| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Transplantation Biology Research Center, Massachusetts General Hospital/Harvard Medical School, Boston, MA 02129
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The contribution of the indirect pathway to allorecognition and to both acute (6) and chronic (7, 8) forms of allograft rejection has also been validated by numerous studies. One pertinent example is that of Auchincloss et al. (9), who reported the rapid rejection of skin allografts derived from MHC class II-knockout mice, thereby demonstrating that indirect recognition alone can be an effective pathway of alloreactivity in this small animal model. The Ag-specific, tolerogenic effect of MHC allopeptide administration in the rat model, reflected by suppression of proliferative and cytoxic responses in vitro, provided additional experimental evidence for the importance of indirect allorecognition (10).
Recent evidence indicates that processed MHC allopeptides may be more potent stimulators of an indirect immune response than minor Ags based on their capacity to elicit primary indirect proliferative responses in vitro (11). Using a water lysis technique to remove viable stimulating cells and thus eliminate direct allorecognition in vitro, Gould and Auchincloss (11) compared the indirect response to MHC-derived allopeptides with those from minor histocompatibility Ags in MLR culture assays. A primary indirect response (proliferative) against allogeneic MHC peptides, but not against minor histocompatibility Ags, was observed.
Understanding further the mechanism of indirect recognition and the underlying immunogenicity of MHC alloantigens may facilitate the development of interventional strategies to interrupt this pathway and prolong allograft survival. Therefore, in the present study, we have chosen a partially inbred miniature swine model to examine further, via MLR culture assays, the contribution of indirect responses to alloreactivity across selective MHC (swine leukocyte Ag; SLA)5 and minor histocompatibility barriers. Massachusetts General Hospital miniature swine used in these experiments provide a unique opportunity to reproducibly perform studies across selective MHC disparities in a preclinical large animal model (12). By selectively depleting APCs from the responder and stimulator cell populations, we were able to compare and contrast the proliferative responses observed in the direct vs indirect pathway, respectively, across various combinations of SLA-mismatched responder/stimulator combinations. However, given that IL-2 secretion by stimulating cells is partially radio-resistant (13, 14), several experimental measures were taken to avoid backward stimulation of responder cells which could lead to falsely elevated stimulation indices for the indirect pathway.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MGH MHC-inbred miniature swine were used in these experiments. The immunogenetic characteristics of this herd and intra-MHC recombinant haplotypes have been described previously (12, 15).
Isolation of PBMC
Heparinized whole blood was drawn from miniature swine and diluted with HBSS (Life Technologies, Gaithersburg, MD), and mononuclear cells were obtained by gradient centrifugation using lymphocyte separation medium (Organon Teknika, Durham, NC). The mononuclear cells were washed once with HBSS before the contaminating red cells were lysed with ACK Lysing buffer (BioWhittaker, Walkersville, MD). Cells were then washed in HBSS again and resuspended in MLR medium consisting of RPMI 1640 (Life Technologies) supplemented with 6% fetal porcine serum, 100 U/ml penicillin (Life Technologies), 135 µg/ml streptomycin (Life Technologies), 50 µg/ml gentamicin (Life Technologies), 10 mM HEPES (Fisher Scientific, Pittsburgh, PA), 2 mM L-glutamine (Life Technologies), 1 mM sodium pyruvate (BioWhittaker), and 5 x 10-5 M 2-ME (Sigma-Aldrich, St. Louis, MO). Cell suspensions were kept at 4°C for up to 48 h before use in cellular assays.
APC depletion
APCs were depleted from the PBMC by a two-step procedure. During the first step, 120 x 106 cells were incubated at 37°C in 20 ml standard MLR media (16) on 75-cm plastic flasks (Falcon no. 3023; Beckinson Labware, Lincoln Park, NJ) for 3–4 h. Nonadherent cells were then collected and replated at 37°C for an additional 3–4 h. The remaining nonadherent cells were then adjusted to 3 x 108 in 2 ml MLR media and incubated on 1.2 g of sterile nylon wool (Fenwal Laboratories, Deerfield, IL) at 37°C for 1 h. The cell-containing nylon wool was then rinsed with 8 ml of 37°C MLR medium and nonadherent cells were collected and stored at 4°C for up to 24 h until usage.
