|
|
||||||||
Transplantation Biology Research Center, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02129
| Abstract |
|---|
|
|
|---|
chain complementary-determining region 3 (CDR3).
Results show that T cells infiltrating accepted kidneys
(n = 5) express a restricted polymorphism of the
CDR3 length, whereas PBL from the same animal have the polymorphic
distribution of CDR3 lengths found in naive animals; that the skewed
V
repertoire in accepted grafts involved distinct V
subfamilies
in otherwise MHC-identical recipient animals; that GITC clonal
dominance is not caused by immunosuppression because a second kidney,
accepted without drug treatment, exhibits the same TCR V
CDR3
profiles than those detected in the first graft; and that intragraft
clonal dominance intensifies with time, indicating progressive
preeminence of nonaggressive GITC clones. Collectively, these data
represent the first example, in a preclinical model, of the emergence
of nonaggressive intragraft clones, which may be involved in the
induction/maintenance of local tolerance to allogeneic
tissues. | Introduction |
|---|
|
|
|---|
Recent studies from our Research Center have demonstrated that
thymus-dependent cells are essential for rapid and stable tolerance
induction because thymectomized recipients of kidney allografts
experienced rejection episodes following the 12-day course of
cyclosporine (7). Other findings have pointed to the
importance of grafted kidney, and potentially to graft-infiltrating T
cells (GITC)3, in the
establishment of tolerance in vivo by demonstrating a "protective"
effect of this organ over cardiac transplants (8).
Phenotyping of surface markers and molecular analysis of in situ
cytokine gene expression profiles have indicated that intragraft cells
from accepted and rejected kidneys belonged mainly to the
CD8+ T cell subset (1). Differential
genetic programs were, however, involved in GITC from rejected and
accepted organs, as exemplified by low levels of IFN-
and high
levels of IL-10 mRNA within the tolerant GITC population and the
opposite distribution in rejected tissues (9).
The role of the kidney graft itself has been suspected in local regulation of antidonor reactivity because skin grafts bearing kidney-donor class I plus third-party class II Ags were promptly rejected by animals still tolerant to the kidney (10). The role of local regulation in peripheral T cell tolerance has been further strengthened by in vitro experiments comparing the functional properties of PBL and GITC from tolerant animals. Peripheral cells from tolerant animals could generate antidonor lymphocytotoxicity with appropriate stimulation, whereas under similar conditions, GITC from the same animal displayed no antidonor CTL reactivity (11). These findings suggested that GITC included cell subsets able to down-regulate T cell activity, whereas PBL lacked such regulatory properties.
Collectively, these data suggest that GITC play a role in the tolerance
process, which may involve activation of GITC subsets. They also imply
that the selective activation of some intragraft GITC subsets may
correlate with a skewed V
repertoire as measured by
complementary-determining region 3 (CDR3) length polymorphism or
spectratyping (reviewed in Ref. 12). Clonal dominance of
some V
subsets in rejected grafts has been reported in several
models as a manifest expression of alloaggressive T cell clones
(13). No data are presently available, in a preclinical
model, on the function and specificity of nonaggressive population of
/
T cells persisting in accepted grafts. To address this
question, 16 porcine V
segments were cloned and sequenced to derive
V
-specific primers (14). These tools were used to
analyze the TCR
repertoire of PBL and GITC by determining the
distribution of CDR3 lengths. We demonstrate that tolerance to kidney
allografts is associated with the expression of an oligoclonal TCR
repertoire in intragraft nonaggressive T cells. Similar clonal
dominance was also observed in GITC from a second accepted graft, which
was swine leukocyte Ag (SLA)-matched to the first. Finally,
evidence for the importance of the graft in shaping this oligoclonal
repertoire is presented.
| Materials and Methods |
|---|
|
|
|---|
Five- to 7-mo-old pigs were selected from a herd of partially inbred miniature swine raised at our institution. The following strains were used: SLAg recombinant haplotype (class Ic and class IId) as kidney donors for SLAd (class Id and class IId) recipients. Five recipient animals were studied following an immunosuppressive therapy, which started on the day of kidney transplantation and ended in induction of long-term tolerance. Three of these animals received 1013 mg/kg cyclosporin A (animals 11560, 11561, and 11574) administered daily for 12 days through a central venous catheter. Animals 13285 and 13274 were treated with a continuous i.v. infusion of FK506 (0.3 mg/kg/day) for 12 days. Orthotopic kidney grafts were implanted following unilateral nephrectomy of the recipient, as described (15). Open serial kidney biopsies were performed post-transplantation between postoperative day (POD) 8 and POD 30. In one of the recipient animals, the accepted kidney graft was removed at POD 157 and a second SLAg-matched kidney was implanted on the same day without additional immunosuppressive therapy.
