ERRATUM
The reprinted article appearing below was originally published in the March 15, 2006 issue of Transplantation
(Transplantation 2006; 81(5): 726 –735). Due to an error on the part of the publisher, the original figures appeared in black
and white rather than color. The article is being reprinted here in its entirety, with the correction of the figures now
appearing in color. We regret any difficulties this has caused to the authors of the paper and to the scientific community.
The Replacement of Graft Endothelium by RecipientType Cells Conditions Allograft Rejection Mediated
by Indirect Pathway CD4⫹ T Cells
Yota Kapessidou, Claude Habran, Sofia Buonocore, Véronique Flamand, Luc Barvais,
Michel Goldman, and Michel Y. Braun
Background. Whereas the participation of alloreactive T cells sensitized by indirect allorecognition in graft rejection is
well documented, the nature of recipient antigen presenting cells recognized by indirect pathway CD4⫹ T cells within
the graft has yet to be identified. The purpose of this study was to determine the role played by graft endothelium
replacement in the immune recognition of cardiac allografts rejected by indirect pathway CD4⫹ T cells.
Methods. Transgenic RAG2⫺/⫺ mice expressing I-Ab-restricted male antigen H-Y-specific TcR were studied for their
capacity to reject H-2k male cardiac allografts. Chronic vascular rejection in this model was due to the indirect recognition of
H-Y antigen shed from H-2k male allograft and presented by the recipient’s own I-Ab APC to transgenic T cells.
Results. Immunohistochemical analysis of rejected grafts revealed the presence of numerous microvascular endothelial
cells (EC) that expressed recipient’s I-Ab MHC class II molecules. This observation suggested that graft endothelium
replacement by I-Ab-positive cells of recipient origin could stimulate the rejection of male H-2k graft by I-Ab-restricted
H-Y-specific T cells. To investigate further this possibility, hearts from H-2b-into-H-2k irradiation bone marrow (BM)
chimera were transplanted in transgenic recipients. A direct correlation was observed between the presence of I-Abpositive EC within myocardial microvessels and the induction of acute rejection of chimeric H-2k male cardiac allografts transplanted in transgenic recipients.
Conclusions. We conclude that graft endothelium replacement by recipient-type cells is required for the rejection of
cardiac allograft mediated by indirect pathway alloreactive CD4⫹ T cells.
Keywords: Endothelium, Vasculogenesis, Rejection, Inflammation, Cardiac vasculopathy.
(Transplantation 2006;81: 582–591)
he pathogenesis of chronic vascular rejection in solid organ transplantation remains ill-defined. Current opinion
suggests that it is primarily an immunological process, which
may be enhanced by nonimmunological injury caused by
ischemia and reperfusion (1–3). There is mounting evidence
suggesting that graft vasculopathy is primarily driven by
CD4⫹ T lymphocytes (4 –7). The majority of alloreactive
CD4⫹ T cells can directly recognize alloantigens expressed by
the cells of transplanted organs. CD4⫹ T cells can also recognize and respond to donor antigens processed by recipient
T
This study was supported by the Roche Organ Transplantation Research
Fund. Y.K. is recipient of a bursary from the Fondation Erasme.
Institute for Medical Immunology, Faculty of Medicine, Université Libre de
Bruxelles, Brussels, Belgium.
Address correspondence to: Michel Y. Braun, Ph.D., Experimental Transplantation Unit, Institute for Medical Immunology, 8 rue Adrienne Bolland, 6041 Gosselies, Belgium.
E-mail: mbraun@ulb.ac.be
Received 16 November 2004. Revision requested 20 December 2004.
Accepted 19 August 2005.
Copyright © 2006 by Lippincott Williams & Wilkins
ISSN 0041-1337/06/8105-582
DOI: 10.1097/01.tp.0000184444.93108.d1
582
antigen-presenting cells (APC) and indirectly presented to T
cells as peptides in the binding groove of recipient class II
major histocompatibility complex (MHC) molecules (8, 9).
The recent observation that the absence of secondary lymphoid tissue results in the failure to produce an alloimmune
response in fully vascularized, fully allogeneic cardiac transplants, implies that Ag presentation by dendritic cells (DC)
migrated in the host lymphoid tissue is required for alloreactive T cells to be activated (10). Direct activation of alloreactive T
cells is thought to arise predominantly during the acute rejection
of allograft, early after transplantation, and is probably limited
by the longevity of donor “passenger” APC that migrate from the
graft in the recipient lymphoid organs (11). On the contrary,
presentation of donor antigens by recipient DC (indirect pathway) would be carried out permanently, at least as long as the
allograft persists. Thus, it appears probable that alloreactive T
cells sensitized by the indirect pathway be mostly responsible for
the mediation of chronic rejection (12–14).
Following antigenic priming within secondary lymphoid organs, alloreactive T lymphocytes will carry out their
effector functions into tissues by migrating through postcapillary venules. Although the sequence of molecular events
Transplantation • Volume 81, Number 5, March 15, 2006
Kapessidou et al.
© 2006 Lippincott Williams & Wilkins
leading to transendothelial migration and tissue infiltration
have been relatively well established for T cells (15), the mechanisms that allows accumulation of donor specific T lymphocytes into the allograft has yet to be clarified. Endothelial cells
(EC) are known to express MHC class II molecules in vitro
and in vivo upon activation (16 –18) and recent observations
indicate that antigen presentation by the endothelium enhances T lymphocyte diapedesis at the site of inflammation,
leading to localized recruitment of antigen-specific T cells
(19, 20). Thus, during allograft rejection, graft EC would play
a major role in T cell trafficking allowing alloreactive CD4 T
cells to exit the blood and accumulate at the site of rejection.
Though such mechanism explains graft recruitment of direct
pathway alloreactive CD4 T cells, it should not apply to indirect pathway T cells because of the molecular nature of their
antigen recognition.
Until recently, it was believed that the pathogenesis of
transplant arteriosclerosis was initiated by immune damage
to the endothelium followed by smooth muscle cell migration
and proliferation in the intima (21). According to this concept, endothelium of the graft remained of donor origin and
vascular rejection was considered to originate from grafted
tissue (22). However, this conventional hypothesis is challenged by recent studies that have addressed the question of
endothelial cell (EC) replacement in solid-organ transplantation. It has been shown in the recent past years, that bonemarrow– derived precursor elements have the potential to induce vasculogenesis of ischemic tissues (23–25), including
transplanted organs, and to participate in the neoendothelium and neointima in postangioplasty arteries (26). EC precursors are present in adult bone marrow and have properties
of hemangioblasts (24, 27). In human transplantation, part of
EC in the blood vessels of transplanted organs were found to
be of recipient origin (28 –30). These cells were observed in
poorly functioning and severely damaged grafts and were
particularly detected among patients whose graft suffered
chronic vascular rejection. It was postulated that, in vascularized allografts, EC of the recipient can replace those of the
donor and that this replacement was associated with rejection. This concept was strengthened by Heeger and colleagues
recent observation, though made in a murine model of skin
grafting, involving adoptive transfer of TcR-transgenic H-Yspecific CD8⫹ T cells, that indirect effector pathways required
processing and presentation of the donor H-Y antigen by recipient endothelium present within the graft (31). We undertook the present study to investigate whether recipient-derived EC that have replaced donor cells within the graft could
represent the main cell target recognized by indirect pathway
alloreactive CD4⫹ T cells for mediating the rejection of vascularized allografts.