Mixed lymphocyte reaction
MLR assays, to test for proliferative response to alloantigen,
have been described previously (17). Briefly, 4 x
105 responders and an equal number of irradiated
(25 Gy) stimulators were incubated in 200 µl of standard MLR media
using flat-bottom 96-well plates (Costar, Cambridge, MA). After 5 days
of incubation, one µCi of [3H]thymidine was
added to each well, followed by an additional 5-h incubation. For time
series experiments, MLR plates were incubated for 1, 2, 3, 4, and 5
days before harvest. [3H]-incorporation was
determined in triplicate samples by liquid scintillation. Stimulation
indices were calculated by dividing the average cpm detected for a
particular responder-stimulator pair by the average cpm for the same
responder stimulated by self stimulators. Each experiment was performed
at least three times with reproducible results. Background counts were
uniformly low, ranging from 100–300 cpm with the exception of the
experiment represented in Fig. 5
A. Background counts for
this experiment were 583 ± 80 cpm, still <10% of the relevant
back stimulation demonstrated by that figure.
|
PEDSV.15, an SV40-transformed SLAd PAEC
clone, was derived and maintained in culture as previously described
(18). Retroviral constructs used in this study included
the pPBM14 recombinant genome which contained cDNA sequences for the
pig DR
and -
polypeptide chains of the SLAc
allele together with that of the neomycin resistance. Internal ribosome
entry sequences derived from the Ig H chain binding protein and the
encephalomyocarditis virus were intercalated between class II cDNA to
ensure coordinate translation from a single transcript
(19). The second construct (N2) contained the
neomycin resistance gene cloned into the Moloney retrovirus backbone
(20). Cells were transduced according to published
protocols (21), and selected in G418 (400 µg/ml active;
Mediatech, Herndon, VA) for 10–14 days for further use in in vitro
assays. The presence of retroviral transcripts in transduced cells was
monitored by Northern blot analysis of electrophoresized total RNA
probed with radioactive NeoR or
DRb cDNA fragments.
Mixed lymphocyte-endothelial cell culture (MLEC)
MLECs using SLA class II-transduced PAEC were performed to test for recognition of processed allogeneic SLA class II peptides presented by self-SLA class II molecules. MLECs were performed as described above for MLR cultures, using APC-depleted responders and irradiated, responder-matched PAECs (cell concentration: 1 x 104/well) as stimulators and incubating for a total of 4 days.
In vivo sensitization
For secondary proliferation assays, cells were obtained from animals that were sensitized to allogeneic SLA. Animals were grafted with fresh split thickness skin grafts (40 x 40 x 2.2 mm) that were harvested from donors using a Zimmer dermatome and placed on graft beds, also prepared with a dermatome, on the lateral thorax as previously described (4). Skin grafts were allowed to reject. The day of rejection was defined as the time at which <10% of the skin graft showed signs of viability as judged by color, texture, and warmth to touch. Every 3–5 mo, animals were reimmunized by s.c. injection of 2 x 107 PBMC (from skin-graft donor) in 2 ml HBSS.
In vivo tolerance induction
To avoid backstimulation in some assays, stimulator cells were obtained from animals that had been tolerized to the MHC of the responder used in the MLR. The tolerizing regimen consisted of fully SLA-mismatched kidney allograft along with a 12-day course of continuous i.v. FK506 starting on the day of transplant (22). FK506 trough levels were measured and maintained between 40 and 60 ng/dl. FK506-treatment was discontinued 12 days after the transplant, with no additional immunosuppressive treatment provided. Tolerance was defined on the basis of graft survival beyond 100 days with normal organ function and no histological evidence of rejection.
Flow cytometric analysis
Flow cytometry was used to detect APCs in PBMC before and after APC depletion and for detection of SLA class II on transduced PAECs. APC were detected by one-color, indirect flow cytometry analysis. Cells were stained by incubation for 30 min with a B cell-specific mouse-anti-swine CD21 Ab (BB6-11C9, IgG1) or a mouse Ab specific for swine monocytes, SWC3 (74-22-15, IgG1). Polyclonal, FITC-conjugated goat-anti-mouse IgG (Sigma-Aldrich) was used as a secondary reagent after washing the cells twice. All steps were performed at 4°C using flow cytometry buffer (HBSS containing 0.5% BSA and 0.5% sodium azide).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Proliferative responses of naive animals to a variety of
alloantigens in standard 5-day MLR assays are shown in Fig. 1. No proliferative responses were
detected when responders and stimulators were SLA-matched but differed
at multiple minor histocompatibility Ag loci. Only weak responses were
noted when responders and stimulators were mismatched at SLA class I
loci only. In contrast, SLA class II disparate stimulators elicited
very high responses, similar to those seen to fully mismatched (class I
and II disparate) stimulators (23).
