| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
University of Minnesota Cancer Center and Department of Pediatrics, Division of Bone Marrow Transplantation, Minneapolis, MN 55455;
Division of Clinical Research, Fred Hutchinson Cancer Research Center Department of Medicine, Seattle, WA 98195;
Department of Microbiology, Dartmouth Medical College, Hanover, NH 03756;
§
Department of Ophthalmology, Louisiana State University Medical Center, New Orleans, LA 70112; and
¶
Science Applications International-Frederick and the Laboratory of Leukocyte Biology, National Cancer Institute-Frederick Cancer Research and Development Center, Frederick, MD 21702
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
were more susceptible to DLI-GVHD, whereas those
deficient in IL-12 or p55 TNFRI were not. Collectively, these data
indicate that host T cells, which are capable of generating antidonor
CTL effector cells, are responsible for the impaired ability of DLI to
induce GVHD. These same mechanisms may limit the efficacy of DLI in
cancer therapy under some conditions. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The explanation for the lower risk of GVHD of DLI as compared with
donor lymphocyte infusion at the time of BMT remains unknown. Included
among the possible explanations as to why DLIs are more well-tolerated
include the lower levels of proinflammatory cytokines present as the
time from conditioning therapy increases (13), lower
numbers of host APCs that are gradually eliminated by conditioning
(14), and the existence of regulatory cells (12, 15, 16, 17, 18, 19, 20, 21, 22, 23) that develop later post-BMT and might assist the host in
resisting DLI-GVHD. With respect to the latter mechanism, Johnson,
Truitt, and colleagues have identified a donor bone marrow (BM)-derived
population that suppresses DLI-GVHD (12). These
investigators have conclusively shown that two subpopulations of
thymus-derived cells exist that are
Thy1+TCR
ß+CD8-
and either are CD4+ or
CD4- and are capable of suppressing
GVHD.
In this study, we considered the possibility that residual host BM-derived T cells also contribute to the suppression of DLI-GVHD for several reasons: 1) DLI given to patients that relapse or to patients conditioned with nonmyeloablative therapy are exposed to an environment in which host T cells are readily detectable and may be more abundant than donor BM-derived T cells present at the time of infusion, and 2) studies by several investigators have shown that the tolerance observed in mixed donor-host chimeras is not readily broken by DLI cells (12, 16, 19, 21, 22), indicating that there is a resistance mechanism operative in the recipient at the time of DLI. Our studies have identified the host cell surface Ags required for optimal resistance to DLI-GVHD and have further identified the effector mechanisms responsible for this resistance in a mouse model of DLI-induced GVHD. These data have implications for humans receiving DLI as therapy for relapse post-BMT or to increase donor chimerism levels.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 (termed B6) mice were purchased from the National
Institutes of Health (Bethesda, MD). B10.BR/SgSnJ
(H2k), B6.PL-Thy1a/Cy
(termed B6-Thy1.1), C57BL/6-CD28tm1 Mak, CD28
deletional mutant (termed CD28-/-),
C57BL/6-CD4tm1 Mak, CD4 deletional mutant (termed
CD4-/-), C57BL/6-CD8tm1
Mak, CD8 deletional mutant (termed
CD8-/-), C57BL/6-tnfrsf1atm1
Mak, TNF receptor type I (p55) (termed
TNFR-p55-/-),
C57BL/6-ifngtm1Ts, IFN- mutant (termed
IFN--/-),
B6Smn.C3H-Faslgld, functional Fas ligand
(FasL)-defective (termed gld), and
C57BL/6-Pfptm1Sdz10, perforin deletional mutant
(termed perforin-/-) mice were purchased from
The Jackson Laboratory (Bar Harbor, ME). Perforin-deficient, gld
homozygous mice were established from founder males provided to one of
the authors (P.J.M.), bred, and screened as previously described
(24), and breeder pairs were maintained at the University
of Minnesota. C57BL/6-il12btm1Jm, IL-12
ß-chain-/- (termed
p40-/-) were bred at Science Applications
International, National Cancer Institute-Frederick Cancer Research and
Development Center (Frederick, MD) and maintained at the University of
Minnesota. C57BL/6 4-1BB receptor deletional mutant (termed
4-1BB-/-) mice were generated as described
(25), backcrossed nine generations, and bred at the
University of Minnesota. B10.BR-CD45.1 mice were established at the
University of Minnesota by crossing
B6.SJL-Ly5aPtprcaPep3b
(B6-Ly5.2, CD45.1) mice with B10.BR/SgSnJ mice and then screening
F2 intercrosses for expression of H2 and CD45
alleles. Homozygotes expressing H2k and CD45.1
(termed B10.BR-CD45.1 mice) were maintained at the University of
Minnesota. B10.BR-Thy1.1 mice were established at the University of
Minnesota by crossing B6-Thy1.1 mice with B10.BR/SgSnJ mice and then
screening F2 intercrosses for expression of H2
and Thy1 alleles. Homozygotes expressing H2k and
Thy1.1 (termed B10.BR-Thy1.1 mice) were maintained at the University of
Minnesota. Mice were bred and housed in a specific pathogen-free
facility in microisolator cages. Donors and recipients were used at
8–10 wk of age.