MATERIALS AND METHODS
Animals
The study protocol was reviewed and approved by the
Committee on Ethics on Animal Experiments of the Faculty
of Medicine of the Université Libre de Bruxelles, and the experiments were conducted according to the Guidelines of the
Administration de la Santé Animale et de la Qualité des
Produits Animaux of the Ministère de l’Agriculture of Belgium.
583
Male and femaleCBA/Ca (H-2k) and C57BL/6 (B6) (H2 ) mice between 4 and 8 weeks were purchased from Harlan
Netherlands (Horst, The Netherlands). Female RAG2-/Marilyn mice, which are transgenic for a TCR (V␣1.1, V6)
specific for the H-Y peptide NAGFNSNRANSSRSS presented
by I-Ab, have been previously described (32). Male and female
H-2k or H-2b, recombinase 2 (RAG2)-/- common gammachain (␥c)⫺/⫺ double-deficient mice were obtained from J.
Di Santo (Pasteur Institute, Paris, France). All animals were
maintained in isolators in a pathogen-free animal environment at the Faculty of Medicine (Université de Bruxelles).
b
Adoptive Transfer and Skin Transplantation
TCR transgenic spleen CD4⫹ T cells were purified from
Marilyn mice using anti-CD4 magnetic beads and the VarioMacs system from Miltenyi Biotec (Paris, France). Highly purified (⬎95%) TCR transgenic CD4⫹ T cells (10⫻106 per
mouse) were injected intravenously into female RAG2⫺/⫺
␥c⫺/⫺ H-2b or H-2k recipients. One day later, full-thickness
tail skin allografts from male or female RAG2⫺/⫺ ␥c⫺/⫺ H-2b
or H-2k donors were placed on the lateral flank of the reconstituted recipients as previously described (33). Graft survival
was assessed daily and rejection was defined by complete destruction of the skin. Skin grafting was carried out under strict
sterile condition.
Heart Transplantation and Immunization to
Donor Antigen
Abdominal vascularized heterotopic heart transplants
were performed as previously documented and palpated daily
for evidence of a heartbeat (34). Rejection was defined as a
complete cessation of palpable cardiac contraction and was
confirmed by direct visualization after laparotomy. Heart
grafting was carried out under strict sterile condition. All
grafts were collected at the time of rejection and routine histology and immunohistochemistry were performed.
In some experiments, T cell priming was achieved by
intraperitoneal injection of bone marrow-derived dendritic
cells (DC) as previously described (34). Briefly, bone marrow
cells were collected and cultured for 10 days in medium containing recombinant murine GM-CSF (20 ng/ml) (kindly
provided by K. Thielemans, Vrije Universiteit Brussel, Brussels, Belgium). For the induction of acute rejection of male
H-2b hearts by Marilyn recipients, male B6 DC (4⫻105 per
mouse) were injected in Marilyn mice one day before transplantation (34).
Bone-Marrow Chimeras
Tibia and femora bone marrow cells were isolated from
RAG2⫺/⫺ ␥c⫺/⫺ H-2b female donors as previously described
(35) and injected intravenously (15⫻106 per mouse) in
RAG2⫺/⫺ ␥c⫺/⫺ H-2k male recipients previously exposed to a
single 15-Gy dose of radiation from a 137Cs source (Mark I
gamma irradiator; J. L. Shepherd and Associates, Glendale,
CA). Recipient animals were maintained in pathogene-free
conditions. Full chimerism, was assessed on blood samples
using fluorescent flow cytometry (FACScalibur, Becton Dickinson, Erembodegem-Aalst, Belgium) with FITC-conjugated
mouse antimouse H-2Kb monoclonal antibody (clone AF688.5; BD Biosciences Pharmingen, Erembodegem-Aalst, Belgium) (data not shown). Heart chimerism was assessed at
584
predetermined time points (days 10, 20 and 70) by immunohistochemistry and immunofluorescence. Chimeras were
used as donors of heart transplant 55 days after bone marrow
engraftment.
Histology and Immunohistochemical Staining
Organs were embedded in OCT compound (Sakura
Finetek, Torrance, CA), frozen in liquid nitrogen and cryopreserved at ⫺80°C no more than 48h. Frozen organs were
then sectioned at 8 m sections, fixed in acetone and stored at
⫺20°C until use. Frozen sections were first rehydrated for 5
min with 1% PBS buffer solution (pH: 7.3). After blocking of
endogenous peroxidase activity with 0.3% hydrogen peroxide in 1% PBS, for 10 min, the tissue sections were incubated
with 0.1% biotin (Sigma, Bornem, Belgium) for 20 min and
then with the primary antibody (1:50 dilution) overnight at
4°C in a humidified chamber. On the next day, the sections
were incubated with a biotinylated secondary antibody (1:200
dilution) for 30 min. Development of immunoperoxidase activity was accomplished after signal amplification by HRPconjugated streptavidin (streptavidin-HRP, Dako, Heverlee,
Belgium). Sections were counterstained with hematoxylin,
dehydrated with ethanol and mounted for analysis. Positive
(C57BL/6 normal heart) and negative (CBA/Ca normal
heart) controls were included in every staining experiment.
Monoclonal antibodies used for immunohistochemistry
were purchased from BD Biosciences Pharmingen: rat antimouse CD105 (clone MJ7/18), rat anti-mouse I-A/I-E (clone
M5/114), mouse anti-mouse I-Ab (clone 25-9-17), mouse anti-mouse I-Ek (clone 14-4-4S) and Armenian hamster antimouse CD3 (clone 145-2C11). Secondary antibodies were biotinylated goat anti-mouse Ig, biotinylated mouse-absorbed
goat anti-rat Ig and biotinylated mouse anti-hamster Ig cocktail from BD Biosciences Pharmingen.
Routine hematoxylin and eosin staining was conducted
on paraffin-embedded fragments (5 m) from transplanted
hearts or skins. Criteria evaluated included edema, degree of
infiltration, and myocyte death as evidenced by nuclear disintegration, loss of striation and alteration in color. To study
cardiac vasculopathy of rejected grafts, paraffin embedded
sections were coloured with Weigert solution (Klinipath,
Geel, Belgium) for two hours, and counterstained with hematoxylin and eosin.
