| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Carlos and Marguerite Mason Transplantation Research Center, Department of Surgery, Emory University School of Medicine, Atlanta, GA 30322
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-producing T cells within 3–4 wk
posttransplant, whereas mice with a C3H background generate neither
CTL- nor IFN--producing cells. Thus, differences appear to be in the
generation of the immune response and not T cell homing. Strain
differences in costimulation blockade-induced hyporesponsiveness
persist in the absence of CD4+ T cells, implying a direct
effect on CD8+ T cells. We demonstrate that genetic
differences are important in cells of hemopoietic origin and that the
costimulation blockade-resistant phenotype is dominant. Analysis of BXH
recombinant inbred strains indicates that multiple loci contribute to
the phenotype, and that the blockade resistance loci are preliminarily
linked to 17 markers on four chromosomes. We conclude that strain
variation in allograft MSTs following CD40/CD28 blockade results from
the ability of CD8+ T cells in some strains to use
alternative modes of costimulation to mount an effective
alloresponse. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is thought that optimal activation of T cells requires two types of signals. The first signal, delivered upon prolonged interaction of the MHC with the TCR, lends specificity to the immune response, whereas the second signal, mediated by costimulatory interactions between the T cell and the APC, allows for optimal T cell activation and clonal expansion. Various molecules have been shown to have costimulatory roles for T cell activation, including CD40 and CD28 (6, 7, 8, 9). These two pathways have been shown to be particularly important in a variety of immune responses in vivo, unveiling potential therapeutic applications. For example, blockade of the CD28 pathway can ameliorate the symptoms of experimental autoimmune encephalitis, a mouse autoimmune disease that attacks the central nervous system and is mediated primarily by CD4+ T cells (10). In the transplant community, great interest has been generated by the observation that simultaneous blockade of the CD40 and CD28 costimulatory pathways significantly prolongs allograft survival in both mice and primates in the absence of further immunosuppression (11, 12). Furthermore, blockade of either or both of these pathways has been a central component of several strategies aimed at inducing donor-specific tolerance (13, 14, 15). Nevertheless, CD40/CD28 blockade alone has not been shown to induce indefinite graft survival.
Different strains of mice show greatly disparate responses when receiving skin allografts and blockade of the CD40/CD28 pathways. Our laboratory has previously observed that although C3H/HeJ mice enjoy impressive prolongation of skin allograft survival following costimulation blockade treatment, other strains do not respond as favorably. C57BL/6 (B6)3 mice are particularly refractory to the effects of CD40/CD28 blockade. In this setting, we have identified a population of CD8+ asialo GM1+ T cells that are responsible for costimulation blockade-resistant rejection (16). The disparity between the two strains is independent of the donor strain. B6 mice are similarly resistant to costimulatory blockade following receipt of either a BALB/c or C3H/HeJ skin graft, whereas C3H/HeJ mice accept both BALB/c and B6 skin allografts for >100 days following treatment (A. Bingaman, unpublished observations).
In an effort to better understand the requirements for the generation
of effector T cells capable of rejecting skin allografts, we sought to
explore the underlying genetic causes of the enhanced ability of B6
mice, in comparison to C3H/HeJ mice, to reject skin allografts in a
CD40/CD28-independent manner. In this report we show that this
disparity is neither a result of MHC haplotype nor T cell precursor
frequency. We demonstrate that mice with a B6 background generate
allospecific CTLs or IFN--producing T cells in a
CD40/CD28-independent fashion, unlike mice with a C3H background.
Furthermore, our data point to a direct effect on the activation of
CD8+ T cells, rather than an indirect effect
mediated by CD4+ T cells. By the use of
irradiation chimeras, we demonstrate that the genetic differences
between the two strains assert themselves in blood-derived cells.
Analysis of F1 mice indicates a dominant mode of
inheritance, and analysis of BXH recombinant inbred strains implicates
at least two loci involved in the blockade-resistant phenotype.
Furthermore, preliminary linkage analysis of the BXH strains highlights
17 markers on four chromosomes as candidates for the location of the
blockade resistance loci. Taken together, our data suggest that at
least two H-2-independent loci directly modulate the activation of
CD8+ T cells in the absence of CD40/CD28
costimulatory signals.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Adult male 6- to 8-wk-old BALB/c, B6, C3H/HeJ, C3H.SW-H-2b, B6.AKR-H-2k, and B6.SJL mice were purchased from The Jackson Laboratory (Bar Harbor, ME). F1 mice were generated by mating B6 males to C3H.SW females.
