriched for
lymphocytes by centrifugation over Lympholyte-M (Cedarlane
Laboratories, Hornby, Ontario, Canada). The lymphocytes were then
passed over Cellect immunoaffinity columns (Biotex, Edmonton, Canada)
to effectively remove IgG+ and
CD4+ cells. Purity of eluted cells (<0.2%
CD4+ and 0.6 ± 0.3%
B220+) was assessed by flow cytometry.
Flow cytometry
Freshly isolated lymphocytes, before and after
CD8+ T cell enrichment, T cell blasts obtained
from day 5 cultures (see below), and peripheral blood leukocytes
obtained from reconstituted SCID animals were directly labeled with
FITC-conjugated rat anti-mouse CD4 (clone RM4.4), CD8
(clone
53-6.7), and CD45R/B220 (clone RA3-6B2) (PharMingen, San Diego, CA). To
detect the 2C clonotype, lymphocytes were stained with the biotinylated
Ab 1B2-H6 (36) (a gift from Richard Miller, Toronto,
Canada), which specifically recognizes the clonotype encoded by the 2C
TCR transgene. Streptavidin-FITC (PharMingen) was used as a second-step
reagent. Frequency determinations were calculated from single-parameter
fluorescence histograms on an Elite flow cytometer (Coulter
Electronics, Palo Alto, CA) after gating on viable lymphocytes.
In vitro priming and proliferation assay
Non-TCR transgenic cells were primed in vitro by mixing 2 x 105 CD8+ T cells from control or mutant B6 mice with 3 x 105 3500-rad-treated BALB/c splenocytes in a total of 0.2 ml of Eagle’s MEM supplemented with 10% FCS, 10-5 M 2-ME, and antibiotics (EMEM). Cells were incubated in 96-well flat-bottom plates at 37°C in 10% CO2 in air. Proliferative responses were determined by pulsing on days 3, 4, and 5 of primary culture with 0.6 µCi [3H]thymidine for 6 h, harvesting, and counting samples on a Wallac beta emission counter (Gaithersburg, MD). For CTL assays, primary cultures were established in 24-well plates with 2 x 106 control or mutant B6 CD8+ T cells as responders and 3 x 106 3500-rad-treated BALB/c splenocytes as stimulators in a total volume of 2 ml EMEM. For adoptive transfer, primary cultures were established in 75-cm2 flasks with 20 x 106 control or mutant B6 CD8+ T cells as responders and 30 x 106 3500-rad-treated BALB/c splenocytes as stimulators in a total of 20 ml EMEM.
Immune deviation of 2C CD8+ T cells
Lymphocytes from 2C TCR transgenic donors were isolated as
described above for non-TCR transgenic donors, omitting the
immunoaffinity enrichment step. 2C T cells were then cultured in the
presence (treated) or absence (untreated) of 20 ng/ml anti-IFN-
(clone XMG1.2, rat IgG1) (37) and 1 ng/ml recombinant
mouse IL-4 (mrIL-4) (PharMingen). For proliferation assays, 1 x
105 responder 2C T cells were cultured with
3 x 105 3500-rad-treated BALB/c stimulator
cells. Pulsing, harvesting, and counting were conducted as described
above. For CTL assays, treated and untreated primary cultures were
established in 24-well plates with 1 x 106
responders and 3 x 106 stimulators. For
adoptive transfer, treated and untreated primary cultures were
established in 75-cm2 flasks with 10 x
106 2C responder cells and 30 x
106 irradiated BALB/c stimulator cells.
CTL assay
CTL activity in vitro was assessed by a standard 51Cr release assay. Briefly, primary MLCs were established in 24-well plates as described above. On day 5 of culture, varying numbers of effector T cells were incubated with 104 51Cr-labeled P815 (H-2d) tumor target cells for 4 h at 37°C in 10% CO2. Supernatants were harvested and 51Cr release was detected on a TopCount counter using LumaPlate solid scintillation (Packard, Meriden, CT). Cytotoxic activity was expressed as percent specific lysis, calculated by the formula ((experimental release - spontaneous release)/(maximum release - spontaneous release)) x 100.