Immunofluorescence Staining
Eight-micron cryostat frozen sections of spleens and
hearts (naı̈ve and chimeric) removed from B6 mice, were first
rehydrated in 1% PBS buffer solution (pH: 7,3) for 5 min. In
the different experimental settings, sections were then incubated overnight at 4°C with primary rat anti-mouse CD105
(clone MJ7/18), rat anti-mouse I-A/I-E (clone M5/114), biotin-conjugated mouse anti-mouse I-Ab (clones 25-9-17 et
AF6-120.1), biotin-conjugated mouse anti-mouse I-Ab and
I-Ak, Armenian hamster anti-mouse CD11c (clone HL3) or
rat anti-mouse CD11b (clone M1/70) antibodies, all from BD
PharMingen and in 1:50 dilution. After washing in PBS (twice
for 5 min), the slices were incubated for 30 min at room
temperature with FITC-conjugated goat anti-rat or goat antiArmenian hamster IgG (in 1:100 dilution; both from Jackson
ImmunoResearch Europe, Ltd., Cambridgeshire, UK). Labeling for MHC class II molecule expression was revealed after
Transplantation • Volume 81, Number 5, March 15, 2006
signal amplification of mouse anti-mouse antibodies for 30
min with biotin-conjugated goat anti-mouse polyclonal Ig
(BD PharMingen), followed by TEXAS RED-Streptavidin (in
1:200 dilution, for 30 min; Vector Laboratories). All secondary antibodies were mouse-absorbed and did not bind mouse
heart sections (data not shown). For two color immunofluorescence staining, slices were incubated overnight at 4°C with
rat anti-mouse CD105, or Armenian hamster anti-mouse
CD11c or rat anti-mouse CD11b in combination, for each
experiment, with the three different anti-MHC class II antibodies tested (see above). Control isotype antibodies were
included in every staining experiment (BD Pharmingen).
Coverslips were mounted with antifade VECTASHIELD
HardSet Mounting Medium (Vector Laboratories). Sections
were analyzed with immunofluorescence Nikon Eclipse 80i
microscope and recorded by Nikon Digital Camera
DXM1200F. For two color immunofluorescence imaging recorded fluorophore images were merged using Adobe Photoshop (version 8.0.1).
T-cell Stimulation with Endothelial Cells
Endothelial cells were derived from thoracic aorta (16).
Briefly, neutralized collagen extracellular matrix was prepared of a three-dimensional type I rat tail collagen gel (4,08
mg/ml) (BD Biosciences Pharmingen), aliquoted into six
wells of a 24-well plate (500 l per well), and equilibrated
overnight with complete endothelial cell medium (1 ml per
well) consisting of RPMI medium containing 20% heat-inactivated fetal calf serum (FCS), 2-mercaptoethanol (1 l/ml),
gentamicin (1 l/5 ml) and endothelial cell growth supplement (50 g/ml) (Becton Dickinson). On the next day, the
medium was removed and small pieces of the sectioned thoracic aorta from B6 or CBA/Ca male or female mice were
plated, endothelial side down, onto the collagen gel. After
36h, the cultures were supplemented with 1 ml per well of the
complete endothelial cell medium. Within 3-5 days, endothelial cells were passaged (P1) onto tissue culture–treated plastic flasks using 0.3% collagenase H solution (Sigma, Bornem,
Belgium). Further passaging was carried out with 5 mM
EDTA. Phenotype analysis of cultured endothelial cells was
carried out by flow cytometry. Early passage cells (P2-P5)
were washed with cold PBS/2% FCS and then incubated (4⫻
105 per sample) with rat anti-mouse CD105 (clone MJ7/18)
or rat anti-mouse I-A/I-E (clone M5/114) primary antibodies. FITC-conjugated mouse-absorbed goat anti-rat Ig antibodies were then used as secondary antibodies. Immunolabelled cells were analyzed on a FACScalibur (Becton
Dickinson).
In experiments that tested the T cell stimulatory capacity of cultured endothelial cells, 20⫻104 H-2b or H-2k male or
female endothelial cells (P3-P4) were plated in 48-well plates,
stimulated with recombinant murine IFN-␥ (500 U/ml) for
48h and washed as described (16). Marilyn spleen CD4⫹ T
cells purified from animals preimmunized 7 days earlier with
irradiated (20 Gy) B6 whole spleen cells injected intraperitonealy (10⫻106 per mouse). T cells were then added to the
wells (5⫻105 per well) with and without peptide (final concentration of 10 M). Culture medium consisted of RPMI
1640 (Biowhittaker, Petit-Rechain, Belgium) supplemented
with 20 mM Hepes, 2 mM L-glutamine, 1 mM nonessential
amino acids, 10⫺5 M 2-ME and 5% FCS (Biowhittaker). Cul-
Kapessidou et al.
© 2006 Lippincott Williams & Wilkins
tures were kept at 37°C in 5% CO2 atmosphere. After 72 hr,
they were mixed gently and 100 l aliquots were transferred
to 96-well round-bottomed plates, pulsed with [3H]thymidine for 8 h, collected and counted. Culture supernatants
were assessed for their content of IFN-␥ by ELISA according
to manufacturer’s protocol (R&D Systems, Oxon, United
Kingdom). The peptide H-Y (NAGFNSNRANSSRSS) used in
this study was synthesized by Neosystem (Strasbourg,
France).
Statistical Analysis
Statistical analysis was performed using the two-tailed
Mann Whitney non-parametric U test. Results are presented
as mean ⫾ SEM.