RNA extraction and cDNA preparation
Procedures were performed with RNA derived from either 106 PBMC or from kidney biopsies, according to standard procedures (16). The RNA was purified through a 5.7 M CsCl cushion in 25 mM sodium citrate, and first-strand cDNA was synthesized using 2 µg of RNA, Superscript reverse transcriptase (Life Technologies, Rockville, MD), and poly d(T) primer (Life Technologies) in a final volume of 20 µl, according to the manufacturers recommendations.
Design of V
-specific primers and RT-PCR spectratyping
We have previously characterized 16 different TCR V
sequences
in pig (14). V
primers were designed to anneal
specifically to each V
sequence (Tables I
and II
). Specificity of annealing for
each primer was analyzed by RT-PCR on the 16 V
sequence-containing
plasmids and this showed the expected amplification for each V
(C.
Baron, manuscript in preparation).
|
|
subfamilies
Eleven of the 16 TCR
transcripts tested were amplified from
cDNA templates with a common C
primer in combination with one
V
-specific primer (Table I
). Computer sequence analysis, however,
predicted that the common C
primer would form stable duplexes with
the V
2, 5, 8, 10, and 14 primer sequences. Therefore, distinct
C
-specific primers (C
2, 5, 8, 10, and 14) were designed to
selectively amplify the
transcripts, using the combinations listed
in Table II
. One microliter of the first strand cDNA (corresponding to
0.1 µg of total RNA) was amplified in 25 µl of final volume, which
contained 0.2 mM of each dNTP, 1.5 mM MgCl2, 50
mM KCl, 10 mM Tris-HCl (pH 9), 0.8 µM of each primer, and 0.5 U of
the Taq DNA polymerase (Fisher Scientific, Pittsburgh, PA).
Amplification conditions were as follows: denaturation at 94°C for
30 s, annealing at 58°C for 40 s, and extension at 72°C
during 50 s. Thirty cycles were performed and terminated by a
10-min extension time at 72°C.
Primer extension analysis with 32P-labeled oligonucleotide primers
Sixty picomoles of the primer, 50 µCi of
[32P]ATP (NEN, Boston, MA), 15 U of T4
polynucleotide kinase (New England Biolabs, Beverly, MA), and 2.5 µl
of T4 polynucleotide kinase buffer were mixed in a final volume of 25
µl and incubated at 37°C for 1 h. The labeled oligonucleotide
was isolated over a G25 spin column, following the manufacturers
directions (Boehringer Mannheim, Indianapolis, IN). Primer extension
was conducted on the various V
cDNA templates generated above with a
new, 32P-labeled C
primer
(5'-ATCTCCGCTTCCGATGGTTCAA-3') annealing to a sequence internal to that
covered by the previous C
(Tables I
and II
). Some extensions were
generated with two J
primers (5'-CCAGGTGCCTGGGCCAAAGTG-3' and
5'-AGCACAGTCAGCTTGGAACCGTC-3') specific for the J
2.1 and J
2.2
sequences, respectively (14). Two cycles of primer
extension reactions were performed as follows: 2 µl of each PCR
product was added to a mixture containing 0.1 µM of the radiolabeled
primer, 0.2 mM of each dNTP, 0.2 U of Thermosequenase (Amersham
Pharmacia Biotech, Piscataway, NJ), 26 mM Tris/HCl, pH 9.5, and 6.5 mM
MgCl2 in a final volume of 7 µl. The extension
reaction included 30 s at 94°C followed by two amplification
cycles consisting of a denaturation step at 94°C for 30 s,
annealing at 58°C for 40 s, and extension at 72°C for
40 s.