Cytokine ELISA
Primary MLCs were established in 24-well plates as described
above. For non-TCR transgenic experiments, supernatants were collected
on the fifth day of culture and analyzed for IFN- and IL-4 content
using commercial ELISA kits (PharMingen). For experiments using 2C TCR
transgenic T cells, blasts were harvested after 5 days in primary
culture and then restimulated by mixing 1 x
106 primed cells with 3 x
106 BALB/c stimulator cells in each well of a
24-well plate in a total of 2 ml EMEM. After 24 h of secondary
culture, supernatants were harvested and analyzed for IFN- and IL-4.
Recombinant mouse IFN- and IL-4 (PharMingen) were used as standards.
Color development was measured on a Multiskan PLUS spectrophotometer
(Titertek, Huntsville, AL).
Islet preparation and transplantation
C57BL/6 scid/scid, C57BL/6
IFN--/-, and BALB/c
IFN--/- mice were
rendered diabetic (two consecutive blood glucose readings >20 mM) with
a single i.v. injection of 140–180 mg/kg streptozotocin (Calbiochem,
La Jolla, CA). Four hundred fifty pancreatic islets, isolated from
BALB/c donors by collagenase (Sigma, St. Louis, MO) digestion and
Ficoll purification (38), were then implanted to the left
renal subcapsular space of the diabetic animals (39). To
assess allograft function, blood glucose values were measured every
other day with a precision blood glucose meter (MediSense, Bedford,
MA). Grafts placed into SCID animals were allowed to heal for a minimum
of 10 days before cellular reconstitution; these grafts were considered
established. In all cases, graft rejection was defined as three
consecutive blood glucose values >10 mM, which corresponds to three
SDs above the mean normal reading. Differences in graft survival were
analyzed by a Mann-Whitney U test. In all mice with grafts
functioning >60 days, removal of the graft-bearing kidney confirmed
graft-dependent control of blood glucose levels.
Adoptive transfer of CD8+ T cells
Primary MLCs were established in 75-cm2 flasks as described above. After 5 days in culture, cells were harvested and counted. Samples of these cells were analyzed by flow cytometry to confirm their CD8+ phenotype. For non-TCR transgenic experiments, 10 x 106 primed cells were then injected i.v. into C57BL/6 scid/scid animals bearing established BALB/c islet allografts. For 2C TCR transgenic experiments, cells were primed for 5 days as described above, and were then restimulated in vitro as follows. In 75-cm2 flasks, 10 x 106 primed 2C T cells, either treated or untreated, were mixed with 30 x 106 irradiated BALB/c splenocytes as stimulator cells. On the third day of culture, cells were harvested, counted, and again analyzed by flow cytometry to confirm CD8+ phenotype. A total of 10 x 106 restimulated 2C T cells were then injected i.v. into C57BL/6 scid/scid animals bearing established BALB/c islet allografts. Islet allograft function was monitored as described above.
Histologic examination of grafted tissues
At the time of rejection or after 60 days of graft function, graft-bearing kidneys were removed and fixed in 10% formal saline. Paraffin sections were stained with hematoxylin-eosin, and, in parallel sections, insulin granules were detected by immunoperoxidase staining. Tissue sections were examined to determine the degree of tissue damage and mononuclear cell infiltration of the graft.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-deficient
mice
Experiments to date have failed to demonstrate a requirement for
host-derived perforin or FasL in the rejection of islet allografts.
Therefore, we asked whether such rejection requires host production of
the inflammatory mediator IFN-, as suggested by correlative studies
in vivo (13). To do this, we used immunocompetent C57BL/6
wild-type and IFN--/-
mice as allograft recipients. These animals were rendered diabetic with
streptozotocin and then grafted with allogeneic BALB/c islets. Rapid
rejection was observed in both the wild-type and the
IFN--/- recipients,
indicating that IFN- is not required for rejection of allogeneic
islets by immunocompetent animals. Similar results were obtained when
the strain combination was reversed; i.e., BALB/c
IFN--/- mice were used
as recipients of C57BL/6 islets (Fig. 1
).