RESULTS
MHC Class II Expression by Graft Endothelial
Cells Correlates with the Recruitment of Specific
CD4ⴙ T Cells and Graft Rejection
We have recently developed a unique model of experimental transplantation using transgenic mice (Marilyn mice)
that express a TcR transgene which recognizes the male antigen H-Y (encoded by the Dby gene) in an I-Ab-restricted
fashion. The transgenic T cells are not alloreactive to the H-2k
haplotype, because Marilyn TcR transgene was isolated from
spleen cells of (H-2k X H-2b)F1 female mice and Marilyn T
cells are not activated when adoptively transferred into immunodeficient recombinase-activating gene-2 (RAG2)⫺/⫺
common ␥-chain⫺/⫺ double-mutant H-2k male or female
mice (33, 36). Surprisingly, in addition to rejecting H-2b male
skin, Marilyn female recipients were able to reject allogeneic
skin from H-2k males (33). One explanation for this finding is
that primed CD4⫹ T cells infiltrating the graft could be restimulated by recipient I-Ab-expressing APCs infiltrating the
graft (such as macrophages or DC) that had captured and
processed H-Y antigen produced by the graft. T cells and
recipient APCs would then be able to destroy the graft
through a nonspecific effector pathway such as cytokine-mediated cytotoxicity or delayed-type hypersensitivity reaction
(37, 38). Another explanation for this apparent lack of specificity of the effector response could be proposed from the
observation that subdermal blood vessels supplying the skin
graft derive from the recipient and not from the donor
(39, 40). Antigen presentation by vascular endothelium has
been shown to lead directly to the recruitment of antigenspecific CD4⫹ T cells at the site of inflammation in several
animal models of autoimmunity or infection (41– 43). In human, migration of resting memory CD4⫹ T cells across EC
monolayers expressing the relevant Ag enhanced two- to
4-fold the frequency of Ag-specific T cells in the migrated T
cell population (44). Thus, T cell recruitment and destruction
of H-2k male skin by Marilyn recipients might have occurred
as the result of the direct interactions between Marilyn T cells
and recipient I-Ab-expressing vascular endothelial cells displaying H-Y peptides derived from the H-2k allograft. In order to test whether antigen-specific interaction between T
cells and graft endothelium was sufficient to bring about graft
rejection, we set up experiments where presentation of transplantation antigen was carried out exclusively either by recipient cells (indirect pathway) or by the allograft itself (direct
585
pathway). Highly purified spleen CD4⫹ T cells (⬎95 %) from
RAG2-deficient Marilyn mice were isolated and then injected
iv into immunodeficient RAG2⫺/⫺ ␥c⫺/⫺ double-deficient
female mice expressing H-2k or H-2b molecules. Reconstituted animals were then observed for their capacity to reject
RAG2⫺/⫺ ␥c⫺/⫺ H-2b or H-2k male skin grafts. As expected,
all H-2k male allografts were rejected by Marilyn T cells activated by recipient APCs alone (Table 1). As seen in Table 1, all
the reconstituted H-2k recipients were able to reject acutely
H-2b male skin. In this setting, since i) mouse T cells do not
express MHC class II molecules and ii) Marilyn T cells do not
recognize H-2k molecules (33), rejection was caused by T cell
recognition of graft-derived cells alone and could not be the
result of T cell stimulation by recipient H-2k APC infiltrating
the graft. Graft rejection depended on the expression of H-2b
molecules by the graft since H-2k male skin was not rejected
by T cell-reconstituted H-2k recipients. Moreover, failure to
reject female H-2b skin demonstrated that rejection was antigen-specific. In an attempt to identify the cell population responsible for the antigen-specific T cell recruitment in H-2b
grafts rejected by reconstituted H-2k recipients, we analyzed
the specific expression of donor MHC class II molecules I-Ab
on skin graft sections. As depicted in Figure 1A and B, in H-2b
grafts rejected by H-2k recipients, I-Ab expression was restricted to the endothelium of dermis blood vessels. Despite
the use of sensitive signal-amplifying techniques, we did not
detect I-Ab expression on graft-infiltrating cells of rejected
RAG⫺/⫺ ␥c⫺/⫺ H-2b skin. Numerous cells infiltrating these
grafts, however, expressed I-Ek MHC class II molecules and,
thus, were derived from H-2k recipients and could not be
recognized by Marilyn T cells (Fig. 1C). Taken together these
results supported the concept that cognate recognition of the
graft endothelium by T cells was sufficient for the migration
of antigen-specific T cells into the graft that led to rejection.
TcR Trangenic CD4ⴙ T Cells Respond to
Endothelial Cell Stimulation
There is conflicting evidence about the capacity of endothelial cells to act as antigen presenting cells to stimulate T
cells. Recent data suggest that cytokine-activated EC can act
as “semi-professional” APC and have the ability to enhance T
cell responsiveness and cytokine production without eliciting
full T cell activation (45, 46). In our model, though MHC
class II expression by EC appeared to be sufficient for rejection, evidence for cognate interaction between EC and Marilyn T cells was lacking. We therefore analyzed in vitro the
immune recognition of mouse EC by Marilyn T cells. Mouse
TABLE 1. Graft-dependent (direct) or donor-dependent
(indirect) antigen presentation stimulates the acute
rejection of male skin grafts by Marilyn T cells
n
Recipients
Donors
11
k
b
H-2
H-2 male
8
H-2k
H-2k male
2
2
2
H-2b
H-2b
H-2k
H-2b male
H-2k male
H-2b female
Survival (days)
11, 11, 12, 12, 12, 12,12,
14, 14, 18, 18
47, 53, 69, ⬎100, ⬎100,
⬎100, ⬎100, ⬎100
11, 11
12, 12
⬎100, ⬎100
586
FIGURE 1. Expression of MHC class II molecules in H-2b
male skin grafts rejected by RAG2⫺/⫺ ␥c⫺/⫺ double-deficient H-2b or H-2k female mice reconstituted with H-Y-specific Marilyn CD4 T cells. (A) Sections of skin grafts from
H-2b (a, b) or H-2k (c) male donors that were transplanted
onto H-2b (a) or H-2k (b, c) RAG2⫺/⫺ ␥c⫺/⫺ recipients reconstituted with Marilyn T cells, were immunostained with
antibodies specific for mouse I-Ab MHC class II molecules
(brown staining). Original magnification: ⫻400. (B) Expression of I-Ab MHC class II molecules in H-2b skin grafted onto
Marilyn T cell-reconstituted RAG2⫺/⫺ ␥c⫺/⫺ H-2k recipients was limited to graft endothelium. Original magnification: ⫻1000. (C) Count of infiltrating cells expressing I-Ab
or I-Ek MHC class II molecules in skin grafts from H-2b male
donors and rejected by H-2b (b/b) or H-2k (b/k) RAG2⫺/⫺
␥c⫺/⫺ recipients reconstituted with Marilyn T cells. Controls included H-2k male skin grafts onto reconstituted H-2k
female recipients (k/k).
vascular EC were isolated from thoracic aorta and expanded
in endothelial cell growth factor-containing medium (Fig. 2A
and 2B). Isolated EC expressed the specific angiogenic endothelial cell marker CD105 (endoglin) (47, 48) and cell surface
MHC class II molecules (Fig. 2C). Because EC are poor stimulators for T cell priming (46), in vitro experiments addressing the capacity of EC to stimulate T cells were carried out
with previously primed T cells. Marilyn T cells were purified
from the spleen of mice preimmunized with GMC-SF-derived male I-Ab DC. In mixed cultures with primed Marilyn T
cells, male H-2b, not H-2k, EC specifically stimulated T cell
proliferation (Fig. 2D), demonstrating the specific immune
recognition of EC expressing MHC class II I-Ab molecules by
Marilyn T cells. Proliferation and cytokine production by purified T cells was not the result of antigenic stimulation by
I-Ab DC, or other APC, contaminating the responder cell
populations since adding an excess of H-Y peptides in cultures with H-2k EC did not stimulate proliferation nor cytokine production (Fig; 2D).
Transplantation • Volume 81, Number 5, March 15, 2006
FIGURE 2. Mouse endothelial cells (EC) isolated from
thoracic aorta express MHC class II molecules and can
stimulate antigen-primed T cells for proliferation and production of IFN-␥. (A) EC expand from thoracic aorta explants after five days of culture in conditioned medium.