|
APCs were depleted from PBMC based on plastic and nylon wool
adherence properties. The effectiveness of the APC depletion was
assessed phenotypically by FACS analysis. Representative data for one
experiment are shown in Fig. 2, indicating that a marked reduction of both monocytes and B cells was
achieved following APC depletion. This procedure had no effect on the
relative percentage of normal swine T cell subpopulations, which
included CD4-CD8- and
CD4+CD8+ as well as
CD4+CD8- and
CD4-CD8+ T cells
(24, 25). In addition, surface SLA class II expression on
resting T cells, also normal in swine (26), was unchanged
following APC depletion.
|
As depicted in Fig. 3, the removal
of either class I or II disparate responder APCs to specifically
measure the direct pathway resulted in an increase in the stimulation
index (SI), compared with that observed with bulk PBMC. In contrast,
when the stimulators were APC-depleted to measure indirect responses,
proliferation was reduced from an SI of 302 with bulk PBMC to 118 in a
class II mismatch, and from 18 to 2.7 in a class I mismatch. Removing
both responder and stimulator APCs reduced the SI to background levels,
confirming that APC depletion of both responder and stimulator
populations was complete. This result also demonstrated that following
APC depletion, residual SLA class II-expressing T cells could not
provide sufficient stimulus for a detectable proliferative response.
Therefore, these T cells could serve as an adequate source of SLA class
II Ags without directly stimulating responder cells to proliferate,
thereby enabling the specific measurement of the indirect proliferative
response.
|
To eliminate any effect of backstimulation on the proliferative
response attributed to the indirect pathway in Fig. 4, several experimental measures were
taken to prevent responder Ag recognition by stimulating T cells. The
first set of experiments included stimulators lysed by repetitive
(three times) freezing and thawing (F/T) to remove viable cells
potentially capable of causing a backstimulating
effect. Lysis of stimulator cells was visually confirmed by light
microscopy. No proliferative responses were observed in either class I
or II mismatched combinations when these nonbackstimulating suspensions
of stimulator cell remnants were used as stimulators in MLR assays,
despite the presence of responder APCs (Fig. 4). Again, the depletion
of APCs from the responder cell population resulted in a significantly
augmented proliferative response to nonlysed stimulators compared with
that of bulk PBMC.
|
A). This experiment was also
repeated using PBMC from SLAcc swine tolerized in
vivo to SLAdd (see Materials and
Methods) as responders and stimulators (Fig. 6). As observed with
F1 responders, no MLR responses were
observed when bulk or APC-depleted PBMC from these tolerized animals
were used as responders. However, the effect of backstimulation on the
measurement of the indirect pathway in both of these experiments can be
noted in Figs. 5A and 6A. Despite the absence of
a direct proliferative response given the use of tolerant responders,
APC-depleted parental or nontolerant stimulators induced a
significant MLR response, presumably via recognition by
stimulator cells and subsequent cytokine release which we have termed
backstimulation. No proliferative response of tolerant
SLAdd responders was observed following
lysis of naive SLAcc stimulator cells by repeated
F/T (data not shown).
|
B and 6B, respectively. In both series,
the use of bulk PBMC led to a sizable proliferative response that was
augmented significantly upon isolation of the direct pathway via
removal of the responder APCs. The design of these latter experiments
eliminated any potential contribution from backstimulation, and no
detectable MLR responses were observed in either series when the
indirect pathway was analyzed. Priming was required to detect indirect MLR responses
The objective of the next experiment was to determine whether a
secondary indirect response could be achieved by priming responder
cells. Proliferation of primed and unprimed parental responders against
F1 stimulators are shown in Fig. 7. Similar to the observations presented
in the previous figures (Figs. 5B and 6B), no
indirect response was observed with unprimed responders against
tolerant stimulators. However, primed responders from swine sensitized
with class II mismatched skin grafts and s.c. boosting with
donor-matched PBMC, mounted substantial indirect antidonor MLR
response, with nearly a 6-fold increase in SI compared with unprimed
responders. With bulk PBMC and APC-depleted responders, the MLR
responses of these in vivo-primed responders were also elevated
compared with those observed using naive responders.