|
produced by donor-derived cells within the
graft might be sufficient to overcome the host deficiency and
contribute to rejection. However, when C57BL/6
IFN--/- islets were
transplanted into BALB/c
IFN--/- hosts, rapid
rejection was still observed (Fig. 1
), indicating that islet
donor-derived IFN- is not responsible for the rejection observed in
IFN--deficient hosts. Thus IFN-, like perforin and FasL, does not
appear to be generally required for the overall rejection response, a
result consistent with findings in other models (16, 32, 33). Primed CD8+ T cells are sufficient to cause islet allograft rejection
The results above indicate that the overall process of islet
allograft rejection can occur in the absence of host production of
IFN-, and previous studies have shown that this process is also
independent of perforin and Fas (23). Such results
prompted us to ask whether any of these classical effector pathways are
indeed used by CD8+ T cells, a subset of
lymphocytes that is known to play a major role in islet allograft
rejection. Our first objective was to confirm that primed
CD8+ T cells are indeed sufficient to cause islet
allograft rejection. Purified CD8+ T cells were
cultured in vitro with irradiated BALB/c (H-2d)
stimulator cells to ensure the generation of allospecific
CD8+ effector T cells. After 5 days of culture,
107 primed alloreactive
CD8+ T cells were transferred i.v. into
graft-bearing SCID recipients. To eliminate the contribution to the
rejection process of nonspecific inflammation associated with
transplantation surgery, we allowed all islet grafts to heal for at
least 10 days before proceeding with cellular reconstitution of the
SCID hosts. Thus we were able to examine the ability of a defined
population of cells, i.e., primed CD8+ effector T
cells, to reject established, quiescent islet allografts.
Importantly, all allografts survived indefinitely in SCID mice that
were left unreconstituted, illustrating that an adaptive immune
response is essential for graft rejection. However, reconstitution of
SCID mice with primed CD8+ T cells led to rapid,
efficient rejection of all grafts (Table I), confirming that
CD8+ effector T cells are independently capable
of bringing about rejection of islet allografts. At the time of
rejection, flow cytometric analysis of peripheral blood leukocytes from
reconstituted mice reconfirmed a very high degree of
CD8+ T cell purity (0.3 ± 0.2%
CD4+ and 1.3 ± 0.4%
B220+ contamination). Priming of the
CD8+ T cells was essential for efficient graft
rejection, because adoptive transfer of unprimed, freshly isolated
CD8+ T cells resulted in rejection in only two of
five recipients (Table I), consistent with previous findings
(40).
|
To determine the effector mechanisms used by
CD8+ T cells in islet allograft rejection, we
used four strains of mice as sources of CD8+ T
cells: 1) control C57BL/6 (H-2b) mice, 2)
B6-perforin-deficient
(pfp-/-) mice, 3)
B6-FasL-deficient (gld/gld) mice, and 4)
B6-IFN--deficient
(IFN--/-) mice. As
measured by [3H]thymidine incorporation on days
3, 4, and 5, the proliferative responses of CD8+
T cells from each mutant mouse strain were similar to control responses
(Fig. 2a). As expected,
CD8+ T cells from each strain produced similar
levels of IFN- in primary culture, with the exception of the
IFN--/-
CD8+ T cells, which produced no detectable
IFN- (Fig. 2b). However, when primed
CD8+ T cells from primary cultures were assayed
for in vitro cytotoxicity against target cells bearing
H-2d alloantigens, blasts derived from
pfp-/- mice were found to
have greatly reduced killing activity compared with blasts from the
other strains. In this relatively short (4 h) assay, no significant
impairment of killing activity was observed in the blasts from
gld/gld mice (Fig. 2c).
|
We next sought to assess the function of the mutant
CD8+ T cells in vivo. As before, islet allografts
survived indefinitely in unreconstituted SCID hosts, but reconstitution
with 107 wild-type primed
CD8+ T cells led to rejection of all grafts
within 23 days. Interestingly, a very similar result was obtained with
perforin-deficient CD8+ T cells. Although these
cells showed drastically impaired cytotoxicity in vitro, in vivo they
were able to reject islet allografts acutely, indicating that perforin
is not required for CD8+ T cell-mediated islet
allograft destruction (Fig. 3). Although
the majority of islet grafts (seven of eight) were eventually rejected
in animals reconstituted with primed gld/gld
CD8+ T cells, rejection in these mice was
significantly delayed relative to controls, suggesting that FasL may be
of some importance as a mediator of allograft immunity (Fig. 3).