Original magnification: 200⫻. (B) Confluent culture of EC
after two passages in culture dishes containing conditioned
medium (haematoxylin staining). Original magnification:
400⫻. (C) Isolated mouse endothelial cells express cell surface marker CD105 and MHC class II molecules (I-Ab). Specific immunostaining shown in the figure was analyzed by
flow cytometry on EC harvested from P3 cultures. (D) EC
isolated from H-2b, not H-2k, male mice stimulated cell proliferation (upper panel) and IFN-␥ production (lower panel)
in Marilyn CD4 T cells. Adding H-Y peptide to the coculture
increased T cell proliferation in response to EC stimulation
(upper panel). Results presented in the figures are the
means ⫹ SD of triplicates. They are representative of two
independent experiments.
Chronic Rejection of Vascularized Cardiac
Transplant by Indirect Pathway T Cells
Correlates with the Expression of Recipient MHC
Class II Molecules on the Endothelium of the
Graft
Unlike skin grafts, vascularized heart allografts contain
mostly blood vessels of donor origin. However, recent studies
have established that bone marrow-derived endothelial progenitor cells (EPCs), are mobilized in response to trauma or
ischemia and are able to contribute to tissue repair and new
blood vessel formation (23–25) The development of blood
vessels from circulating endothelial precursors, termed vasculogenesis, was previously thought to be restricted to embry-
© 2006 Lippincott Williams & Wilkins
Kapessidou et al.
587
onic development, but is now accepted to play a role in postnatal processes including the repair of ischemic tissues (24,
27). Thus, this raises the possibility that recipient-derived endothelial cells could represent the APC recognized by indirect
pathway CD4⫹ T cells within the graft and thereby promote
rejection of cardiac grafts. To test this hypothesis, we analyzed
Marilyn recipients for their capacity to reject H-2k heart
grafts. On the contrary to skin grafts, hearts from H-2k males,
but not from H-2k females, were rejected by unprimed Marilyn mice in a chronic way (mst ⬎50 days). Weigert staining of
heart-beating grafts by day 50 showed typical signs of cardiac
vasculopathy including arterial wall thickening and luminal
obliteration (49, 50) (Fig. 3A and B). Immunostaining of rejected grafts demonstrated the presence of numerous recipient I-Ab-expressing cells within capillary vessels (Fig. 3C–E).
Remarkably, infiltrated Marilyn T cells, identified by cell surface expression of CD3, were found in close contact with capillaries (Fig. 3F and G). Taken together, these results suggested that replacement of H-2k graft endothelium by
recipient-derived I-Ab EC rendered heart graft susceptible to
rejection mediated by Marilyn T cells.
We also observed that numerous EC within control
normal H-2b hearts expressed I-Ab molecules (Fig. 3E).
Whereas cell surface up-regulation of MHC class II antigen
expression on EC upon stimulation has been demonstrated in
murine and human systems (46, 51), constitutive expression
in normal heart has been reported in humans (52), but is
highly controversial in the mouse (46, 53). In an attempt to
clarify this issue, we analyzed the pattern of expression of
MHC class II molecules in normal heart by immunostaining
of frozen tissue sections. Two color immunostaining revealed
numerous endoglin (CD105)-positive capillary EC strongly
stained by antibodies to mouse MHC class II molecules
(Fig. 4A). Specific expression of MHC class II molecules was
confirmed by using one rat anti-mouse (clone M5/114) and
two different mouse anti-mouse monoclonal antibodies to
I-A/I-E MHC class II molecules (clones 25-9-17 and AF6120.1). Antibodies from 25-9-17 and AF6-120.1 stained specifically capillary EC from H-2b, not H-2k, hearts (Fig. 3; ref.
35), demonstrating that immunostaining in our study was
not the result of unspecific binding of the primary antibodies
to cardiac EC. Moreover, the absence of immunostaining
with isotype-matched control primary antibodies confirmed
this conclusion (Fig. 4A). Other cell types expressed MHC
class II molecules in freshly explanted native hearts. Morphogically, they could be identified as dendritic cells (DC) and
were CD11b- or CD11c-positive cells (Fig. 4B). The number
of CD11c⫹ cells and CD11b⫹ cells per heart section was low
and did not account for all MHC class II-expressing cells
present on a single tissue section (Fig. 4B).
FIGURE 3. H-2k male hearts are chronically rejected by
Marilyn recipients. (A) Extensive vasculopathy was observed on H-2k hearts that survived more than 50 days
in Marilyn recipients. Original magnification: 400⫻. (B)
This pathology was absent in non-transplanted H-2k hearts.
Original magnification: 400⫻. (C–E) The expression of
MHC class II molecule I-Ab (brown staining) was analyzed
by immunolabelling with specific antibodies (clone 25-917) in non-transplanted H-2k male hearts (C), in H-2k male
hearts that survived more than 50 days in Marilyn recipients
(D), and in non-transplanted H-2b hearts (E). Original magnification: ⫻400. (F) Immunostaining with anti-mouse CD3
antibodies revealed numerous TcR-transgenic T cells infiltrating H-2k male hearts chronically rejected by Marilyn
recipients. Original magnification: 400⫻.
Presence of Recipient-Derived EC within
Cardiac Allograft Stimulates Rejection Mediated
by Indirect Pathway CD4 T Cells
Subsequently to the observation that chronic rejection
of male H-2k heart correlated with the presence of recipientderived I-Ab EC within rejected graft, our working hypothesis
was that male H-2k heart allograft could be rejected by Marilyn recipients if it contained sufficient number of I-Ab endothelial cells able to recruit and stimulate male-specific T cells
within the graft. Total body irradiation has been shown to
induce microvascular endothelial apoptosis in tumors and
normal tissues (54 –56) Moreover, lethal irradiation of mice
followed by bone marrow transplantation leads to functional
complete replacement of host hemopoietic APC in spleen and
heart (57), as well as to variable expression of host class II
molecules on endothelial cells (51). Thus, to investigate
whether endothelium replacement by recipient EC precursors could trigger rejection by indirect pathway T cells, we
created chimeric hearts by reconstituting lethally irradiated
(15 Gy) male H-2k RAG⫺/⫺ ␥c⫺/⫺ mice (n⫽3) with bone
marrow (BM) isolated from female H-2b RAG⫺/⫺ ␥c⫺/⫺ donors. Double immunostaining of frozen heart sections with
anti-I-Ab MHC class II molecule (25-9-17) and anti-CD105
monoclonal antibodies showed BM-derived I-Ab-positive EC
within myocardial capillaries of chimeric mice as well as in
some of major vessels (Fig. 5A). I-Ab EC were detected from
day 10 after bone marrow grafting and their number increased with time thereafter (Fig. 5B). We then tested whether
I-Ab endothelium-containing H-2k chimeric hearts were susceptible to acute rejection once transplanted in Marilyn recipients, demonstrating the essential role played by recipient
endothelium in grafts recognized by indirect pathway T cells.