|
In addition to cytokine production by stimulator cells,
backstimulation may also be caused by improved Ag presentation by
responder APCs. The following experiments were performed to determine
whether the indirect proliferative response of primed cells against
tolerant and heterozygous stimulators involved responder T cells rather
than enhanced Ag processing/presentation by responder APC. To this end,
the PAEC cell line PEDSV.15 (SLAd) was transduced
with a recombinant retrovirus for the pig DRc
allele (Fig. 8A) to generate
G418-resistant transduced cells named PED-DR. Control transduction
involved a retroviral construct only containing the neomycin resistance
gene (NeoR), which produced the PED-Neo transduced cells
(see Materials and Methods). As shown in Fig. 8B,
both the NeoR and DRc proviral genomes
were correctly transcribed and led to the two expected genomic (Fig. 8, A and C) and spliced (Fig. 8, B and
D) transcripts. However, PED-DR cells did not express the
transduced DR heterodimer at the cell surface as shown in the flow
cytometry analysis presented in Fig. 8C (bottom histograms).
This result suggested that transduced DR and --chains in PED were
fully degraded or presented as peptides on PAEC MHC molecules. The
second hypothesis was confirmed by showing that the stimulation of
primed, but not naive, T cells with PED-DR led to marked proliferative
responses as compared with stimulation with PED-Neo cell controls (Fig. 9). This response was primarily blocked
by anti-class II DR, to a lesser extent by anti-DQ, but not by
anti-class I mAbs, indicating that class II peptides were mainly
presented through the endogenous class II pathways (data not
shown).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Appropriate experimental measures were taken to avoid the phenomenon of backstimulation on the observed proliferative responses. Backstimulation refers to the ability of irradiated stimulator cells to recognize the responder APC and react via release of cytokines into the culture media, leading to a proliferative effect on responder T cells (13, 14, 27). Alternatively, backstimulation may also occur through a mechanism involving activation or maturation of responder APCs, leading to more efficient presentation of processed peptide. The activation or maturation of responder APCs may also cause them to express important secondary signals that could enable direct recognition of stimulator T cells (trans-costimulation). The stimulating T cells themselves may also undergo activation-induced expression of secondary signals that allow direct recognition. Preliminary data from our laboratory to be presented elsewhere suggest that activated or PHA-blasted T cells may serve as effective APCs, increasing the likelihood of this latter proposed mechanism of backstimulation.
The use of stimulator cells lysed by repeated F/T effectively abolished
backstimulation and failed to evoke any detectable indirect response
among native responders (Fig. 4). However, to more closely mimic
clinical transplantation, where intact allogeneic cells or tissues are
transferred, stimulator cells in the MLR culture assays were kept
structurally intact in subsequent experiments. In these experiments,
backstimulation was avoided by using stimulators that were unresponsive
to the responder, including F1 progeny and PBMC
derived from animals tolerized in vivo specifically to the SLA
haplotype of the responder. When responders tolerant to the (parental)
stimulators were used, a sizable proliferative response was observed in
the absence of a direct response. This indirect response was eliminated
when parental responders and F1 stimulators
(Figs. 5B and 6B) were used, suggesting that the
response previously attributed to the indirect pathway was actually
caused by backstimulation.