Skin grafting
Full thickness skin grafts (
1 cm2) were
transplanted on the dorsal thorax of recipient mice and secured with a
band-aid for 7 days. Graft survival was then followed by daily visual
inspection. Rejection was defined as the complete loss of viable
epidermal graft tissue. Statistical analyses were performed using a
Mann-Whitney U test.
Treatment protocols
Skin graft recipients were treated with 500 µg each of hamster anti-murine CD40 ligand (CD40L) Ab (MR1) and human CTLA4-Ig (provided by Diane Hollenbaugh, Bristol-Myers Squibb, Princeton, NJ) administered i.p. on the day of transplantation (day 0) and on postoperative days 2, 4, and 6. CD4-depleted experimental groups received 100 µg rat anti-mouse CD4 (GK1.5) i.p. on days -3, -2, -1, and weekly until harvest. Treated irradiated recipients of CFSE-labeled T cells received 500 µg each of hamster anti-murine CD40L and human CTLA4-Ig i.p. on days 0 and 2 posttransfer.
Flow cytometry
Analyses of splenocytes of irradiated recipients of CFSE-labeled T cells were conducted using fluorochrome-conjugated Abs (rat IgG2a-PE, rat IgG2b-APC, anti-CD4-PE, and anti-CD8-APC; PharMingen, San Diego, CA). Peripheral blood was analyzed by staining with fluorochrome-conjugated Abs (CD45.1-PE, CD45.2-FITC, CD4-APC, CD8-APC, mouse IgG2a-FITC, and mouse IgG2a-PE) followed by RBC lysis and washing with a whole blood lysis kit (R&D Systems, Minneapolis, MN). Flow cytometry was performed using a FACSCalibur, with CFSE fluorescence data being collected on the FL1 (FITC) channel. Data were analyzed using Cellquest software (Becton Dickinson, Braintree, MA).
CFSE labeling and adoptive transfers
Splenic and mesenteric lymph node cells were harvested from B6 or C3H.SW mice. After RBC lysis and nylon wool passage, the cells were incubated in 10 µM CFSE in RPMI 1640 (Molecular Probes, Eugene, OR). After 10 min the staining was halted by the addition of cold RPMI 1640. Irradiated (1800 rad) BALB/c, B6, or C3H.SW mice then received 2 x 107 CFSE-labeled cells i.v. by penile vein injection. After 66–72 h, splenocytes were harvested from the recipients, the RBC were lysed, and the remaining cells were stained for flow cytometry as above.
Determination of precursor frequencies
Precursor frequency was determined as previously described (17). In short, peaks were numbered for the number of times the cells had divided (n). A T cell that divides n times generates 2n daughter cells. Therefore, to obtain a number of precursors for each individual peak, the number of cells in that peak is divided by 2n. For allogeneic transfers, the precursor numbers of peaks 4–8 were added together and divided by the total number of initial precursors to generate a precursor frequency for divisions 4–8. These divisions were chosen based on the observation that syngeneic transferred cells do not proceed detectably beyond three divisions.
Cytotoxicity assays
BALB/c CL.7 cells were used as targets. Target cells were
suspended in saline and 5% FCS (1 x
107/ml) with 750 µCi 51Cr
(sp. act. 470 mCi/mg; NEN Life Science Products, Boston, MA) for 90 min
at 37°C. Target cells were washed three times and plated in 96-well
round-bottom plates at 1 x 104 targets/well
in R10 medium (RPMI 1640 supplemented with penicillin, streptomycin,
2-ME, and 10% FCS (HyClone)). Effectors were prepared by harvesting
splenocytes, lysing RBC, and passing them through nylon wool. Effectors
were plated at the appropriate ratios in quadruplicate. Four wells were
set aside for total lysis by addition of 2% Triton X-100 to the
targets, and four wells were set aside for spontaneous lysis by the
addition of R10 medium without effector cells. After 5 h, the
supernatant was harvested and analyzed by gamma counting. Percent
specific lysis was determined by use of the following formula: 100
x (cpm unknown - cpm spontaneous)/(cpm total - cpm
spontaneous).