However, the most striking finding was obtained when animals were
reconstituted with primed
IFN--/-
CD8+ T cells. In these animals, 75% of the islet
grafts continued to function beyond 60 days, the designated endpoint of
the study (Fig. 3). In mice with long-term graft function, surgical
removal of the graft-bearing kidney was always followed by a sharp rise
in blood glucose, demonstrating that the graft was responsible for the
maintenance of euglycemia (not shown). Thus, although
IFN--/-
CD8+ T cells display normal proliferative and
cytotoxic activity in vitro, they are severely impaired in their
ability to mediate allograft rejection in vivo.
|
-/- CD8+ T cells
trigger mononuclear cell infiltration of islet allografts
We next addressed whether
IFN--/-
CD8+ T cells were able to traffic to the sites of
the islet allografts, as failure to do so might explain the long-term
graft survival observed in animals reconstituted with these cells.
Islet grafts from both unreconstituted and reconstituted SCID animals
were harvested by nephrectomy of the graft-bearing kidney either at the
time of graft failure or after 60 days of graft survival. The grafts
were then examined histologically for evidence of lymphocytic
infiltration. In sections of long-term functioning allografts from
unreconstituted SCID hosts, normal islet architecture was apparent
without detectable leukocyte infiltration (Fig. 4A). However, in sections of
long-term (>60 days) functioning grafts from animals reconstituted
with IFN--/-
CD8+ T cells, distinct islets were surrounded by
a pronounced cellular infiltrate (Fig. 4B). In contrast,
only scar tissue and residual mononuclear cells remained at the sites
of rejected allografts from SCID animals reconstituted with wild-type
CD8+ T cells (Fig. 4C), with similar
results for pfp-/- or
gld/gld CD8+ T cells (not shown). Thus
the inability to produce IFN- does not appear to prevent primed
IFN--/-
CD8+ T cells from homing to and persisting at the
site of the allograft.
|
production inhibits islet
allograft rejection by 2C TCR-transgenic CD8+ T cells
The experiments described above relied on the use of knockout mice
to demonstrate the importance of IFN- for CD8+
T cell-mediated islet allograft rejection. We sought to extend our
results through a parallel study in which the IFN- gene remained
intact while IFN- production by alloreactive
CD8+ T cells was inactivated by incubation with
anti-IFN- and mrIL-4 during the priming phase. To increase the
yield of allospecific T cells through this priming step, we chose to
use lymphocytes obtained from C57BL/6 mice carrying the 2C TCR
transgene, because the vast majority of the T cells from these mice are
CD8+ T cells specific for a BALB/c alloantigen
(Ld)) (35). Preliminary studies
revealed that 2C mice vigorously reject BALB/c islet allografts,
demonstrating the relevance of the 2C specificity to allogeneic islet
tissue (data not shown).
To examine the role of IFN- in graft destruction by 2C
CD8+ T cells, spleen and lymph node cells from 2C
mice were isolated and cultured in vitro with irradiated BALB/c
splenocytes for 5 days in the presence or absence of anti-IFN-
and mrIL-4. Treatment in this manner was previously shown to cause
naive CD8+ T cells to differentiate toward an
IL-4-producing "Tc2" phenotype, rather than the default
IFN--producing "Tc1" phenotype (11, 41, 42).
Following priming, the 2C CD8+ T cells were
washed, restimulated for 24 h with alloantigen, and then assayed
for cytokine production, proliferation, and cytotoxicity in vitro. Upon
restimulation, the untreated cells produced very high levels of
IFN-, while the cells primed in the presence of anti-IFN- and
mrIL-4 failed to produce any detectable IFN- (Fig. 5a). In contrast to earlier
reports, the inactivation of IFN- production observed here was not
accompanied by expression of IL-4, indicating that the treated cells
had ceased IFN- production but had not fully switched to a Tc2
phenotype. Nonetheless, the treated cells proliferated normally in
secondary responses to alloantigenic stimulation (Fig. 5b)
and exhibited only a slight decrease in cytotoxicity against
Ld-bearing targets (Fig. 5c)
|
-producing) vs treated
(IFN--nonproducing) 2C T cells in vivo, primed cells were harvested
after 5 days of primary culture, restimulated for three additional
days, and then adoptively transferred into C57BL/6 scid/scid
animals bearing BALB/c islet grafts. Flow cytometry confirmed that the
adoptively transferred lymphocyte blasts were <0.5%
CD4+, 94.5 ± 4%
CD8+, and 98.0 ± 1% 2C
clonotype+. Untreated 2C
CD8+ cells caused very rapid rejection of islet
allografts in all recipients. In contrast, the treated 2C
CD8+ T cells, which failed to produce detectable
IFN- upon restimulation in vitro, showed impaired capacity for islet
allograft destruction, with three of six grafts functioning beyond 60
days and the remaining grafts being rejected significantly later than
controls. No graft rejection was observed in unreconstituted animals
(Fig. 6). These results further support
the conclusion that IFN- production is essential for efficient
CD8+ T cell-mediated islet allograft
rejection.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
upon restimulation. A more likely possibility is that graft-destructive
CD8+ T cells are partially reliant on the
Fas/FasL apoptotic pathway, so that in the absence of this pathway,
rejection is mediated by alternate means and proceeds more slowly.