GVHD
To determine whether GVHD potential early post-BMT differs from that generated later post-BMT, B10.BR mice were lethally irradiated with 8 Gy total body irradiation (TBI) by x-ray (39 cGy/min) on day -1 followed on day 0 by the i.v. infusion of B6 BM T cell-depleted (TCD) by anti-Thy1.2 (clone 30-H-12; provided by Dr. David Sachs, Massachussetts General Hospital, Cambridge, MA) + baby rabbit complement (Nieffenegger, Woodland, CA) treatment. For early post-BMT studies, donor splenocytes (25 x 106) were infused on day 0. To assess GVHD potential of the same splenocyte inoculum later post-BMT studies, cohorts of mice that had been transplanted 3 wk earlier were given donor splenocytes on day 21 post-BMT.
DLI-mediated GVHD
Our DLI procedure has been described in detail (26). For most experiments, unless otherwise indicated, B6, B6-Thy1.1, or B6 deletional mutant mice undergoing BMT for GVH assessment were conditioned with 800 cGy irradiation from an x-ray source on day -1 and infused with TCD BM from B10.BR or B10.BR-Thy1.1 donors on day 0. Donor B10.BR or B10.BR-CD45.2 or B10.BR-CD45.1 splenocytes were administered i.v. at a dose of either 25 or 50 x 106 on day 21 post-BMT. In one experiment, mice were given either 8 or 9 Gy TBI and reconstituted with TCD BM on day 0 followed by the infusion of 38 x 106 B10.BR-CD45.2 splenocytes on day 21 post-BMT.
Adult thymectomy
Our thymectomy procedure has been previously described (27). In brief, recipient mice were anesthetized using sodium pentobarbital. A 0.5-cm ventral midline incision was made from the suprasternal notch to the second or third rib. The two thin white lobes of the thymus overlying the heart were gently removed by a combination of aspiration and dissection. The incision was closed, and the animal was allowed to recover under a warm lamp. Histological inspection of designated mice for remaining thymic remnants has confirmed the adequacy of this procedure for thymic removal.
Thoracic duct cannulation
For thoracic duct lymphocyte (TDL) isolation, cannulae were inserted, as previously described, in the thoracic duct of recipients at the time of peak proliferation, which is 6 days after donor splenocyte infusion (28).
mAb administration
The anti-Thy1.1 mAb-producing hybridoma, 1A14 (mouse IgG2a), (29) was provided by Dr. Irv Bernstein (Fred Hutchinson Cancer Research Center, Seattle, WA). 1A14 was propagated as ascites, purified by ammonium sulfate precipitation, and then dialyzed. Anti-Thy1.1 mAb was administered at a weekly dose of 400 µg i.p. beginning on day 7 post-BMT and continuing for 3 wk post-BMT.
Flow cytometry
TDL effluent and donor splenocyte phenotype was measured using
mAb directed toward CD4 or CD8, CD19, and Mac1. When indicated, TDLs
were assessed for evidence of activation by forward- and side-scatter
profiles and the coexpression of CD4 and CD8 with activation Ags
including CD25, L-selectin (CD62L), CD54, CD44, and CD11a. All
fluorochrome-labeled mAb, unless indicated, were obtained from
PharMingen (San Diego, CA). The origin of TDL was determined by flow
cytometry using anti-H2 (donor: H2k:11-4.1,
mouse IgG2a; host H2b: EH-144, mouse IgG) mAb. To
determine whether donor cells were derived from the BM or spleen
source, TDLs were analyzed with CD45.2 (clone 104-2, rat IgG2a) and
CD45.1 (clone A20-1.7, rat IgG2a), both provided by Dr. U.
Hammerling (Memorial Sloan-Kettering Cancer Research Center, New York,
NY). Cells were washed three times and resuspended for analysis by two-
or three-color flow cytometry using FITC-, PE-, or biotin (along with
streptavidin-peridinin chlorophyll protein)-conjugated mAb purchased
from PharMingen or Becton Dickinson (Mountain View, CA). Irrelevant mAb
control values were subtracted from values obtained with relevant mAbs.
All results were obtained using a FACScalibur (Becton Dickinson, San
Jose, CA). Forward- and side-scatter settings were gated to exclude
debris. A total of 10,000 gated cells were analyzed for each
determination.
Statistical analyses
Group comparisons of continuous data were made by Student’s
t test. Survival data were analyzed by lifetable methods
using the Mantel-Peto-Cox summary of 
2
(30). Actuarial survival and relapse rates are shown.
Probability (p) values <0.05 were considered
significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In previous studies, we observed that donor splenocytes given on
the day of BMT were more potent inducers of GVHD lethality than those
given 3 wk post-BMT (26). To quantify the degree of
expansion under these two settings, B6 recipients were lethally
irradiated (8 Gy TBI), reconstituted with B10.BR-CD45.1 TCD BM on day
0, and then cohorts of mice, as indicated, were given the same inoculum
of B10.BR-CD45.2 splenocytes (25 x
106/recipient containing a total of 4.5 x
106 CD4+ T cells and
3.3 x 106 CD8+ T
cells) on either day 0 or 21 post-BMT (Table I
). Six days after splenocyte infusion,
the time of peak expansion in vivo, 5–6 mice per group were
cannulated, TDL were collected from individual mice and then pooled for
analysis, and the number of T cells of donor BM, donor spleen, and host
origin were enumerated. Non-BMT controls were cannulated for
comparison. There was a total of 60 x 106
donor spleen-derived CD4+ T cells collected per
mouse over a 24-h period when recipients of supplemental splenocytes
were cannulated on day 6 post-BMT, representing a >13-fold expansion
in mature donor CD4+ T cells produced over input
cell number at this time. A total of 82 x
106 CD8+ T cells derived
from donor spleen were produced per day, representing a >25-fold
increase over input cell number. In marked contrast, the same spleen
cell inocula administered to a cohort of mice transplanted 3 wk earlier
and cannulated on day 27 post-BMT (6 days after DLI) showed that donor
spleen-derived CD4+ T cells expanded 
1.6-fold
(total 7 x 106 per recipient) and donor
spleen-derived CD8+ T cells expanded 1.5-fold
(total 5 x 106 per recipient) as compared
with input cell number. Based on the expression of a panel of
activation Ags (CD11a, CD25, CD44, and CD54) and consistent with the
differences in expansile properties, donor spleen-derived T cells were
observed to have a higher degree of activation when collected on day 6
as compared with day 27 post-BMT (data not shown). Thus, mature
CD4+ T cells expanded >8-fold more and mature
CD8+ T cells expanded >17-fold more when infused
on day 0 rather than day 21 post-BMT.