IFN- ELISPOT assays
Allospecific T cell responses were measured by IFN- ELISPOT
assay by using nylon wool-passed splenocytes obtained from skin-grafted
B6 or C3H.SW mice or Histopaque-separated (Sigma, St. Louis, MO)
peripheral blood leukocytes obtained from B6, C3H.SW, or
F1 mice. The capture Ab for this assay, rat
anti-mouse IFN- (clone R4-6A2; PharMingen), was incubated at 4
µg/ml in PBS (100 µl/well) at 4°C overnight in
ester-cellulose-bottom plates (Millipore, Bedford, MA). After washing
with PBS, 3 x 105, 1.5 x
105, or 0.75 x 105
effector cells were added to the wells. Stimulators were either
irradiated (2000 rad) donor splenocytes added at a 1:1
stimulator-to-effector ratio or mouse dendritic cells obtained through
overnight transient adherence added at a 1:10 stimulator-to-effector
ratio. Effector cells were incubated for 14–16 h at 37°C with or
without stimulators. After the culture period, cells were removed by
washing the plate in PBS/Tween 20 (0.05%), and then biotinylated
anti-mouse IFN- (clone XMG1.2; PharMingen) was added at 4
µg/ml (100 µl per well). After 2–3 h at 4°C, unbound Ab was
removed, and HRP-avidin D (Sigma) was added. Spots were developed with
the substrate 3-amino-9-ethyl-carbazole (Sigma) with 0.015%
H2O2. Each spot represents
an IFN--secreting cell, and the frequency of these cells can be
determined by dividing the number of spots counted in each well by the
total number of cells plated at that dilution. Naive splenocytes
produce IFN- at a frequency of 
3 per 105
cells with or without stimulation.
Irradiation chimeras
C3H.SW mice (CD45.2+) received 1200 rad of irradiation to completely ablate the hemopoietic cells. One day later, irradiated mice were rescued with 1 x 107 B6.SJL (CD45.1+) bone marrow cells injected i.v. Mice were treated with 500 µg MR1 and CTLA4-Ig on days 0, 2, 4, and 6 posttransfer to prevent graft-versus-host disease (GVHD). Mice were monitored over a course of 6 wk for expression of the donor CD45 alloantigen in the peripheral blood by flow cytometry. Skin grafts were given when chimerism exceeded 98%. Surprisingly, attempts to generate B6.SJL chimeras with C3H.SW bone marrow using the same protocol were not successful, precluding analysis in the opposite direction.
Linkage analysis
Each recombinant inbred BXH strain was assigned a phenotype based on median survival times (MSTs) of allogeneic skin grafts following treatment with CTLA4-Ig and anti-CD40L. Three strains were assigned a B6-like phenotype, and three strains were assigned a C3H-like phenotype. Linkage to 330 polymorphic markers (18) was assessed using Map Manager version 2.6.6.1 (19). Linkage was assessed within 95% confidence limits, and the software calculated logarithm of odds scores.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous work in our laboratory has established a significant
difference in the ability of distinct strains of mice to accept skin
allografts following blockade of the CD28/CD40 T cell costimulatory
pathways. In particular, BALB/c skin allograft survival on C3H/HeJ
recipients treated with CTLA4-Ig and anti-CD40L is dramatically
longer than the survival of these grafts on treated B6 recipients. In
these experiments, the MST of BALB/c skin allografts on C3H/HeJ
recipients treated with anti-CD40L and CTLA4-Ig was >100
days vs 11 days in untreated controls (Fig. 1
A), whereas the MST of skin
allografts on B6 mice receiving the same treatment was 22 days vs 10
days in untreated controls (Fig. 1B).
|
C). In contrast, B6.AKR mice acquired
the more resistant phenotype of the B6 parent strain, rejecting their
grafts with a MST of 20 days following treatment (Fig. 1D).
These results indicate that the H-2 haplotype is not the major
determinant of strain-associated allograft rejection responses after
costimulation blockade. The disparity between B6 and C3H.SW mice is not due to differences in T cell precursor frequency
One explanation of the above data is that although the H-2 locus
of the B6 and C3H.SW mice is the same, other background differences may
determine the precursor frequencies of alloreactive T cells, thus
leading to the observed disparity in rejection time following
treatment. For example, in one viral system, strain variation in the
ability to mount an antiviral response is linked to expression of an
endogenous mouse mammary tumor virus (MMTV) superantigen, leading
to selective deletion of major portions of the T cell repertoire
(20). To test this hypothesis, we used a previously
described GVHD model (21). T cell-enriched splenocytes
from B6 or C3H.SW mice were labeled with CFSE (Molecular Probes), a
membrane-permeable ester that, when cleaved by nonspecific
intracellular esterases, becomes fluorescent and remains trapped in
live cells. Cellular proliferation can be monitored due to the
progressive loss of fluorescence with successive cell divisions.