Indeed, perforin-dependent cytotoxicity may be such an alternate
pathway. The question of whether perforin-dependent and FasL-dependent
cytotoxicity represent two independent and compensatory pathways for
graft-destructive CD8+ T cells is currently
under examination, although it should be noted that
perforin-deficient mice do reject Fas-deficient islet allografts
normally (23).
The most striking finding of the present study is that primed
CD8+ T cells from IFN- knockout mice, despite
their normal proliferative and cytotoxic behavior in vitro, were
significantly impaired in their ability to reject islet allografts in
SCID hosts. This was not due to a failure of mononuclear cells to
migrate to and persist at the graft site, a possibility suggested by a
recent study of skin allograft rejection (44), since
transfer of IFN--/-
CD8+ T cells induced a profound cellular
infiltrate in established islet allografts. Similar results were
obtained through the use of 2C TCR transgenic
CD8+ T cells primed in the presence of
anti-IFN- and recombinant IL-4. Our intent was to mimic in vitro
the putative process of "immune deviation," whereby the presence of
regulatory cytokines such as IL-4 during activation inhibits production
of IFN- by T cells and causes these cells to acquire an
IL-4-producing and potentially nondestructive phenotype
(45). In our hands, priming in the presence of
anti-IFN- and IL-4 did lead to the inactivation of IFN-
production by the 2C T cells, but did not cause a corresponding
initiation of IL-4 production. In these respects, our treated 2C
CD8+ T cells were similar to a recently described
population of differentiated Tc1 CD8+ T cells
that, following treatment with IL-4, lost the ability to produce
IFN-, TNF, and IL-10 but did not switch to IL-4 production
(46). In that study, the "cytokine-negative" cells
displayed normal perforin- and FasL-mediated cytotoxicity, although
they did exhibit a marked defect in a long-term tumor cell killing
assay. Thus one possible explanation for our results is that the 2C
cells were already predisposed to the Tc1 phenotype at the time of
their isolation from the donor mice. However, this seems unlikely as we
were also unable to elicit IL-4 from normal (i.e., non-TCR transgenic)
purified alloreactive C57BL/6 CD8+ T cells, while
purified alloreactive CD4+ T cells treated in the
same manner switched readily to an IL-4-producing phenotype (not
shown). Therefore, we surmised that CD8+ T cells,
as compared with their CD4+ counterparts, exhibit
less plasticity during the priming phase and are biased to retain a
default Tc1 phenotype when stimulated with alloantigen. In any case,
when transferred into graft-bearing SCID recipients, these IFN-- and
IL-4-negative CD8+ T cells exhibited a
significant defect in their ability to mediate graft rejection as
compared with untreated, IFN--producing control cells. Thus,
regardless of whether IFN- production by alloreactive
CD8+ T cells was blocked through targeted
disruption of the IFN- gene or through cytokine deviation in vitro,
the end result was a marked abrogation of graft-destructiveness in
vivo.
The exact contribution(s) of IFN- to islet rejection remains
unclear. IFN- is known to up-regulate the expression of MHC proteins
and enhance Ag presentation in many tissues (12).
Therefore, a simple explanation for our findings is that IFN- serves
to increase MHC class I alloantigen expression on islet cells above a
threshold density, so that the islets become susceptible targets for
allospecific CD8+ CTLs (31).