|
In other experiments we have observed that host T cells are not only present in the thoracic duct lymphatics but also can be detected in the peripheral blood of recipients at the time of DLI administration (data not shown). Specifically, lethally irradiated (8 Gy TBI) B6 recipients of B10.BR TCD BM (n = 8) had more host than donor CD4+ T cells (5 ± 1.2 vs 1.8 ± 0.6%, respectively; p = 0.00004) and CD8+ T cells (1.6 ± 0.4 vs 0.6 ± 0.1%, respectively; p = 0.0002). Splenic flow cytometric analysis on day 28 also reveals a persistent host T cell component in recipients that is comparable to the donor BM-derived T cell component. Collectively, these data indicate that a major difference in the recipient environment is greater numbers of donor and host BM-derived T cells that coexist later post-BMT and that this type of environment in the recipient is associated with a diminution of GVHD risk following donor lymphocyte infusion.
Elimination of donor BM-derived and a reduction in the number of host BM-derived T cells present at the time of DLI is associated with an increased severity of DLI-GVHD
To ensure a prolonged time period in which host BM-derived T cells
would not be able to be replenished from the thymus, experiments were
performed in which a cohort of mice was adult thymectomized (ATx).
Euthymic B6-Thy1.1 recipients were given 8 Gy TBI, reconstituted with
TCD B10.BR-Thy1.2 BM, and then given B10.BR-Thy1.2 splenocytes or no
DLI cells on day 21 post-BMT. A cohort of recipients were thymectomized
>1 month before BMT. These latter recipients would not be able to
produce new thymus-derived donor T cells and would have limited numbers
of host T cells that would remain after TBI and BMT. ATx recipients had
severe weight loss, in contrast to the more modest weight loss seen in
euthymic DLI-treated recipients (Fig. 1A). ATx recipients given DLI
had a lower survival rate than euthymic recipients given DLI (0 vs 80%
survival; p < 0.03) (Fig. 1B).
|
The presence of both host CD4+ and CD8+ T cells and the coexpression of CD28 on host T cells are required for optimal DLI-GVHD resistance
Because of the increasing use of DLI administered to recipients with a substantial number of host T cells, we chose to focus on analyzing the effects of host T cells in the GVH resistance of DLI-treated recipients. To avoid effects of ATx on donor BM-derived T cell-mediated DLI-GVHD resistance while ensuring host T cell functional incapacitation or depletion, euthymic recipient mice with genetic deficiencies in molecules that are known to affect T cell expansion or function were examined. Lethally irradiated (8 Gy TBI) B6, B6-Thy1.1, or B6 homologous deletional mutant recipients were reconstituted with B10.BR TCD BM (20 x 106) and given either no supplemental DLI or B10.BR DLI on day 21 post-BMT. A cohort of B6-Thy1.1 recipients was given DLI was also given anti-Thy1.1 mAb to selectively deplete host T cells.
Removal of both host CD4+ and
CD8+ T cell subpopulations with anti-Thy1
allelic mAb resulted in a 20% survival rate, which was significantly
(p = 0.0021) lower than the 90% survival rate
of controls (Fig. 2). These data indicate
that Thy1+ host T cells were important regulators
of the resistance of the recipient to DLI-GVHD. Recipients incapable of
producing host CD8+ T cells had a 90% long-term
survival rate and those incapable of producing host
CD4+ T cells had a 70% long-term survival rate.
Although these data suggest that either host CD4+
or CD8+ T cells are required for optimal
resistance to DLI-GVH, Thy1 is expressed on
CD4-CD8- T/NK cells;
therefore, the lower survival observed in anti-Thy1 allelic
mAb-treated recipients could be due to resistance mediated by
CD4-CD8- cells (17, 31, 32, 33, 34).
|
).
CD28-/- recipients had only a 40% long-term
survival rate (p = 0.015 vs wild-type
controls), which was comparable (p = 0.28) to
anti-Thy1.1 mAb-treated recipients. Consistent with the survival
data, CD28-/- recipients had mean weights that
were at or below those recorded for anti-Thy1.1 mAb-treated
recipients (data not shown). Collectively, these data indicate that
host T cells require CD28 Ag expression to resist DLI-GVHD. Recipients deficient in the generation of functional host antidonor CTLs due to the deletion of perforin and FasL or 4-1BB but neither TNFRI (p55 TNFR) nor CD40 ligand (CD40L) are highly susceptible to DLI-GVH
To determine which pathways might be responsible for host T cell
suppression of DLI-GVHD, recipients were selected with deficiencies in
one or more molecules known to be critical for supporting CTL
generation or function (Fig. 3). These
data were obtained from the same experiment as depicted in Fig. 2.