Labeled cells were injected i.v. into lethally irradiated syngeneic (B6
or C3H.SW) or allogeneic (BALB/c) hosts. Spleens were harvested 66–72
h later, resuspended, and stained for expression of CD4 and CD8 vs CFSE
fluorescence. As shown in Fig. 2B, T cells transferred into
syngeneic hosts underwent only minimal replication, with no cells
proceeding beyond three divisions. Conversely, allogeneic T cells
transferred into BALB/c hosts divided extensively, replicating up to
eight times in 3 days (Fig. 2A).
|
C). Among the
CD8+ subpopulation, 4.8% of donor C3H.SW and
4.1% of donor B6 T cells divided in BALB/c recipients, respectively
(Fig. 2C). CD4+ and
CD8+ cells dividing four or more times in
syngeneic hosts were undetectable above background. Precursor
frequencies in either T cell subpopulation were not significantly
different between strains, leading us to conclude that differences in
the initial T cell precursor frequency do not account for the distinct
responses to alloantigen after costimulation blockade.
Donor B6 and C3H.SW T cells were also tested for their ability to
replicate in irradiated BALB/c mice in the face of costimulation
blockade. Following transfer, mice were treated with anti-CD40L and
CTLA4-Ig on days 0 and 2. As seen in Fig. 2C, the percentage
of B6 and C3H.SW T cells able to divide in the face of costimulation
blockade did not vary markedly in either the CD4+
or the CD8+ subpopulations. After blockade, 2.1%
of CD4+ B6 T cells and 1.7% of
CD4+ C3H.SW T cells divided four or more times in
response to alloantigen, resulting in a 62 and 75% decrease from
untreated mice, respectively. Following treatment, 1.8% of B6
CD8+ T cells and 2.0% of C3H.SW
CD8+ T cells divided in response to alloantigen,
decreasing from untreated mice 55 and 58%, respectively. No striking
differences in the ability to proliferate in response to alloantigen in
the face of costimulation blockade were observed, and we concluded that
differences in the ability to mount a rejection response between the
two strains could not be accounted for by differences in the frequency
of T cells able to proliferate in response to alloantigen in the
presence or absence of costimulation blockade.
B6 mice generate allospecific CTL and IFN--producing T cells
following costimulation blockade, whereas C3H.SW mice are unable to do
so
Although no early differences in proliferative responses between
the two strains could be detected, we hypothesized that B6 and C3H.SW
mice might vary in their ability to generate and sustain functional T
cell responses following skin engraftment and treatment with CTLA4-Ig
and anti-CD40L. In particular, it seemed likely that differences
would be seen in the CD8+ subpopulation, as our
previous work has established a crucial role for
CD8+ T cells in costimulation blockade-resistant
rejection in B6 mice (16). To test this hypothesis,
splenocytes from B6 or C3H.SW mice receiving BALB/c skin allografts
were assessed for direct ex vivo CTL activity and frequency of
IFN--producing cells at various days posttransplant.
As shown in Fig. 3, untreated C3H.SW and
B6 mice receiving BALB/c skin allografts mounted CTL responses in the
spleen with similar kinetics. Ex vivo CTL responses were detectable by
day 8, peaked at or near day 12, and remained measurable even at 26
days posttransplant. The frequency of IFN--producing T cells, as
measured by ELISPOT assay, was also similar between the two strains
(Fig. 4). Substantial numbers of
IFN--producing T cells in the spleen could be measured on day 8 in
both strains. The frequency of these cells peaked between days 12 and
18 and remained significant at 26 days posttransplant. These
experiments showed no marked difference between the B6 and C3H.SW
strains in their ability to mount allospecific IFN- and CTL
responses following skin engraftment.