When the effect of backstimulation was carefully eliminated from the
MLR culture, indirect proliferative responses to SLA class II-derived
alloantigens occurred only following in vivo or in vitro priming,
suggesting that the immunogenicity of MHC allopeptides parallels that
of processed nominal Ags (Fig. 7). To test whether the secondary
indirect responses were due to a change in the activation properties of
the T cells and not of the responder APCs, we removed the APCs from the
primed responders and added irradiated, MHC-matched stimulators that
presented processed allogeneic MHC. For this purpose, we used
responder-matched PAECs that were transduced with an MHC gene matched
to the priming cells. Resting PAECs express nearly undetectable levels
of endogenous MHC class II by FACS (Fig. 8C). Upon
activation with IFN-
, their class II expression was up-regulated and
detectable at high levels on the surfaces of all PAECs (data not
shown). Only primed T cells responded specifically to the
transgene-expressing PAECs (Fig. 9A). This response did not
require up-regulation of endogenous MHC class II, a result consistent
with the inability of the transgene to be regulated by IFN- (Fig. 9B). We hypothesize that the total amount of processed
transgene expression on endogenous MHC of resting PAEC was equal to the
expression after up-regulation of endogeneous MHC class II. A possible
response to processed transgene expressed on MHC class I could be ruled
out by selective blockage of MHC molecules with mAbs to class II but
not class I Ags. Although unlikely, a direct response by primed T cells
to low-level surface expression of the transgene, undetectable by flow
cytometry or immunoprecipitation, could not definitively be ruled out
in this system.
Our results confirm previous data showing that recognition of SLA-class
II alloantigens via the direct pathway causes vigorous proliferative
responses that do not require Ag priming (16, 28).
Similarly, SLA class I-mismatched responder/stimulator pairs also
resulted in a primary, although diminished, proliferative response,
compared with the nonexistent primary response observed across minor
histocompatibility barriers. Similar to class II-induced human T cells
(23, 29), a regulatory or tolerizing effect of naive class
II-positive swine T cells may underlie the augmented proliferative
responses observed following removal of responder APCs (direct pathway;
Figs. 3–7). Given the possibility of nylon wool adherence by some
class II-expressing T cells, this increase in the SI suggests that
these accessory T cells (with concomitant nylon wool adherence
properties) in the responder population may have had a suppressive
effect on the proliferative responses observed with bulk PBMC.
In summary, these results demonstrate that an in vitro proliferative response to MHC Ags through the indirect pathway requires priming of the responder T cells. Therefore, in the MLR culture assay, MHC alloantigens are indistinguishable from minor Ags. These results contradict those obtained in the mouse model (11) in which MHC-derived allopeptides exhibited an apparently greater immunogenicity in the MLR assay, compared with nominal/minor Ags. Continued study is necessary to determine the mechanism of indirect allorecognition, or more importantly, why and how this process is so important in allograft rejection. A better understanding of indirect mechanisms may eventually enable the development of effective therapeutic strategies to prevent or interrupt these immunological responses to foreign tissues.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Klinikum Mannheim,
Universitätsklinikum, Chirurgishe Klinik, Mannheim, Germany.
3 Current address: Department of Surgery,
Massachusetts General Hospital/Harvard Medical School, Boston, MA
02114.
4 Address correspondence and reprint requests to Dr. David H. Sachs, Transplantation Biology Research Center, Massachusetts General Hospital/Harvard Medical School, MGH-East, Building 149-9019, 13th Street, Boston, MA 02129. E-mail address: sachs{at}helix.mgh.harvard.edu
5 Abbreviations used in this paper: SLA, swine leukocyte Ag; PAEC, porcine aortic endothelial cell; MLEC, mixed lymphocyte-endothelial cell culture; SI, stimulation index; F/T, freezing and thawing.
Received for publication October 18, 2001. Accepted for publication July 25, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ heterodimers. Xenotransplantation 4:161.
This article has been cited by other articles:
|
|
 |
|
  P. S. Cho, D. J. Messina, E. L. Hirsh, N. Chi, S. N. Goldman, D. P. Lo, I. R. Harris, S. H. Popma, D. H. Sachs, and C. A. Huang Immunogenicity of umbilical cord tissue derived cells Blood, January 1, 2008; 111(1): 430 - 438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Almeciga, Z. C. Wang, J. Zuniga, M. Fernandez-Vina, O. Clavijo, H. Araujo, V. Romero, J. Henry, S. Ferrone, and E. J. Yunis Allorecognition of an HLA-A*01 Aberrant Allele by an HLA Identical Family Member Carrying the HLA-A*0101 Allele J. Immunol., December 15, 2006; 177(12): 8643 - 8649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Potian, H. Aviv, N. M. Ponzio, J. S. Harrison, and P. Rameshwar Veto-Like Activity of Mesenchymal Stem Cells: Functional Discrimination Between Cellular Responses to Alloantigens and Recall Antigens J. Immunol., October 1, 2003; 171(7): 3426 - 3434. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