Perforin-deficient/gld-homozygous recipients that lack the two dominant
pathways for CTL effector mechanisms, perforin and FasL, were found to
be highly susceptible to DLI-GVHD with only 10% of recipients
surviving long term (p = 0.0008 vs wild-type
controls). At 2 mo post-BMT, mean body weights showed a 27% weight
loss as compared with pre-BMT body weights (data not shown). In
contrast, wild-type control had lost only 13% of pre-BMT mean body
weight at 2 mo post-BMT. As was the case for the other groups,
perforin-deficient/gld-homozygous recipients given no DLI had a 100%
long-term survival rate, indicating that the low survival in
perforin-deficient/gld-homozygous recipients was due to DLI cells and
not the transplant procedure (data not shown).
|
). 4-1BB-/-
recipients had a lower survival rate than wild-type controls
(p = 0.00048), which was similar to that
observed with perforin-deficient/gld-homozygous recipients
(p = 0.20). As observed with
perforin-deficient/gld-homozygous recipients, mean body weights were
27% lower than pre-BMT weights at 2 mo post-BMT (data not shown).
Previously, we have shown that CD40L binding to CD40 promotes GVHD by
donor splenocytes infused into lethally irradiated (on day 0) post-BMT,
which was associated with increased donor antihost CTL generation
(39). However, we could not identify a critical role for
this pathway in DLI-GVH resistance because 90% of
CD40L-/- mice survived long term (Fig. 3), and
mean weights at 2 mo post-BMT averaged only 14% lower than pre-BMT
body weights (data not shown). Recent evidence indicates that TNFRI
plays a critical role in in vivo donor T cell alloreactivity
(40). However, we could not identify a role of TNFRI (p55
TNFR) expression on host T cells in DLI-GVH resistance.
TNFRI-/- had an 80% long-term survival rate,
which was not significantly different (p =
0.27) from wild-type controls (Fig. 3).
TNFRI-/- recipients had mean weights of only
17% lower than pre-BMT body weights (data not shown). Cumulatively,
these data indicate that both the perforin-deficient/gld-homozygous and
4-1BB pathways but neither the CD40L nor TNFRI (p55) pathways were
critical regulators of host T cell-mediated DLI-GVH resistance.
Having established that perforin-deficient/gld-homozygous recipients
were especially poor in suppressing DLI-GVH, we sought to determine
whether a deficiency of either the perforin or FasL pathway alone was
sufficient to reduce DLI-GVHD. As controls for DLI-GVHD, wild-type and
CD28-/- B6 recipients were studied. Lethally
irradiated recipients were reconstituted with B10.BR TCD BM and given
either no DLI or B10.BR DLI cells. All recipients given BM without DLI
had a 100% survival rate at 5 mo post-BMT (data not shown).
CD28-/- recipients of DLI had a significantly
(p = 0.0023) lower survival rate as compared
with wild-type controls (40 vs 100%, respectively) (Fig. 4). Mean weights had a nadir 24% lower
than pre-BMT body weights 50 days post-BMT as compared with 98% for
wild-type controls and 105% for CD28-/-
recipients not given DLI (data not shown). In contrast to data
indicating that perforin-deficient/gld-homozygous deficient recipients
had an impaired ability to suppress DLI-GVH, recipients only deficient
in perforin had a 100% survival rate, indicating that this pathway
alone was not critical for suppressing DLI-GVH. When gld
FasL-/- recipients were used, 75% of mice
survived long term. In the studies shown in Fig. 3, the survival curves
for CD28-/- and
perforin-deficient/gld-homozygous recipients were overlapping. Although
the long-term survival outcome for CD28-/-
recipients was virtually identical with that shown in Fig. 3, recipients deficient in either perforin or FasL had survival rates of
75%. Together, these data would indicate that suppression of
DLI-GVHD requires either perforin or FasL expression by host T
cells.
|
but not IL-12p70 protein is involved in
the suppression of DLI-GVH
T cytotoxic type I cytokine production is needed to facilitate CTL
generation. Two cytokines known to be important in CTL generation are
IFN- and IL-12. To explore the potential role of IFN- cytokine
production by host T cells in DLI-GVH suppression, BM-reconstituted
IFN--/- recipients were given DLI (Fig. 4).
These recipients had a significantly (p =
0.0075) lower survival rate (50% at 5 mo) as compared with controls.
Consistent with the survival data, IFN--/-
recipients given DLI had mean weights of 82% of pre-BMT body weights
at 3 mo post-BMT vs 117% for IFN--/-
recipients given no DLI (data not shown).
To determine whether the production of IL-12 would affect DLI-GVHD, we
used as recipients IL-12p40-/- mice that are
incapable of producing functional IL-12p70 protein consisting of a
p35/p40 heterodimer. IL-12p40-/- recipients had
an 88% long-term survival rate (Fig. 4) and mean body weights that
were 
7% lower than pre-BMT body weight (data not shown). Therefore,
the production by the host of IFN- but not IL-12p70 protein
influenced the host suppression of DLI-GVH.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production by the host is an
important factor that regulates DLI-GVHD lethality, neither IL-12p70
protein production by nor p55 TNFR signaling of host was required.
Collectively, these data provide important insights as to why GVHD
lethality is reduced when donor lymphocytes are infused later rather
then earlier post-BMT.