|
|
-producing cells were measured in B6 and C3H.SW
mice receiving CTLA4-Ig and anti-CD40L at the time of skin
engraftment. Fig. 3 details the ex vivo CTL activity of T cell-enriched
splenocytes at various points postengraftment. At 8 and 12 days
postengraftment, no CTL activity could be detected in either B6 or
C3H.SW spleens. However, by days 18 and 26, allospecific CTL activity
could be detected ex vivo in B6 spleens. In contrast, C3H.SW mice
showed no increase in CTL activity above naive controls at any time
point, and neither strain showed an ex vivo response to syngeneic
targets (data not shown). The generation of IFN--producing cells
exhibited a similar pattern. B6 mice mounted a costimulation
blockade-resistant IFN- response in the spleen beginning at day 18
postengraftment, whereas C3H.SW mice failed to generate detectable
numbers of IFN--producing cells above background (Fig. 4). These experiments demonstrate a fundamental difference in the ability of these two strains to generate effector T cells independently of the CD28/CD40 costimulatory pathways. They also imply that genetic differences defining the disparate phenotypes assert themselves in the priming and generation of the immune response and not during the effector phase, such as in T cell homing to the graft.
Inhibition of costimulation blockade-independent rejection in C3H.SW mice is at least partly a direct effect on CD8+ T cells
Although CD8+ T cells appear to be the
determining factor in costimulation blockade-resistant rejection of
skin allografts in B6 mice (16),
CD4+ T cells have been shown to play a regulatory
role in prolonging allograft survival in strategies using
anti-CD40L and CTLA4-Ig (Ref. 22 and our unpublished
observations). Another possible explanation for the inability of some
strains to overcome costimulation blockade at early time points may
include the generation of CD4+ regulatory cells.
We hypothesized that perhaps the genetic differences between the two
strains resided in the ability to generate regulatory
CD4+ T cells to quench the
CD8+ T cell response. To test this hypothesis, B6
and C3H.SW mice received BALB/c skin grafts with and without
costimulation blockade. Half of the mice from each group were also
depleted of CD4+ T cells in vivo with the GK1.5
mAb. Responses were assessed by measuring ex vivo CTL activity and
frequency of IFN--producing T cells at 21 days posttransplant.
BALB/c skin allografts induced substantial direct ex vivo CTL responses
in both B6 and C3H.SW mice by day 14 posttransplant (data not shown).
Ex vixo CTL activity persisted at day 21 (Fig. 5A). Following treatment with
CTLA4-Ig and MR1, B6 mice maintained the ability to generate a readily
detectable CTL response by day 21, whereas C3H.SW mice failed to
develop measurable CTL responses above background (Fig. 5A).
The frequency of IFN--producing cells, as measured by ELISPOT
analysis, followed a similar pattern. Both mice generated high numbers
of IFN--producing cells at days 14 and 21. Treatment with
costimulation blockade completely abrogated this response in C3H.SW
mice. In contrast, B6 mice were able to generate large numbers of
IFN--producing cells at day 21 (Fig. 5B). Following CD4
depletion alone, both IFN- and CTL responses remained strong in both
strains, whereas costimulation blockade combined with CD4 depletion
inhibited all detectable IFN- or CTL responses in the C3H.SW mice.
In contrast, B6 mice generated both ex vivo cytolytic activity and
IFN--producing cells by day 21, even in the absence of
CD4+ T cells (Fig. 5, A and
B).
|
The genetic disparities that lead to costimulation blockade-resistant rejection are not relevant in the parenchyma
At least two possibilities could explain the differing ability of the two strains to generate costimulation blockade-resistant immune responses. One was that the disparate phenotypes are an intrinsic property of hemopoietic cells, and that genetic differences between the two strains manifested themselves specifically in immune cells. Alternatively, we considered the possibility that strain-associated differences in parenchymal tissue may determine the generation of costimulation blockade-resistant effector T cells. Irradiation chimeras were used to better define the respective roles of parenchymal and blood-derived immune cells. C3H.SW mice, which express the CD45.2 alloantigen, were lethally irradiated, then rescued with bone marrow cells from B6.SJL mice expressing the CD45.1 alloantigen. Mice were treated with CTLA4-Ig and anti-CD40L at the time of bone marrow infusion to facilitate engraftment and prevent GVHD. Chimerism was measured by tracking the CD45 alloantigen in the peripheral blood by flow cytometry. At 6 wk posttransplant, peripheral blood leukocytes were >98% CD45.1+. Chimeric mice then received a BALB/c skin allograft with or without costimulation blockade and were assessed for allograft survival.
Untreated chimeric C3H.SW mice rejected BALB/c skin allografts in
normal control time (MST = 10 days, data not shown). When treated
with CTLA4-Ig and anti-CD40L at the time of engraftment, these mice
acquired the more aggressive costimulation blockade-resistant phenotype
displayed by B6.SJL mice receiving the same treatment (Fig. 6). Treated chimeric C3H.SW mice had a
MST of 17 days, whereas treated B6.SJL and C3H.SW controls had MST
values of 17 and >70 days, respectively. Unfortunately, we were unable
to generate the reciprocal B6.SJL chimeras. Nonetheless, these results
strongly support a pivotal role for cells of hemopoietic origin in
determining the strain-associated response to allogeneic skin grafts
following costimulation blockade.
|
We next sought to further characterize inheritance of the
costimulation blockade resistance phenotype by determining its
expression in F1 mice. B6 mice were mated to
C3H.SW mice to generate F1 hybrids that were of a
single H-2 haplotype. F1 mice received BALB/c
skin grafts with or without costimulation blockade at 6–10 wk of age
and were assessed for allograft survival. As seen in Fig. 7A, F1
mice readily rejected their skin grafts following costimulation
blockade (MST = 20 days). This is similar to the rate of rejection
of skin grafts following costimulation blockade in B6 mice (MST =
23 days). To characterize the responses functionally, recipient mice
were analyzed at various time points postengraftment. Peripheral blood
leukocytes were assessed for frequencies of IFN--producing cells by
ELISPOT assay. As predicted by graft survival, F1
and B6 mice both generated high numbers of IFN--producing cells.
Conversely, C3H.SW mice failed to generate a substantial number of
IFN--producing cells in the peripheral blood by day 19 (Fig. 7B). These experiments suggest that costimulation blockade
resistance is inherited in a dominant fashion.
|
To further investigate the genetics of this phenotype, we assessed the ability of 12 BXH recombinant inbred strains to reject BALB/c skin grafts following costimulation blockade. Because the genomes of these strains represent a patchwork of inheritance from the B6 and C3H/HeJ parent strains, we hypothesized that analysis of their rejection phenotypes would allow us to make use of available linkage maps to identify areas of the genome that potentially contribute to costimulation blockade-resistant rejection. BXH, B6, and C3H/HeJ mice were tested for significant differences in survival of BALB/c skin grafts after costimulation blockade.
BXH recombinant inbred strains segregated into at least three distinct
phenotypes. BALB/c skin graft survival in three strains, BXH-10, -11,
and -14, was not significantly different from that in B6 mice. Graft
survival in three other strains, BXH-3, -9, and -19, was not
significantly different from that in C3H/HeJ mice. The remaining
strains (BXH-2, -4, -6, -7, -8, -9, and -12) showed an intermediate
phenotype, as graft survival was significantly distinct from that in
both the B6 and C3H/HeJ mice. It remains unclear whether the
intermediate phenotypes can be further grouped and distinguished, as no
significant differences between them were evident with the numbers of
mice tested. Survival scores and MST values are outlined in Table I.
|
Analysis of known polymorphic markers among BXH RI strains shows preliminary linkage to 17 markers on four chromosomes
Recombinant inbred strains bearing the distinctive B6 or C3H/HeJ
phenotypes were analyzed using Map Manager software (19).
The blockade resistance phenotype was tested for linkage to 330 known
polymorphic markers (18) in each of the RI strains.
Preliminary linkage was demonstrated to 17 markers on four different
chromosomes within 95% confidence limits. All linked markers had
logarithm of odds scores of 1.8 and are as follows: chromosome 5,
En2, Emv1, D5H4S43, Pmv5, Qdpr; chromosome 9, D9 Mit8,
D9Nds2, D9Nds1, Gtpc, Kfo1, Ltw3; chromosome 13, Pmv9;
chromosome 15, Xltr5, Ly6, Ly6d, Iapls1–22, Pmv-17. This
low-resolution linkage analysis is limited by the relatively small
number of recombinant inbred strains available. However, it provides an
important starting point for further studies using large numbers of
experimental backcrosses and intercrosses, allowing us to focus on
5% of the mouse genome.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and CTL responses following
receipt of a BALB/c skin allograft, anti-CD40L, and CTLA4-Ig.
Although B6 mice are able to mount delayed but substantial IFN- and
CTL responses to alloantigen, C3H.SW responses are almost completely
suppressed. Furthermore, we have shown that costimulation blockade
directly blunts CD8+ T cell activation and
expansion in C3H.SW but not B6 mice, regardless of the presence of
CD4+ T cells. Our analysis of allogeneic skin
graft survival on BXH recombinant inbred strains following
costimulation blockade implies that multiple loci contribute to the
phenotype, and we propose that the phenotype distribution among the 12
strains is consistent with a two-gene hypothesis. Preliminary linkage
mapping has been performed using data available from the BXH strains to
identify candidate chromosomal segments for the location of the
blockade resistance loci. Using this type of analysis, we have excluded the genes encoding CD40, CD154, CD28, CD80, and CD86, showing that resistance to Ab therapy is not simply due to polymorphisms in the target molecules themselves. Furthermore, we are able to exclude numerous other molecules that have been shown to play a costimulatory role in T cell activation or as growth factors for activated T cells, including ICAM-1, 4–1BB, 4–1BB ligand, Fas, Fas ligand, TRANCE, IL-2, IL-15, CD25, IL-15R, etc. However, it should be noted that we cannot exclude the possibility of differences in downstream events in one or more of these pathways. These results set the stage for further analysis of large numbers of experimental backcrosses or intercrosses to establish definitive linkage and provide a high resolution map.
Eventual identification of the loci encoding this phenotype will
provide important information regarding the activation and
costimulatory requirements of alloreactive CD8+ T
cells. Several molecules have previously been implicated as important
in the costimulation of CD8+ T cells. Among these
is 4-1BB, a member of the TNF superfamily that is important in
CD28-independent costimulation of CD8+ T cells
during antiviral responses (23). However, disruption of
the 4-1BB costimulatory pathway alongside CD28/CD40 blockade in B6 mice
does not significantly prolong skin allograft survival (our unpublished
observations). However, it should be noted that at least one other
report has observed delayed skin allograft rejection upon simultaneous
disruption of the CD28 and 4-1BB pathways (24). Likewise,
the Fas-Fas ligand pathway has been proposed as a
CD8+ T cell costimulatory factor
(25). Recent data from our laboratory using B6.gld mice
has implicated this pathway in the costimulation blockade-resistant
CD8+ T cell response in mice with a B6 genetic
background.4 Other molecules may play a role in
CD8+ T cell costimulation, including IL-6 and
TNF-
(26), ICAM-1, LFA-3 (27), Ly-6C
(28), and NO generated by inducible NO synthase
(29).
The role of CD40 and CD28 in the costimulation of CD8+ T cells is not clearly understood. CD40 and CD28 themselves do not appear to be crucial elements of CD8+ T cell activation in B6 mice in alloresponses to skin. Our lab has characterized a population of CD8+ T cells in these mice able to overcome blockade of both the CD40 and CD28 pathways (16). Newell et al. have also reported the rejection of small bowel grafts by CD8+ T cells independently of CD28 (30). In addition, antiviral responses to LCMV and viral clearance are impaired only minimally following disruption of the CD40 and CD28 pathways (31, 32, 33). However, other antiviral responses, such as to vesicular stomatitis virus (VSV), seem to have a much greater requirement for CD28 and CD40 costimulation in the activation and expansion of CD8+ antiviral T cells (34). In various experiments, CD40L has been shown to play a central role in conditioning APCs for the priming of CD8+ T cell responses (35, 36, 37). Still other groups, using tetramerized MHC molecules in vitro, have shown that CD8+ T cells can potentially be activated without any second signal whatsoever, requiring only signaling via the TCR (38). Clearly, the conditions that dictate the requirements for CD8+ T cell activation in vivo in these widely varied models are poorly understood. The identification of alternative and relevant means of CD8+ T cell costimulation in vivo is essential to a proper understanding of their biology, activation requirements, and interactions with other cell types. We propose that during alloresponses to skin grafts in some strains, including B6, CD8+ T cells have distinct costimulatory requirements from that of CD4+ T cells and are able to mediate skin graft rejection through alternative routes of T cell activation. The results in this paper provide new insights into CD28/CD40-independent rejection of skin allografts by CD8+ T cells and represent a major step toward identifying the precise pathways used in vivo by B6 mice in overcoming costimulation blockade that are unavailable to C3H/HeJ mice.
We do not believe these results are unique to these particular strains,
as differences in skin graft survival following costimulation blockade
are also seen in other strain types. For example, DBA/2 mice enjoy
impressive prolongation of skin graft survival following costimulation
blockade, much like the C3H/HeJ mice, whereas BALB/c mice reject their
grafts in a similar time course to
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Drs. Christian P. Larsen or Thomas C. Pearson, Emory University Transplantation Immunology Laboratory, Suite 5105, Woodruff Memorial Building, 1639 Pierce Drive, Atlanta, GA 30322.
3 Abbreviations used in this paper: B6, C57BL/6; MST, median survival time; CD40L, CD40 ligand; GVHD, graft-versus-host disease.
4 J. Trambley, A. Lin, E. Elwood, A. W. Bingaman, F. Lakkis, T. C. Pearson, and C. P. Larsen. FasL is important in costimulation blockade-resistant skin graft rejection. Submitted for publication.
Received for publication July 26, 2000. Accepted for publication September 22, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and CTLA4. J. Clin. Invest. 101:2446.[Medline]
This article has been cited by other articles:
|
|
 |
|
  H. Xu, J. Yan, Y. Huang, P. M. Chilton, C. Ding, C. L. Schanie, L. Wang, and S. T. Ildstad Costimulatory blockade of CD154-CD40 in combination with T-cell lymphodepletion results in prevention of allogeneic sensitization Blood, March 15, 2008; 111(6): 3266 - 3275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Kemball, E. Szomolanyi-Tsuda, and A. E. Lukacher Allogeneic Differences in the Dependence on CD4+ T-Cell Help for Virus-Specific CD8+ T-Cell Differentiation J. Virol., December 15, 2007; 81(24): 13743 - 13753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Yu, X. Xu, M. D. Vu, E. D. Kilpatrick, and X. C. Li NK cells promote transplant tolerance by killing donor antigen-presenting cells J. Exp. Med., August 7, 2006; 203(8): 1851 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Kemball, E. D. H. Lee, E. Szomolanyi-Tsuda, T. C. Pearson, C. P. Larsen, and A. E. Lukacher Costimulation Requirements for Antiviral CD8+ T Cells Differ for Acute and Persistent Phases of Polyoma Virus Infection J. Immunol., February 1, 2006; 176(3): 1814 - 1824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sho, K. Kishimoto, H. Harada, M. Livak, A. Sanchez-Fueyo, A. Yamada, X. X. Zheng, T. B. Strom, G. P. Basadonna, M. H. Sayegh, et al. Requirements for induction and maintenance of peripheral tolerance in stringent allograft models PNAS, September 13, 2005; 102(37): 13230 - 13235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Davidson, B Diamond, D Wofsy, and D Daikh Block and tackle: CTLA4Ig takes on lupus Lupus, March 1, 2005; 14(3): 197 - 203. [Abstract] [PDF]
|
|
|
|
|
|
M. J. Nathan, J. E. Mold, S. C. Wood, K. Csencsits, G. Lu, E. J. Eichwald, and D. K. Bishop Requirement for Donor and Recipient CD40 Expression in Cardiac Allograft Rejection: Induction of Th1 Responses and Influence of Donor-Derived Dendritic Cells J. Immunol., June 1, 2004; 172(11): 6626 - 6633. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. van Maurik, B. F. de St. Groth, K. J. Wood, and N. D. Jones Dependency of Direct Pathway CD4+ T Cells on CD40-CD154 Costimulation Is Determined by Nature and Microenvironment of Primary Contact with Alloantigen J. Immunol., February |
,
n = 6) and C3H/HeJ (
, n = 7)
mice receiving CTLA4-Ig and anti-CD40L had MSTs of >100 days,
compared with 11 and 12 days, respectively, in untreated controls (
,
, n = 7). D, B6.AKR
(H-2k) mice maintain the phenotype of the B6 parent strain
following costimulation blockade independently of the H-2 locus.
Treated B6.AKR (H-2k) (
and 
, Naive B6
and C3H.SW mice, respectively. The x-axis displays the
E:T ratio. Percent specific lysis was calculated as described in
Materials and Methods. Error bars are given as the SEM
(n = 3 for all groups).