To exclude the possibility that failure to reject acutely H-2k
male grafts could be due to the lack of T cell priming, Marilyn
recipients were immunized with bone marrow-derived I-Ab
588
Transplantation • Volume 81, Number 5, March 15, 2006
FIGURE 4. Two-color immunofluorescence analysis of MHC class II antigen expression on cardiac EC cells. (A)
CD105 and MHC class II I-Ab molecules
are both expressed by the endothelium
of some, but not all, cardiac capillaries
in H-2b hearts. Right top panel shows
double immunostaining for CD105 and
I-Ab. Original magnification: 1000⫻. (B)
In addition to cells expressing CD105
and/or I-A, the myocardium of nontransplanted heart contain a very low
number of cells expressing CD11c. The
number of cells expressing CD11b are
also less frequent than cells expressing
I-A molecules. Original magnification:
400⫻.
FIGURE 5. The presence of recipient-type EC within the graft conditions
the rejection of heart allograft by indirect-pathway T cells. (A) Double immunostaining with anti-CD105 (green fluorescence) and anti-I-Ab (clone 25-9-17)
(red fluorescence) antibodies revealed
that several CD105-positive EC in hearts
isolated from H-2b-into-H-2k BM chimera (day 60 after BM transplantation)
expressed I-Ab molecules. Control immunostaining with isotype antibody is
shown in the lower left corner of each
picture. Original magnification: 400⫻.
(B) In hearts of H-2b-into-H-2k BM chimera, the number of capillaries containing I-Ab-positive cells increased with
time after BM transplantation. (C) The
presence of capillaries containing I-Abpositive cells induced, in Marilyn recipients, the acute rejection of heart graft
isolated from H-2b-into-H-2k BM chimera. H-2b (filled circles; n⫽4)) and
H-2k (open circles; n⫽4) male heart
grafts survived indefinitely in unprimed
Marilyn recipients. Preimmunization
with H-2b-positive male dendritic cells
(DC) induced the acute rejection of H-2b
(filled triangles; n⫽5), not H-2k (open
triangles; n⫽3)), male cardiac grafts in
Marilyn mice. Cardiac grafts from H-2binto-H-2k BM male chimera (taken 60
days postBM transplantation) are
acutely rejected by immunized Marilyn
recipients (closed squares; n⫽4).
male DC prior to transplantation as described by Buonocore
et al. (34). One day after immunization, Marilyn recipients
were transplanted with normal or chimeric (55 days after BM
engraftment) organs and observed for signs of rejection. As
expected from previous experiments on rejection of MHC
class II-disparate hearts (34, 58), immunization by I-Ab male
DC induced the acute rejection of H-2b RAG⫺/⫺ ␥c⫺/⫺ male
cardiac grafts but not that of H-2k RAG⫺/⫺ ␥c⫺/⫺ male hearts
(Fig. 5C). On the contrary, chimeric H-2b-into-H-2k male
cardiac grafts were acutely rejected by immunized Marilyn
© 2006 Lippincott Williams & Wilkins
mice (mst⫽19 days; Figure 5). Thus, the presence of EC cells
expressing I-Ab within H-2k male cardiac allograft conferred
susceptibility to rejection mediated by indirect pathway H-Yspecific I-Ab-restricted CD4⫹ T cells.
DISCUSSION
In the present study, we investigated the role played by
endothelial cells (EC) in the immune recognition of grafted
tissues. One of our main findings is that graft endothelial
expression of MHC class II molecules targeted by recipient’s
T cells appears to be by itself sufficient to recruit antigenspecific T cells within the graft and to bring about rejection.
This was initially shown in a highly antigen-specific TcR
transgenic transfer system where transferred T cells could not
recognize recipient’s MHC class II molecules. In this context,
T cell recognition of target MHC class II molecules on skin
allograft EC alone appeared sufficient to promote rejection.
Subsequently, we also observed, in a transplantation model of
vascularized cardiac graft, that hearts containing endothelial
cells expressing MHC class II molecules targeted by T cells
suffered rejection.
The fact that rejection appears to depend on T cell immune recognition of graft EC raises the question of how
CD4⫹ T cells sensitized by indirect recognition of graft antigens could alone promote allograft rejection. Serial BM transplantation studies have revealed that adult BM-derived stem
cells give rise to functional EC that rapidly integrate into
blood vessels of several organs, including heart (23–25). Our
results indicate that recipient-derived EC, integrated into the
blood vessels of the graft, could well represent the main targets recognized by indirect pathway CD4⫹ T cells within
transplanted tissues. Marilyn T cells were indeed found close
to microvessel cells that expressed recipient’s I-Ab molecules
in rejected H-2k male cardiac grafts. Moreover, induction of
acute rejection of chimeric hearts containing recipient-derived I-Ab EC also provided strong evidence in support of this
concept. These results have an important implication in the
pathogenesis of chronic rejection. They indeed suggest that
the participation of indirect pathway alloreactive T cells in the
process is conditioned by the presence of recipient-derived
EC within the transplanted tissues.
In a previous report, we observed that H-2k male skin
grafts were acutely rejected by Marilyn mice (33). We show
here that, unlike skin, vascularized H-2k male cardiac grafts
placed in primed Marilyn recipients suffer from slower inflammatory processes and develop signs of chronic rejection.
Thus, skin appears to be more susceptible than heart to acute
rejection mediated by indirect pathway CD4⫹ T cells. The
same conclusion has also been reached in other systems
where the failure to reject heart from donors with genetic
deficiency in interstitial DC supports the importance of the
direct pathway of allorecognition in acute vascularized organ
rejection (59) Similarly, in the case of direct pathway deficiency, kidney grafts depleted of DC are rejected significantly
later (by the indirect pathway) than nondepleted grafts (9).
Since healing of skin graft is known to be the site of active
neovascularization shortly after grafting (25), we propose
that the higher susceptibility of skin graft to rejection mediated by indirect recognition could result from T cell recognition of recipient’s EC precursors incorporated in newly
Kapessidou et al.
589
formed blood vessels. On the contrary, incorporation of recipient type EC in vascularized grafts is probably limited during postischemia/reperfusion periods, thus restraining indirect pathway T cells to recognize the graft and to mediate the
process of acute rejection.
The localization of MHC class II molecule expression
in non-transplanted heart has been studied in some detail by
others. Some studies, essentially performed on human heart,
indicated that cardiac MHC class II molecule expression was
essentially restricted to capillaries of the myocardium (52).
Our own observation confirms this pattern of expression in
the mouse heart. Immunocytochemical analyses in situ with
three types of monoclonal antibodies (clones 25-9-17 and
AF6.120.1 (mouse anti-mouse I-Ab) and M5/114 (rat antimouse I-A/I-E)) revealed indeed high expression of MHC
class II antigens by microvascular EC. These findings oppose
previous studies that found essentially no detectable MHC
class II molecule expression in nontransplanted murine
hearts (53). We do not have at this point any experimental
data that could explain these conflicting results. Since it has
been proposed that microvascular EC expression of MHC
class II molecules in human heart is maintained by circulating
factors, including IFN-␥ produced by NK cells (60), we suppose that EC MHC class II expression reported in the present
study could result from the activity of unidentified circulating
mediators. However, these molecules are unlikely to derive
from NK or T cell activity since most of our studies were
performed with RAG⫺/⫺ ␥c⫺/⫺ mice, known to be deficient
in these cell subsets.
It is important to mention that our results do not exclude the possibility that recipient’s MHC class II moleculeexpressing leukocytes infiltrating the graft do participate in
the process of rejection and interact in an antigen-dependent
fashion with T cells within the graft, as suggested by others
(61). However, these interactions do not seem sufficient to
promote rejection since, as shown by Chen et al., non-chimeric H-2k male hearts are not rejected acutely by Marilyn T
cells despite extensive infiltration of the graft by I-Ab CD11bpositive leukocytes (61).
In conclusion, the outcome of transplantation of grafts
that contain recipient-derived EC in our model supports the
importance of the indirect pathway of allorecognition in
chronic vascularized organ rejection. The development of animals exhibiting endothelium-specific genetic deficiency of
MHC class II molecule expression will be of great value to
ascertain this process. Vascular rejection is a severe complication of transplantation leading to a very poor prognosis that
warrants aggressive treatment— even then a high percentage
of grafts are lost. Our findings suggesting the role of endothelial cell replacement in triggering chronic rejection by indirect
pathway CD4⫹ T cells offer some insight into this complication and appeal new therapeutic targets.
ACKNOWLEDGMENTS
We thank Christelle Sequaris for excellent assistance with
cryostat sections, Laetitia Cuvelier for helpful suggestions concerning two-color immunofluorescence staining, and David
Perez-Morga for helpful assistance in two-color immunofluorescence recorded imaging.
590
Transplantation • Volume 81, Number 5, March 15, 2006
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Orbaek Andersen H. Heart allograft vascular disease: an obliterative
vascular disease in transplanted hearts. Atherosclerosis 1999; 142: 243.
Day JD, Rayburn BK, Gaudin PB, et al. Cardiac allograft vasculopathy:
the central pathogenetic role of ischemia-induced endothelial cell injury. J Heart Lung Transplant 1995; 14: S142.
Tullius SG, Tilney NL. Both alloantigen-dependent and -independent
factors influence chronic allograft rejection. Transplantation 1995; 59:
313.
Lee RS, Yamada K, Houser SL, et al. Indirect recognition of allopeptides
promotes the development of cardiac allograft vasculopathy. Proc Natl
Acad Sci USA 2001; 98: 3276.
Pietra BA, Wiseman A, Bolwerk A, et al. CD4 T cell-mediated cardiac
allograft rejection requires donor but not host MHC class II. J Clin
Invest 2000; 106: 1003.
Vella JP, Vos L, Carpenter CB, Sayeg MH. Role of indirect allorecognition in experimental late acute rejection. Transplantation 1997; 64:
1823.
Honjo K, Xu XY, Bucy RP. Heterogeneity of T cell clones specific for a
single indirect alloantigenic epitope (I-Ab/H-2Kd54-68) that mediate
transplant rejection. Transplantation 2000; 70: 1516.
Talmage DW, Dart G, Radovich J, Lafferty KJ. Activation of transplant
immunity: effect of donor leukocytes on thyroid allograft rejection.
Science 1976; 191: 385.
Lechler RI, Batchelor JR. Restoration of immunogenicity to passenger
cell-depleted kidney allografts by the addition of donor strain dendritic
cells. J Exp Med 1982; 155: 31.
Lakkis FG, Arakelov A, Konieczny BT, Inoue Y. Immunologic ‘ignorance’ of vascularized organ transplants in the absence of secondary
lymphoid tissue. Nat Med 2000; 6: 686.
Braun MY, McCormack A, Webb G, Batchelor JR. Mediation of acute
but not chronic rejection of MHC-incompatible rat kidney grafts by
alloreactive CD4 T cells activated by the direct pathway of sensitization.
Transplantation 1993; 55: 177.
Benichou G, Valujskikh A, Heeger PS. Contributions of direct and
indirect T cell alloreactivity during allograft rejection in mice. J Immunol 1999; 162: 352.
Vella JP, Spadafora-Ferreira M, Murphy B, et al. Indirect allorecognition of major histocompatibility complex allopeptides in human renal
transplant recipients with chronic graft dysfunction. Transplantation
1997; 64: 795.
Valujskikh AD, Matesic A, Gilliam D, et al. cells reactive to a single
immunodominant self-restricted allopeptide induce skin graft rejection in mice. J Clin Invest 1998; 101: 1398.
Steeber DA, Tedder TF. Adhesion molecule cascades direct lymphocytes recirculation and leukocyte migration during inflammation.
Imunol Res 2000; 22: 299.
Kreisel D, Krupnick AS, Szeto WS, et al. A simple -method for culturing
mouse vascular endothelium. J Immunol Methods 2001; 254: 31.
Marelli-Berg FM, Hargreaves RE, Carmichael P, et al. Major histocompatibility complex class II-expressing endothelial cells induce allospecific nonresponsiveness in naive T cells. J Exp Med 1996; 183: 1603.
Pober JS, Collins T, Gimbrone MA Jr., et al. Inducible expression of
class II major histocompatibility complex antigens and the immunogenicity of vascular endothelium. Transplantation 1986; 41: 141.
Karin N, Szafer F, Mitchell D, et al. Selective and nonselective stages in
homing of T lymphocytes to the central nervous system during experimental allergic encephalomyelitis. J Immunol 1993; 150: 4116.
Kawai T, Shimauchi H, Eastcott JW, et al. Antigen direction of specific
T-cell clones into gingival tissues. Immunology 1993; 93: 11.
Weis M, von Scheidt, W. Cardiac allograft vasculopathy: a review. Circulation 1997; 96: 2069.
Briscoe DM, Alexander SI, Lichtman AH. Interactions between T lymphocytes and endothelial cells in allograft rejection. Curr Opin Immunol 1998; 10: 525.
Shi Q, Rafii S, Wu MH, et al. Evidence for circulating bone marrow–
derived endothelial cells. Blood 1998; 92: 362.
Takahashi T, Kalka C, Masuda H, et al. Ischemia- and cytokine-induced mobilization of bone marrow– derived endothelial progenitor
cells for neovascularization. Nat Med 1999; 5: 434.
Asahara T, Masuda H, Takahashi T, et al. Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in
physiological and pathological neovascularization. Circ Res 1999; 85:
221.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
Walter DH, Rittig K, Bahlmann FH, et al. Statin therapy accelerates
reendothelialization: a novel effect involving mobilization and incorporation of bone marrow– derived endothelial progenitor cells. Circulation 2002; 105: 3017.
Rafii S. Circulating endothelial precursors: mystery, reality, and promise. J Clin Invest 2000; 105: 17.
Lagaaij EL, Cramer-Knijnenburg GF, van Kemenade FJ, et al. Endothelial cell chimerism after renal transplantation and vascular rejection.
Lancet 2001; 357: 33.
Sinclair RA. Origin of endothelium in human renal allografts. BMJ
1972; 4: 15.
Williams GM, ter Haar A, Parks LC, Krajewski CA. Endothelial changes
associated with hyperacute, acute, and chronic renal allograft rejection
in man. Transpl Proc 1973; 5: 819.
Valujskikh A, Lantz O, Celli S, et al. Cross-primed CD8(⫹) T cells
mediate graft rejection via a distinct effector pathway. Nat Immunol
2002; 3: 844.
Lantz O, Grandjean I, Matzinger P, Di Santo JP. Gamma chain required
for naive CD4⫹ T cell survival but not for antigen proliferation. Nat
Immunol 2000; 1: 54.
Braun MY, Grandjean I, Feunou P, et al. Acute rejection in the absence
of cognate recognition of allograft by T cells. J Immunol 2001; 166:
4879.
Buonocore S, Flamand V, Goldman M, Braun MY. Bone marrow-derived immature dendritic cells prime in vivo alloreactive T cells for
interleukin-4-dependent rejection of major histocompatibility complex class II antigen-disparate cardiac allograft. Transplantation 2003;
75: 407.
Eiref SD, Zhang W, Popma SH, et al. Creation of chimeric hearts via
bone marrow transplantation: a new tool for testing the “passenger
leukocyte” hypothesis. Ann Thorac Surg 1997; 64: 628.
Scott D, Addey C, Ellis P, et al. Dendritic cells permit identification of
genes encoding MHC class II-restricted epitopes of transplantation antigens. Immunity 2000; 12: 711.
Vella JP, Magee C, Vos L, et al. Cellular and humoral mechanisms of
vascularized allograft rejection induced by indirect recognition of donor MHC allopeptides. Transplantation 1999; 67: 1523.
Valujskikh A, Matesic D, Gilliam A, et al. T cells reactive to a single
immunodominant self-restricted allopeptide induce skin graft rejection in mice. J Clin Invest 1998; 101: 1398.
de Waal RM, Bogman MJ, Cornelissen IM, et al. Expression of donor
class I major histocompatibility antigens on the vascular endothelium
of mouse skin allografts. Transplantation 1986; 42: 178.
de Waal RM, Bogman MJ, Maass CN, et al. Variable expression of Ia
antigens on the vascular endothelium of mouse skin allografts. Nature
1983; 303: 426.
Offner H, Buenafe AC, Vainiene M, et al. Where, when, and how to
detect biased expression of disease-relevant V beta genes in rats with
experimental autoimmune encephalomyelitis. J Immunol 1993; 151:
506.
Karin N, Szafer F, Mitchell D, et al. Selective and nonselective stages in
homing of T lymphocytes to the central nervous system during experimental allergic encephalomyelitis. J Immunol 1993; 150: 4116.
Kawai T, Shimauchi H, Eastcott JW, et al. Antigen direction of specific
T-cell clones into gingival tissues. Immunology 1998; 93: 11.
Marelli-Berg FM, Frasca L, Weng L, et al. Antigen recognition influences transendothelial migration of CD4⫹ T cells. J Immunol 1999;
162: 696.
Marelli-Berg FM, Jarmin SJ. Antigen presentation by the endothelium:
a green light for antigen-specific T cell trafficking? Immunol Lett 2004;
93: 109.
Kreisel D, Krasinskas AM, Krupnick AS, et al. Vascular endothelium
does not activate CD4⫹ direct allorecognition in graft rejection. J Immunol 2004; 173: 3027.
Miller DW, Graulich W, Karges B, et al. Elevated expression of endoglin, a component of the TGF-beta-receptor complex, correlates with
proliferation of tumor endothelial cells. Int J Cancer 1999; 81: 568.
Duff SE, Li C, Garland JM, Kumar S. CD105 is important for angiogenesis: evidence and potential applications. FASEB J 2003; 17: 984.
Hayry P, Isoniemi H, Yilmaz S, et al. Chronic allograft rejection. Immunol Rev 1993; 134: 33.
Hruban RH, Beschorner WE, Baumgartner WA, et al. Accelerated arteriosclerosis in heart transplant recipients is associated with a T lymphocyte–mediated endothelialitis. Am J Pathol 1990; 137: 871.
Kapessidou et al.
© 2006 Lippincott Williams & Wilkins
51.
52.
53.
54.
55.
56.
Gao EK, Kosaka H, Surh CD, Sprent J. T cell contact with Ia antigens on
nonhemopoietic cells in vivo can lead to immunity rather than tolerance. J Exp Med 1991; 174: 435.
Rose ML, Coles MI, Griffin RJ, et al. Expression of class I and class II
major histocompatibility antigens in normal and transplanted human
heart. Transplantation 1986; 41: 776.
Hasegawa S, Becker G, Nagano H, et al. Pattern of graft- and hostspecific MHC class II expression in long-term murine cardiac allografts: origin of inflammatory and vascular wall cells. Am J Pathol 1998;
153: 69.
Van der Meeren A, Vandamme M, Squiban C, et al. Inflammatory
reaction and changes in expression of coagulation proteins on lung
endothelial cells after total-body irradiation in mice. Radiat Res 2003;
160: 637.
Boerma M, Kruse JJ, van Loenen M, et al. Increased deposition of von
Willebrand factor in the rat heart after local ionizing irradiation. Strahlenther Onkol 2004; 180.
Mouthon MA, Vereycken-Holler V, Van der Meeren A, Gaugler MH.
57.
58.
59.
60.
61.
591
Irradiation increases the interactions of platelets with the endothelium
in vivo: analysis by intravital microscopy. Radiat Res 2003; 160: 593.
Krasinskas AM, Eiref SD, McLean AD, et al. Replacement of graftresident donor-type antigen presenting cells alters the tempo and
pathogenesis of murine cardiac allograft rejection. Transplantation
2000; 70: 514.
Sayegh MH, Wu Z, Hancock WW, et al. Allograft rejection in a new allospecific CD4⫹ TCR transgenic mouse. Am J Transplant 2003; 3: 381.
Wang Z, Castellaneta A, De Creus A, et al. Heart, but not skin, allografts
from donors lacking Flt3 ligand exhibit markedly prolonged survival
time. J Immunol 2004; 172: 5924.
McDouall RM, Batten P, McCormack A, et al. MHC class II expression on
human heart microvascular endothelial cells: exquisite sensitivity to interferon-gamma and natural killer cells. Transplantation 1997; 64: 1175.
Chen Y, Demir Y, Valujskikh A, Heeger PS. Antigen location contributes to the pathological features of a transplanted heart graft.
Am J Pathol 2004; 164: 1407.