Under conventional fully myeloablative therapy in humans, few host T
cells remain early post-BMT because patients are profoundly lymphopenic
for several weeks post-BMT. Following nonmyeloablative conditioning and
in situations in which host hemopoiesis has returned with relapse of a
hemological malignancy, host T cells can be present in substantial
numbers. Our murine DLI model in which host T cells are present at the
time of DLI administration despite lethal TBI at levels that
approximate those of BM-derived T cells simulates such situations (see
Table I). Reisner and colleagues have reported that clonable host T
cells are present in the spleen of mice following lethal (9 Gy) TBI
(41). Similarly, we have shown that the administration of
9 Gy TBI by x-ray to B6 recipients does not completely eliminate the
host T cell compartment as measured by thoracic duct cannulation (data
not shown). Under conditions of lethal (8 Gy TBI) irradiation, we have
observed in our model that host T cells are present in the peripheral
blood and thoracic duct lymphatics at the time of DLI administration.
These host T cells gradually diminish in number as the new donor
BM-derived T cells are educated in the thymus. A kinetic analysis of
TDL number in lethally irradiated B6 recipients of TCD B10.BR BM has
shown that approximately equal numbers of donor and host BM-derived T
cells are present on day 27 post-BMT, whereas on day 41 post-BMT, there
are 4-fold fewer host T cells present in the thoracic duct
lymphatics than were present on day 27 post-BMT (data not shown). The
administration of DLI on day 21 post-BMT resulted in a complete
elimination of host but not donor BM-derived T cells as assessed on day
41 post-BMT (data not shown). We hypothesize that DLI-GVH resistance is
mediated by residual host T cells that are capable of eliminating
alloreactive donor T cells. In this scenario, donor BM-derived
regulatory cells would become the dominant and perhaps sole resistance
mechanism at time periods when host T cells have been eliminated or
markedly reduced in frequency.
As reported by Johnson et al. (12), the purposeful depletion of donor and host BM-derived T cells by creating ATx recipients removes resistance mechanisms of the host to DLI-GVH with all mice succumbing to GVH-induced lethality within 6 wk after DLI administration, a time course and outcome consistent with GVHD induction by the same number of cells infused on the day of BMT. Adult thymectomy has a number of effects, which include the depletion of donor and host BM-derived T cells. In older humans in whom thymus function has often been found to be impaired, the incidence of severe GVHD due to the infusion of T cells in the donor graft is increased (42). We do not know whether DLI-GVH is worse in such instances or in situations in which recipients are older or heavily pretreated with chemotherapy, also known to diminish thymic function (43, 44). Whether our DLI data in euthymic or ATx mice better simulate such conditions is unknown.
In a series of elegant studies, Johnson and Truitt have demonstrated that donor BM-derived T cells contribute to DLI-GVH suppression (12). The new findings of our study are that host T cells also contribute to DLI-GVH resistance. On the surface, this may seem predictable. However, host T cells have coexisted with donor BM-derived cells and, therefore, could have been presumed to be tolerant of donor alloantigens. It is particularly intriguing that a sustained antidonor response is capable of being generated in these mice despite the continuous availability of donor alloantigens to the host, which occurs post-BMT. Instead, the selective elimination of Thy1+ host T cells removed DLI-GVH resistance. Tolerance to alloantigens is a process that can take some time to develop (45). Host T cells present at 3 wk post-BMT would likely encounter fewer donor lymphohemopoietic cells than later post-BMT. Therefore, it is possible that host T cells are not initially tolerized by donor cells until a later time post-BMT than when DLI cells are given. In that event, host T cells would retain the capacity to resist DLI cells. If this were the case, then there would be a risk of host antidonor-mediated graft rejection in these chimeras. However, long-term chimeras are healthy and have donor cell reconstitution. Because host T cells are in a quiescent state in the absence of DLI administration, there would be no impetus for a host antidonor T cell-mediated graft rejection process to occur. With DLI administration, host antidonor effector cells are generated but these cells may not migrate in sufficient numbers to eliminate a BM compartment that is highly cellular at this time and that is capable of self-renewal. Thus, even if host T cells were to inhibit transiently donor BM-derived hemopoietic cells, this process is apparently not of sufficient magnitude to impact survival or eventual hemopoietic reconstitution in this model system.
Alternatively, suppressor cells may regulate host-mediated DLI-GVH resistance. Consistent with this latter hypothesis, Strober’s group reported that total lymphoid irradiation generates a population of Thy1+ suppressor cells that inhibits allogeneic responses in vitro and showed that such cells can be adoptively transferred and inhibit acute GVHD lethality in vivo (15). Sykes and Sachs have found Thy1+ host natural suppressor cells that develop in lethally irradiated BM-reconstituted mice that have been reconstituted with a mixture of donor and host BM, and they have found that such recipients were highly resistant to the effect of recipient strain lymphocytes injected in an attempt to break tolerance (16, 17, 18). Tutschka has shown in a rat BMT model of histoincompatible BM that splenic suppressor cells could be adoptively transferred into secondary recipients, which suppressed GVHD responses in these rats and resulted in resistance to small numbers of DLI cells given 250 days post-BMT (19, 20). Active resistance mechanisms that suppress GVHD responses have been observed in canine radiation chimeras (21, 22) and in humans (23). Our studies and these literature reports indicate that host resistance mechanisms to allogeneic lymphocyte infusion can develop in mice conditioned for BMT with either total lymphoid irradiation or TBI-containing regimens. In addition to the existence of regulatory and suppressor cells, host veto cells (46, 47) may be required for elimination of donor T cells capable of becoming antihost CTL effector cells (24, 48, 49, 50, 51, 52, 53, 54). Host veto cells can inactivate or eliminate donor T cells that have specificity for these host T cell CTL precursor cell populations. Donor veto cells have been hypothesized to inactivate host T cells that are capable of resisting donor lymphohemopoietic grafts (55). Conversely, host veto cells have been hypothesized to down-regulate GVHD caused by donor T cells. Whether host T cells in our system function to down-regulate DLI-GVHD via regulatory/suppressor cells, direct host antidonor CTL-mediated elimination or a veto cell mechanism is not currently known.
Either CD4+ or CD8+ host T
cells appeared to be required for optimal resistance because
anti-Thy1 allelic mAb-treated recipients had a higher incidence of
DLI-GVH lethality than either CD4-/- or
CD8-/- recipients in the same experiment (Fig. 2). Because Thy1+ cells include a subpopulation
of cells that are
CD4-CD8-
Thy1+ T/NK and are known to produce high
levels of IL-4 that may affect GVHD responses, experiments in which
anti-Thy1 allelic mAb infusion is used to deplete host
CD4+ and CD8+ T cells could
be difficult to interpret because anti-Thy1 mAb would have effects
on other cell populations (17, 31, 32, 33, 34). We do not favor
the possibility that anti-Thy1 mAb inhibited DLI-GVH resistance by
eliminating an IL-4-producing T/NK cell population for two reasons.
First, we have shown that IL-4 production can increase GVHD lethality
in lethally irradiated B10.BR recipients of B6 donor grafts
(56). Second, our data with recipients that have a genetic
deficiency in one of several molecules involved in facilitating host T
cell expansion and/or CTL generation indicated that such recipients had
similar survival curves as compared with anti-Thy1 allelic mAb
treatment. An alternative approach would be to target
CD4+ and CD8+ T cells with
mAbs directed against these determinants. However, anti-CD4 and
anti-CD8 mAb infusion would eliminate DLI-derived cells, along with
donor and host BM-derived T cells if circulating levels of mAb were
present post-BMT. In the absence of such data, we cannot formally
exclude the possibility that a Thy1+
CD4-CD8- host T/NK cell
was responsible for DLI-GVH resistance.
With respect to regulatory cells, the survival rate was lower in
CD4-/- but not CD8-/-
recipients given DLI as compared with wild-type controls, possibly
suggesting that host CD4+ T cells are critical to
the suppression process. The mechanisms by which host
CD4+ T cells contribute to DLI-GVH resistance
could be through direct regulatory mechanism(s) or by facilitating the
expansion of CD8+ T cells. In regard to the
former mechanism, ample evidence exists that there are
CD4+ regulatory T cells capable of suppressing
GVHD lethality. For example, Beschorner, Hess, and coinvestigators have
reported that the infusion of autoregulatory CD4+
T cells into recipients at high risk for developing syngeneic GVHD can
effectively prevent this disease process (57, 58).
Sakaguchi (59, 60) and Shevach (61, 62, 63) both
have shown that there is a population of thymus-derived
CD4+CD25+ regulatory T cells that suppress
autoimmune diseases. Bluestone and coworkers (64) have
shown that this population suppresses diabetes generation in
diabetes-prone nonobese diabetic mice. Powrie, Coffman, and
coinvestigators have shown that
CD4+CD45Rbhigh cells
suppress the generation of inflammatory bowel disease (65, 66), which may be relevant to GVHD-induced gastrointestinal
injury. Johnson and Truitt have documented the existence of a
subpopulation of donor BM-derived
Thy1+CD4+ T cells along
with a population of donor BM-derived
Thy1+TCRß+CD4-8-
T cells that suppress DLI-GVHD. In preliminary studies, we have
confirmed the contribution of donor Thy1+
BM-derived T cells to DLI-GVH resistance but have found that both donor
and host BM-derived T cells are required for optimal resistance because
selection depletion of neither population alone impaired resistance to
the same extent as observed with ATx recipients (B. R. Blazar and
P. A. Taylor, unpublished data). CD4-/-
recipients appeared to have a modestly reduced resistance to DLI-GVH.
In our study, anti-Thy1 allelic mAb, which eliminates host T cells,
causes a marked reduction in DLI-GVH lethality. We hypothesize that
under heavy irradiation conditions, host BM-derived regulatory cells
may be incapable of expanding or are more effectively eliminated by
heavy irradiation, thereby rendering the recipient more susceptible to
GVHD lethality than when the same number of donor lymphocytes are given
later post-BMT.
We have shown that the population of host T cells that participates in DLI-GVH resistance expresses the CD28 Ag. Previously published data from our group have demonstrated that CD28 expression is required to permit optimal expansion of alloreactive T cells at the time of GVHD induction (35). The requirement for CD28 expression is diminished under situations of heavy irradiation, perhaps due to the fact that proinflammatory cytokines are increased and more substantial tissue injury with release and/or up-regulation of alloantigens occurs. In the DLI setting, 3 wk have passed between lethal TBI and DLI infusion. Under these conditions, host T cell expansion is more likely to be dependent upon CD28/B7 costimulation (35). Our previous data have shown that CD28-/- donor T cells are capable of generating donor antihost CTLs to a similar extent on a per cell basis as wild-type cells (35). However, because expansion of CD28-/- T cells is severely impaired, the net result would be a failure of host CD28-/- T cells to generate sufficient numbers of CTLs to respond to DLI. Based in part upon the CD28-/- data, we hypothesized that there must be a critical number of host CTLs needed to effectively resist DLI cells. Several lines of evidence support this hypothesis. For example, although CD28/B7 costimulation was required for optimal DLI-GVH resistance, CD40L/CD40 costimulation was not, despite the fact that this pathway affects both the in vivo expansion of alloreactive cells and the generation of alloreactive CTLs in the same BMT setting as described above for CD28-/- cells (39).