However, if this is the full extent of the contribution of IFN- to
the rejection process, then the question of how
CD8+ T cells actually kill the islets remains
unanswered. As mentioned earlier, it is possible that for
CD8+ T cells, perforin and FasL operate
redundantly, i.e., killing through one pathway compensates for an
experimental defect in the other. In such a scenario, the role of
IFN- would be to up-regulate islet cell expression of Fas, as seen
in other cell types (47, 48, 49, 50, 51), as well as expression of MHC
class I, thereby enhancing the activity of both killing pathways.
Alternatively or perhaps additionally, IFN- may act by inducing the
expression of apoptosis-associated proteins other than Fas/FasL, such
as TNF, TRAIL, and their respective receptors. In human monocytes,
IFN- promotes killing of intracellular pathogens by inducing
expression of TNF, which in turn triggers apoptosis by binding to its
receptors, TNFR1 and TNFR2 (52). Similarly, IFN- has
been shown to up-regulate the expression of TRAIL, while
down-regulating the level of the survival factor NF-
B, to induce
killing of virus-infected human fibroblasts (53). Whether
any of these pathways also contribute to CD8+ T
cell-mediated islet allograft rejection is the subject of current
studies in our laboratory.
There is also evidence to suggest that IFN- may directly contribute
to islet damage. For example, IFN- induces the expression of several
components of the respiratory burst machinery in many cell types
(12). Consequently, IFN- can, in conjunction with other
cytokines such as TNF and IL-1ß, cause measurable damage to
pancreatic islets by inducing the formation of NO and oxygen free
radicals in ß cells (reviewed in Ref. 54). IFN- may
also regulate some functional apoptotic events within islets, as
suggested by the report that IFN- can directly induce apoptosis of
immature mouse monocytes (55). However, our previous
results indicate that islets can be cultured in very high
concentrations of IFN- without compromising subsequent function in
vivo (56). Thus the toxicity of IFN- alone as a direct
cytopathic agent is unlikely.
Of course, the relevant effects of IFN- in islet rejection need not
be limited to CD8+ T cells and their islet cell
targets. IFN- exerts profound effects on endothelial cell
permeability, probably through the induction of chemokines (such as
IFN--inducible protein-10, macrophage inflammatory protein-1ß,
RANTES, and monocyte chemoattractant protein-1), to promote the
formation of a nonspecific monocytic infiltrate (12).
IFN- is also an extremely potent activator of macrophages and NK
cells, whose function is not impaired in the SCID reconstitution system
(12). It is possible that IFN- triggers the migration
and activation of such inflammatory cells that in turn promote
nonspecific tissue injury. In support of this hypothesis, it was
recently shown that T cell-derived IFN- is essential for the
rejection of tumor allografts, and that this requirement is due to the
ability of IFN- to activate cytotoxic macrophages at the graft site
(57). Similarly, islet-specific autoimmune pathogenesis by
CD8+ T cells in a transgenic model has been shown
to be macrophage dependent (58).
In our studies, we observed a requirement for IFN- only when we
examined CD8+ effector T cells in isolation.
However, we obtained a very different result when we used
immunocompetent IFN--deficient mice as graft recipients. These mice,
which carry a normal complement of lymphoid cells, promptly rejected
allogeneic islet grafts, even when the islets were obtained from
IFN--/- donors. Such
results are complementary to the finding that IFN-
receptor-deficient hosts acutely reject islet allografts
(59). Clearly, then, there must be an IFN--independent
mechanism for islet allograft rejection. The existence of such a
mechanism is also indicated by the results of the heart and skin
allograft studies mentioned previously. Why are IFN--deficient
activated CD8+ T cells apparently unable to use
this IFN--independent pathway to reject established islet
allografts? There are several conceivable answers to this question.
First, there is the possibility that in the absence of
CD4+ T cells, IFN- contributes to
differentiation and/or expansion of effector CD8+
T cells. We have attempted to control for this by demonstrating normal
proliferative and cytotoxic behavior in our primed
IFN--/-
CD8+ T cell populations. In fact, it is generally
believed that IFN- has an anti-proliferative effect on T cells,
rather than a growth-promoting role (34, 60). Nonetheless,
it may be the case that during priming, IFN- impacts properties of
CD8+ T cells not reflected by standard MLR or CTL
assays, especially when those cells differentiate in the absence of
CD4+ T cells as described in our study. This
important caveat must be considered in the interpretation of our
results.