V
CDR3 length polymorphism analysis
Each of the primer extension reactions (7 µl) was terminated by adding 4 µl of a 30-mM EDTA formamide solution. After heating at 75°C for 3 min, each reaction mix was loaded on a 6% polyacrylamide/8-M urea sequencing gel and electrophoresed at 50 W for 25 h depending on the length of the expected band. Autoradiography was performed with exposure times of 624 h. Spectratyping bands were quantified from autoradiographies of the sequencing gels using the Molecular Analyst software (Bio-Rad, Hercules, CA).
| Results |
|---|
|
|
|---|
CDR3 length polymorphism in normal PBL
follows a Gaussian distribution
The TCR
chain repertoire expressed by PBL from four naive
inbred miniature swine was first analyzed. These samples comprised PBL
from tolerant animals 13285, 11560, and 11561 collected before kidney
transplantation, as well as PBL from naive animal 10715. Study of the
CDR3 length polymorphism for the 16 V
subfamilies revealed a
polyclonal distribution of CDR3 lengths as exemplified by the
representative profiles presented in Fig. 1
. The repertoire of the 16 V
subfamilies followed a Gaussian distribution similar to that previously
reported for PBL from other species (12, 17, 18). Similar
Gaussian diversity was observed in samples collected from three
additional naive swine (data not shown).
|
repertoire
Kidney biopsies were harvested 30 days post-transplantation in
five animals that subsequently became tolerant to the class I
mismatched allograft. At that time, accepted allografts still
demonstrated a marked infiltration of recipient-derived mononuclear
cells, as shown in previous studies (19). The pattern of
the TCR
chain CDR3 length polymorphism of GITC from animal 13285 is
shown as a representative example in Fig. 2
. All of the 16 V
subfamilies studied
were successfully amplified, suggesting that cells entering the
accepted kidney were not selected on the basis of the type of TCR V
they expressed. Contrary to the fairly uniform Gaussian profile
observed in PBL, several V
subfamilies showed restricted patterns of
CDR3 length polymorphism (V
8, 14, 20, 24, and 100 in Fig. 2
). This
finding indicated that there was clonal dominance among T cells
expressing particular V
subfamilies within the graft. The other V
subfamilies infiltrating this biopsy sample presented typical
polyclonal profiles similar to those observed for normal PBL. Biopsies
from the four other accepted grafts also demonstrated clonal dominance
among cells infiltrating grafts at POD 29 and 30 (Fig. 3
). This clonal dominance did not,
however, affect the same subset V
subfamilies in each accepted
graft. Infiltrating T cells from the five accepted kidneys exhibited
clear TCR dominance only for a limited number of V
(6 of the 16
studied). All of the GITC samples showed selected TCR profiles for
V
14 at days 29 and 30 posttransplantation as compared with that of
PBL, with a marked oligoclonal dominance for animals 11574 and 13274
(Fig. 3
, bottom). Other V
CDR3 lengths such as V
10
were restricted in three animals, whereas they were polymorphic in the
other grafts (Fig. 3
).
|
|
Kidney samples collected at POD 8, 18, and 29 for animal 11574 and at 8 days posttransplantation and 29 for animal 13274 were analyzed by spectratyping to determine the evolution of clonal dominance.
The evolution of the CDR3 length polymorphism profiles is presented in
Fig. 4
. V
dominance was clearly
obvious as early as POD 8 in animal 13274 for the V
1, 6, and 14
subfamilies. The same CDR3 length profile distribution, with similar
dominance, was detected in day 29 GITC. A similar trend was observed in
the GITC of recipient animal 11574, which demonstrated comparable
skewed profiles at days 18 and 29 for V
100 (Fig. 4
). To confirm the
progressive T cell clonal dominance among GITC using the V
14 gene
segment in the kidney of animal 13274, V
14-J
2.1 and
V
14-J
2.2 spectratypings were performed (Fig. 4
, last two
columns). As expected, the CDR3 length profile in those particular V-J
combinations was restricted to few peaks by day 8 posttransplantation,
a trend that accentuated with time and became in essence
"monoclonal" by day 30 for V
14-J
2.1.
|
To evaluate the importance of the microenvironment surrounding the
GITC in the establishment of clonal T cell dominance, we compared TCR
repertoires expressed by PBL and GITC collected on the same day. Fig. 5
shows the CDR3 length profiles obtained
for PBL and kidney samples at POD 30 in three different tolerant
animals. The CDR3 length polymorphism expressed by PBL and by GITC
exhibited different patterns: most of the V
subfamilies that showed
a restricted CDR3 length polymorphism in GITC showed a normal Gaussian
distribution as in PBL that were collected on the same day. This
finding indicated that GITC expressed a V
repertoire distinct from
that of the peripheral T cell pool.
|
Previous studies from our laboratory have demonstrated that
animals tolerant to class I mismatched kidney grafts also accepted
second kidney grafts, SLA-matched to the first, without the requirement
for additional immunosuppression (19). In this setting,
the second transplant always contains lower levels of GITC as compared
with those observed in the first kidney (reviewed in Ref.