Perforin-deficient/gld-homozygous recipients were highly susceptible
hosts for DLI-GVHD lethality. Braun et al. demonstrated that donor
splenocytes obtained from the perforin-deficient/gld-homozygous, but
not mice deficient in either pathway alone, were incapable of causing
GVHD lethality when infused into BALB/c recipients (67).
Alloreactive T cells generated from perforin-deficient/gld-homozygous
mice were unable to kill Con A-stimulated lymphoid targets or a B cell
lymphoid-malignant cell line (67), suggesting that the
absence of both perforin and functional FasL prevents alloreactive T
cells from eliminating lymphoid cells such as those contained in DLI.
It has been shown previously that the infusion of donor
perforin-deficient/gld-homozygous CD8+ T cells
was unable to reverse the BM graft rejection capabilities of recipient
CD8+ T cells, whereas the infusion of donor
CD8+ T cells deficient in either pathway alone
are sufficient to reverse graft rejection (24), further
supporting a critical role of either cytolytic pathway in eliminating
lymphoid cells. Studies by Levy and colleagues further indicate that
FasL/Fas interactions are an important mediator of GVHD-induced
lymphoid hypoplasia, whereas the effect of perforin is less substantial
(68, 69). Therefore, it would be expected that
gld-homozygous recipients might have more difficulty resisting DLI-GVHD
than perforin-deficient recipients. Our data are consistent with this
hypothesis but also provide direct evidence that DLI-GVHD resistance
requires either perforin or FasL but that absence of both pathways
severely impairs this resistance mechanism. In addition to identifying
the intracellular machinery required for host CTL resistance to
DLI, we have identified some of the cell surface molecules that
regulate CTL generation, which are critical for resisting DLI.
Recipients that lack 4-1BB were at least as susceptible as
perforin-deficient/gld-homozygous mice to DLI-GVH lethality.
Signaling via 4-1BB receptor enhances the in vivo CTL generation
against alloantigen-bearing targets (36) and is needed for
optimal GVHD lethality of splenocytes infused on the day of BMT (our
unpublished data). Despite the fact that Hill, Ferrara, and coworkers
have shown that p55 TNFR expression by alloreactive T cells facilitates
GVHD induction when T cells are given on the day of BMT
(40), our data indicate that signalling via p55 TNFR is
dispensable for DLI-GVH resistance. A potential difference in the
biology between the lack of effect of p55 TNFR on DLI-GVH as compared
with day 0 BMT is that release of proinflammatory cytokines such as
TNF- would be more closely linked to day 0 of BMT and would have
dissipated by day 21 post-BMT so that TNF- is not present in
sufficient concentrations at the time of DLI to drive the alloresponse.
Although it is unlikely that TNF- contributes to DLI-GVHD, we cannot
exclude a role for TNF- in DLI-GVHD because we have not precluded
the binding of TNF- to the p75 subunit of TNFRII.
T cytotoxic type I cytokines support CTL generation, and our data
clearly indicate that IFN- production by the host is a prerequisite
for optimal DLI-GVHD resistance. We have previously reported that B6
IFN--/- donor splenocytes infused into
lethally irradiated B10.BR recipients cause an accelerated GVHD
lethality process, suggesting that IFN- can function as a negative
regulator of T cell responses. Sykes and coworkers have shown that high
levels of IFN- produced in the very early post-BMT time period
suppress GVHD responses (70). The lack of IFN-
production by the host can have other potential sequences. For example,
insufficient numbers of host antidonor alloreactive CTLs may be
generated. Alternatively, the production of IFN- by the host could
render DLI cells more susceptible to host-mediated elimination by
up-regulating cell surface molecules (e.g., Fas; Ref. 71)
on DLI cells that render these cells better targets. In contrast to the
role of IFN- in DLI-GVHD resistance, IL-12p70 is not required
despite the fact that IL-12 is known to be a key regulator of
alloreactivity as has been shown in acute GVHD responses by several
investigators (72, 73, 74, 75). Thus, signaling via IFN- but
not TNF- or IL-12 appears to regulate DLI-GVH resistance.
In summary, we have found that host BM-derived
Thy1+ cells can suppress the DLI-GVH response.
Host T cell resistance requires CD28 and 4-1BB receptor cell surface
expression. Mice lacking both perforin and FasL cytolytic pathways are
highly susceptible to DLI-GVH. IFN- production but not IL-12 or
TNF- signaling of host T cells is needed to inhibit DLI-GVH
lethality. Despite this host resistance mechanism, DLI therapy is an
effective approach to treat some types of relapse post-BMT and to
increase donor chimerism in patients receiving nonmyeloablative
conditioning. The findings that host T cells can resist DLI-mediated
GVHD have potentially important clinical implications.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Bruce R. Blazar, University of Minnesota Hospital, Box 109 Mayo Building, 420 Southeast Delaware Street, Minneapolis, MN 55455.
3 Abbreviations used in this paper: BMT, bone marrow transplantation; DLI, delayed lymphocyte infusion; ATx, adult thymectomized; BM, bone marrow; GVHD, graft-vs-host disease; TBI, total body irradiation; TCD, T cell-depleted; TDL, thoracic duct lymphocyte(s); FasL, Fas ligand; CD40L, CD40 ligand.