A second possibility is that in immunocompetent mice, cells other
than CD8+ T cells, such as
CD4+ T cells or B cells, compensate for the
absence of IFN- in the effector phase of the response. This
compensation could occur either through the production of mediators
that mimic the action of IFN- or through a discreet
CD8+ T cell-independent killing mechanism. A
third potential explanation for our data lies in the fact that in our
SCID reconstitution system, grafts are established in the absence of
adaptive immunity, whereas in the immunocompetent
IFN--/- mice, a full
complement of B and T cells is present at the time of engraftment.
Perhaps, in the latter case, a transient burst of inflammatory
mediators other than IFN- (as a nonspecific consequence of islet
transplantation surgery) is sufficient to bypass the need for IFN-
and initiate T cell-dependent destruction of the graft. However, if
such postsurgical inflammation has fully resolved before reconstitution
of a SCID host, then CD8+ T
cell-derived IFN- is required to increase expression of target
structures and to recruit nonspecific effector cells to the quiescent
graft site, thereby recreating the conditions in which graft rejection
can take place. We are currently examining these alternate
possibilities.
In summary, we have made use of a SCID reconstitution system to
characterize the relative contributions of CD8+ T
cell-derived perforin, FasL, and IFN- to the process of islet
allograft rejection. We have shown that neither perforin nor FasL is
independently required for this process, because
CD8+ T cell-mediated allograft rejection proceeds
normally in the absence of perforin and is only slightly impaired in
the absence FasL. We have also shown that although islet allograft
rejection can occur in immunocompetent mice lacking IFN-, production
of this cytokine nonetheless appears to be a rate-limiting step in the
destruction of established allografts by primed
CD8+ T cells. In further studies, we hope to
clarify the conditional roles for IFN- in islet allograft immunity
and tolerance.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ronald G. Gill, Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center, Campus Box B-140, 4200 East Ninth Avenue, Denver, CO 80262.
3 Abbreviations used in this paper: FasL, Fas ligand; TRAIL, TNF-related apoptosis-inducing ligand; EMEM, Eagle’s MEM; mrIL-4, recombinant mouse IL-4.
Received for publication February 4, 2000. Accepted for publication April 14, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Annu. Rev. Immunol. 15:749.[Medline]
monoclonal antibody. J. Immunol. 144:4648.[Abstract]
, and CTLA4. J. Clin. Invest. 101:2446.[Medline]
is critical for long-term allograft survival induced by blocking the CD28 and CD40 ligand T cell costimulation pathways. J. Immunol. 160:2059. is essential for destruction of ß cells and development of insulin-dependent diabetes mellitus. J. Exp. Med. 185:531.-interferon delays but does not prevent diabetes in NOD mice. Diabetes 45:812.[Abstract]
impacts at multiple points during the progression of autoimmune diabetes. Proc. Natl. Acad. Sci. USA 94:13844. is necessary for initiating the acute rejection of major histocompatibility complex class II-disparate skin allografts. Transplantation 67:1362.[Medline]
. Transplantation 62:1908.[Medline]
: implications for long-term allograft acceptance. Transplantation 68:124.[Medline]
and IL-4 regulate the growth and differentiation of CD8+ T cells into subpopulations with distinct cytokine profiles. J. Immunol. 155:2928.[Abstract]
-induced chemokine Mig. J. Immunol. 163:4878. and tumor necrosis factor and potentiates cytokine-mediated hematopoietic suppression in vitro. Blood 85:3183.-/- mice, but not in TNF--/- mice: role for IFN- in activating apoptosis of hepatocytes. J. Immunol. 159:1418.[Abstract]
(TNF-) and/or interferon- (IFN-), resulting from down-modulation of Bcl-2 expression. Clin. Exp. Immunol. 111:211.[Medline]
and IFN- render microglia sensitive to Fas ligand-induced apoptosis by induction of Fas expression and down-regulation of Bcl-2 and Bcl-xL. Eur. J. Immunol. 28:4398.[Medline]