5). To determine whether the T cell clonal dominance we
observed in GITC could have been in part due to pharmacological
immunosuppression with cyclosporin A, we analyzed the V
repertoire
of T cells infiltrating a second graft in animal 13285 and we compared
it with the one expressed in the first kidney transplant. The second
kidney was implanted at the time of the removal of the first graft on
POD 157. Kidney biopsies from the first and second grafts taken on POD
30 were compared. The results of spectratyping for the three V
subfamilies that showed a clear clonal dominance in the first graft are
presented in Fig. 6
A.
Strikingly, there was clonal dominance in the same three V
subfamilies in the GITC from the second graft. These findings were
confirmed by V
14-J
2 spectratyping of the same samples (Fig. 6
B), which clearly demonstrated clonal dominance. The
results also emphasized the striking resemblance between the profiles
of GITC from a first graft and those of cells infiltrating a second
accepted graft in a long-term tolerant animal.
|
| Discussion |
|---|
|
|
|---|
chain
repertoire usage under normal physiologic conditions (24)
and various pathological situations, including graft-vs-host disease
(25), allografts (17, 26, 27), response to
viral Ags (28), rheumatoid arthritis (29),
and malignancies (18). In the present study, this
technology was used to evaluate, in a preclinical model, the TCR V
repertoire of intragraft T cells and to assess potential correlates
with in situ T cell tolerance. Standard approaches to T cell
specificity of GITC have often involved cellular purification and/or in
vitro cultures that have frequently biased the initial
phenotype/function of these cells (30). In contrast, the
use of V
spectratyping, also known as immunoscope (12),
appears more appropriate because it allows a fine RT-PCR analysis of
intragraft T cell V
dominance without modifying their intrinsic
properties.
The present spectratyping study demonstrates V
CDR3 length
polymorphism dominance of cells infiltrating accepted kidney grafts
(Figs. 2
and 3
) that is independent of the repertoire of circulating T
cells (Fig. 5
). V
TCR dominance of GITC has been previously reported
in rats in which tolerance-inducing protocols involved donor-specific
transfusion of blood cells (17). In this model, however,
donor-specific transfusion performed before allogeneic heart
transplantation induced similar clonal expansion in both GITC and
circulating T cells collected at the same time points. This later
finding prevents further conclusions on the direct involvement of these
cells in the induction/maintenance of local tolerance in that model. In
contrast to the rat model, our tolerance-induction protocol in swine is
particularly suited to studying potential correlations between
preponderant V
CDR3 length polymorphism in GITC and in situ
tolerance because it relies only on transient immunosuppressive
treatment (6) that does not affect the overall TCR
repertoire of circulating cells.
The clonal ascendance observed in GITC from accepted kidney transplants
may be due to trivial causes such as small biopsy sizes or immune
reactivity to alloantigens. However, our results demonstrate that the
same clonal-specific TCR
transcripts, permanently overexpressed in a
first graft (Fig. 4
), reappeared in a second graft accepted by a
long-term tolerant animal (Fig. 6
). These data strongly argue against
an effect of the sampling size and instead suggests that the GITC
clonal dominance is present throughout the kidney graft parenchyma.