Received for publication June 8, 2000. Accepted for publication August 2, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß TCR+NK1.1+ cells. J. Immunol. 145:3209.[Abstract]
on graft-versus-host disease and graft versus leukemia. J. Clin. Invest. 104:459.[Medline]
and IL-4 in acute graft-versus-host disease after allogeneic bone marrow transplantation in mice. J. Clin. Invest. 102:1742.[Medline]
is required for inhibition of acute graft-versus-host disease by interleukin 12. J. Clin. Invest. 102:2126.[Medline]
This article has been cited by other articles:
|
|
 |
|
  H.-J. Kolb Graft-versus-leukemia effects of transplantation and donor lymphocytes Blood, December 1, 2008; 112(12): 4371 - 4383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Chakraverty, B. Flutter, F. Fallah-Arani, H.-S. Eom, T. Means, G. Andreola, S. Schwarte, J. Buchli, P. Cotter, G. Zhao, et al. The Host Environment Regulates the Function of CD8+ Graft-versus-Host-Reactive Effector Cells J. Immunol., November 15, 2008; 181(10): 6820 - 6828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Miller and B. R. Blazar Response: The role of G-CSF on the risk of graft-versus-host disease after donor lymphocyte infusions Blood, May 15, 2008; 111(10): 5256 - 5257. [Full Text] [PDF]
|
|
|
|
|
|
S. Mielke, K. Rezvani, B. N. Savani, R. Nunes, A. S. M. Yong, J. Schindler, R. Kurlander, V. Ghetie, E. J. Read, S. R. Solomon, et al. Reconstitution of FOXP3+ regulatory T cells (Tregs) after CD25-depleted allotransplantion in elderly patients and association with acute graft-versus-host disease Blood, September 1, 2007; 110(5): 1689 - 1697. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Durakovic, V. Radojcic, M. Skarica, K. B. Bezak, J. D. Powell, E. J. Fuchs, and L. Luznik Factors governing the activation of adoptively transferred donor T cells infused after allogeneic bone marrow transplantation in the mouse Blood, May 15, 2007; 109(10): 4564 - 4574. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Rezvani, S. Mielke, M. Ahmadzadeh, Y. Kilical, B. N. Savani, J. Zeilah, K. Keyvanfar, A. Montero, N. Hensel, R. Kurlander, et al. High donor FOXP3-positive regulatory T-cell (Treg) content is associated with a low risk of GVHD following HLA-matched allogeneic SCT Blood, August 15, 2006; 108(4): 1291 - 1297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Fukushima, T. Yamaguchi, W. Ishida, K. Fukata, R. S. Mittler, H. Yagita, and H. Ueno Engagement of 4-1BB Inhibits the Development of Experimental Allergic Conjunctivitis in Mice J. Immunol., October 15, 2005; 175(8): 4897 - 4903. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Rettig, J. K. Ritchey, J. L. Prior, J. S. Haug, D. Piwnica-Worms, and J. F. DiPersio Kinetics of In Vivo Elimination of Suicide Gene-Expressing T Cells Affects Engraftment, Graft-versus-Host Disease, and Graft-versus-Leukemia after Allogeneic Bone Marrow Transplantation J. Immunol., September 15, 2004; 173(6): 3620 - 3630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. E. Anderson, J. M. McNiff, C. Matte, I. Athanasiadis, W. D. Shlomchik, and M. J. Shlomchik Recipient CD4+ T cells that survive irradiation regulate chronic graft-versus-host disease Blood, September 1, 2004; 104(5): 1565 - 1573. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Y. Mapara and M. Sykes Tolerance and Cancer: Mechanisms of Tumor Evasion and Strategies for Breaking Tolerance J. Clin. Oncol., March 15, 2004; 22(6): 1136 - 1151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Young, L. S. Kay, M. J. Phillips, and L. Zhang Antitumor Activity Mediated by Double-Negative T Cells Cancer Res., November 15, 2003; 63(22): 8014 - 8021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Billiau, S. Fevery, O. Rutgeerts, W. Landuyt, and M. Waer Transient expansion of Mac1+Ly6-G+Ly6-C+ early myeloid cells with suppressor activity in spleens of murine radiation marrow chimeras: possible implications for the graft-versus-host and graft-versus-leukemia reactivity of donor lymphocyte infusions Blood, July 15, 2003; 102(2): 740 - 748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hummel, D. Wilms, M. Vitacolonna, and M. Zoller Donor T cell and host NK depletion improve the therapeutic efficacy of allogeneic bone marrow cell reconstitution in the nonmyeloablatively conditioned tumor-bearing host J. Leukoc. Biol., November 1, 2002; 72(5): 898 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. D. Billiau, S. Fevery, O. Rutgeerts, W. Landuyt, and M. Waer Crucial role of timing of donor lymphocyte infusion in generating dissociated graft-versus-host and graft-versus-leukemia responses in mice receiving allogeneic bone marrow transplants Blood, August 13, 2002; 100(5): 1894 - 1902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Y. Mapara, Y.-M. Kim, S.-P. Wang, R. Bronson, D. H. Sachs, and M. Sykes Donor lymphocyte infusions mediate superior graft-versus-leukemia effects in mixed compared to fully allogeneic chimeras: a critical role for host antigen-presenting cells Blood, August 13, 2002; 100(5): 1903 - 1909. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