and tumor necrosis factor- and potentiates apoptotic death. Exp. Hematol. 27:433.[Medline]
-mediated control of Coxiella burnetii survival in monocytes: the role of cell apoptosis and TNF. J. Immunol. 162:2259. mediates a novel antiviral activity through dynamic modulation of TRAIL and TRAIL receptor expression. J. Immunol. 163:920.. J. Exp. Med. 181:127. released from T cells is essential for induction of effector macrophages in the rejection site. J. Immunol. 160:6012. receptor signaling is not required in the effector phase of the alloimmune response. Transplantation 65:1649.[Medline]
in immune regulation. I. IFN- inhibits the proliferation of Th2 but not Th1 murine helper T lymphocyte clones. J. Immunol. 140:4245.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  S. M. Coley, M. L. Ford, S. C. Hanna, M. E. Wagener, A. D. Kirk, and C. P. Larsen IFN-{gamma} Dictates Allograft Fate via Opposing Effects on the Graft and on Recipient CD8 T Cell Responses J. Immunol., January 1, 2009; 182(1): 225 - 233. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Yolcu, X. Gu, C. Lacelle, H. Zhao, L. Bandura-Morgan, N. Askenasy, and H. Shirwan Induction of Tolerance to Cardiac Allografts Using Donor Splenocytes Engineered to Display on Their Surface an Exogenous Fas Ligand Protein J. Immunol., July 15, 2008; 181(2): 931 - 939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Plesner, P. Liston, R. Tan, R. G. Korneluk, and C. B. Verchere The X-Linked Inhibitor of Apoptosis Protein Enhances Survival of Murine Islet Allografts Diabetes, September 1, 2005; 54(9): 2533 - 2540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-i. Seino, T. Yamauchi, K. Shikata, S. Kobayashi, M. Nagai, M. Taniguchi, and K. Fukao Prevention of acute and chronic allograft rejection by a novel retinoic acid receptor-{alpha}-selective agonist Int. Immunol., May 1, 2004; 16(5): 665 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Abdi, T. K. Means, T. Ito, R. N. Smith, N. Najafian, M. Jurewicz, V. Tchipachvili, I. Charo, H. Auchincloss Jr., M. H. Sayegh, et al. Differential Role of CCR2 in Islet and Heart Allograft Rejection: Tissue Specificity of Chemokine/Chemokine Receptor Function In Vivo J. Immunol., January 15, 2004; 172(2): 767 - 775. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Makhlouf, A. Yamada, T. Ito, R. Abdi, M. J. I. Ansari, C. Q. Khuong, H. J. Winn, H. Auchincloss Jr., and M. H. Sayegh Allorecognition and Effector Pathways of Islet Allograft Rejection in Normal versus Nonobese Diabetic Mice J. Am. Soc. Nephrol., August 1, 2003; 14(8): 2168 - 2175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Nicolls, M. Coulombe, J. Beilke, H. C. Gelhaus, and R. G. Gill CD4-Dependent Generation of Dominant Transplantation Tolerance Induced by Simultaneous Perturbation of CD154 and LFA-1 Pathways J. Immunol., November 1, 2002; 169(9): 4831 - 4839. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Feng, D. Wang, R. Yuan, C. M. Parker, D. L. Farber, and G. A. Hadley CD103 Expression Is Required for Destruction of Pancreatic Islet Allografts by CD8+ T Cells J. Exp. Med., October 7, 2002; 196(7): 877 - 886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Abdi, R. N. Smith, L. Makhlouf, N. Najafian, A. D. Luster, H. Auchincloss Jr., and M. H. Sayegh The Role of CC Chemokine Receptor 5 (CCR5) in Islet Allograft Rejection Diabetes, August 1, 2002; 51(8): 2489 - 2495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. F. Halloran, L. W. Miller, J. Urmson, V. Ramassar, L.-F. Zhu, N. M. Kneteman, K. Solez, and M. Afrouzian IFN-{{gamma}} Alters the Pathology of Graft Rejection: Protection from Early Necrosis J. Immunol., June 15, 2001; 166(12): 7072 - 7081. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
); BALB/c islets into C57BL/6 IFN-
); C57BL/6 islets into BALB/c hosts
(n = 6; •); C57BL/6 islets into BALB/c
IFN-
);
C57BL/6 IFN-
).
) are also shown. Error bars represent SD of results
from quadruplicate wells. b, IFN-