It is reasonable to assume that the antigenic triggers to clonal
dominance in the tolerant GITC population are the allogeneic MHC class
I Ags because our transplantation model makes use of MHC class
I-mismatched, class II-matched renal grafts. This common antigenic
disparity did not, however, promote dominance of the same V
TCR
clones in each accepted graft, a possible consequence of the genetic
diversity of the TCR loci in these animals. Selective TCR V
repertoires have been detected on alloreactive T cells in graft
rejection responses (26, 27, 31). Likewise, preliminary
data from our laboratory indicate that clonal dominance occurs among
GITC-invading class I disparate kidneys rejected by untreated recipient
animals. However, such clonal selection implicates different V
clones from those expending in tolerated grafts. Therefore, clonal
dominance among GITC is not exclusive to either tolerance or
rejection, but rather is a consequence of the recipient immune response
to alloantigens. Compelling evidence, however, indicates that the
selected sets of GITC in rejected and accepted grafts are functionally
different. First and foremost, MHC class I-mismatched kidneys,
implanted in animals treated with the short course of cyclosporin A,
are permanently accepted without any histopathological sign of chronic
rejection, although they are infiltrated by recipient T cells (reviewed
in Ref. 5). Such enduring survival demonstrates that cells
infiltrating tolerated grafts are not alloaggressive, a feature in
agreement with the absence of T cell activation and cytotoxicity
against donor-type Ags in these animals (6). Secondly, the
time course study provides evidence on the progression of clonal
dominance resulting in the predominance of few GITC subsets (Fig. 4
),
suggesting that the elected T cell clones are progressively recruited
for their nonalloaggressive function(s). Finally, the reappearance in a
second graft implanted in a long-term tolerant animal of similar GITC
V
profiles as those detected in the first graft (Fig. 6
) strongly
supports the view that the clonal emergence of these particular T cell
clones is related to the intragraft tolerance process. Considering
these data all together, we would suggest that although initially
recruited by allogeneic activation, T cell subsets invading accepted
grafts develop into nonalloaggressive cells that are likely recruited
in tolerance-associated processes. Because previous studies have shown
that tolerance, in our model, is established through nondeletional
mechanisms (10, 32), we may envision that part of the
activated GITC pool includes a regulatory subset dedicated to the
down-modulation of alloaggressive cytotoxic clones, as described in
other species (4, 33, 34).
Clonal expansion and/or up-regulation of certain TCR-rearranged V
transcripts may account for the preponderance of some tolerogenic GITC.
Expansion of T cell clones has been reported in correlation with the
induction of oral tolerance (35). Experiments are in
progress to test this hypothesis in pigs, although the in vivo labeling
of dividing cells by standard dye or nucleoside substitutes appears
difficult to assess in this large animal model. Alternatively, T cell
dominance may result from transcriptional up-regulation of particular
rearranged V
genes in a way similar to that reported for
nonproliferating T cells that activated TCR
gene transcription upon
TCR triggering (36, 37).
To our knowledge, this is the first report, in a preclinical model, demonstrating that accepted grafts develop a selective and persistent pool of activated T cells. Further studies will be required to establish the functional impact of this cell population(s) on the induction and/or maintenance phases of immune tolerance to vascularized grafts.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Christian LeGuern, Transplantation Biology Research Center, Massachusetts General Hospital, MGH-East, Building 149-9019, 13th Street, Boston, MA 02129. E-mail address: leguern{at}helix.mgh.harvard.edu ![]()
3 Abbreviations used in this paper: GITC, graft-infiltrating T cells; CDR3, complementary-determining region 3; SLA, swine leukocyte Ag; POD, postoperative day. ![]()
Received for publication February 14, 2001. Accepted for publication August 3, 2001.
| References |
|---|
|
|
|---|
variable region used by T cells infiltrating kidney transplants. J. Immunol. 166:2589.
rearrangement in heart allografts. J. Immunol. 157:1250.[Abstract]
-chain genes by polymerase chain reaction (PCR) DNA sequencing and automated high-resolution PCR fragment analysis. Blood 86:3930.
TCR repertoire of naive mouse splenocytes. J. Immunol. 164:5782.
T cell receptor repertoires against hen egg white lysozyme (HEL) in nontransgenic versus HEL transgenic mice. J. Exp. Med. 180:861.
T cell receptor diversity. Science 286:958.
repertoire of T cells infiltrating long-term rejected kidney allografts. J. Immunol. 164:1553.This article has been cited by other articles:
![]() |
A. Miyagawa-Hayashino, T. Tsuruyama, H. Egawa, H. Haga, H. Sakashita, T. Okuno, S. Toyokuni, K. Tamaki, H. Yamabe, T. Manabe, et al. FasL Expression in Hepatic Antigen-Presenting Cells and Phagocytosis of Apoptotic T Cells by FasL+ Kupffer Cells Are Indicators of Rejection Activity in Human Liver Allografts Am. J. Pathol., November 1, 2007; 171(5): 1499 - 1508. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |